mozilla.org releases Mozilla Milestone 0.9.4Friday September 14th, 2001mozilla.org today released Mozilla Milestone 0.9.4. New to this release is the ability to disable the JavaScript window.open() method during page load and unload events. You can find more information on what's new at the release notes.
No more pop-up windows! Woohoo! Scott Is there a way to turn it off on a site-specific basis? It looks like a global setting. For some sites, disabling popup windows makes the site not work (e.g., Outlook web client). Oh, I get it. It's just for page load and unload events. That probably covers most ad cases today, but advertisers adapt rapidly to this kind of thing.... This one's great. It sets up a local proxy and it will filter anything you want (many filters are provided, but you can write your own). If it's in HTML, it can be filtered! http://proxomitron.cjb.net/ What does PDT+ mean? Thanks, Netvigator PS Mozilla rocks! PDT is the Netscape Product Delivery Team (I think that's the name) and they are responsible for deciding what goes into a Netscape release. When Mozilla was done with a code branch like 0.9.2 (after we released the 0.9.2 Milestone and went back to working on the trunk) some Netscape developers, and a few other contributors, continued to check into that branch working toward a release (which became Netscape 6.1). The managers that decided what got checked into that branch for 6.1 were called PDT and PDT+ is the flag they use for marking bugs they want checked in. They are going to be directing some work on the 0.9.4 branch now that Mozilla is done with it an back working on the trunk towards 0.9.5. Hope this helps some. --Asa I'm using the Mozilla 0.9.4 build now and it seems rather nice now. Although I have no idea what those keywords mean in that message, I take it that Netscape will be using the 0.9.4 branch for their Enterprise component. On a side note, the HTML and DevEdge sidebars Netscape created are really neat. Netscape seems to be really pushing the sidebar as a touted feature. Anyone else noticed how sometime after 0.9.3 the spacing between bookmark menu items got larger? It's pretty annoying because it occupies a lot more room on the screen now, and requires more scrolling. I think this is the correct bugzilla entry for it: http://bugzilla.mozilla.org/show_bug.cgi?id=96817 (In case you'd like to vote for it :) Any ideas yet, mozilla people ? As many as needed :) Probably a lot, since we all want an extremely good product. I already use mozilla all day long as my browser. I am waiting for an IMAP bug to get fixed before I use it for mail too. is 1.0 when most people can use it without many problems? basically I do everything I can to get everyone here at work to use it already. all of our company web sites are tested to be sure they work with mozilla. so if 1.0 means its time to switch to moz-zilla, I think it has arrived. just my rambling... I am just curious, as I use IMAP mail everyday without any problems. Brian! 18266 http://bugzilla.mozilla.org/show_bug.cgi?id=18266 which is a problem where sub folders which have mail placed in them from server side filtering such as procmail don't get their message counts updated. I currently use the workaround of searching all messages for a nonsense string to update the message count, but even then subfolders stop working at all after awhile. I think that might be related to this bug: 90494 http://bugzilla.mozilla.org/show_bug.cgi?id=90494 though. > is 1.0 when most people can use it without many problems? No, 1.0 is a version that is stable/usable and has all the debugging menu items disabled and does not point you to a "thank you for helping to test this development build" page when you start the thing. It would be nice to have a mozilla build for non techie users that has been well tested and packaged for end users (i.e. without the alpa version warnings and debuging/testing related menu-items). So far the milestones have been increasingly better. I wouldn't have much problems with the current milestone being labeled 1.0. The question of the original poster was when this is going to happen. I think it is a reasonable question considering the quality of the current milestone. A 1.0 would mean: - a clear deliverable by the mozilla project - a signal to end users that they can go ahead and install the thing - a signal to third parties that there's something to target their mozilla extensions at - an opportunity for mozilla developers to start the possibly destabilizing work on next generation features Erm, no, you're wrong. That's not what 1.0 will be. Mozilla is a development project, and will always be for testing and development purposes. There won't be a build that doesn't thank you for helping test it. There are builds packaged for end users - Netscape, Beonex, etc. If you want to produce your own, you're welcome to. What 1.0 I think signifies is the all-important API freeze. What that means is that bits can be developed