Mail/News Performance Effort Underway
Friday November 2nd, 2001
Seth Spitzer today sent an update out about what the Mail/News team is doing to meet their goals for Mozilla 1.0. They'll be focusing almost 100% of their effort for at least the next two milestones (0.9.7, 0.9.8) on Footprint and Performance of the various parts of Mail, News, and Addressbook. Not only will they be focusing on it, they'll also be locking down their part of the tree and only accepting performance or footprint fixes. Very few exceptions will be made. Click the Full Article link to see Seth's full post.
Full Article...
#1 Mozilla 1.0 feature freeze
by niner
Friday November 2nd, 2001 7:01 PM
As in the Mozilla 1.0 manifesto was pointed out that there will be a feature freeze maybe at 0.9.8 or 0.9.9 could this collide with the MailNews performance effort? There it sounds that new features and bug should be set to 0.9.9 but when there is a Mozilla 1.0 freeze does that mean that new features don't have a real chance until post 1.0?
Or short question: does PGP encryption in Mail have a chance to make it in the 1.0? I think that would be an important feature especially now that the framework for mailencryption has been done with S/MIME.
#2 Fix the critical IMAP bug first
by navneet
Friday November 2nd, 2001 9:42 PM
Hello,
I use IMAP to check multiple accounts. Everytime there is a new mail in the second account, I need to exit the browser and come back to read the mail since get messages doesn't get it. This seems to be a known bug. This is an important bug and exists even on the Netscape builds.
I think this should be fixed before getting into any performance fixes.
- Navneet
#27 Re: Fix the critical IMAP bug first
by schapel
Monday November 5th, 2001 8:22 AM
Here are the IMAP bugs from Bugzilla: http://bugzilla.mozilla.org/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&email1=&emailtype1=substring&email2=&emailtype2=substring&bugidtype=include&bug_id=&changedin=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&product=MailNews&short_desc=imap&short_desc_type=allwordssubstr&long_desc=&long_desc_type=allwordssubstr&bug_file_loc=&bug_file_loc_type=allwordssubstr&status_whiteboard=&status_whiteboard_type=allwordssubstr&keywords=&keywords_type=anywords&field0-0-0=noop&type0-0-0=noop&value0-0-0=&newqueryname=&order=Reuse+same+sort+as+last+time
The appropriate way to indicate that you want a bug fixed is to vote for it.
#29 Re: Re: Fix the critical IMAP bug first
by saltheart
Monday November 5th, 2001 9:37 AM
Sheesh, that\'s a lot of IMAP bugs...however, I didn\'t see 61428 (IMAP subfolder message counts not getting updated).
What does RFE mean on some of the bugs?
Also how many votes does it normally take to get someone to notice things?
#35 Re: Re: Re: Fix the critical IMAP bug first
by schapel
Monday November 5th, 2001 11:03 AM
RFE means "request for enhancement". There's been some debate about how to handle enhancements in Mozilla. Because there is a Severity level named "enhancement", there is no way to indicate that some enhancements are more severely needed than others... there are currently some RFEs that assigned Critical Severity! The [RFE] is used to indicate that the bug is really just an enhancement.
There's no obvious mapping of votes for a bug to effort expended getting that bug fixed. However, if someone comes here and says that a bug is absolutely critical to fix, and there are 3 or fewer votes for it, you can bet that it's really not a major problem for the vast majority of Mozilla users.
That's why I often post links to bugs and bug lists that people complain about -- so we can vote and see which ones really are the critical ones to fix. At the very least, clicking on the link is an introduction to Bugzilla that gets visitors here more involved than simply reading and posting to MozillaZine!
#3 PGP is not even considered
by nick
Friday November 2nd, 2001 9:42 PM
If you follow the newgroups and have read the "Encryption and Signing" specs you'll find out that PGP is not even considered. Of course, most current Mozilla users want PGP, not S/MIME. S/MIME is a feature that Netscape wants in order to seduce corporate users into switching from NS4 or Outlook to Mozilla.
I'm not against Netscape persuing their own objectives, but PGP should be at least mentioned in a document that supposedly set the future of Mozilla's encryption and signing features. The spec talks all over the place about S/MIME and there's no provision for other encryption schemes. One example of this is that they use the word "certificate", which is an S/MIME term.
The lack of an architecture which included both encryption schemes was the reason given to stop PGP integrationm, which was already been written (bug 22687).
#4 Re: PGP is not even considered
by gwalla
Friday November 2nd, 2001 9:47 PM
If you read recent comments in the main PGP bug < http://bugzilla.mozilla.org/show_bug.cgi?id=22687 >, you will see that the S/MIME landing implemented a plugin archetecture for encryption/signing in mailnews. Someone could write a PGP plugin for Mozilla now (it just couldn't go in the tree just yet because of the lockdown).
#5 Re: PGP is not even considered
by nick
Friday November 2nd, 2001 10:36 PM
The arch document I found in that bug are some rough notes about adding some header and handling the encryption/signing with "converters".
I haven't checked the implementation which is being cheked in. But reading through this "Mail: Encryption and Signing" document (which is hosted at "http://www.mozilla.org/mailnews/specs/security/", could it be more official/canonical? ).
This document is clearly S/MIME only, I quote:
"If the client has found a certificate for all the current addressees, the 'Send' button is modified to incorporate a "will be encrypted" icon (closed lock)."
What if those certificates (keys?) are from different encryption schemes? The issue is not mentioned in "Issues", nor it's mentioned as one of the cases whoen with screenshots. The spec is very detailed about visual UI feedback, but it doesn't specify how it will show to which scheme the keys for a recipient belong.
But there's no need to look for this details. The summary of this document's content clearly says it all:
"Summary
The S/MIME features of Mail enhance the security of your emails. Specifically, these features:
* Ensure the privacy of your email while it travels from your machine to your recipient's machine using high-grade encryption.
* Enable you to identify yourself and others in email messages by way of digital signatures.
* Ensure that no one can modify the email while it travels from your machine to your recipient's machine."
#6 engmail in mozdev handle PGP/GPG for mail/news
by tseng_mike
Saturday November 3rd, 2001 7:52 AM
at http://enigmail.mozdev.org/
#7 Awesome
by jedbro
Saturday November 3rd, 2001 11:01 AM
Has anyone tried this out???
COuld someone post some feedback, on it's functionality with Moz?
I would love to use it, but since I DO use Mozmail as my primary client, I can't afford to screw it up right now buy installing somthing that doesn't even work.
Thanks for the input!
Cheers
--Jedbro
#17 Enigmail
by michaelg
Sunday November 4th, 2001 7:49 AM
I've played with it a bit. It's not too bad, you can sign, encrypt and decrypt messages.
It doesn't use the new plugin api that is landing w/ S/MIME (obviously), so the implementation is a but klugey, but maybe it will get re-written to use the crypto plugin.
#8 PGP is not even half-there
by Gerv
Saturday November 3rd, 2001 12:57 PM
> PGP integrationm, which was already been written (bug 22687).
This is not true. NAI wrote a PGP implementation which was not judged technically suitable. The back end was a closed-source PGP module which has never been released. Now NAI have sold off PGP, the future of that module is uncertain, and the developer assigned to working on it is off doing something else.
Of course the current documents are all about S/MIME - because that's the only mail encryption scheme which has a working implementation. As far as I can see, the UI looks general enough to be used for either - and if not, could be generalised easily when a PGP implementation actually turns up.
The situation with PGP is this: the architectural barriers are now out of the way because the Mail/News team have written the interfaces they asked the PGP guy to write. So, anyone could write a PGP plugin along the lines of S/MIME. (You'd need to find or write PGP code that is tri-licenseable.) Once someone comes up with one, _then_ we can talk about getting it into the tree. If it doesn't get in, it might well be possible to make a few tweaks so it can be an XPI. But speculating is pointless. The people who want PGP have to go away and _write_ it.
Gerv
#12 What about enigmail?
by jedbro
Saturday November 3rd, 2001 3:37 PM
Hey Gerv? What about enigmail? http://enigmail.mozdev.org/ (-->enigmail.mozdev.org)
I haven't tried it myself, but I've heard it works just fine.. has anyone talked to the author about implementing it into the tree?
cheers
<>< Jedbro
#47 PGP has more votes than ANY other bug
by caseyperkins
Monday November 5th, 2001 9:29 PM
I don't understand why implementing PGP is not a a major priority. The PGP plugin for MailNews bug (http://bugzilla.mozilla.org/show_bug.cgi?id=22687) has 223 votes, according to Bugzilla's query page. That makes it THE most voted for bug in Bugzilla, period. It has 117 more votes than the next most "popular" bug, bug #49141. The next highest MailNews bug (#28237) only has 61 votes. Do our votes not count for anything?
Hopefully, they do. Everybody who wants this, take a few minutes to further increase the vote count. Mozilla ***needs*** PGP.
#57 Does it?
by jedbro
Tuesday November 6th, 2001 6:47 AM
Sure, you think it needs it, most "testers and developers" think in needs it.
But is PGP really used by the masses at large? I think the answer is a "No".
Don't get me wrong, I would love to see it implemented.. but I think that should be the work of a 3rd-party, not of Mozilla Coders, they have enough to do these days..
#66 Yes, it does
by caseyperkins
Tuesday November 6th, 2001 1:11 PM
I accept your argument that the masses at large don't have much use for this, but consider this:
Many technically-adept persons do want this feature, and it is among those persons that Mozilla is most likely to achieve some penetration into the market first. Those people are the ones who will advocate it to other, less technically-adept persons. I was speaking to someone just the other day, for instance, who said that he doesn't have any use for Mozilla mail until PGP is implemented.
The bug is one of the dependencies in the tracking bug "Bugs that make Mozilla advocacy harder," and rightly so.
#81 Re: Yes, it does
by asa
Tuesday November 6th, 2001 8:17 PM
And when someone who cares about PGP comes up with an implementation that can pass code review then we'll have PGP. It's as simple as that. Also, when you think about it, saying that something has 200+ votes isn't saying much. There are over 30,000 Bugzilla accounts, aboug 15,000 of them active. If a bug had 10,000 votes then I'd say that it's something that a lot of Mozilla contributors care about. Even then that's a small number compared to the people who download Milestone builds which is nearing 200,000 or so per milestone. A few hundred votes says something, and I happen to pay pretty close attention to bugs with lots of votes, but it doesn't say as much as you seem to suggest.
--Asa
#86 Re: Re: Yes, it does
by caseyperkins
Tuesday November 6th, 2001 8:40 PM
Asa,
If 227 votes (the revised count) doesn't mean anything (as you suggest), then votes in Bugzilla don't have any meaning at all, since there is no other bug with even close to that many votes.
While it's valid to say that "someone who cares" should code PGP, my question is why do not the Moz coders care about it?
Anyhow, I don't mean to be stubborn or critical; I think you guys are doing a great job, and I've been using Moz as my browser and mail client for over a year. But I do think the lack of PGP is a fatal flaw for many people, at least as far as MailNews is concerned. Since I'm not capable of implementing it myself, I'm just trying to raise the profile of the bug so maybe someone will.
Casey
#178 I agree.. but..
by jedbro
Thursday November 8th, 2001 6:03 PM
True.. I also would LOVE to see it implemented...
But I still stand firm that it isn't a 1.0 requirement.
1st, I would like to see a performace increase, nasty bugs out, API freeze, and then we can work on PGP...
Enigma, already allows you to use PGP in mozilla for those that are blood thirsty, and once the S/MIME get's in an freezes, plugins for PGP should me easy to implement.
Let's hope someone does =)
Cheers
--Jedbro
#184 Re: Re: Re: Yes, it does
by asa
Thursday November 8th, 2001 11:43 PM
Casey, don't misquote me. I said "A few hundred votes says something, and I happen to pay pretty close attention to bugs with lots of votes, but it doesn't say as much as you seem to suggest." Nowhere in that statement do I suggest that 227 votes doesn't mean anything. I specifically said that is does mean something and that it means something to me. What it means is that a very small minority of of bugzilla bug reporters and commenters care more about that one issue than they do about other issues and took the time to note that concern with a vote (and probably a comment). If all of mozilla's major bugs were fixed and I had the time to implement a new feature, had the ability to implement features and had no particular itch to scratch I would probably implement PGP. Unfortunately I don't have the time or the ability to implement Mozilla features. What I can do is try to focus resources where they are most needed and right now that certainly isn't working on new features. PGP would certainly make some number of people happy but I'd wager that more people using Mozilla and Mozilla-based browsers care about silly bugs like 59108 than about PGP.
--Asa
#82 A friend of...
by pbreit
Tuesday November 6th, 2001 8:23 PM
A friend of a friend of my cousin's boss's wife said he had no use for Mozilla until it can do gausian blurs!
#85 the votes speak for themselves, smarty
by caseyperkins
Tuesday November 6th, 2001 8:32 PM
You can mock my story, but it's true. Furthermore, because of the overwhelming abundance of votes for PGP, I believe it can be taken to be somewhat representative.
Don't you have anything sustantive to say?
#225 Re: Does it?
by benb
Friday November 9th, 2001 9:17 PM
> But is PGP really used by the masses at large?
Yes, outside of enterprises (companies with thousands of employees), PGP is *the* encryption scheme.
#230 Re: Re: Does it?
by strauss
Saturday November 10th, 2001 2:49 PM
> > But is PGP really used by the masses at large? < <
> Yes, outside of enterprises (companies with thousands of employees), PGP is *the* encryption scheme. <
It probably is dominant in that sense, but in absolute terms it's still not widely used. One reason is that it's badly designed. Probably a few people here have seen the C-MU usability study on PGP at http://reports-archive.adm.cs.cmu.edu/anon/home/ftp/usr0/ftp/1998/CMU-CS-98-155.pdf . It demonstrates that even computer science professionals often find it hard to understand PGP, and that non-gearheads almost always have big problems. Unfortunately, the usability hasn't improved significantly since that study.
#231 Re: Re: Re: Does it?
by roc
Saturday November 10th, 2001 5:10 PM
Alma picked Mac PGP because she thought it had a decent UI. IIRC I think she suspects other mail encryption tools have at least as many usability problems as she found in her study of PGP.
But you're right that S/MIME and other encryption schemes (e.g., Lotus Notes' scheme) are dominant in large organisations. I think that's because CAs and certificate hierarchies are naturally suited to large hierarchical organisations, and PGP's web of trust is more suited to loose-knit groups and networks of individuals.
#235 Re: Re: Re: Re: Does it?
by gwalla
Sunday November 11th, 2001 5:11 PM
AIUI, there's nothing really preventing S/MIME from being used in a web-of-trust fashion, or PGP from being used with the CA hierarchy model, specification-wise. It's just a matter of how existing tools have been designed.
#224 PGP *is* there, since long
by benb
Friday November 9th, 2001 9:12 PM
> NAI wrote a PGP implementation which was not judged
> technically suitable.
No, it as good as the rest of Mozilla.
> Of course the current documents are all about S/MIME -
> because that's the only mail encryption scheme which has
> a working implementation.
You write the spec first and then implement. And we have working backend implementations for S/MIME (PSM) just as we have for PGP (GnuPG).
> the architectural barriers are now out of the way because
> the Mail/News team have written the interfaces they asked
> the PGP guy to write.
No, they didn't! Neither did the PSM folks.
> Once someone comes up with one, _then_ we can talk about
> getting it into the tree
Just to have it rejected *again*? I have better things to do.
> The people who want PGP have to go away and _write_ it.
They did, and all they faced was opposition by the Mailonews owners.
#222 Re: PGP is not even considered
by benb
Friday November 9th, 2001 9:09 PM
> NAI wrote a PGP implementation which was not judged
> technically suitable.
No, it as good as the rest of Mozilla.
> Of course the current documents are all about S/MIME -
> because that\\\'s the only mail encryption scheme which has
> a working implementation.
You write the spec first and then implement. And we have working backend implementations for S/MIME (PSM) just as we have for PGP (GnuPG).
> the architectural barriers are now out of the way because
> the Mail/News team have written the interfaces they asked
> the PGP guy to write.
No, they didn\\\'t! Neither did the PSM folks.
> Once someone comes up with one, _then_ we can talk about
> getting it into the tree
Just to have it rejected *again*? I have better things to do.
> The people who want PGP have to go away and _write_ it.
They did, and all they faced was opposition by the Mailonews owners.
#223 PGP *is* there, since long
by benb
Friday November 9th, 2001 9:10 PM
ops, reply to wrong post.
#9 Moz Mail - Planned compatibility w/Norton AV 2001?
by JayeshSh
Saturday November 3rd, 2001 1:01 PM
Hello All,
I am using Netscape 6.1 (based on the 0.9.2 build), and have noticed that the email program does not function with Norton Anti-Virus\\\'s \\\"POPproxy\\\" scanning system.
Since viruses are often sent through email, would Mozilla consider ensuring compatibility with such anti-virus scanning programs for the 1.0 release?