separately, but are not allowed to break conventions on how they communicate with each other. What 1.0 also means is that work can begin on the 2.x series, so people will once again be able to commit MAJOR changes, which alter the way mozilla is used and presented, and works, significantly, as a separate development stream from the now stable tree. I think it would be very beneficial to have a Mozilla 1.0 with the debugging/QA stuff taken out. I agree that Mozilla is *primarily* a development project, but I think their are some great benefits to having a "standard" Mozilla browser release for general users. Brian! I'm simply talking about a tagged version mozilla 1.0 for end users. Probably somebody who knows what he/she is doing could prepare on in a matter of days. The current version number and way of packaging mozilla (i.e. with all the debugging stuff etc.) just screams to end users that it isn't for them. With most open-source projects there is a clear distinction between stable versions, testing versions and development versions. Mozilla has not had a stable end-user version yet (milestones are explicitly meant for testing). You might claim netscape 6.1 is such a stable version but I disagree. Netscape 6.1 is a derived product, just like forte is based on a netbeans release (a stable release) and star office will be based on a stable release of open office. We all agree that the current milestone is pretty good. No crashes, everything seems to work as it should, it is fast. Basically there's no reason an end user shouldn't use this version. So why not communicate this to these end users by labeling this build 1.0 and remove some menus that the ende user won't need. The work that is needed for this step is not much but the message communicated is of great value. If you would like to do a jillesZilla distribution or something like that it would be great. See Beonex http://beonex.com/ for an example of how this can be done. mozilla.org does not manage a distribution. The linux kernal isn't a distribution, it's a technology that is used in derivative works which are branded, distributed and supported by companies like RedHat. Mozilla provided binaries are for testing and development purposes and should not have useful testing items removed. See http://www.mozilla.org/build/distribution.html for info on building your own distribution. --Asa You recently wrote: "If you are involved enough in Mozilla to have an active Bugzilla account I recommend you skip Milestone builds alltogether." If this is the case, I really doubt the debugging stuff is needed in Milestone builds that are mostly going to be used by "end users". (I'm not saying 0.9.4 should be tagged 1.0 or anything like that. I merely think that the Milestones shouldn't have debugging menus etc. Wasn't this the way it used to be?) All Mozilla builds should have these menus. Just because they are less likely to be used in Milstone builds doesn't mean they won't be used. I use them in the Milestone build for testing that build before it is pushed to FTP. The browser buster gets a lot of hits from Milestone builds. The QA menu has items extremely useful to Milestone testers wanting to file bugs. mozilla.org makes binaries for the explicit and exclusive purpose of providing developers and testers with testing builds so they don't have to build the source themselves. mozilla.org does not provide builds with the purpose of attracting end users. Just because a few of them happen to get our builds and use them doesn't mean we should cripple our testing binaries to make them a little more palatable for those end users. --Asa Asa, I really think it is important to have a 1.0 mozilla release. Here are my reasons, I am sure others would agree with me: 1) I have no intention of using Netscape 6.x. I like Mozilla and I use it for all my day to day browsing. 2) I still use NS 4 for email, because I can't afford even the slightest risk of losing email messages. My business depends on it. 3) I'd like a version of mozilla that has all the debugging stuff removed, which is rock solid as far as the mail app goes, which can support java and helper apps properly, is optimised as far as possible without more major codebase changes, and which has a stable api for third party developers to use. When this happens, it should be called 1.0 4) Once 1.0 is released I will remove NS 4 completely from my system and replace it with Mozilla 1.0. I'll still continue testing nightlies of mozilla, but 1.0 will be the benchmark with which to determine if something works or if it is broken. 5) 1.0 is a psychological point for Mozilla. It says to people: the mozilla project has achieved something really solid. It says that Netscape are basing their new browsers on a stable product. It says to companies that they can develop third party extensions without worrying about having the rug pulled from under them. It validates the open source development model. It will be a great headline on Slashdot (I'm not joking). Finally, it might encourage more people to dip into the code and find out how it works. I am willing to bet you that sourcecode downloads for a mozilla 1.0 would go through the roof. All of these things would surely be of benefit to the mozilla community. >>Asa, I really think it is important to have a 1.0 mozilla release. Here are my reasons, I am sure others would agree with me: Not the people that started this whole thing. From the original release mission statement: "So, Mozilla is a set of technologies, but not a specific (in biologic terms, Mozilla is a genus; a particular product is a species)." >>1) I have no intention of using Netscape 6.x. I like Mozilla and I use it for all my day to day browsing. No one is going to stop you. What's wrong with Netscape or Beonex? >>2) I still use NS 4 for email, because I can't afford even the slightest risk of losing email messages. My business depends on it. I've been using Mozilla mail (I'm an extreme user with tens of thousands of emails in dozens of folders, multiple accounts and a lot of traffic with hundreds of mails a day) exclusively since M10 (or maybe M9 or M11, can't remember now) about two years ago and I've never lost a single message. >>3) I'd like a version of mozilla that has all the debugging stuff removed, which is rock solid as far as the mail app goes, which can support java and helper apps properly, is optimised as far as possible without more major codebase changes, and which has a stable api for third party developers to use. When this happens, it should be called 1.0 >>Those wants are not out of line with my 1.0 thinking either. I would like the codebase to be rock solid, support java and helper apps, optimized as far as possible (including major codebase changes if necessary), and stable/frozen APIs. The difference here is that I'm talking about code any you're talking about binaries. Mozilla isn't in the binary business and probably won't be anytime soon. >>4) Once 1.0 is released I will remove NS 4 completely from my system and replace it with Mozilla 1.0. I'll still continue testing nightlies of mozilla, but 1.0 will be the benchmark with which to determine if something works or if it is broken. Great. I'm not going to try to discourage you. If you want a 1.0 based product without debug and QA menus, however, you're going to have to build it yourself, hack those items out with userChrome.css or get it from someone else like Netscape or Beonex. >>5) 1.0 is a psychological point for Mozilla. It says to people: the mozilla project has achieved something really solid. It says that Netscape are basing their new browsers on a stable product. It says to companies that they can develop third party extensions without worrying about having the rug pulled from under them. It validates the open source development model. It will be a great headline on Slashdot (I'm not joking). Finally, it might encourage more people to dip into the code and find out how it works. I am willing to bet you that sourcecode downloads for a mozilla 1.0 would go through the roof. All of these things would surely be of benefit to the mozilla community. I agree completely (I think, unless you said something about end user binaries in there). --Asa 1) I too plan to keep using mozilla. Netscape adds some fluff to the overall process. 2) I am also using n4 for email still. IMAP in 0.9.4 is still not usable on a daily basis. 3-4) I am fine with debug and QA menus. I don't see why mozilla wouldn't have a 1.0 build without debug, but I guess that is up to the people who would have to build it. Asa, is it just an extra work thing about the debug? being a genus doesn't mean you can't take out the debug code for a 1.0 release. especially if it turns out that most intelligent people would prefer to just use mozilla instead of some branded version. kind of a support the community kind of thing. OR, is taking the debug stuff out a useless endeavor? will it not affect anything? and if so why do peple want it out? thanks! michael p.s. how come you seem annoyed by these suggestions? >1) I too plan to keep using mozilla. Netscape adds some fluff to the overall process. What about Beonex? You see too much fluff there? What about AsaZilla ;) >2) I am also using n4 for email still. IMAP in 0.9.4 is still not usable on a daily basis. I've been using IMAP mail in Mozilla exclusively for more than a year and a half and have never had any serious problems. You should give it a try again. It's getting darned good. >3-4) I am fine with debug and QA menus. I don't see why mozilla wouldn't have a 1.0 build without debug, but I guess that is up to the people who would have to build it. Asa, is it just an extra work thing about the debug? being a genus doesn't mean you can't take out the debug code for a 1.0 release. It's not just debug code (test cases really). It's testing links and options for testing different features. > especially if it turns out that most intelligent