Additionally, the IMAP function for Netscape Mail does not work either. Any information on why that could be occuring (related to Norton AV?) would be appreciated. (I also have Norton Personal Firewall 2002 running - would that cause problems with POP or IMAP email access?)
Thanks,
- Jayesh
#10 Re: Moz Mail - Planned compatibility w/Norton AV 2
by niner
Saturday November 3rd, 2001 1:34 PM
for such problems you can determine very easy wheres the problem. Just shut down the firewall try IMAP again and see if it works. If then maybe in your Firewall the IMAP port 143 is not open.
#11 Re: Moz Mail - Planned compatibility w/Norton AV 2
by tny
Saturday November 3rd, 2001 3:03 PM
Jayesh,
On my XP box (XP Pro) Norton AV2002 seems to be working with Mozilla Mail/News. It is possible that a reinstall of Norton might resolve your problems.
#13 Re: Moz Mail - Planned compatibility w/Norton AV 2
by Tanyel
Saturday November 3rd, 2001 4:35 PM
I think the Norton Antivirus documentation explains how to configure e-mail clients that are not automatically configured. Even if it did not, I think the software would catch the virus after it was downloaded.
#16 Re: Moz Mail - Planned compatibility w/Norton AV 2001?
by taxexile
Sunday November 4th, 2001 7:35 AM
The Nortons AV email protection works by using a local POP proxy server. This is a service that is running on your machine, that is listening on POP's port.
To use it, change the following settings to get your mail via the POP proxy:
incoming POP server: pop3.norton.antivirus [which incidentally, is your machine, or localhost or 127.0.0.1]
account id/name: this is comprised of your account name [on the Real POP server], a forward-slash, and then the address of the real POP server you want to connect to. i.e taxexile/pop3.myisp.co.uk [if my id were 'taxexile' and my ISPs POP server was pop3.myisp.co.uk
best o'luck
#14 please
by ihxo
Saturday November 3rd, 2001 5:23 PM
for christ sake can mozilla be just mozilla ?? I only want a browser that works. and have otehrstuff like calendar mail news be some addon item.
#15 Then..
by jedbro
Saturday November 3rd, 2001 6:31 PM
Then don't bitch and download the browser-only K-meleon.
http://kmeleon.sourceforge.net/
It's incrediby fast/stable/ and did I mention fast???
Cheers
--jedbro
#23 please
by pbreit
Sunday November 4th, 2001 11:43 PM
that's not it. the problem is that because mozilla was scoped to provide all that extraneous stuff (mail, news, edit, skins, chat, etc) they have been disasterously late at getting to 1.0 (and have thus lost the browser battle). rule #1 of software development is scope management and mozilla has failed at that pretty miserably. there are dozens of mail/news/chat/edit clients on every platform. there's no reason for mozilla to build them.
if they had concentrated on building a high performance, standards-compliant, stable, small footprint *browser* there might have been hope. unfortunately, they did the opposite and we're stuck with IE.
#25 Re: please
by macpeep
Monday November 5th, 2001 2:49 AM
Knowing what the response will be, I'll answer some of your comments before the mob gets you. (It can be ugly.. I've been there..)
Some of the stuff (the IRC client is a great example) is coded by volunteers who are only doing this because they cared about that particular feature. It's not like a lack of IRC client would have meant more resources to work on the web browser part. On the other hand, added features ALWAYS mean added risk and more testing, so stuff like that DO have a resource cost and you are definitely right about the scope management. Keep in mind though that the IRC client isn't even in Netscape 6.x (at least I don't think it is...) so the cost of that hasn't been too tremendous.
As far as "being stuck with IE" goes.. I don't know.. I don't care about all the politics of "Microsoft sucks and should be destroyed" and all that crap. To me, there are a number of web browsers all with their positive and negative sides. Currently in my score card, IE has the #1 spot bug Mozilla is closing in fast. I don't feel that I'm "stuck" with IE. I kinda like it.. The UI is simple, the browser is VERY fast and extremely stable. I have always had the latest version of IE as soon as they got out, ever since 5.0 and I really have only had a handful of browser crashes on all three computers (all Win2K boxes) that I use on a daily basis at work and at home (plus Pocket IE on my Compaq iPAQ) over the time span of something like 2-3 years. And never have the crashes taken down the entire OS or even shell like so many people claim. I've seen that on Win98 though.. but that's a toy OS anyway.
by pbreit
Monday November 5th, 2001 10:53 AM
Oh, I definitely agree that IE is a pretty nice browser. It's still quite a bit better than Moz/Netscape as far as I can tell. I've been trying to use K-meleon, moz and netscape as my primary browser and they just don't seem to work right. I get some pretty bizarre rendering on fairly normal sites and quite frequently things just don't even work (javascript? dhtml?). At a complete minimum, Moz/Netscape would need to render as reliably as IE and right now, I don't believe that it does. And having the basics not work right makes all the extraneous features even more irritating.
Also, I do understand that reallocating all the follks working on chat, news, mail, skins, edit, etc. wouldn't necessarily make Gecko dev faster ("programmers pardox"), but I've got to think it would help. For example, just having a look at bug-zilla and the weekly status updates makes it seem that like 50% of the total effort is being devoted to superfluous components.
#64 Rendering and JS issues
by SubtleRebel
Tuesday November 6th, 2001 1:07 PM
I do not know which websites you are referring to, so I can not really comment too much, but I will say that some of the websites that do not look right or have JS errors with Mozilla,but are ok in IE, are a result of non-standard HTML and JS. If a web designer utilizes some proprietary MS code in their website, Mozilla will not necessarily support it. IMO, this is a good thing; it definitely should not change until after everything else is done.
#67 Re: IE
by SubtleRebel
Tuesday November 6th, 2001 1:14 PM
Who are you suggesting should do this reallocation? I can not think of anyone in the Mozilla organization who has that kind of authority.
Who are the programmers responsible for the 50% of effort devoted to the "superfluous components" that you mention? You can pretty much guarantee that very few, if any, of those programmers are employed by the Mozilla organization.
#32 that wouldn't serve their customer base.
by joschi
Monday November 5th, 2001 10:35 AM
Netscape's product has been a full Inernet browsing sweet for many major releases now. If they produced just a browser, they would not have fullfilled their commuitment to their customer base.
#39 customer base?
by pbreit
Monday November 5th, 2001 12:23 PM
I think the commitment to remain viable would out-weigh and feature commitments to the existing small and dwindling customer base of which only a sub-set actually used the extraneous features.
One major hope is tha AOL switches to Gecko. But this cannot happen until Gecko matches IE in performance, reliability, rendering expectations, etc. And AOl has no need whatsoever for mail, news, chat, skins and edit.
#135 Re: please
by Ugg
Wednesday November 7th, 2001 11:04 PM
"because mozilla was scoped to provide all that extraneous stuff "
Mail/News and Composer are required to replace/duplicate Netscape 4 functionality, so even if outside Mozilla volunteers didn't want to work on them, Netscape would anyway. Skinning is a side effect of the extensible cross-platform UI. (And I award you one Clueless Point for not understanding that even after all this time.)
"they have been disasterously late"
You can't be late without a deadline, and AFAIK, Mozilla has never had one.
"lost the browser battle"
Yawn. That's two CPs. And while I'm at it, who made you judge and jury?
To lend some substance to my remarks, I've seen sites reporting that hits from Mozilla browsers have increased significantly since around Netscape 6.2 or so. That doesn't sound to me like a lost battle.
"if they had concentrated on building a high performance, standards-compliant, stable, small footprint *browser*"
They did.. well, except for the footprint, but that's the price of progress, isn't it? Have you tried using Mozilla to surf the web? Works pretty well. RAM is dirt cheap now; go buy yourself a gig and stop whining. (To those who argue that they shouldn't have to upgrade their hardware to use new software, I say, "fine, then keep using the old software", as well as "wtf are you on?".)
"we're stuck with IE."
I guess there's no escape from a prison that only exists in your mind.
#164 Re: Re: please
by strauss
Thursday November 8th, 2001 1:25 PM
> Skinning is a side effect of the extensible cross-platform UI. <
Which is to say, this unneeded feature is a side effect of failing to respect platform user experience standards, itself a terrible design mistake, and one which now makes Mozilla seem distinctly "previous generation" on Mac OS X and Windows XP.
#166 Re: Re: Re: please
by Ugg
Thursday November 8th, 2001 1:44 PM
"Which is to say, this unneeded feature is a side effect of failing to respect platform user experience standards, itself a terrible design mistake..."
This "unneeded" feature is the reason why Mozilla is cross-platform.
I have an experiment for you, strauss, since you're unfailingly critical of Mozilla, and since you don't seem to think that Mozilla should exist on more than one platform: pretend that there is no Mozilla for your computing platform of choice. Use some other browser, because Mozilla is so hopelessly flawed that even if there were a version you could use - which, remember, there isn't! - you still wouldn't want to. And then, realizing that Mozilla's successes or failings will never affect you, kindly go away and STFU, since your only reason for coming here must be to troll.
My hypothesis is that everyone will be pleased. Care to test?
#168 Re: Re: Re: Re: please
by strauss
Thursday November 8th, 2001 2:02 PM
> This "unneeded" feature is the reason why Mozilla is cross-platform. <
Entirely false. It is quite feasible to do cross-platform development and use native widgets. I've done it, and it's considered one of the coolest features of the new Eclipse/SWT platform from IBM as well. You just set up a thin cross-platform widget layer and code to that. The reason you do it is because platform standards are the most important component of user-friendliness on a platform.
> since you don't seem to think that Mozilla should exist on more than one platform <
As usual, pro-Mozilla trolls have no qualms about resorting to distortion.
#169 Re: Re: Re: Re: Re: please
by roc
Thursday November 8th, 2001 2:24 PM
Ah, the old native widgets debate.
Eclipse/SWT is cool, but it does not prove anything relevant to Mozilla.
Here are some things that are very hard to do using native widgets:
-- Fully implement the DOM event model (e.g., allow capturing mouse clicks in widgets).
-- Fully implement CSS styling (e.g., support background images in edit controls, clip and z-order widgets properly, render widgets covered by overlapping translucent images, support HTML I18N features).
-- Make it all robust (e.g., create 500 comboboxes instantly, make sure you can put arbitrary amounts of Unicode text into a text widget).
Eclipse doesn't have to do any of these things.
I think the ultimate proof that native widgets are not the way to go is that Microsoft themselves don't use native widgets in IE! Even though it would be much easier for them to use native widgets than for Mozilla. (They only have to support one platform, they have the source code to the native widgets and they could even improve the native widgets to meet their needs if they wanted to.) But they still chose to implement their own special widget set just for IE.
#177 Re: Re: Re: Re: Re: Re: please
by strauss
Thursday November 8th, 2001 4:31 PM
> Fully implement the DOM event model (e.g., allow capturing mouse clicks in widgets). <
I've written plenty of event handling code in Windows and Mac and don't know what problem you're referring to here. The system hands the event to the application first.
> support background images in edit controls <
Background images _in_ widgets is not a CSS feature. The background of a widget is what it's drawn above, not what's drawn inside it.
> clip and z-order widgets properly <
Widgets are drawn in a z-ordered fashion on both Mac and Windows. Please note, however, that overlapping widgets is almost always a design mistake.
Clipping is fully controllable through the GrafPort or drawing context. However, please note that clipping to only part of a widget is also a design mistake -- even more so on a theme-switching system where you can't anticipate what will be shown and what removed. Clipping out the whole thing is fine, and trivial to do.
> render widgets covered by overlapping translucent images <
What problem do you anticipate there? If you have translucency, you have to draw to an offscreen pixmap in layer order, but that's the same with or without widgets, and has long ben a standard drawing mode on both platforms.
> create 500 comboboxes instantly <
Instantly? You sure you're a Mozilla user? Anyhoo, I don't see what's faster about creating 500(!) XUL widgets as opposed to native widgets.
> make sure you can put arbitrary amounts of Unicode text into a text widget <
Text editing is a special case and most applications do roll their own when they have to do this kind of thing.
There aren't any CSS requirements that are broken by native widgets. In fact CSS3 rerquires using the platform standard appearance.
> I think the ultimate proof that native widgets are not the way to go is that Microsoft themselves don't use native widgets in IE! ... They only have to support one platform... <
Microsoft does use native widgets in IE, and they support more than one platform.
#191 Re: Re: Re: Re: Re: Re: Re: please
by roc
Friday November 9th, 2001 9:44 AM
> The system hands the event to the application first. <
So on Windows you want to capture the messages in the GetMessage()/DispatchMessage() loop, run the Web page's Javascript in there, and use the results to futz with the messages before dispatch? Do you have any idea how error-prone that would be? Do you have any idea what Windows might do if the Web page decided to (for example) cancel all the mouseup (but not mousedown) messages for a control? Plus it simply won't work if Windows decides to SendMessage() some of the messages you need to capture.
> Background images _in_ widgets is not a CSS feature.
That is technically true, but people do it and IE and Mozilla support it. (CSS borders on edit controls get used a lot.)
> Widgets are drawn in a z-ordered fashion on both Mac and Windows. <
True, but their z-order support is not sufficient for our purposes. Suppose I have a text element (with transparent background) that the Web page author has positioned overlapping an edit widget, and above the edit control in the z-order. How would you render this with native widgets in Windows?
> Please note, however, that overlapping widgets is almost always a design mistake. <
Tough luck, Mozilla still has to render it right. Welcome to the real world.
> Clipping is fully controllable through the GrafPort or drawing context. <
On Windows (dunno about Mac) you can't clip native widgets this way because you don't get to set the clip region on their DC when they paint. (How much do you really know about Windows programming anyway?. VB doesn't count.)
> However, please note that clipping to only part of a widget is also a design mistake ... Clipping out the whole thing is fine, and trivial to do. <
That would not be "fine". Mozilla has to render all pages correctly, even if the results are ugly.
> If you have translucency, you have to draw to an offscreen pixmap in layer order, but that's the same with or without widgets, and has long ben a standard drawing mode on both platforms. <
Do you have any idea how hard it is to get a Windows native widget to render into an off-screen buffer, or anywhere except the screen? The code that tried to do this for Netscape 4 printing (with less than complete success) was sprinkled with nasty comments about "Jedi mind tricks". If you think this is so easy, show me some sample code that renders a simple Windows edit control into an off-screen buffer. Please use only documented Windows APIs.
> Anyhoo, I don't see what's faster about creating 500(!) XUL widgets as opposed to native widgets. <
Well, for example, on Windows 95/98 this would probably hang your system by running it out of "System Resources".
> Text editing is a special case and most applications do roll their own when they have to do this kind of thing. <
Ah, HERE we go. If you're going to do your own text editing, which is the most complex widget by far, then you might as well go the whole hog.
> In fact CSS3 rerquires using the platform standard appearance. <
I think not. Citation?
> Microsoft does use native widgets in IE <
WinIE 5.x certainly does not (I haven't tried IE6). Just fire up IE, bring up a page with text edit controls, and use Spy++ to try to select the native edit controls in IE's window. You can't because they're not there.
> and they support more than one platform. <
MacIE is a completely separate code base from WinIE, so having WinIE use Windows native widgets wouldn't have created any cross-platform coding problems.
#212 Re: Re: Re: Re: Re: Re: Re: Re: please
by strauss
Friday November 9th, 2001 2:04 PM
Things you're saying are dangerous are in my experience things that are normal. In many cases they're things that have long been encapsulated in COM objects in the Platform SDK, like offscreen drawing. You seem to be talking about an implementation strategy that uses only the base Windows API from 3.x rather than the current state of the API.
I don't use VB, thank you very much, and yeah, that is fairly insulting. All my Windows development from 3.1 to the present has been in C++.
I will, on reflection, modify a couple of points. Partial clipping of widgets could be important to animation effects such as wipes. You woudln't want to leave them in that state but during intermediate states you'd definitely want to show a clean line. I don't see that you've demonstrated a problem doing that, though. Ditto for z-ordering, since during animation widgets might fly over each other. Again, no problem with doing that has been demonstrated.
GDI heap overflow was a serious bug in Windows in past versions, to be sure. Theoretically you can render more complex web pages on those older versions w/o using native widgets because that bypasses the OS bug. However, I don't think that a 500-widget edge case on older OS versions justifies a complete abandonment of platform standards.
The CCS3 user interface proposal is at http://www.w3.org/TR/css3-userint . It calls plainly for "system default look and feel." Looks like they got some UI guys involved in this one. (Finally....)
Windows IE uses native widgets. They may very well roll their own text editing -- most serious applications do, on either platform. The base text editing capabilities on either Mac or Windows are suitable mostly for small text fields. Why don't you try looking for buttons, scroll bars, etc. in addition to text fields, though?
Oh, and BTW, there is a code fork between the Windows and Mac IE platforms, but not a "completely separate code base." I know this from working directly with the Mac IE engineers on a project at a previous company. It's more of a fork than Mozilla would like to have, and so it doesn't bear directly on the feasibility of a platform-independent widget layer, but it does show that they were able to use native widgets on both platforms.