people would prefer to just use mozilla instead of some branded version. kind of a support the community kind of thing. They're free to use our testbuilds. It doesn't make a lot of sense to me when there are great alternatives like Beonex out there (how about supporting distributions in the community). >OR, is taking the debug stuff out a useless endeavor? will it not affect anything? and if so why do peple want it out? Taking this stuff out is taking out the options that make the build useful to the people it was created for. Mozilla provides binaries for testing. If you want something that's not made for testing you have a couple of nice choices, Netscape and Beonex are the two big ones now and I really really hope that others come along and do distributions. >thanks! michael >p.s. how come you seem annoyed by these suggestions? mozilla.org has said from the very beginning that it was not an end-user distribution. Builds are made for one purpose, to make life easier for folks that want to help out with testing and development. Why is no one here interested in supporting the distributions that must be successfull for this project and community to be successful. Beonex is a great browser, free, with user support, no testing menus, and a number of privacy and security tweaks that should appeal to many people. Why would you want to use testbuilds when there are solid distributions available? --Asa #146 Re: Re: 1.0 is in the eye of the beholder ; >by strauss Thursday September 20th, 2001 10:34 PM Why do you keep bringing up Beonex? It's for Linux. Nobody uses Linux for anything but servers. It's got well under 1% of the desktop computer installed base. >Why do you keep bringing up Beonex? It's for Linux. Nobody uses Linux for anything but servers. I use linux and not for a server. >It's got well under 1% of the desktop computer installed base. Beonex is a distribution for linux and will probably have windows builds soon. If that distribution doesn't cover enough of the desktops to satisfy you then try Netscape 6 which is a distribution for linux, windows, mac and solaris. Why do you keep reading and posting to mozillazine when you clearly don't think Mozilla is a useful technology? --Asa #150 Re: Re: Re: Re: 1.0 is in the eye of the beholderby strauss Monday September 24th, 2001 1:41 PM Linux on the desktop is irrelevant. Even Linux advocates are waking up to that fact -- see this month's Wired. The membership of the open source cult is tiny, despite the fact it has a loud online presence. I read Mozillazine because I have to try to gauge risks to our DHTML-based application and it is one of a handful of useful source on Mozilla. Mozilla is a risk in case it ever achieves significant market share and complicates our development and test procedures. I participate in Mozillazine because (1) lurking is boring and (2) seeing what kinds of responses I get tells me a lot about the technical maturity of the project, which helps me estimate its risk level. #151 Re: Re: Re: Re: Re: 1.0 is in the eye of the beholby SmileyBen Monday September 24th, 2001 5:00 PM Well glad to hear you dismissing any illusions that you're actually around here because you want the project to succeed, or even support its aims! #153 Re: Re: Re: Re: Re: Re: 1.0 is in the eye of the bby strauss Monday September 24th, 2001 5:46 PM It would be great if there were a better and non-Microsoft web browser. If I had any belief that Mozilla was becoming a better web browser than IE, then I'd be cheerleading for it. As it is, it's lateness and slow improvement cycle are making it a me-too project which at best will be only as good as IE, and all that does is complicate my test matrix and development process. Sad but true. Two years ago I had a different set of expectations.... #154 Re: Re: Re: Re: Re: Re: Re: 1.0 is in the eye of tby SmileyBen Monday September 24th, 2001 8:20 PM So what exactly do you want from this browser? As a web developer (and I thought you were a serious one too), I'm aware that Mozilla already is better that IE. Sure it's slower on many systems (but faster on some others), but it does the job of a web browser - rendering HTML - a damn sight better than IE. It boasts feature parity, and a good deal more with it. I'm not exactly sure what it is you feel it's missing. If it's really only that it's kinda buggy, then please, please go back to a closed source world, where you'd be none the wiser for a few months, by which time most of the bugs will be ironed out. I'm not just cheerleading - I understood a few months ago the moans about Mozilla, but now I really don't. I for one am genuinely impressed with what mozilla.org has achieved - and the fact they did it in about half the time the world's richest software company could do it - that's gotta say something. But I guess I don't count in your eyes, as evidenced by Sawmill titlebars at the top and Gnome panel at the bottom of my screen... #155 Re: Re: Re: Re: Re: Re: Re: Re: 1.0 is in the eyeby strauss Monday September 24th, 2001 8:46 PM It doesn't have feature parity. M$ extensions to the DOM may be proprietary but they are darn powerful. And there's nothing like Visual InterDev for Mozilla. What do i want from it? What I used to want was a clean, well-factored, reliable browser on which I could implement various ideas about improving the browsing experience as a weekend project. Now that it's become obvious that Mozilla is a vast, lumbering beast that would require three coders and ten QA people to add the simplest significant feature (instead of just fixing other people's bugs), I no longer have that hope. I guess right now all I want is for Mozilla not to hurt me too much if at some point we have to support it in our application. Think I'm SOL on that one, though -- we haven't been able to get our developers to stay off the IE proprietary features, since those save them a lot of time.... #157 Re: Re: Re: Re: Re: Re: Re: Re: Re: 1.0 is in theby dman84 Tuesday September 25th, 2001 2:50 AM I say then there are several w3c standards beond MS - propreitary DOM that Mozilla supports. Check in with your peoples and see if they can switch.. its people like them that insist that propreitary is better way to move the world forward that use standards and make everyone happy. Its sounds to me that your all MS is the best way to go, despite they cant even fix the IIS worth a crap. Netscape browser/Mail client over MS IE/OE is way better on security and virus protecting. It supports all that MSIE/OE standards and more. So why is it so hard your that you cannot tell your peoples to get with standards, dont your voice count at your company. I guess they like to spend lots of expensive money on proprietary crap! when you can spent the less money and support all the standards like everyone else and still have a better browser experience. Look at SGI - gone, they thought that their proprietary Workstations at $100,000 a pop were better than a PC workstation costing 5-10 G's. Look where they ended up, I'd sure hate to work for a company that insists that management is right despite the employee's imput for suggesting better things which would/could/may get you fired for trying something that makes sense and is intuitive. Tell them to wake up! supporting proprietary formats don't get many companies very far. They either stay where they are at: small percentage or the others usually die trying. #158 Re: Re: Re: Re: Re: Re: Re: Re: Re: 1.0 is in theby SmileyBen Tuesday September 25th, 2001 7:11 AM Erm. The argument that it doesn't have feature parity because MS adds extra non-standard stuff doesn't hold water. Feature parity for a browser is about what web functions it supports. We don't say that IE is far lagging mozilla in features because it doesn't support mozilla skins, or XUL, or Search plug-ins, etc. etc. I guess in not supporting P3P Mozilla could be considered to be lagging, though I'm not sure if that's actually a W3C recommendation yet. After all, Mozilla *does* support all the tags and attributes MSIE supports - the fact that it chooses to interpret ones that feature nowhere in the standards (largely by ignoring them) is just as valid an interpretation as IE rendering them in its own bizarre way. Mozilla certainly can't hurt you in the way you imagine, at least not as much as Mozilla's failure would. The moment IE wipes out any competition, there goes your companies ability to create the best application on an even playing field. >> After all, Mozilla *does* support all the tags and attributes MSIE supports... by ignoring them... << OK, first, that's a pretty strange statement, but anyway, I'm not talking tags and attributes -- I'm talking DOM. Do you do any hard-core, GUI-application-level DHTML work? I hate to say it, I'm an old M$-hater from way back, but IE's extensions are sweet for DHTML, and the W3C stuff is totally inadequate. What is Mozilla going to give me for fine-grained control of text editing or for drag and drop between windows, for instance? >> The moment IE wipes out any competition, there goes your companies ability to create the best application on an even playing field. << I don't know what you mean, or what you think that translates into in business terms. How does having to support multiple browsers help my application's competitiveness? Mostly what it does is make the test matrix twice as large, which slows down development by a factor of 50% or so due to the extra unit tests and bugs, and QA by a factor of 2 or so. Cross-platform development is hard. If the browsers have lots of obscure little bugs all over the place, it becomes _extremely_ hard, and it starts to limit the feature set to a least-common-denominator basis. What does support for two browsers give back to us? Not freedom of choice -- we have to support every browser that our customers might use -- we have no choice. It's really pure overhead from the developer perspective. Multiple Browsers on