#215 Re: Re: Re: Re: Re: Re: Re: Re: Re: please
by roc
Friday November 9th, 2001 3:29 PM
> In many cases they're things that have long been encapsulated in COM objects in the Platform SDK, like offscreen drawing. <
Great. Then show me the code, or even just give me an MSDN URL to something that will let me render a native widget into an offscreen buffer.
> Again, no problem with doing that has been demonstrated. <
I have asked you how to render text overlapping a native widget, and you haven't answered it. Until you do, I consider the problem demonstrated.
> However, I don't think that a 500-widget edge case on older OS versions justifies a complete abandonment of platform standards. <
That "edge case" happens every time a Slashdot user gets moderation points and views a popular article. "Older OS versions" includes last year's release of Windows ME. And this was just one example of how you can get burned by relying on platform widgets. Anyway, put down your straw man --- no-one's advocating a complete abandonment of platform standards, and nothing rests solely on this particular case.
> The CCS3 user interface proposal is at (LINK) . It calls plainly for "system default look and feel." <
Mozilla's custom widgets are designed to provide look and feel that matches the system. This does not mean, however, than they have to be implemented as actual native widgets.
> Windows IE uses native widgets. They may very well roll their own text editing -- most serious applications do, on either platform. The base text editing capabilities on either Mac or Windows are suitable mostly for small text fields. Why don't you try looking for buttons, scroll bars, etc. in addition to text fields, though? <
OK. I just checked IE5.5. Radio buttons, scrollbars and push buttons are all custom, not native. Ironically the only native widgets I can find inside an IE content area are comboboxes. So what you should have said is, "Windows IE uses native widgets --- except most of the time, they don't." Your credibility is dropping fast, I'm afraid.
Incidentally, Microsoft Office also doesn't use native widgets.
#216 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: please
by strauss
Friday November 9th, 2001 5:29 PM
Well, that's fine. I don't have a Windows development system at present (thank goodness), and I couldn't spend days working on feasibility studies for free to settle a BBS argument if I did. But you have not done anything more to demonstrate the supposed impossibility of these things than I have to demonstrate their feasibility. I personally have always found that engineers conjure a lot of fake impossibilities as excuses for what they want to do, and I have no reason to think you are not doing that here. I am much more of a Mac engineer than a Windows engineer (again, thank goodness), but I frequently run into Mac people telling me things are impossible when all they need to do is, for instance, install a QD bottleneck procedure in the GrafPort, or at very worst patch a trap. I know that Windows has many similar bottlenecks and intercepts, and based on your attitude, I really doubt that you are trying to figure out how to use them in these cases. One particular reason I have for doubting that in this case is your weird statement that something would go horribly wrong with handling events in the main event loop of the application, which I know is a standard practice.
The one issue that is easily settled here is about CSS3. Mozilla widgets do not emulate platform widgets. They are not allowed to imitiate Aqua for intellectual property reasons. That means that Mozilla will never look like a Mac application without using the native control set, and will never be able to implement CSS3 recommendations about default system look and feel. On Mac OS 9 it's bad enough to have something that looks like Windows 95; on Mac OS X, it's distinctly ugly and regressive.
#220 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: please
by SmileyBen
Friday November 9th, 2001 8:00 PM
And, gosh, I bet Mozilla developers lie awake at night (as you seem to) worrying about this particular feature of CSS3, a standard set the browser doesn't yet support.
Oh an by the way, Apple did climb down a bit about Aqua - they've said that it is okay to develop Moz widgets to mimic native ones, but, I believe, only if they only work on the mac. Whilst a fork to deliberately break one theme's cross-platform compatibility doesn't look like the first thing people will work on, to say 'never' is a bit extreme - and you can be pretty sure that Netscape will be at least considering this for their closed-source mozilla-based browser.
#228 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: pl
by choess
Friday November 9th, 2001 11:35 PM
In fact, Mozilla shouldn't be supporting CSS3; most or all of it is still in working draft stage, and subject to change. (See the MS XSLT fiasco for why supporting working drafts is bad). All CSS3 support in moz uses -moz-name values, rather than the official values.
#229 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re
by strauss
Saturday November 10th, 2001 1:18 PM
> In fact, Mozilla shouldn't be supporting CSS3 <
Correct, at this point. However, the user interface proposal is a pretty good indication of what it will need to support in the future.
by SubtleRebel
Monday November 12th, 2001 9:48 AM
I am sure that when CSS3 is finalized, Mozilla will be among the first in line to support it.
Until then, lets focus on the things that need to be done now instead of future features. ;-)
#227 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: please
by roc
Friday November 9th, 2001 10:50 PM
> I know that Windows has many similar bottlenecks and intercepts <
Name one. Come on, you're an experienced Windows programmer. Just name one. (Hint: I'm thinking of TWO now, and I didn't have to look them up. (But they're not good enough for what's needed.))
Honestly, I have presented many facts that invalidate your arguments, and you have ignored them. Then when I ask you for specific facts to back up your arguments, you can't, and instead resort to name-calling. This is disappointing.
> One particular reason I have for doubting that in this case is your weird statement that something would go horribly wrong with handling events in the main event loop of the application, which I know is a standard practice. <
Not in Windows. In Windows events are dispatched through a very standard GetMessage()/TranslateMessage()/DispatchMessage() loop to be processed by each window's WndProc. In most frameworks, including MFC, you don't even write that loop, it's given to you. In Windows, you do all your event processing in the WndProc attached to each window. The reason for that is that only posted messages go through the application main event loop, and someone might use SendMessage to send an event directly to your window, and those messages don't go through the application's main event loop. (Well, threading complicates the picture a bit, but that's the basic idea for single-threaded applications.)
#217 Time setting
by SubtleRebel
Friday November 9th, 2001 5:43 PM
Your comments about the current state of Windows APIs versus the state of Windows APIs in the past fails to take one thing into consideration - when Mozilla made the decision to go with non-native widgets was back when the Windows APIs were not as good. So even if it were possible to use native widgets now, it was definitely not possible then.
Of course, anyone who wants to work on developing a Mozilla based browser using native widgets can still do so.
#188 pro-Mozilla trolls ?
by SubtleRebel
Friday November 9th, 2001 9:09 AM
What exactly is your definition of a troll?
#18 Re: please
by tny
Sunday November 4th, 2001 11:02 AM
I suppose that you didn't notice that one can choose to install only the browser from the installer?
Some of us DO need the mail client . . . I can't use Outlook Express anymore, too many virus problems; and Eudora doesn't handle everything I want it to. I also like having the same mail client on all three platforms. Mozilla is designed to replace all the functionality of the old Netscape Communicator package - mail, composer, and browser.
#19 Re: please
by asa
Sunday November 4th, 2001 12:34 PM
"for christ sake can mozilla be just mozilla ??"
You mean you want us to stop making binaries avialable and let mozilla just be mozilla (a collection of sourcecode available from cvs.mozilla.org)? I'd rather not do that. We get too much value from the testing community, a big part of which would probably not be contributing if they were required to make their own test builds. If you don't like the binaries we provide for testing then feel free to build your own or to find someone that makes builds to your liking.
--Asa
#49 I think he talks about a browser only Mozilla (nt)
by dipa
Monday November 5th, 2001 11:50 PM
#59 Re: I think he talks about a browser only Mozilla
by Tanyel
Tuesday November 6th, 2001 9:53 AM
I think that is easily understood. I think Asa just wants to argue.
#69 My attempt to guess Asa's motivation
by SubtleRebel
Tuesday November 6th, 2001 1:35 PM
I think Asa fully understood what the guy was saying, but Asa also knows that Mozilla is not just a browser. Obviously anyone saying "why can't Mozilla just be Mozilla" does not have any real comprehension of what Mozilla is. So Asa posted about an aspect of Mozilla that the guy had probably never even considered before.
It is pointless to argue with someone who does not know what they are talking about. Asa has plenty of better things to do than argue.
By posting the message that he did, Asa hopefully caused some people to think for a moment and maybe gain a better understanding of what Mozilla is really about.
#24 Re: please
by macpeep
Monday November 5th, 2001 2:38 AM
Mozilla is Mozilla. Duh! From Mozilla we have various products like Sea Monkey, the "browser suite" as Mike Angelo would say. That's in theory a reference implementation, tho it's a little silly to say that since Netscape 6.x and Mozilla have about 99.999999% in common. There are also other products based on Mozilla tho. K-Meleon and Galeon are lightweight browser-only products based on Mozilla - not on Sea Monkey. Also, if and when AOL will at some point ditch IE in favor of Mozilla, you'll likely to see a very "different" incarnation of Mozilla code than Sea Monkey.
Personally, I think Sea Monkey is very, very close to the ideal browser. I still use Netscape 4.7x for email even though I've ditched Communicator for web browsing years ago in favor of IE. Sea Monkey seems like both a great browser and a great email & news client if only some of the bugs would be ironed out. I don't care all that much about the editor part but then again, since it's needed for other than just a stand alone editor, I don't mind really. And hey! If it becomes good enough, it could even be used as a word processor.
For me, the "form factor" (yes, I'm using this term outside of its actual meaning) is great. I'm sure many agree. I don't mind if the web browser is stand alone either, but then I need a separate good email client and at the moment, the best one I've found is Netscape Communicator 4.7x. If the two would be in the same app, that would just be an added bonus IMHO since it would mean faster switching between the two apps and more seamless integration for things like clicking on mailto: links etc.
The only problem I have with Mozilla at the moment is the lack of UI quality. Before, it was a performance, stabilty and polish problem. Now, I think mostly it's just a polish problem.
Maybe 5 more months... Maybe 5 more milestones...
#41 exactly!
by pbreit
Monday November 5th, 2001 1:41 PM
It's guaranteed that a group focused on making a browser or a group focused on making a mail client is going to have better results than a group developing both (and news, and edit, and chat, etc. etc.). The benefit of a combined browser/email client is almost zero. All that needs to happen is for a mailto link to open up a compose window in your default email client. If the email client wants to use gecko, fine. The task switching shouldn't be an issue. IE and Outlook are separate programs and the switching between them is more than OK.
If AOL ever switches to Moz, they won't use anything beyond Gecko. And they simply won't be able to switch to it until it's at parity with IE. Since getting bundled in AOL would be a massive, massive victory, I'm surprised they would risk it by not focusing on Gecko only.
#45 combined browser/email client has major benefits
by SubtleRebel
Monday November 5th, 2001 7:21 PM
I suppose it all depends on how you use your web browser and email client, but I find have the integration extremely useful.
Also, regardless of how OK you think it is to switch back and forth between IE and Outlook, what advantage do you find in having them as separate apps?
#48 many advantages to specialization
by pbreit
Monday November 5th, 2001 9:53 PM
the obvious advantage is due to the age old reality that specialization leads to superior results. it's not so much that "dis-integrated" apps are "better", it's that it's extremely difficult to do one thing well, much less 5. nowhere is this more evident than with the mozilla project. unfortunately, building a world-class browser is really hard work, made all the more difficult by also attempting to build a mail client, a news client, a chat client, an html editor and skins. None of which are crucial, necessary or even important to *browsing*.
i'd like to hear one (one!) example of how integrating an email client into a browser is "extremely useful" (note: handling mailto tag doesn't count as this is simply done with two non-integrated clients like IE and Outlook). then ask yourself, "would i prefer to have moz 1.0 in march 2002 with this 'extremely useful' feature or in august 2000 (a generous 2.5 years after mozilla's birth) without it?"
#51 Re: many advantages to specialization
by gwalla
Tuesday November 6th, 2001 12:07 AM
"the obvious advantage is due to the age old reality that specialization leads to superior results."
This argument would carry more weight if Outlook (your example) wasn't a horrible, useless piece of junk.
Besides, the Mozilla browser and mail/news projects are specialized: different teams work on each.
by pbreit
Tuesday November 6th, 2001 1:03 AM
last i checked, microsoft made more than email clients! try again.
that's what everyone says, each component has its own group of developers. the problem with that argument is that bug-zilla is littered with their bugs, weekly status reports packed with their updates, precious management energy to guide progress, megabytes of extra code, etc. etc. etc.
#54 Re: come on guys
by macpeep
Tuesday November 6th, 2001 4:49 AM
Come on.. It's no use arguing over this. It's obvious that doing two things is harder than doing just one. It's simply a matter of how much harder it is and how this weights against the benefits of integration.
Someone said, earlier in this thread, that specialization yields superrior results. That's not always true. I work for a company that develops a 3D game engine. The very value of a game engine is that it solves several things - combined - in one piece of software. Instead of game developers having to get a 3D game engine from one place, a physics engine from another place, a sound engine from a third place, tools from a fourth and so on, they can get it all in one package. Integration is the value here. In the same way, in a "web browser suite", I see some value of having the apps rolled into one. Not huge benefit, but some. For example shared resources and libraries result in smaller code. The extra footprint added by the editor in Mozilla, I would guess, is absolutely minimal. Same goes for the IRC client and address book. The mail & news client is more complex but even here, I'm sure it's much smaller than a separate, stand alone one with the same features.
Another big benefit is unity of interface and the things you can do through integration. The AOL instant messenger client can run in the sidebar and you there's a high level of integration between it and the mail & news client. It could integrate even further to, for example, allow XUL based collaboration tools to run on top of Mozilla while you talk with your co-workers on AIM..
#68 I agree with macpeep.
by SubtleRebel
Tuesday November 6th, 2001 1:25 PM
Very well said macpeep.
(normally I would not post a superfluous agreement post like this, but since I have been accused of only posting "attacks" on idiots, I thought I'd make an effort to point out when I agree too)
#55 Re: many advantages to specialization
by AlexBishop
Tuesday November 6th, 2001 6:42 AM
"i'd like to hear one (one!) example of how integrating an email client into a browser is 'extremely useful'"
You can check mail in the background while browsing the Web.
Alex
#56 Re: Re: many advantages to specialization
by AlexBishop
Tuesday November 6th, 2001 6:44 AM
Wait... I've got more. What about being able to render advanced HTML inside email messages?
Alex
by pbreit
Tuesday November 6th, 2001 12:57 PM
that's exactly right but that's embedding mozilla in a mail client. it's the reverse which is a bad idea. that's the beauty of creating a base browsing component. any app can use it if it's designed right.
not sure why you couldn't read email and browse at the same time with non-integrated clients.
keep tryin'!
#70 Re: yep
by SubtleRebel
Tuesday November 6th, 2001 1:57 PM
"that's exactly right but that's embedding mozilla in a mail client. it's the reverse which is a bad idea. that's the beauty of creating a base browsing component. any app can use it if it's designed right."
If you have a Mozilla web browser app and a separate email app with Mozilla embedded into it, then you have duplicate code using extra resources. If you combine the two into a suite the way it is now then much much less code is duplicated.
(BTW, the Mozilla email client is not embedded in the browser. It is integrated.)
Also if I am browsing the web and want to send a quick email message to someone, Mozilla allows me to do it, but IE makes me launch a separate app.
Likewise Mozilla makes it much easier to forward a web page to someone; with IE, the email app has to be launched and the web page info has to be transferred to the email app, but with Mozilla it is already right there.
Another benefit is that if I want to work offline, I only have to do it in one place for both email and web browser.
How about if I want to hide all my email and web browser windows at once? Mozilla enables me do it in one step.
When I am reading an email with a link to a web page, Outlook would have to launch a web browser app, but Mozilla does not.
Although this has not been an exhaustive list of benefits by any means, I have to stop for now. I'll be back later.
#91 Re: Re: yep
by rgelb
Tuesday November 6th, 2001 11:53 PM
>>If you have a Mozilla web browser app and a separate email app with Mozilla embedded into it, then you have duplicate code using extra resources. If you combine the two into a suite the way it is now then much much less code is duplicated. <<
Not really, the rendering engine (ie. gecko) is a dynamic lib (DLL in Windows), so there would be less duplicate code. When it is combined, you are still using memory resources that you would with a DLL.
I really liked the integration in NS 4.7x. That product had charisma and that "just right" feeling, especially the mail client. Moz integration just doesn't have that for some reason. Maybe it is the slow speed, but it just doesn't feel right. This is difficult to quantify, of course. Maybe it is the 2 second pause before Mozilla brings up the mail client, fired from the browser.
#149 Re: Re: Re: yep
by SubtleRebel
Thursday November 8th, 2001 7:28 AM
Shared libraries between separate browser and email clients can definitely help reduce the amount of duplicate code, but it would not eliminate it completely. If the web browser and email client were developed totally apart from one another then that makes it harder for then to coordinate the data sharing.
#200 Re: Re: Re: Re: yep
by rgelb
Friday November 9th, 2001 11:25 AM
>>Shared libraries between separate browser and email clients can definitely help reduce the amount of duplicate code, but it would not eliminate it completely. If the web browser and email client were developed totally apart from one another then that makes it harder for then to coordinate the data sharing. <<
Theoretically, yes. I use a proxy object to communicate between two separate programs. I've been doing this in COM since 1997. Works great.
#71 Re: many advantages to specialization
by AlexBishop
Tuesday November 6th, 2001 2:58 PM
Here's another advantage. Many organisations use Netscape Communicator 4.x simply because of its LDAP support. They choose Netscape just for Messneger but use Navigator as well. Therefore, a compelling argument for using one component leads them to use the whole suite. In fact, a lot of corporate users are waiting for Mozilla's LDAP support to improve before upgrading.
Alex
#80 Address book
by WillyWonka
Tuesday November 6th, 2001 8:02 PM
I like being able to press F9 and then clicking on a name in the sidebar.
If I see something interesting on the web I can fire off an email with the link and a message to a specific person in a couple seconds. It's great having the address book there.
#123 Should become useful when idle
by solhell
Wednesday November 7th, 2001 3:04 PM
Short answer:
- I won't have had used mozilla or netscape at all if it didn't have the mail/new component.
#129 short response
by pbreit
Wednesday November 7th, 2001 5:43 PM
short response: millions more would be using mozilla if it didn't include mail/news and came out a year or more ago.
also, usage of mozmail indicates that you do not work at a big company, do not use aol, do not use hotmail, ms exchange or lotus notes and thus, no offense, you do not matter.
#136 Re: short response
by gwalla
Wednesday November 7th, 2001 11:06 PM
"also, usage of mozmail indicates that you do not work at a big company, do not use aol, do not use hotmail, ms exchange or lotus notes and thus, no offense, you do not matter."
Okay, this makes about as much sense as saying "eating marmalade on toast indicates you do not eat English muffins, do not use jam, do not use jelly, preserves, or peanut butter and thus, you do not matter" in a debate about whether marmalade is good or not. In a debate about whether Mozilla mailnews should exist (why are we even having this discussion? It's not going to magically get completely deleted from existence), it's ridiculous to on the one hand argue that nobody uses it and then when somebody says they do to say that they don't count precisely because they use it. I guess you can claim that "everybody uses something other than Mozilla Mailnews" as long as you define "everybody" as "everybody who doesn't use Mozilla Mailnews", but if that's your only argument then you're in trouble.
Furthermore, "usage of mozmail indicates that you do not work at a big company" is demonstrably false. I used Mozilla Mailnews back when I worked for the Hearst Corporation and nobody complained. Not every big company is as draconian in their computer use policies as yours apparently is.
Your criteria for what doesn't count as a person manages to completely ignore anybody who might actually get their email through their non-AOL ISP (what a concept!), including just about every college student whose college provides net access.
As one final point, what is it with people saying something deliberately insulting and then tacking on "no offense" as if suddenly that makes it not an insult? I see it happening more and more, and it drives me up a wall. Is it meant to add a litle extra stab by making the recipient of the insult feel bad for being offended?
#137 short reply
by gwalla
Wednesday November 7th, 2001 11:10 PM
short reply: Mozilla wouldn't exist if Mozilla mailnews didn't.
When Netscape opened the Mozilla source, they expected to get a full replacement to Netscape Communicator out of it. If suddenly a major component that used to be in Communicator was scrapped, you'd better believe they would have pulled their people off of the project.
#142 Re: short response
by SubtleRebel
Thursday November 8th, 2001 4:01 AM
"usage of mozmail indicates that you do not work at a big company,"
AOL Time Warner is pretty big company and I think it is safe to say that they have employees using the Netscape email client.
I also know of several people employed by big companies who are using Mozilla or Netscape for email, but I think the above examples is more than sufficient proof that your statement was undeniably wrong.
#152 Re: short response
by tny
Thursday November 8th, 2001 7:47 AM
> short response: millions more would be using
> mozilla if it didn't include mail/news and
> came out a year or more ago.
Bull. Maybe if it came out a year or more ago; but most (not all) Netscape users use Netscape mail, too, and had mail/news been lacking, it would have been one more reason to turn to IE (because whether they're really separate or not, Outlook Express is downloaded with IE, so it is perceived as being a part of IE.
#20 Not even 4.x features yet.
by johann_p
Sunday November 4th, 2001 2:15 PM
The sad thing about MailNews is that some very basic features that have been in 4.x are still missing
in mozilla. I do think that these features are essential for a relevant part of potential users,
but obviously effort and priorities were allocated for the more .... uhhhm *cool* features.
What missing features? Quote current message, movemail, open additional mail windows without loosing your
current view and more (unable to look up bug numbers because i get the funny red "dont use this build with
bugzilla" message -- is this a joke?)
#21 Re: Not even 4.x features yet.
by AlexBishop
Sunday November 4th, 2001 3:21 PM
"(unable to look up bug numbers because i get the funny red "dont use this build with bugzilla" message -- is this a joke?)"
That's no joke. Your build of Mozilla has a nasty bug http://bugzilla.mozilla.org/show_bug.cgi?id=108175 that causes it to clear the contents of text fields. As you can imagine, that really screws up Bugzilla.
Alex
#44 Re: Re: Not even 4.x features yet.
by schapel
Monday November 5th, 2001 2:17 PM
You can still use the latest nightlies to QUERY Bugzilla, just don't try to UPDATE bugs. If you want to do anything besides looking at bug reports, charts, and graphs in Bugzilla, you can still download the Nov. 1 build: http://www.mozillazine.org/build_comments/
#22 Re: Not even 4.x features yet.
by Ben_Goodger
Sunday November 4th, 2001 9:09 PM
Don\'t forget newsgroup filters!
#26 Re: Re: Not even 4.x features yet.
by dave532
Monday November 5th, 2001 6:07 AM
Newsgroup filters were never in 4.x
#40 unbelievable
by pbreit
Monday November 5th, 2001 12:28 PM
it's just unbelievable that people cannot see the nearly-fatal result of trying to build so much more than just the browser. imagine how excellent it would have been if moz had released a killer, high performance, very stable, standards-compliant, small footprint *browser* a year ago. instead, we're 6 months away from a huge piece of bloatware that's sure to remain inferior to IE.
#43 Re: unbelievable
by strauss
Monday November 5th, 2001 2:00 PM
Your tone is becoming overexcited, but I agree with what you are saying. The strategic importance of an email client is far less than that of a browser, for one thing. The economic network effects of an email client are much less than those of a browser -- an email client is not a development platform, while a browser is. And yes, there is a definite loss of focus from extra features of this type that are not critical to the core strategic mission. Finally, we have the fact that the mail/news client and the HTML editor are mediocre at best, certainly much worse than the competition, and are likely to be rejected by the market.
Imagine that the resources committed to the mail/news client and HTML editor had instead been dedicated to QA or bug fixing, and imagine what a difference that would have made to the current state of the defect curve and the core browser. A reallocation of this sort may still be necessary to eventual release of 1.0.
#46 Re: Re: unbelievable
by SubtleRebel
Monday November 5th, 2001 7:54 PM
How can Mozilla 1.0 ever be released if resources are allocated away from the mai/news client? There are several issues with the mail/news client that have to be addressed before the 1.0 goals can be achieved. Regardless of what certain people may want, the Mozilla project is not only about a web browser.
Besides the Mozilla organization can not exactly allocate resources anyway. Netscape can decide that they want their programmers to focus on one particular aspect or another, but the Mozilla organization can not tell them what to do (well, I suppose they could tell them, but no one at Netscape has to abide by it). Likewise for IBM and the various other companies who contribute code. As for the many individual programmers who are helping out, you are dreaming if you think they are going to be told what code that they are supposed to be working on.
BTW, what "strategic importance" and "economic network effects" are you talking about? Whose strategy and whose economic network?
(although I still disagree with you about a lot of things, I will say that your posts have been much better lately. The only time that I cringed reading this one was when you said "defect curve" near the end.)
#60 Re: Re: Re: unbelievable
by strauss
Tuesday November 6th, 2001 10:01 AM
> BTW, what "strategic importance" and "economic network effects" are you talking about? Whose strategy and whose economic network? <
The strategic importance of the browser is that the browser has become a primary development platform. Microsoft is willing to give away their browser for free because it is important to them that they dominate the market for development platforms. In particular, they were afraid that the rise of the browser as platform would make Windows irrelevant, and so they have been developing Explorer both as a fallback against that possibility and as a way of tying the browser back to Windows -- for instance, by undersupporting other platforms such as Mac and Linux, and by tying content to proprietary Windows-only technologies like ActiveX.
Network effects are an economic phenomenon of importance to devlopment platforms in which the value of each unit is increased by the value of every other unit deployed. The value of a copy of Internet Explorer or Windows or any other platform (including Mozilla) grows exponentially with the number of other units deployed because of issues like the greater probability of software being developed for that platform and the ease of interchanging data with other users of the same platform.
The strategic importance of Mozilla/Netscape 6 to AOL Time Warner is that it prevents them from being completely at Microsoft's mercy when it comes to their content platform. If they use their installed base to drive adoption of Mozilla, Mozilla will benefit from network effects ropughly at the point where it achieves a threshold market share at which many major sites decide they have to tune their content to work optimally under Mozilla. At this point the race becomes more even and AOL Time Warner is happy because it has more control over the content platform. That's their strategic interest.
Personally I do not like Microsoft dominance at all, but I would be much more enthusiastic about Mozilla as a competitor if not for the fact that (1) AOL Time Warner is at least as evil as Microsoft and (2) it doesn't seem that Mozilla is actually superior to Explorer, only different, and so far inferior from a development perspective w.r.t. bugs and feature completeness.
As for email clients, email is much simpler and much less of a development platform than web pages, and since these days you can exchange email perfectly well between any of the clients, network effects don't really come into play. That's why an email application doesn't have the same strategic importance. If email compatibility between user agents was as impaired as web page compatibility between user agents, it would be a different story. For instance, if people were stupid enough to turn on the "Use Word format for email" option in Outlook, that would start to feed network effects, but they're not that stupid for the most part.
Similarly, an HTML editor could theoretically be strategically important, much like Microsoft's development tools for Windows are strategically important, but since in fact they're all roughly equivalent and none is more compatible than any other (FrontPage is specifically tuned for Explorer to try to feed network effects, but it does not achieve that goal), there's not an actual strategic play there. The market for HTML editors will remain dominated by tools like DreamWeaver, HomeSite, BBEdit, etc. Netscape's mediocre HTML editor does nothing to further AOL Time Warner's strategic goals.
#61 Re: Re: Re: Re: unbelievable
by SubtleRebel
Tuesday November 6th, 2001 11:18 AM
Ok, so if I understand you correctly, you are primarily talking about the strategic importance and economic network effects for AOL Time Warner.
The thing is that the software project under discussion was Mozilla.
Now if the argument were that the Netscape developers should focus on the browser only then perhaps your statements would have a little more merit, but that is not what was being said. Besides, even in regards to Netscape, I do not believe that dropping the email client and HTML editor would be a good move. The majority of Netscape's current install base make use of one or both of those parts of the suite.
#62 Re: Re: Re: Re: Re: unbelievable
by strauss
Tuesday November 6th, 2001 12:56 PM
Mozilla is owned by and paid for by AOL Time Warner for strategic reasons. It is not independent, although it falsely tries to present itself as such.
#75 UNBELIEVABLE!
by SubtleRebel
Tuesday November 6th, 2001 6:38 PM
Ummmm, how can I say this. . . you are just plain wrong.
#78 Re: UNBELIEVABLE!
by strauss
Tuesday November 6th, 2001 7:06 PM
#93 Re: UNBELIEVABLE!
by rgelb
Wednesday November 7th, 2001 12:01 AM
Dude, you need to stop living in lala land.
#95 Re: UN-UN-BELIEVABLE!
by macpeep
Wednesday November 7th, 2001 2:18 AM
Strauss is quite right. Even though Mozilla is used by Nokia, and other companies too have put efforts and resources into it, the dropping out of Netscape from the Mozilla project would mean such a serious decrease in resources and expertice that there is no way in hell Nokia (or Apple, or IBM etc.) would continue supporting it. My guess is that they would all move to Opera in a flash.
And then there\'s the operational issues. Where is the Mozilla HQ located physically? Where are all / most of the meetings held? Where is Bugzilla? Who pays for its bandwidth?
Without Netscape, the Mozilla project would grind to a halt and cease to exist in all other sense except that the source code would still be out there. It wouldn\'t DIE but it would cease to be RELEVANT.
Strauss mentioned the network effect earlier, how the value of a network is exponentially proportional to the number of nodes in the network. Well, apply that to market share.
#112 Re: Re: UN-UN-BELIEVABLE!
by strauss
Wednesday November 7th, 2001 12:40 PM
> Mozilla is used by Nokia <
Actually, I don't think Nokia does use Mozilla. Look at this link http://news.cnet.com/news/0-1005-200-4520937.html which says AOL is licensing a Nokia browser. Gecko is way too big for most handhelds.
> the dropping out of Netscape from the Mozilla project would mean such a serious decrease in resources and expertice that there is no way in hell Nokia (or Apple, or IBM etc.) would continue supporting it. <
Yes, exactly, which is why although the project would theoretically survive without Netscape, in reality without their constant infusion of cash and resources, it would never achieve release-ready quality levels. It's having a hard enough time with their sponsorship -- it would get nowhere without it.
> Where is the Mozilla HQ located physically? <
At the Netscape headquarters complex.
> Where are all / most of the meetings held? <
At the Netscape headquarters complex.
> Where is Bugzilla? <
On a Netscape server. Remember Gerv complaining last week Netscape wouldn't give them money to upgrade it?
> Who pays for its bandwidth? <
Netscape Communications Corporation, a wholly owned subsidiary of AOL Time Warner.
#114 Re: Re: Re: UN-UN-BELIEVABLE!
by AlexBishop
Wednesday November 7th, 2001 1:21 PM
"Netscape Communications Corporation, a wholly owned subsidiary of AOL Time Warner."
Actually, it's a wholly owned subsidiary of America Online Inc., which is a wholly owned subsidiary of AOL Time Warner.
Though you told us where Mozilla and Bugzilla are, you failed to state where The Amazing Netscape Fish Cam http://home.netscape.com/fishcam/ is located. It's in JavaScript land at Netscape, apparently. Pressing Ctrl+Alt+F in Netscape 6.2 doesn't bring up the Fish Cam though, which is a bug that should be fixed immediately.
Alex
#118 Re: Re: Re: UN-UN-BELIEVABLE!
by macpeep
Wednesday November 7th, 2001 2:12 PM
"Actually, I don't think Nokia does use Mozilla. Look at this link (LINK) which says AOL is licensing a Nokia browser. Gecko is way too big for most handhelds."
It's used by Nokia, but not for handhelds. It's on a prototype "media terminal" which may or may not be actually ever sold. The device is still far from a hypothetical release so if something was to happen to Mozilla (Netscape dropping their support), the likely move would be to switch to Opera rather than throwing 150 more people on the project. It's just a browser for them, not a core technology.
The browser licensed to AOL was a WAP microbrowser.
#119 Re: Re: Re: Re: UN-UN-BELIEVABLE!
by strauss
Wednesday November 7th, 2001 2:47 PM
If it's a network appliance, it will probably not reach market. All such devices that were released have flopped, and I think they have all been withdrawn now.
#115 Re: Re: Re: Re: Re: Re: unbelievable
by AlexBishop
Wednesday November 7th, 2001 1:30 PM
"Mozilla is owned by and paid for by AOL Time Warner for strategic reasons. It is not independent, although it falsely tries to present itself as such."
Does it? The press release announcing the creation of mozilla.org http://home.netscape.com/newsref/pr/newsrelease577.html makes it clear that it's "a dedicated team within Netscape".
Alex
#117 Re: Re: Re: Re: Re: Re: Re: unbelievable
by strauss
Wednesday November 7th, 2001 1:44 PM
Good link, thanks. It seems this fiction of separation has sprung up since then. At the time of its formation the "Mozilla organization" was clearly stated to be part of Netscape.
#65 well stated
by pbreit
Tuesday November 6th, 2001 1:08 PM
well stated, strauss.
the usage of the browser part of the mozilla project is critical to the mozilla effort since without it, mozilla has no relevance whatsoever. whereas without mail, news, edit, chat and skins, mozilla would still be relevant if the browser was good.
the majority of Netscape's current (small) installed base defeinitely does not use news, edit, chat or skins. i suppose it's possible that the majority uses mail, but even that i doubt since so many users do not have much of a choice in the matter nor could set up POP/IMAP on a reliable basis. AOL users use AOL. Coporate users use Outlook, Notes, etc. Mac users use Eudora. Hotmail users use Hotmail or Outlook. Not sure what that leaves.
#76 Mac users use Eudora??
by SubtleRebel
Tuesday November 6th, 2001 6:46 PM
Why do you say that Mac users use Eudora? I know a lot of Mac users and only one of them uses Eudora. Several use Netscape 4.x, some use Apple's Mail program, and a few use Outlook.