the market give you nothing. The world would be so much better if there had always been just one browser from one company on one plattform, wouldn't it? Okay, maybe there wouldn't even be a DOM now or any way of using webtechnologies for creating real applications, cause why should this only company develop such things?? Just because they like you and want to give you these tools? Or may it just be that they did all this to be better than their competition? But you're right, all these browsers on all these plattforms with all their proprietary crap are dificult to support. Pretty sad that there aren't any standards. Ooops there are even some. So my only question is, why are you complaining about just the one browser that does support these standards that make developing easier best and not the others? Btw. you have a choice which browsers you support. You have just to live with the fact that you may lose some customers, but that is your choice. If you want these customers you have to support their browsers... >> After all, Mozilla *does* support all the tags and attributes MSIE supports... by ignoring them... << >OK, first, that's a pretty strange statement, but anyway, I'm not talking tags and attributes -- I'm talking DOM. It's not a strange statement in the slightest. You may not have noticed this but HTML is written in ASCII. It is a set of characters and symbols. That's what mark-up is. For mark-up to be useful for anything, each arrangement of symbols must be interpreted in a particular way. A browser is free to interpret them in any way they like. Mozilla, in fact, as well as IE, is able to interpret every single possible combination of these symbols. However, there's an additional part of the story. That part is where, rather than just randomly interpreting them, or have someone write the mark-up and the interpret it themselves (presumably back to what they had intended when they wrote it) people make agreements about how such things should be interpreted. By agreeing, anyone can write mark-up and they don't need to write a program to interpret it - if they did, the exercise would be one in futility, and wouldn't benefit anyone. These agreements are called standards - and in fact MS helped created those standards as much as anyone. Any mark-up that doesn't fit these standards can be interpreted any way a browser likes. In fact, mozilla does a much better job of this, since it interprets any additional tags and attributes the same (ignores them), rather than in various different ways for a small set of them as IE does. But anyway, that's not what you are talking about. It is why Mozilla is better those, so I thought a relevant reply. > What is Mozilla going to give me for fine-grained control of text editing or for drag and drop between windows, for instance? And what is IE giving me for fine-grained control of text editing or drag and drag? Exactly zilch, I can tell you. I'm sure you can work out why. >> The moment IE wipes out any competition, there goes your companies ability to create the best application on an even playing field. << >I don't know what you mean, or what you think that translates into in business terms. How does having to support multiple browsers help my application's competitiveness? There are a few points here. Firstly, standards are always better for business than following one company's lead. If you want to know why having one monopoly controlling things might be bad, and having them break things in any way they like, even undocumented, you could either be paranoid (rather silly) or just Real, AOL, the Kerberos team, or whoever, the list goes on. Secondly, I *seriously* hope you don't just pop up things you develop into one browser window and then decide they work on IE. Have you ever noticed the differences in how IE 4 and IE 4, IE 4 and IE 5, IE 5 and IE 5, IE 5 and IE 6, IE 6 and IE 6, not to mention IE3 and IE 6...... Where on Earth do you get the idea you're only supporting a single browser? >What does support for two browsers give back to us? Not freedom of choice -- we have to support every browser that our customers might use -- we have no choice. No you don't. Sounds to me like management have decided you should use proprietary extensions rather than simply using proper web standards (correct me if I'm wrong). And if that's true, the company deserves to fail should someone actually bother to support standards, as is correct. Now that is, of course, just my attitude, and many people I'm sure would want to distance themselves from it, taking a less hardline stance, so don't go with 'Well if that's mozilla's attitude don't be surprised if businesses don't love it'. But, well, I don't really care if businesses love it, or if it makes good business sense for IE to gain 100% share - the users won't benefit, and bizarrely, I put what's good for users ahead of what's good for companies... Wow. An Essay! I think someone else put it best when they said you don't need to