#83 thanks, subtlerebel
by pbreit
Tuesday November 6th, 2001 8:25 PM
for making my argument for me! the fact is, people use a lot of different email clients because there are a lot of decent ones available and/or they have no choice. in fact, i bet from a pure number of users perspective, the email client in aol is the one most used by mac users.
#90 Re: thanks, subtlerebel
by SubtleRebel
Tuesday November 6th, 2001 11:46 PM
No I did not support your point. I questioned you about your comment about mac users and Eudora, but for some reason you totally failed to even acknowledge the question - let alone answer it.
As for what your point was, you were saying that you did not think many people used Netscape's email client and gave a bunch of unsubstantiated reasons - some of which were just plain absurd.
The Netscape 4.x email client is one of the best and a lot of people use it. Since many people who use IE still use the Netscape 4.x email client, it is possible that more people use Netscape's email client than the number of people who use the web browser.
Your argument that no one uses Netscape's email client is totally unfounded. You have provided no real facts to support your argument and I definitely have not said anything to support your claims.
#105 whatever
by pbreit
Wednesday November 7th, 2001 10:36 AM
my point was that there are good email client options available on all platforms and that there is little need fo mozilla to supply one and that many people have already picked their email client and quite a few people (in corporate america) have no say in the matter. how is that so hard to understand?
you say i'm not backing up my comments? find one credible person that would agree with you that "more people use Netscape's email client than the number of people who use the web browser."
sorry, dude.
#143 Misquote
by SubtleRebel
Thursday November 8th, 2001 4:14 AM
What I actually said is in the post that you replied to. Why attempt to discredit me by blatantly misquoting me?
In your previous post you claimed that what I had said supported your argument when it obviously did not.
You can have your own opinion, but stop misrepresenting mine.
#155 too subtle for subtle
by pbreit
Thursday November 8th, 2001 10:19 AM
subtle, you were inadvertantly supporting my argument. by noting all the different good and popular mail clients in use, you support my argument that we do not need another.
#160 who's we?
by joschi
Thursday November 8th, 2001 11:43 AM
You may not need another mail client, but Netscapes customers do, and that's why they are working on it. I love it when outside observers think they know everything about a company and its market.
by pbreit
Thursday November 8th, 2001 4:09 PM
Please let me know one or two reasons why netscape users need another mail client. I've iterated the reasons they don't several times.
They are working on a mail client because 1) they felt it necessary to reproduce the Netscape 4.x suite and 2) there is a small but vocal minority requesting it.
It's dangerous to eschew outsider perspectives. This invariably leads to an overly internal focus and a misunderstanding of market needs. Mozilla's a good case study.
#189 Getting real tired of this
by SubtleRebel
Friday November 9th, 2001 9:26 AM
Go back and read my post, I was commenting on your insane implication that Eudora was the email client of choice for Mac users. As part of pointing out that most Mac users do NOT use Eudora, I mentioned the Netscape email client and others. How in the world does that support your argument that "we" do not need the Netscape email client?
I mentioned multiple email clients that are being used and one of them was Netscape. You claim that that supports the idea that Netscape's email client is not needed. I am sorry but that is not logical at all.
Some people use email clients other than Netscape, but there are many people for whom Netscape/Mozilla fills a need that NO OTHER email client does.
by pbreit
Friday November 9th, 2001 12:35 PM
a moz mail client is not needed because there are other decent options you freaking idiot!
#208 Not too subtle
by SubtleRebel
Friday November 9th, 2001 1:05 PM
A Mozilla mail client is needed because the other options do not provide the same functionality.
#232 Re: subtle
by Ugg
Saturday November 10th, 2001 10:38 PM
"a moz mail client is not needed because there are other decent options you freaking idiot!"
Television is not needed. Sushi is not needed. But I want television and sushi. A lot of people want television and sushi.
Mail and news in Mozilla might not be needed, but I want them, and so do a lot of other people. So STFD and STFU already. Geez.
#233 tv and sushi
by pbreit
Sunday November 11th, 2001 1:20 PM
so why doesn't mozilla make tv's and sushi? seriously?
#234 Re: tv and sushi
by macpeep
Sunday November 11th, 2001 2:03 PM
"so why doesn't mozilla make tv's and sushi? seriously?"
If you really wonder that, maybe you better go and think about it somewhere else instead of coming here and annoying people. I mean, if stuff is THAT unclear to you. It's getting pretty philosophical here anyway. Why does anyone do anything, really?
People work on Mozilla for two reasons:
1) Because companies (AOL, Netscape, IBM etc.) need it for some reason and they pay people to work on it.
2) Because people find it fun.
Why does Mozilla have the feature set it has:
1) Traditional reasons. Netscape 4 had a browser, email, news and editor -> Mozilla has the same set.
2) Because the companies that have people working on Mozilla find that they need an email application and that the resources they are putting on the project are worth what they are getting out of it.
3) Because contrary to what you think and think you've proven one way or the other, people do like, want and need Mozilla mail & news. Or.. Netscape mail & news, which is basically the same thing.
I use email a *LOT* and I've tried just about every serious email app for Windows and Netscape 4.7x is the one I use now, even though I use IE 6 for web browsing. Why? Because I think it's the best email app around. If / when Mozilla produces one of [at least] the same quality, I'll happily switch - in particular if the web browser is good enough to replace IE 6.
#238 Nicely said
by SubtleRebel
Monday November 12th, 2001 10:14 AM
I agree with everything that macpeep said. The only difference is that I have already switched to using the Mozilla email client. I agree that it is still lacking somewhat as compared to the 4.x client, but since I have switched to Mozilla as my primary browser, it is just more convienent to use Mozilla for email too. I have high hopes that after this round of focus on mail/news that the team will be able to bring it up 4.x or better.
#150 Re: whatever
by tny
Thursday November 8th, 2001 7:36 AM
There are two mail clients that do what I need for Windows: Mozilla and Outlook Express. Not only is Outlook Express a sieve, but on my XP box it doesn't work (odd bug). So now the ONLY mail client I can use on my XP box is Mozilla. (BTW, XP is smart enough to tell me how many unread messages I have in Mozilla - not Netscape 6.x, Mozilla - on the login screen). Eudora doesn't cut it for me; though Mail.app on OS X does.
I just wish there were some way of mass-saving email to txt files; I get 50-75 emails a day and have to save about half of them.
#153 Re: Re: whatever
by AlexBishop
Thursday November 8th, 2001 9:20 AM
"BTW, XP is smart enough to tell me how many unread messages I have in Mozilla - not Netscape 6.x, Mozilla - on the login screen"
Netscape advertise this as a feature of 6.2. Mail clients have to be written to support this feature and the most recent versions of Netscape 6 and Mozilla have this support.
Alex
#157 Re: Re: Re: whatever
by tny
Thursday November 8th, 2001 11:12 AM
I stand corrected - s/XP is smart enough to tell me
how many unread messages I have in Mozilla/Mozilla is
smart enough to tell XP to display how many unread
messages I have in Mozilla/x
#84 re news: are you kidding?
by pbreit
Tuesday November 6th, 2001 8:28 PM
c'mon! hardly anyone at all reads news groups anymore! net news went down-hill pretty fast in about '91!
#89 Re: re news: are you kidding?
by SubtleRebel
Tuesday November 6th, 2001 11:36 PM
Millions of news messages are posted every day. Someone is using it.
#104 plus internal use..
by macpeep
Wednesday November 7th, 2001 9:18 AM
many companies and organizations also use newsgroups internally
by pbreit
Wednesday November 7th, 2001 10:42 AM
25 million aol users are *not* using news. millions in non-technical positions in corporate america are *not* using news. discussions have clearly shifted away from the caos of netnews towards focused communities (mozillazine being an example).
no offense, but you guys are living in some sort of dream world of your own creation. your view-points show very little grounding in reality. i'd suggest you take a step back and see what's going on in the real world.
#144 So what?
by SubtleRebel
Thursday November 8th, 2001 4:26 AM
What is the point of telling us some people who you do not think use a news client?
Your claim was that "hardly anyone at all reads news groups anymore!"
There are millions of people who read news groups. I think that that qualifies as more than "hardly anyone."
Besides you have no basis for your statements about who is not using news.
There are AOL users who ARE using news. I see them post messages to newsgroups all the time.
There are also people in non-technical positions in corporate America who use news clients for accessing newsgroups on the Internet, on intranets, and on extranets.
It is becoming more and more apparent that you are the one living in your own dream world where you create statistics out of thin air.
#156 bad proxy
by pbreit
Thursday November 8th, 2001 10:25 AM
subtle, i imagine that you an i are more alike than not in how we use computers. problem is, you and I and many of my co-workers and friends are horrible proxies for internet users in general. i understand this, you do not. when developing a product, there is nothing more critical than understanding the market you wish to serve. i do not believe that mozilla/netscape wishes to limit itself to the niche of high-tech professionals.
#190 Rather presumptuous
by SubtleRebel
Friday November 9th, 2001 9:39 AM
From reading your posts here, you imagine a lot of things, but once again your imagination does not coincide with reality.
I do not know why you imagine that you and I are alike in how we use computers. Can you provide a detailed description of how you think I use computers and how it is like the way that you use computers?
#204 i imagine
by pbreit
Friday November 9th, 2001 12:41 PM
i imagine you are a professional in the high tech industry. i imagine you use a few different platforms, have several email accounts, try out different browsers, have used opera, are technically adept, etc.
what i have been proposing is not at all what i want for myself but what i believe mozilla needs to do to be viable in the marketplace.
#192 Re: bad proxy
by SubtleRebel
Friday November 9th, 2001 9:50 AM
You claim that I am a horrible proxy for Internet users in general.
You claim that I am unaware that I am not a good proxy.
You claim that you are a bad proxy, but you are aware of it.
My response :
Since you know you are a bad proxy then you should stop trying to be one.
Some day I may come to the conclusion that I am no longer a good proxy, but until then I will continue.
#173 Re: news
by gwalla
Thursday November 8th, 2001 2:47 PM
Um, no. Lots of AOL users are on Usenet. I don't know where you get this stuff from.
#176 aol news users
by pbreit
Thursday November 8th, 2001 4:14 PM
my guess would be that fewer than 5% (about 1 million) of aol users read or post to newsgroups on a regular basis. do you think that's in the ball-park? and that the percentage is going down. and that the vast majority of them read and post through aol's proprietary interface.
apologize for not having the data. i'd be happy to be proven wrong.
#194 Your guess
by SubtleRebel
Friday November 9th, 2001 10:18 AM
At least now you admit that you are only guessing. Now if we could only get you to realize that making guesses does not result in statistics. Especially uneducated guesses.
Whether or not AOL users use AOL's proprietary interface for reading and posting to newsgroups is irrelevant. If they are using AOL's news interface then they are using news. The point being discussed was your claim that AOL users do not use news; you made that statement as part of your claim that hardly anyone uses news and therefore Netscape should not develop a new client.
The reality is that AOL users and non-AOL users both use news. AOL's software provides an integrated news client to its users. Netscape's software provides an integrated news client to its users.
#205 so what's your point?
by pbreit
Friday November 9th, 2001 12:47 PM
what i stated was that the number of AOL users who read news is small, who post to news even smaller and that for both of these endeavours they generally use the AOL news client. and thus, a moz news client is of little relevance to the 30 million aol users.
i believe is is fairly clear in my posts when i am presenting an opinion and when i do not have factual support. i actually spend a lot of time trying to find the data and am not usually successful. on the occasions i do present data, you seem to disregard. i don't recall you presenting much data either so i find your criticism in quite absurd.
#236 Re: so what's your point?
by gwalla
Sunday November 11th, 2001 5:15 PM
The point is that an opinion formulated on a complete lack of evidence is just an opinion, and should not be used as a basis for making real decisions.
#187 Re: news
by bcwright
Friday November 9th, 2001 8:11 AM
Usenet participation has been declining for years, in terms of percentage of Internet users if not in absolute numbers (though quite likely in absolute numbers as well). This decline is across the board, both technical and non-technical, AOL and non-AOL. But it's a decline from a very large number base. Even if it includes only, say, 10% of all Internet users, it's still quite a large number of users, and the decline is not precipitous.
In other words, I think that all of you are like the blind men and the elephant, each finding some tidbit about the elephant such as his tail or his trunk or his ears that he thinks respresents the essense of "elephantness" - you're only looking at part of the picture.
--Bruce
#196 I see the whole elephant...
by SubtleRebel
Friday November 9th, 2001 10:37 AM
Pbreit is the only one in this thread who is speaking with limited perspective.
#211 uhmm...i don't think so
by pbreit
Friday November 9th, 2001 1:57 PM
my perspective is clearly much, much broader than yours. it's obvious you're carrying a ton of netscape/mozilla baggage.
#213 Re: uhmm...i don't think so
by strauss
Friday November 9th, 2001 2:05 PM
It's bad enough having one side slinging a dozen insults a day. Don't hurt your case by lowering yourself to that level, please.
#219 Re: Re: uhmm...i don't think so
by SubtleRebel
Friday November 9th, 2001 5:53 PM
Yes I agree, having both of you constantly slinging low level insults is just too much. Decide which one of you is going to do it or trade off days or something.
#218 Re: uhmm...i don't think so
by SubtleRebel
Friday November 9th, 2001 5:51 PM
Please provide even one example of how your perspective is broader.
Keep it brief and to the point.
by pbreit
Friday November 9th, 2001 8:09 PM
you think news is a big opportunity. i think it is a small opportunity. done.
#226 Is that an example?
by SubtleRebel
Friday November 9th, 2001 9:53 PM
What was that? In absolutely no way does it indicate that your perspective is broader than mine.
BTW, I also never said it was a big opportunity.
#96 Re: re news: are you kidding?
by gwalla
Wednesday November 7th, 2001 2:26 AM
alt.comics.sluggy-freelance is still going strong!
#77 Netscape's install base does not use news???
by SubtleRebel
Tuesday November 6th, 2001 6:49 PM
How in the world did you come up with the conclusion that Netscape users don't use news?
#102 unfortunately, you're right :-)
by dipa
Wednesday November 7th, 2001 8:05 AM
Otherwise, "signature below quoted text" bug wouldn't exist at all.
#151 Re: Netscape's install base does not use news???
by tny
Thursday November 8th, 2001 7:44 AM
Someone said that News went downhill in 1991. He's right: it went downhill when it became popular.
#181 usenet stats
by pbreit
Thursday November 8th, 2001 7:56 PM
well, usenet stats are virtually impossible to come by. all i could figure was by checking out some of the newsgroups i used to read religiously in the late 80's and early 90's. there seemed to be a couple new sub-groups and maybe 25-50% more article volume. not sure if this indicates that usenet is shrinking or if it just has really slow growth.
in the same time period, i'd guess that the web and email has doubled several times over.
#195 Regardless
by SubtleRebel
Friday November 9th, 2001 10:35 AM
Whether or not usenet usage is growing or shrinking, there are enough users to justify development of a news client. Plus, as macpeep has said, there are a lot of people using private newsgroups that are not part of the public usenet.
#206 uhm...no.
by pbreit
Friday November 9th, 2001 12:49 PM
"Whether or not usenet usage is growing or shrinking, there are enough users to justify development of a news client."
uhm...no.
please tell me that whether or not a market is growing or shrinking is an important factor when making a product decision. please?
#210 Re: uhm...no.
by SubtleRebel
Friday November 9th, 2001 1:44 PM
Yes, whether or not a market is growing or shrinking can be an important factor when deciding whether or to produce a product.
However, it is not a signifigant issue in this case.
There are millions of people currently using some form of news client. If that number is slowly growing or slowly shrinking is not important because it is very slow and could easily start going the opposite direction at any time. The key thing is that the market is not going to disappear any time soon.
So if someone can develop a news client product that is better than other news clients then there is a reason to develop it.
The Mozilla news client has the potential to be better news client than the others (at least for some people if not for all) and therefore there is reason to develop it regardless of whether or not usenet usage is growing or shrinking.
And in case you forgot, usenet usage does not account for all usage of a news client. Mozilla's news client is also well suited for non-usenet newsgroups.
#98 Re: Re: unbelievable
by bcwright
Wednesday November 7th, 2001 3:45 AM
Mozilla has been trying to have it both ways - it doesn't support external email clients, but at the same time its own email client is seriously lacking when compared to 4.7x. This is a recipe for disaster; if they aren't going to be able to achieve at least feature parity with 4.7x, then at the very least it is essential to get it to play nicely with external email clients.
My personal feeling is that performance and footprint of the mail/news client is far less important than either getting feature parity with 4.7x or better support for external clients. The current client is not unbearably slow, but its woeful lack of familiar features is extremely painful.
Not that I'm disagreeing with you, but more with the priorities of the project. It is starting to appear that we'll get the worst of both worlds - lots of effort down the drain, and a client that may be fast but that nobody will want to use.
--Bruce
#53 i disagree
by johann_p
Tuesday November 6th, 2001 3:34 AM
I disagree: it is good that mailnews is part of mozilla. the thing
is - all i want *please* is a netscape that doesnt creash all the
time but has what 4.x had. I am happy with 4.x except for the
crashes, but I am not happy with mozilla because of stuff that
is missing -- despite all the additional stuff that is in it.
I think it is *very* silly that providing everything that 4.x had
was not top priority from the start, because a huge portion of
potential users for mozilla/ns6 are 4.x users -- many of them
will be disappointed.
#30 I wonder ...
by johann_p
Monday November 5th, 2001 9:39 AM
I wonder if people e.g. within Netscape actually *use* mozilla MailNews for
their daily tasks. While mozilla navigator is *much* more stable than 4.x
and apart from the problems that arise from webmasters not sticking to standards
is a clear step forward from 4.x for people who need to work with it,
MailNews is a big step backwards. Its just much more cumbersome and awkward
to do many things with MailNews than it was with the 4.x version.
Another example I have to think of is that it doesnt seem to be possible now
to search the message list and then jump to that message in the open Mailnews
window, within its thread-context. I do believe that even worked some time
ago ...
When I look at the bugs that describe all these quirks I see no activity,
though some of them date back to 2000. Instead, again, people hack on new
features like "labels". I think that's unfortunate.
#31 Re: I wonder ...
by AlexBishop
Monday November 5th, 2001 10:18 AM
I think it's pretty much mandatory for Netscape people to use Mozilla Mail & Newsgroups (well, a Netscape 6 build of it) for their daily work.
Alex
#58 Re: I wonder ...
by tny
Tuesday November 6th, 2001 6:56 AM
Johann_p, the only way this argument will prove anything
> When I look at the bugs that describe all these
> quirks I see no activity, though some of them date
> back to 2000. Instead, again, people hack on new
> features like "labels".
is if you can demonstrate that people who were assigned bugs with no activity since 2000 were working on new features in the interim. One issue with a volunteer project is that sometimes people get assigned work and never do it - stake a claim, but never fulfill their promise; but often on such projects reassignment has a lower priority than new assignment. I don't know how Mozilla handles this problem (on the project I work on, we run through the DB every 6 months or so and remove any assignments that are stale; but we only have a fixed number (35,000) of non-programming tasks to assign), but it's possible that this might be a contributing factor to stale bugs.
#92 Re: Re: I wonder ...
by rgelb
Tuesday November 6th, 2001 11:56 PM
>>One issue with a volunteer project is that <<
I know this is heresy, but this is NOT a volunteer project. If 90% of people didn't get paid by Netscape, there would be no moz.
#145 Source of your statistic?
by SubtleRebel
Thursday November 8th, 2001 4:32 AM
Where did you come up with the information indicating that 90% of people get paid by Netscape?
I am assumming that you meant 90% of the people working on Mozilla and not 90% of people everywhere, but either way, I think you are wrong.
#167 Re: Source of your statistic?
by strauss
Thursday November 8th, 2001 1:59 PM
> Where did you come up with the information indicating that 90% of people get paid by Netscape? <
Take a look at the last 48 hours' checkins at http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=SeaMonkeyAll&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Change+Size&date=hours&hours=48&mindate=&maxdate=&cvsroot=%2Fcvsroot and you'll see that Netscape contributions are clearly at or beyond the 90% level. Of other contributors, most appear to be paid contributions from other open source companies such as IBM, Red Hat, and ActiveState. There are also a few academic contributors such as roc, dbaron and our own bzbarsky -- I don't know if they're contributing their time or working for their schools on staff or as part of an assistantship. (dbaron has a false swell due to the fact that he's making lots of small string changes across the program.)
All together the non-Netscape contributions are less than 10%, and of the non-Netscape conttributions, most are from people who appear to be paid to work on Mozilla at other open source companies. I would be interested to learn more about the role of roc, dbaron and bzbarsky, who are obviously putting in a lot of time.
#171 Re: Re: Source of your statistic?
by roc
Thursday November 8th, 2001 2:38 PM
It's not all about checkins. There is massive QA effort by volunteers. Ian Hickson has done a massive amount of work on Mozilla and has checked in hardly any code.
I'm contributing my time and always have, although I'm not actually a student anymore.
OTOH it doesn't worry me in the slightest if most people working on Mozilla are paid to do so.
#193 Re: Re: Source of your statistic?
by SubtleRebel
Friday November 9th, 2001 10:03 AM
90% of checkins does not equate to 90% of the contributors. For one thing, not everyone is allowed to check in code.
There are also a large number of people who are part of the Mozilla organization who do not write code at all.
Your comment about other contributors being paid by other companies is rather irrelevant, if not contradictory, to rgelb's claim that 90% of the people are paid by Netscape.
If rgelb had said that 90% of checkins are done by people paid by Netscape then perhaps he (?) would be right, but that is definitely not what he said before.
#158 Re: Re: Re: I wonder ...
by tny
Thursday November 8th, 2001 11:15 AM
> but this is NOT a volunteer project.
Not exclusively, no; but some of the bugs are being worked on by volunteers, and I wouldn't be surprised if some (not all, not most, but some) of the taking-20-years-to-fix bugs were assigned to lax volunteers.
#28 I just hope 92111 will get "fixed"...
by arsa
Monday November 5th, 2001 8:30 AM
http://bugzilla.mozilla.org/show_bug.cgi?id=92111
#34 Over-zealous Automatic linking of alleged URLs
by SubtleRebel
Monday November 5th, 2001 10:56 AM
Bug #79064 http://bugzilla.mozilla.org/show_bug.cgi?id=79064
This behavior has been around for a while in the Mozilla email client, but it has been annoying me more and more lately. I also think that all of the extraneous links would be very confusing for the average consumer.
I do not know if varada wants help or not, but if I had a current version of CodeWarrior for my Mac, I'd be happy to work on a patch to this hyper text parsing routine.
#36 Unable to drag email attachment to desktop
by SubtleRebel
Monday November 5th, 2001 11:06 AM
Bug #83803 http://bugzilla.mozilla.org/show_bug.cgi?id=83803
For some reason this bug is listed as an enhancement, but after using 4.x email for years, extracting attachments from email via drag and drop seems like expected standard behavior. Since other email clients also offer this functionality, I would think it would be important for Mozilla 1.0 to include it as well.
#38 Re: Unable to drag email attachment to desktop
by schapel
Monday November 5th, 2001 11:42 AM
Right. The 4xp keyword means that fixing the bug will put Mozilla on parity with Netscape 4.x features in this one area. Here are all the open 4xp bugs: http://bugzilla.mozilla.org/buglist.cgi?bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&email1=&emailtype1=substring&email2=&emailtype2=substring&bugidtype=include&bug_id=&changedin=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&short_desc=&short_desc_type=allwordssubstr&long_desc=&long_desc_type=allwordssubstr&bug_file_loc=&bug_file_loc_type=allwordssubstr&status_whiteboard=&status_whiteboard_type=allwordssubstr&keywords=4xp&keywords_type=anywords&field0-0-0=noop&type0-0-0=noop&value0-0-0=&newqueryname=&order=Reuse+same+sort+as+last+time
You can mark which of these many bugs really is important by voting for them!
#37 Hiding email headers..
by macpeep
Monday November 5th, 2001 11:25 AM
I tried to collapse the email headers and found no way to return them. The entire email header area was gone and no matter what options I tried to toggle, I couldn't return them. Any idea if this is a bug (couldn't find anything in Bugzilla) or maybe there's just a very well hidden way to return them?
#42 Re: Hiding email headers..
by schapel
Monday November 5th, 2001 1:45 PM
That sounds like bug 107707: http://bugzilla.mozilla.org/show_bug.cgi?id=107707 It's listed as Priority 1, Severity Critical, currently has 9 votes, and is targetted for 0.9.6. It should be fixed soon.
#50 Decision is right
by dipa
Tuesday November 6th, 2001 12:05 AM
Although I would be extremely happy to see some nasty 4xp bugs fixed (lack of context menu in compose window, for instance), I think the decision to go for performance and footprint is the right one at this moment.
Latest nightlies (despite their instability) show a nice startup time reduction in browser. I hope they will achieve same or better results in mail client perf issues (outliner). The footprint issue also has to be adressed. From what I see in cvs, there is a lot of current work on these fields (many important bugs getting touched again after a long time of ceased activity), making 0.9.6 more of a performance milestone. It seems that the next one will be dedicated to performance and footprint, too. Very nice.
#72 Some FAQs for you to read
by mike_hearn
Tuesday November 6th, 2001 3:48 PM
OK, I get a little sick of people constantly saying "Mozilla should be this, mozilla should be that, mozilla is doomed because they didn't do this" etc. Here are some replies to some FAQs:
1) Mozilla is bloated - maybe in your opinion, but not in mine. I use MozMail because I don't like Outlook Express. I've got too many viruses from it this year, and it won't let me download attachments separately from the email. I'm using ChatZilla right now, because it saves me installing another (harder to use) IRC client. I use Composer all the time, not just in writing emails, but for basic word processing. It's faster to start than Word because I use the quickloader, and it produces good HTML which means my Linux using friends can read what I write. At school, we use Netscape 4.7 or Opera : IE and Outlook Express don't agree with our network setup, and Netscape/Opera does. It's easy to assume there is no need for competition, but believe me, there is.
2) IE renders pages better than Mozilla - depends on where you're coming from. From the users perspective, more pages work in IE than Mozilla for 2 reasons - web page designers write code that only works in IE and this is often due to the second reason, which is that IE is incredibly lax about the rules. Anyone who has written pages and tried to make them render on both Gecko and IE will know what I mean. You can miss end tags in IE and tables will still render. Often, when a page doesn't work in Mozilla, it's the page authors fault, not the browsers. If you file a bug in Bugzilla, somebody will hopefully look at the source and determine what the problem is. You can then email the web admin and ask them to fix the (usually trivial) faults in their pages.
3) Mozilla is taking too long - too long for what or who? If you mean, too long to reach 1.0 that's because the standard for 1.0 is very very high compared to other commerical projects. IE version 1 was unusable. Mozilla v1 will be equivalent to having redeveloped years of work on Netscape 1-4.
4) Mozilla has lost the browser wars - it never fought in them. For people who think IE will always be dominant, think again. I know people who still use Netscape 4.x even though it's less reliable and feature-complete than IE just because they prefer it. For people who think one day Mozilla will rule the world, think again. The thing we should hope for is approximately 50-50 share IE and non-IE browsers. That way, the web will be forced back into line where standards are concerned, but people use the browser they prefer.
5) Mozilla should have feature X/1.0 can't happen without feature X - well, you know the answer to that one really. At the end of the day, there is only so many people working on the source and due to the very stringent quality control procedures involved patches take a while to percolate through Bugzilla. Mozilla will never have some features, will probably one day have many more features, and right now has some features I'd bet you don't know about. Voting for bugs helps determine where the effort goes, but it's not the only thing. The PGP bug may have lots of votes, but until somebody implements it, it won't happen.
6) Mozilla's independance is a farce, it's really owned and controlled by AOL - simply not true. Technically Mozilla is independant. Most of the Mozilla programmers work for Netscape this is true, but there are others that work for IBM/Sun/Apple and other companies, and yet more who are independant (like me). If Netscape pulled out, that'd be a big blow, but it wouldn't stop the project. Nothing can. It's bigger than the sum of its parts.
7) Mozilla will never catch up with IE - it already has in many respects. It's rendering engine in particular is light years ahead of Trident (the IE engine). IE does XML by turning it into HTML internally. This is a hack, and one which prevents it doing things like XUL/SVG/MathML etc. Microsoft like to say that .NET is built on XML, but if that's the case then why does it use IDE generated source code for the gui, instead of a more logical XUL equivalent? Because IE is incapable of it. Why does IE have such an inaccurate and buggy renderer? Because it uses a renderer that is almost a decade old. You may not think it, but technically IE already looks out of date.
8) Mozilla is too slow - yes, Mozilla is slower than native browsers, mainly due to its XUL based interface. I often see people saying that XUL was a mistake both here and on Slashdot. Here is the reason it was designed and built: at the start, the Mozilla Organisation designed and wrote a first class rendering engine, which was called Gecko. It could render almost anything, and was especially good at XML. On each platform it was being tested with little platform specific apps, that had to be maintained individually. I know this because I have been following Mozilla since M3. They didn't have the manpower to do a native front end for each and every platform - if they'd done that Mozilla would run on one platform and that would be Windows. So they said, "Hey, we can write our front end in XML!". XHTML of course isn't a good language for front ends, though that hasn't stopped Microsoft attempting to hack it (yes, windows uses HTML for some of its GUIs). So XUL was created. XUL is a powerful and complete language to describe user interfaces. It uses CSS to do the skinning, so it looks "native" on each platform. That's why it's used. Don't like it? Use something like K-Meleon or Galeon, but don't complain about it here!
In conclusion - Mozilla isn't perfect, because a "perfect" product is by definition impossible to achieve. Everyone has a different idea of perfect. For some, perfection is a small and light browser without anything else. For others, like me, features and reliability is what makes a perfect product. Mozilla will never replace IE entirely, because IE is at the end of the day (poor standards compliance apart) a pretty good product. And I like OS integration.
Until then, you have a choice. Instead of venting feelings here (which has zero effect, let me tell you) do something about them. If you think Mozilla is beyond salvation, then use IE or Opera and never look back. Please don't post your list of favourite bugs, vote for them instead. Otherwise, try and help constructively. Bug triage in particular is easy, requires almost no special skills and is tremendously helpful.
#73 Re: Some FAQs for you to read
by niner
Tuesday November 6th, 2001 4:56 PM
I think the only sad thing about your comment is that there seems to be no way to have a link to just it for future reference :)
But on complaint I have to make: Why shouldn't people post their favourite bugs here? It is an interesting way for me to discover new things about Mozilla in this huge bug database. So there's even a good thing about it.
#79 Re: Re: Some FAQs for you to read
by basic
Tuesday November 6th, 2001 7:36 PM
try this: http://www.mozillazine.org/talkback.html?article=2079&message=72#72
#74 Re: Some FAQs for you to read
by strauss
Tuesday November 6th, 2001 6:36 PM
Lots to argue here, especially your weird assertions about XML in IE, but to address just one point:
> Technically Mozilla is independant. <
Technically it is not. Find me the corporate filing for "the Mozilla organization." It doesn't exist. That "organization" is owned by AOL Time Warner and is completely accountable to them. When AOL Time Warner coughs, Mozilla has no choice but to jump. We've seen this several times this year. I think most of the people who insist on this illusion of separation are sincerely deluded rather than simply trying to trick people into volunteering to help AOL Time Warner's strategic interests for no pay, but they're just as mistaken.
#99 Re: Re: Some FAQs for you to read
by SmileyBen
Wednesday November 7th, 2001 5:32 AM
>> Technically Mozilla is independant. <<
>Technically it is not. Find me the corporate filing for "the Mozilla organization." It doesn't exist. That "organization" is owned by AOL Time Warner and is completely accountable to them. When AOL Time Warner coughs, Mozilla has no choice but to jump. We've seen this several times this year. I think most of the people who insist on this illusion of separation are sincerely deluded rather than simply trying to trick people into volunteering to help AOL Time Warner's strategic interests for no pay, but they're just as mistaken.<
That's such a moronic argument. I wish they made philosophy, or at least logic, compulsory in all school.
Where are the corporate filings for Lee Strauss? Or for cheese? Or for art theory lessons? Well obviously there aren't any, because they aren't corporations. You'd have to have a particularly bizarre ontology to believe there were only corporations.
You're still in the mad, straw man mentality where businesses aren't allowed to contribute to open source projects, and people aren't allowed to do so for monetary gain. Will mozilla.org no longer be independent from OEOne if they merge their calendaring stuff, especially when it's for the sake of making money? Of course it will still be independent, and obviously open source projects *love* large contributions from corporations. It's only closed source projects that make these sorts of distinctions, and won't take contributions from (other) corporations.
#101 Re: Re: Re: Some FAQs for you to read
by dipa
Wednesday November 7th, 2001 7:51 AM
Following his logic, probably he thinks that Linux contributors are just unpaid suckers working for RedHat and Suse.
I wouldn't call someone's arguments moronic. But he doesn't deserve your answer anymore because he's completely out of context. I suppose you have more important tasks to do for Mozilla.
"I think most of the people who insist on this illusion of separation are sincerely deluded rather than simply trying to trick people into volunteering to help AOL Time Warner's strategic interests for no pay"
#111 Re: Re: Re: Re: Some FAQs for you to read
by strauss
Wednesday November 7th, 2001 12:20 PM
> probably he thinks that Linux contributors are just unpaid suckers working for RedHat and Suse. <
Most Linux contributors are paid, and almost all Linux contributions are reviewed and approved by people who are paid to do so. Similarly for Mozilla. The myth of volunteerism in open source projects has been mostly obsolete for a few years. Volunteer effort is not what makes things like Mozilla, the Linux kernel, StarOffice, and GCC run. They advance because people are paid to advance them by companies who feel it's in their economic interest to do so.
#120 prove it.
by joschi
Wednesday November 7th, 2001 2:54 PM
that's a mighty strong assertion with no backing whatsover (read: bull)
#125 Re: prove it.
by strauss
Wednesday November 7th, 2001 3:22 PM
We've all seen that the vast majority of Mozilla work is done by paid contributors from Netscape, and that noteworthy outsiders have almost all been paid employees of companies like Red Hat. Recent explanations of the Linux kernel process on /. showed that almost all the core kernel maintainers were Red Hat employees. When Sun has tried to throw non-core features in OpenOffice to volunteers, such as Mac OS X support, they have not gotten anywhere. GCC fixes were largely from paid employees of Cygnus (a compiler consultancy, now part of Red Hat) for years. GNOME is almost entirely done by employees of Ximian, and they also took a lot of work from paid employees of the now defunct Eazel.
Tell you what, why don't you prove that large open source projects are primarily done by volunteer labor? Gerv would love it -- he was just begging for evidence that open source development was bringing in lots of free outside work so he could show it to Netscape management.
#127 Re: Re: prove it.
by roc
Wednesday November 7th, 2001 3:49 PM
Matthew Thomas recently posted a rather large laundry list of work done by non-Netscape people, and most of that was done by people not paid by anyone to work on Mozilla. Plus, as I've repeatedly observed before, you'd see a lot more non-Netscape and non-corporate Mozilla contributions if Netscape would just stop hiring all the good contributors :-).
Your thesis that volunteer-driven open source projects are not successful is wrong, because you have misinterpreted the evidence. GCC, GNOME, and Linux started out as volunteer projects and were so successful that businesses sprung up and started paying people to work on them. In other words, volunteer-driven projects can and do succeed, but one result of that success is that people can get paid instead of volunteering. (BTW, KDE is an excellent example of a project that has little corporate involvement and yet is very successful.)
StarOffice and Mozilla belong to a different breed of project than any of the above. These are projects of larger scope than most other open source projects, where there is an existing closed source code base and a large established closed-source development organization. This is definitely a lot harder to make work. It seems to take a long time to adjust processes, tools, culture and code to make it easy for contributors to ramp up and for a community to develop. I think Mozilla is still gradually improving there, and no doubt OpenOffice still has a long way to go along that road. Anyway, you can't easily generalize lessons from OpenOffice and Mozilla to volunteer-born projects, nor vice versa I suspect.
#131 Re: Re: Re: prove it.
by strauss
Wednesday November 7th, 2001 8:52 PM
> GCC, GNOME, and Linux started out as volunteer projects and were so successful that businesses sprung up and started paying people to work on them. <
Exactly. As an initial, "toy" project, the volunteer mode may be fun, and it may "work" in the sense of improving a toy project. Once a certain complexity threshold is reached, though, the startup cost of participating is so high, and the need for central control so great, that volunteer work becomes far less significant than coordinated and compensated work.
#132 Re: Re: Re: Re: prove it.
by joschi
Wednesday November 7th, 2001 9:23 PM
It's an interesting theory, but there are just too many cases that prove the opposite. A partial list: Apache, KDE, Gnome (the major leg work was done pre-commercial involvment), the linux kernel, most of the Gnu tools, free/open/net-BSD, and many many more.
To get back on topic, the notion that you are trying to assert, that the rise in paid open source developers must mean the decline of volenteers remains complete heresay. Citing O.S.S projects that have commercial contributors does not prove an absense of volenteer based projects. It's just faulty logic. I'm still waiting for proof (against all evidence) that volenteer based OSS projects are a thing of the past.
#134 Re: Re: Re: Re: prove it.
by roc
Wednesday November 7th, 2001 10:11 PM
Red Hat was not formed to "save" Linux from obscurity or to manage its complexity. Nor was Cygnus formed to save GCC. They were formed to profit from already successful projects --- NOT "toys".
I don't know if you realize this or care, but you come across as terribly condescending. You're telling the people who built (for example) the GIMP, KDE and Apache that, regardless of what they've achieved, their projects must be "toys" because they are only volunteers and not blessed by corporate management and compensation.
#163 Re: Re: Re: Re: Re: prove it.
by strauss
Thursday November 8th, 2001 1:23 PM
> You're telling the people who built (for example) the GIMP, KDE and Apache that, regardless of what they've achieved, their projects must be "toys" because they are only volunteers and not blessed by corporate management and compensation. <
GIMP and KDE are still toys -- they're not ready for prime time.
As for the actual open source successes and near-successes such as GNOME/Eazel, GCC, the Linux kernel, Apache and (perhaps soon) Mozilla, I'm pointing out that the point when they ceased to be toys was also the point at which most of the contributors came to be paid to work on them. Given that most of the contributors are paid, I don't see why they'd be complaining about what I'm saying. I'm sure they're glad to be paid or have been paid to work on open source projects.
#170 Re: Re: Re: Re: Re: Re: prove it.
by roc
Thursday November 8th, 2001 2:31 PM
"Toys" is a weasel word. You can stretch it to mean whatever you want. Fact is, lots of people (including me) use KDE and the GIMP to get work done every day, and in my dictionary, that means they're not "toys".
If you want to ignore the distinction between cause and effect, that's OK, but please avoid making the claim that volunteer-driven projects can't be successful. The record clearly shows that they can be.
#172 Re: Re: No pop-up :)
by joschi
Thursday November 8th, 2001 2:47 PM
He uses the word "toys" because it is indended to be insulting. Since he is absent any evidence for his position he trolls to produce emotional reactions. Once he get's those emotional reactions, he switches the argument to name calling. Pretty standard distraction techniques.
#179 Interesting arguments
by jedbro
Thursday November 8th, 2001 6:35 PM
Wow, this thread has keep me occupied for a while =) hehe.
Nice to see to intelligent people discuss something civilized.
As both of you have many points that I agree with, and other that I don't, I do lean towards roc's point of view. Basically volenteers and paid coders are both KEY elements to Open Source development, this is why Microsoft cringes with the words O.S (no, I do not want to explain, nor prove my point with facts, text and etc., pefer to "waste" my time doing other things.. hehe sorry)..
Secondly, over the last 6 months, reading comments by Strauss, he seems to be very negative and un-objective (in no way am I dissing Strauss, You rock man.. but it's just my point of view).
Anyhow.. I truly enjoyed this thread, thank you!
#180 Re: Interesting arguments
by strauss
Thursday November 8th, 2001 7:42 PM
> Basically volenteers and paid coders are both KEY elements to Open Source development <
Sure. What I'm saying is that volunteers are much more important at lower levels of system complexity, and become much less relevant than paid professional contributors at higher levels of complexity.
Again, you can look at the checkins to see who does the lion's share of work on Mozilla -- it's not just my opinion. If Netscape employees were pulled off the open source trunk, as Gerv suggested they might be before he vanished from sight, Mozilla development would nearly stop.
#185 Re: Re: Interesting arguments
by choess
Friday November 9th, 2001 12:24 AM
I don\'t think this is an either-or question. Given the amount of work NS employees do on major modules, it\'s clear that the project would pretty much go dead in the water without their involvement. OTOH, if Netscape were suddenly to face the responsibility of doing all their own QA + bug triage + maintaining small-but-useful-non-core-feature foo, bar, and baz, it would also throw a serious cog in the pace of development.
I think of the organization of individuals in the developer community as being a loose hierarchy: you have your uber-coders, most of whom are employed by Netscape, your demi-uber-coders (both volunteers and NS intern-type-people), and Teeming Millions like me of volunteer QA\'ers. The top of the hierarchy are the people who are fixing involved, nasty bugs and keeping the core functioning. The next level down does some of that, but they also engage in things like large-scale non-programming projects (helping keep Bugzilla running, etc.), or will often tend their own enclave somewhere in a module (i.e., bz is responsible for getComputedStyle(), bbaetz has ftp viewer and gopher, rbs is coordinating MathML). The Teeming Millions triage bugs, make testcases, resolve the 30% or so (guesstimate) of bugs filed daily that become duplicate/invalid/wontfix, write small patches, and so on. Obviously, the uber-coders are key to keeping the project running, but they need the rest of the hierarchy to run interference for them, complete features that might otherwise be pushed off indefinitely (see getComputedStyle(), above), and generally take care of bugs that it would be a waste of their paid time to have to go through.
In short, I\'d find it hard to call volunteers \"much less relevant\"; losing volunteers would result in a tremendous price in terms of the completeness and polish of the application.
#197 Volunteer value
by SubtleRebel
Friday November 9th, 2001 10:44 AM
Without the outside involvement, the product appeal could suffer. AQ closed development team tends have a much more narrow perspective than an open source development team.
#138 Re: Re: prove it.
by gwalla
Wednesday November 7th, 2001 11:26 PM
Ben Bucksh is one major individual contributor (primarily on the mailnews component). While you could argue that he's employed by Beonex to work on Mozilla, that's because he founded Beonex for just that reason.
#165 Re: Re: Re: prove it.
by strauss
Thursday November 8th, 2001 1:29 PM
> While you could argue that he's employed by Beonex to work on Mozilla, that's because he founded Beonex for just that reason. <
I did not exempt people who are employed by other organizations to work on Mozilla. I used to manage one such outside paid engineer myself. It was full-time job for him for months to bring up one feature, which the browser desperately needed, and which we thought we would need in 2001 because we believed what we were being told about the release schedule. That's why we paid for it. If it had been left to part-time volunteer labor, it wouldn't have gotten done.
#198 Which feature
by SubtleRebel
Friday November 9th, 2001 10:48 AM
Just out of curiousity, what feature did he add to Mozilla?
#103 Re: Re: Re: Some FAQs for you to read
by macpeep
Wednesday November 7th, 2001 9:05 AM
"That's such a moronic argument."
Look, you're at it again..
Strauss just gave good reasons in several posts in the comment threads in this article and you just flat out ignore them as if you hadn't even read them. Or maybe you didn't comprehend them.
Mozilla can't be equated with cheese or a person. It's an organization, which at least in Europe is usually is a legal entity with rights and obligations, very much like a company. In order for Mozilla to do things like sign contracts or obligate others to follow contracts by the Mozilla organization, it has to be a legal entity. Unlike cheese.
Instead of calling people or their arguments morons, you might want to actually look past your own extremely narrow viewpoint and examine the arguments other people are making. There's a reason why there are two opposing views here and it's most likely not because people are stupid or out to annoy you. Have an open mind.
by joschi
Wednesday November 7th, 2001 3:00 PM
an organisation does not have to be legally registered to exist, that is a silly assertion.
by macpeep
Thursday November 8th, 2001 6:45 AM
"an organisation does not have to be legally registered to exist, that is a silly assertion."
Are you sure? In Finland, an organization needs to be legally registered to exist. Otherwise it can't do things like have a bank account, own property, make contracts, hire people, etc. I'd be surprised if this was very different elsewhere. Who owns the domain mozilla.org and the trademark "Mozilla"? If I break a Mozilla license, who have I offended against (who sues me?)? You understand what I mean, I'm sure..
#154 Re: Re: no.
by AlexBishop
Thursday November 8th, 2001 9:23 AM
"Who owns the domain mozilla.org and the trademark 'Mozilla'?"
Netscape own them both (though I believe the trademark on the word 'Mozilla' dates back to when it was Netscape's mascot).
Alex
#110 Re: Re: Re: Some FAQs for you to read
by strauss
Wednesday November 7th, 2001 12:14 PM
> That's such a moronic argument. I wish they made philosophy, or at least logic, compulsory in all school. <
Thanks, I'm quite well read in philosophy. one thing one learns in epistemology is that when people resort to emtoional arguments and ad hominem attacks, it often indicates a lack of evidence and reasoning supporting their viewpoint. You don't gave a scrap of evidence for a separate Mozilla organization (since, after all, there isn't one), so you decided instead to post three paragraphs of personal attacks on me. Forgive me if I'm not impressed.
#122 you mean like this?
by joschi
Wednesday November 7th, 2001 3:02 PM
http://mozillazine.org/talkback.html?article=2079&message=78#78
#139 Re: Re: Re: Re: Some FAQs for you to read
by Ugg
Wednesday November 7th, 2001 11:31 PM
"one thing one learns in epistemology is that when people resort to emtoional arguments and ad hominem attacks, it often indicates a lack of evidence and reasoning supporting their viewpoint."
And one thing that one learns in real life is that most of what you learned in college is worthless. Another thing one learns is that emotional arguments and ad hominem attacks may indicate that the one making them has such contempt for the arguments and/or position of his opponent that he sees no value in attempting to address them in a more logical manner.
When someone insists that 2 + 2 = 3, it is easier to call him a dumbass than it is to use advanced mathematics to prove him wrong, and one can hardly be said to lose the argument by doing so.
#159 Re: Re: Re: Re: Re: Some FAQs for you to read
by tny
Thursday November 8th, 2001 11:22 AM
> one can hardly be said to lose the argument by doing so.
Actually, one can. If an argument lacks a logical foundation, it's not an argument. Argumentum ad hominem, argumentum ex cathedra, etc. are equivalent to a forward pass beyond the line of scrimmage in football - you might win if you don't get caught, but it's not worth the risk trying them.
#199 Specific example
by SubtleRebel
Friday November 9th, 2001 11:08 AM
Taking things out of context really is not fair. Ugg made a statement as part of a specific example. You then quoted part of it and attempted to disprove it in a global context. You misrepresented what he was saying.
Perhaps if I restate Ugg's example in different terms:
If Person #1 makes a statement with no logical foundation and Person #2 responds by saying that that Person #1 is a moron, then how can it be said that Person #2 lost the argument which never really existed?
#109 Independence
by roc
Wednesday November 7th, 2001 11:47 AM
There is some amount of independence. For example: Mitchell Baker was fired by AOL but she still runs mozilla.org --- which would be strange if AOL called all the shots. Drivers sometimes reject or postpone requests to land changes desired by Netscape management. New Netscape employees don't get CVS access until they've persuaded at least one non-Netscape person that they deserve it.
It is true that mozilla.org does depend on Netscape an awful lot. But one major reason for that is that Netscape keeps hiring all the non-Netscape contributors! If that stopped, or had never happened, you would see a lot more non-Netscape activity.
#113 Re: Independence
by strauss
Wednesday November 7th, 2001 12:46 PM
> Mitchell Baker was fired by AOL but she still runs mozilla.org --- which would be strange if AOL called all the shots. <
Her leadership is titular only, and part of the fiction of independence. Remember Gerv complaining last week that the Mozilla leadership were being called on the carpet every week to justify to netscape management why they should keep the project open source? that tells you what the actual reporting structure is.
> Drivers sometimes reject or postpone requests to land changes desired by Netscape management. <
I'd need to see examples and documentation.
> New Netscape employees don't get CVS access until they've persuaded at least one non-Netscape person that they deserve it. <
That's a weird claim. Basis?
#126 Re: Re: Independence
by roc
Wednesday November 7th, 2001 3:27 PM
> Remember Gerv complaining last week that the Mozilla leadership were being called on the carpet every week to justify to netscape management why they should keep the project open source? <
If, insane as that would be, Netscape chose to devote themselves to a closed-source fork of Mozilla, that would still not mean that mozilla.org is not "independent". If Netscape ordered mozilla.org to disband completely, then that would be a real test --- although I suspect non-Netscape folks @mozilla.org like Mitchell Baker, Frank Hecker and Chris Blizzard would have something to say about that.
Netscape employees still do most of the development (partly, as I said, because they've hired (or tried to hire) most of the good non-Netscape contributors), so Netscape carries a lot of weight with mozilla.org --- maybe too much. But to say there is NO degree of independence is simply wrong.
> I'd need to see examples and documentation. <
I can't republish drivers email, sorry. But if you search Bugzilla, you should be able to find bugs where PDT approved a patch to be checked into Netscape's branch before it was approved for the Mozilla trunk (if it was approved for Mozilla at all).
> That's a weird claim. Basis? <
http://bugzilla.mozilla.org/show_bug.cgi?id=103081 and others.
#128 Re: Re: Re: Independence
by strauss
Wednesday November 7th, 2001 3:51 PM
I appreciate the information. However, I don\'t see how you have arrived at your conclusions. There\'s nothing at http://mozilla.org/hacking/getting-cvs-write-access.html that says Netscape employees have to get non-Netscape employees to give them write access. It may sometimes be that one of the approval requests goes outside Netscape, but the way you said it, it sounded like it was required to. And I don\'t see how Netscape-requested fixes getting into the Netscape 6.x branch before the main Mozilla trunk demonstrates that Mozilla people have turned down fixes requested by Netscape -- why would Netscape managers care if the fix got checked into the main trunk if all they\'re doing is prepping for a new 6.x release? Nobody has overridden anyone from Netscape there.
#133 Re: Re: Re: Re: Independence
by roc
Wednesday November 7th, 2001 9:44 PM
Try reading your link again:
>> If super-review is required, then at least three "super-reviewers" (at least one of whom must not be employed by the same entity as is the person seeking CVS access) must approve in writing. <<
>> This entire process applies to everyone who wants CVS write-access to mozilla, including new Netscape employees. <<
#239 Re: Some FAQs for you to read
by ed_welch
Wednesday November 14th, 2001 1:41 AM
I'd like to correct point 2:
"2) IE renders pages better than Mozilla - depends on where you're coming from. From the users perspective, more pages work in IE than Mozilla for 2 reasons - web page designers write code that only works in IE"
IE decided to be 100% backwardly complient, which means they implemented the old IE specific DOM methods (i.e. document.all), while Mozilla took the more puratin approach and only implemented the w3 DOM standard, refusing to support layers.
The result?
Some web sites, particularly Banks use javascript navigation (not a good idea you may say, but there you go). And they have spent a lot of time and effort developing their system to be accessed by both Netscape 4.x and IE, by severing out different javascript code depending on the browser type. Now effectively there is a new DOM standard, naturally they are not too enthusastic about re-writing the whole system again so it can be accessed by Mozilla.
So the truth is that a lot of web sites can only be accessed by IE OR Netscape 4.7 and that's one of the reasons I am still using IE, because accessing my Bank is, like, sort of important.
#240 Banks supporting Gecko
by eiseli
Saturday November 17th, 2001 12:28 AM
Well, believe it or not, but my bank has captures of Netscape 6 in its ads. And yes, they did a redesign. Just a way to show us clients that we are the king. Maybe you're at the wrong bank? http://www.ubs.com/
by pbreit
Tuesday November 6th, 2001 8:45 PM
Mike, those are excellent points and I agree with many of them.
For me, what it comes down to are:
1) Feb 2002 will be the 4th anniversary of Mozilla and it doesn't look like 1.0 will be ready by then. That's a long time to get to 1.0.
2) because IE is so thin (especially when compared to Moz), it feels like it's everywhere. Lots of apps embed it as a rendering engine. When you're looking at a directory listing in windows you can just type in a URL in the address bar, etc. Moz, on the other hand continues the old paradigm of a monolithic app. There should be no concept of "start up time". A window should just open with a view into the net.
3) Moz's biggest opportunity is replacing IE in AOL client. This won't happen until it is performing extremely well. Further, when/if it happens, it will be gecko only. No mail. No news. No chat. No compose. No skins. My beleif is that Mozilla's insistence on including these superfluos components suggests that it really is independent from AOL/TW.
So while I disagree on a number of points, your message was refreshing.
#88 Re: FAQs
by AlexBishop
Tuesday November 6th, 2001 9:04 PM
"1) Feb 2002 will be the 4th anniversary of Mozilla and it doesn't look like 1.0 will be ready by then. That's a long time to get to 1.0."
February? March 31st, surely. October 26th if you start counting from the rewrite (in which case Mozilla is only just over three years old now).
Alex
#94 Re: Re: FAQs
by SubtleRebel
Wednesday November 7th, 2001 12:02 AM
I would think that you would have to start with the rewrite because for all practical purposes that was really the beginning of what will become Mozilla 1.0 in the near future. What came before was a totally different set of applications.
Regardless 4.0 years is not really very long to develop a software suite - especially one that will be so advanced. The fact that Mozilla hs not yet released a 1.0 version is irrelevant. IE is now at version 6.0 (for Windows only) after 7 (or is it 6?) years of development. Mozilla 1.0 will soon pass IE in quality and IE will always have had more years of development.
BTW, if I remember right the codebase for IE never even had a 1.0 release.
#97 Re: Re: Re: FAQs
by AlexBishop
Wednesday November 7th, 2001 3:19 AM
"IE is now at version 6.0 (for Windows only) after 7 (or is it 6?) years of development."
I think it's 6. Microsoft claim they started developing IE in 1994 but I don't believe them - 1995 sounds about right (that's when they licensed the NCSA Mosaic technology from Spyglass.
"BTW, if I remember right the codebase for IE never even had a 1.0 release."
Yeah, they did. It came with the Plus! pack for Windows 95. It was rubbish though so version 2.0 was quickly brought out later that year. That was still no good, so they released 3.0 in late 1996 (about a week before Netscape Navigator 3.0), 4.0 in late 1997, I think (a month or so after Netscape Communicator 4.0), 5.0 in early 1999 and 6.0 in August 2001.
Alex
by SubtleRebel
Thursday November 8th, 2001 4:55 AM
Wasn't IE 1.0 a totally different codebase than 2.0 though?
My imperfect memory is that they totally scrapped 1.0 because it sucked so bad and then came out with 2.0 which was actually derived from a web browser that they had acquired (along with the company that developed it).
#162 Re: IE 1.0
by AlexBishop
Thursday November 8th, 2001 12:54 PM
"Wasn't IE 1.0 a totally different codebase than 2.0 though?
"My imperfect memory is that they totally scrapped 1.0 because it sucked so bad and then came out with 2.0 which was actually derived from a web browser that they had acquired (along with the company that developed it)."
No, 2.0 sucked as well! :-) AFAIK, IE 1.0 was based on NCSA Mosaic, the browser everyone used before Netscape (that was, for the most part, made by the same people). Microsoft licensed the technology in late 1994/early 1995 from Spyglass, who I believe were given the exclusive right to license the Mosaic codebase by the NCSA. The licensing agreement stated that Spyglass would get a percentage of the income from each browser sale. So Microsoft decided to give IE away for free.
Version 2.0 improved on 1.0 but was still rubbish. As I understand it, Microsoft rewrote substantial portions of IE for version 3.0, which was the first decent version (it usually takes Microsoft three versions to get it right). And I believe they rewrote the rendering engine for IE4. So, over the years, Microsoft has probably rewritten most of IE (and the most recent Mac versions are rewrites rather than ports of the Windows code), but there's still some NCSA Mosaic code in there, as the acknowledgement in the About box states.
Alex
#182 Re: Re: IE 1.0
by vcs2600
Thursday November 8th, 2001 8:31 PM
IE 1.0 was so bad that they barely distribted it (I only saw used as a navigator on some misc. MS marketing CDs). I don't know if it was the same codebase or not, but it was significantly worse than even IE 2.0 (based on Mosaic).
#201 Re: Re: IE 1.0
by SubtleRebel
Friday November 9th, 2001 11:36 AM
After reading your post, I think that it was Spyglasss that I was thinking of, but I still think that the Spyglass/Mosaic code might not have been part of the IE 1.0 release. I seem to recall reading somewhere (back at the time) that MS was abandoning the 1.0 codebase and starting over with a newly acquired codebase, but that they were keeping the IE moniker.
I could be totally wrong though because that was a long time ago and was not something that I made any effort to remember. So my recollection is sketchy at best on this topic.
#202 Re: Re: Re: IE 1.0
by AlexBishop
Friday November 9th, 2001 12:18 PM
According to this http://www.zdnet.com/devhead/resources/tag_library/history/ie.html page, IE "was originally based on the Spyglass Mosaic code." As IE6's About box still mentions that it's based on NCSA (Spyglass) Mosaic, I think the same fundamental codebase has always been used, though extensive parts of it have been rewritten.
Alex
#209 Re: Re: Re: Re: IE 1.0
by SubtleRebel
Friday November 9th, 2001 1:31 PM
Based on the info at that link:
1.0 was released in Aug 1995
2.0 Beta released in Oct 1995
2.0 Final released in Nov 1995
Seems to me that that almost supports my view that Version 1.0 and 2.0 were from 2 separate projects running at the same time. Otherwise MS could not have gotten a 2.0 Beta out the door so shortly after the 1.0 release.
The link says that IE was originally based on the Spyglass/Mosaic code, but that does not necessarily mean it is right. ( I also recall claims that Windows 95 did not use DOS. ) The origins of IE 2.0 could be Spyglass/Mosaic code which would then be the origins of the current IE, but it does not explicitly say that 1.0 was based on the Spyglass/Mosaic code.
I am not about to suggest that my knowledge of IE history is flawless, but I also know better than to believe everything that I read, especially from ZiffDavis.
#214 Re: Re: Re: Re: Re: IE 1.0
by AlexBishop
Friday November 9th, 2001 2:33 PM
"1.0 was released in Aug 1995 2.0 Beta released in Oct 1995 2.0 Final released in Nov 1995
"Seems to me that that almost supports my view that Version 1.0 and 2.0 were from 2 separate projects running at the same time. Otherwise MS could not have gotten a 2.0 Beta out the door so shortly after the 1.0 release."
I wouldn't discount it. Version numbering is pretty arbitary. What was there to stop Microsoft claiming that a minor update was actually a whole integer upgrade?
"I am not about to suggest that my knowledge of IE history is flawless, but I also know better than to believe everything that I read, especially from ZiffDavis."
I don't believe everything that I read. I'm not trying to argue with you either. You just asked if IE1 and IE2 had seperate codebases and I tried to answer that question to the best of my knowledge. I did a Google search to check if what I thought was true actually is and it brought up the ZDNet article.
If you want to be really picky, ZDNet isn't even part of Ziff Davis anymore. Ziff Davis sold their online division to CNET (seems like a bit of a silly thing for a technology magazine publisher to do, which is probably why Ziff Davis have spent the time since then rebuilding their online presence with sites such as ExtremeTech).
Alex
by pbreit
Wednesday November 7th, 2001 11:19 AM
dude, you're living in a reality distrotion field!
fact: 2/23/98 "open for business!" http://mozilla.org/news.html
opinion: "I would think that you would have to start with the rewrite"
4 years is an excruciatingly long time to develop a browser. it's even more painful when you recognize that the product requirements were virtually completed prior to 2/23/98!! they had a perfectly functioning prototype of exactly what they were building!! and a complete code-base! if anything, suggesting that it all only got started on 2/23/23 is being very, very generous.
and again, you make my case for me: the sole reason it is taking so long is because they are building a "suite" and not simply a browser. and it is clear that a suite is wholly unnecessary to everyone except those still living in the mid-90's.
#116 Re: wrong!
by AlexBishop
Wednesday November 7th, 2001 1:35 PM
The Mozilla Organization may have been formed in February 1998, but the code wasn't released until March and the rewrite didn't occur until October. That's where I'm counting from. You may as well say that the development of Internet Explorer began when the Microsoft Windows team was formed, even though it required several changes in policy between the formation of the team and the start of the writing of IE.
Alex
#130 the point
by pbreit
Wednesday November 7th, 2001 5:58 PM
right. and the result of making that the goal is the non-release of 1.0 4+ years after beginning with a fully functioning prototype and an entire code-base. it's difficult to exagerate how bad a decision it was not to focus on just a browser.
#140 Re: the point
by gwalla
Wednesday November 7th, 2001 11:38 PM
"fully functional prototype" is overstating a bit. The codebase that was scrapped was awful. Besides, that doesn't support your argument: the original codebase *was* for an internet suite (with mailnews, etc. as well as a browser).
Anyway, you're comparing apples to oranges. Maybe if Mozilla had equivalent functionality to Netscape 1 or IE2, but it's roughly equivalent to IE6 (and superior in some respects) browser-wise. How long did it take for Microsoft to get to the point IE is at now from when they started?
#141 Re: the point
by Ugg
Wednesday November 7th, 2001 11:38 PM
I'm really beginning to think that you're a troll.
"a fully functioning prototype"
...which was crap, btw...
"and an entire code-base."
...which was likewise crap.
"it's difficult to exagerate how bad a decision it was not to focus on just a browser."
Had they done it your way, the browser would be crap as well. Is that what you wanted?
by pbreit
Thursday November 8th, 2001 12:19 PM
subtle, you really are dense!
1. fact: moz was organized on feb 23, 1998; fact: current objective is 1.0 within 6 months = fact: 4+ years
2. navigator/communicator 4.x = full, functioning prototype.
3. code-base/prototype was a browser with various other components tacked on.
4. fine. starting from scratch generally not a bad idea when re-architecting. it's usually nice to be able to refer to existing code that does what you will be building.
5. re-writing gecko from scratch was a good decision. including mail, news, edit, chat was not.
6. that's an opinion that i strongly disagree with. if gecko were completed 1-2 years ago, was high performance, well-architected, stable, standards compliant and easily embeddable, it would be doing better in terms of share and future prospects than it is now, imo.
7. my position is rather simple: moz should have focused on the browser and not concerned itself with miscellaneous other clients. doing so would have produced a strong browser 1-2 years ago (hard to argue with) and enhanced mozilla's viability considerably (arguable).
#174 Re: dense
by gwalla
Thursday November 8th, 2001 2:59 PM
> 2. navigator/communicator 4.x = full, functioning prototype.
But that wasn't what the Mozilla project started with. The code to Netscape 4 was never opened.
#183 Re: dense
by WillyWonka
Thursday November 8th, 2001 10:10 PM
I'm not saying I agree or disagree... the fact is they didn't focus on only the browser and lots of software projects take more than 4 years to produce a good product (Sometimes a bad one too... daikatana). What is the sense in bickering about it?
By the way when is the next duke nukem game coming out? ;)
#186 Re: Re: dense
by macpeep
Friday November 9th, 2001 5:03 AM
Well, Max Payne, another product that was on the schedule of "when it's done", was finally released. :) I know and work closely with lots of people at Remedy Entertainment and I know that schedules like that can have *very* bad effects on people's morale...
#207 You are misinformed
by SubtleRebel
Friday November 9th, 2001 1:02 PM
1> The date that mozilla.org was organized has no relevance to your claim of 4+ years of development of Mozilla 1.0 because development of Mozilla 1.0 did not start on that day. Netscape did not release any code on that day. The code that was eventually released is not really the code that Mozilla 1.0 is being built from.
2> Mozilla.org was not given Netscape 4. Mozilla.org was given Netscape 5. Netscape 5 was not fully functional. If you want, I think you can still download it and check it out for yourself.
3> Codebase was for a suite.
4> The point of conception for the rewrite is when development of Mozilla 1.0 really began.
5> Without the other components, it would not have been a project that I would be very interested in. Several other people in this forum and elsewhere have expressed similar opinions. Really this is more related to #6 though.
6> If Mozilla had chosen to only develop a browser then it would not have received as much support from various companies and individuals. Without that support, the project would have been diminished. Ultimately, the quality of the browser itself would suffer. Part of the reason that you disagree is because you have this idea that no one uses the mail/news/chat/etc features of Mozilla; however, I believe that it is because of the inclusion of these features that Mozilla has been able to survive this long.
7> It is not at all hard to argue with the either point. You are basing your first claim on the assumption that if Mozilla scrapped the non-browser components then there would be signifigantly more people working on the browser component, but I do not see evidence to support that. People who have been working on the other components would not necessarily want to work on the browser. Some people who have worked on the browser might not have done so if it were just a stand alone browser. Also there are a limited number of people who can work on a given section of code at once anyway.
The second part about enhanced viability for Mozilla is definitely unfounded. The current pre 1.0 Mozilla is extremely viable; people have been declaring it dead for years but yet it survives. It is virtually indestructable. How could a stand alone browser be more viable?
#147 Your "point" is beyond irrelevant
by SubtleRebel
Thursday November 8th, 2001 5:21 AM
First of all, your calculation of 4+ years is wrong.
Secondly, if the "prototype" were "fully functioning" then it would not have been a prototype, it would have been a released product.
Thirdly, the prototype and codebase that you are referring to were for a suite from the beginning.
Fourthly, even if the decision had been made to focus on just a browser, the previous codebase would have still been scrapped in favor of the rewrite.
Fifthly, if the previous codebase had not been scrapped in favor of the rewrite then Mozilla would really suck and more than likely would have totally died by now.
Sixthly, if Mozilla were only a browser with none of the other components then it would be used by far fewer people than it is now and it would have received far less support from various companies and individuals.
Seventhly, it is dificult to squeeze so many false premises and conclusions into your posts, but somehow you accomplished it anyway. So it is not really surprising that you are able to exaggerate how bad a good decision was.
#124 the suite is the point.
by joschi
Wednesday November 7th, 2001 3:06 PM
The stated goal of Mozilla is to build a complete networking suite, that's not some rambling tangent they got sucked into.
#100 The many uses of Mozilla
by emlyn
Wednesday November 7th, 2001 5:42 AM
It seems that the people @mozilla.org treat the Mozilla binaries as a tool for finding defects in the source code hosted by the mozilla.org cvs thingy. And while they can be used for exactly that purpose, and serve that purpose well enough, I beleive it should be brought to their collective attention that Mozilla binaries can *also* be used to browse the web, or even to read usenet posts.
Perhaps they already realise this, but I can't recall anyone @mozilla.org publicly acknowledging this feature, so I have to wonder.
#107 What's your point?
by gwalla
Wednesday November 7th, 2001 11:18 AM
It's widely acknowledged (and it says on mozilla.org) that the Mozilla project supplies binaries for testing and development purposes *only*. They *are* for finding defects in the source code, and the source code is for something that can browse the web, read mail and newsgroups, etc.
|