support every browser, you could just support IE, but don't whine if you lose business that way. Many of us, and it's mostly developers in this camp, want it to be simple to develop web applications - which is why we support standards, not proprietary single platforms... I'm going to have to ignore a lot of irrelevant comments here. I also note that your tone has turned snide again, so don't be surprised if I get a little snippy in response. Sorry, I don't see any response in there on the main question: how does having to support multiple browsers help my application's competitiveness "on an even playing field"? Was there a meaning to that comment? As for standards, I haven't seen any reason to conclude that the W3C DOM standards are better than the Microsoft standards. Is it just that it's not Microsoft? The pragmatic fact is that you can do a lot more with DHTML in IE than you can in Mozilla. If there were a bunch of browsers that were all W3C compliant and IE was the oddball, then yes, the standards would have some weight -- they'd have some pragmatic impact. When there is only one browser adopting the so-called "standard," though, it becomes just one product against another. There's no actual standard. We're in business. We're not on a crusade. What do you have to offer us in business terms? User benefit is business benefit; however, your statement that W3C standards are better for users seems strange given the greater richness of DHTML interfaces that can be developed under Microsoft standards. It seems to me there is more an idealism than a pragmatism in your stance, and I am not sure it is an idealism that has any significant real-world benefits. Hmm. My 'irrelevant comments' are snide, are they. Interesting. I guess that speaks for itself. Another thing that speaks for itself is your idea that Mozilla (and netscape, and beonex, and galeon and k-meleon - a single browser) is the only browser that follows standards, which shows ignorance at best, or at worst just trying to back up your case with something you know is wrong. No doubt Opera doesn't count because it currently has little market share. No doubt Konqueror doesn't count because it runs on Linux which doesn't count. No doubt browsers that run on platforms such as webpads, handhelds, even mobile phones, that are unlikely to be IE, will in your world be based on Microsoft standards. To imagine that you will benefit from a Microsoft only world is naive in the extreme (and that's not snide, there are many worse words that could also be applied, that would be rude, naive is simply the truth). And in fact, that answers your next question. You want business benefit not user benefit. Apparently business benefit is user benefit. Well I'll tell you very simply what will benefit 'business' (inasmuch as anything can benefit an abstract entity) - and that's certainly mozilla.org packing up, you're right. The reason you're right, however, is that for IE to have a 100% share would generate bucketloads of cash for Microsoft, and able to push anyone else of the market for internet applications they will succeed in ways that no computer business ever has before. Unfortunately, I really don't know what 'business benefit' really means. Trouble is I'm not American. This isn't a question of idealism vs. pragmatism. Either one of those has to work out what you're aiming for. What mozilla.org is aiming for is a browser that is best for users (whatever that means, I think a good approximation is a browser that empowers its users) and makes life easiest for developers, partially by adopting well accepted standards. Microsoft certainly aren't aiming for those things, and the question of whether they're coincidentally achieved those things more than mozilla up to now is pretty much irrelevant, since overall a project aiming for things is always going to achieve them better in the long run than something going for other things. And yes, I think people involved in mozilla are on a crusade (though I wouldn't call it that, we've already had one moronic use of the word in the last two weeks that's gonna end up killing thousands), as is everyone in the free software / open source movement. It's about software empowering people - user and developers - and being valued for that, and not simply its ability to make people money. I think that's perhaps why you don't come across as having the first sign of understanding of this whole debate, since you talk in monetary terms as if they're all that there is, even though (since you seem intelligent, and to have thought about such things) you probably do. Opera and Konqueror have roughly zero percent market share, so they don't justify the large investment it would take to support them. Ditto network appliances, all of which have failed. That's how business works. If you do things that blatantly waste money, your shareholders can sue you. Sucks but that's how it is. Mozilla's possible success has some chance of preventing MS hegemony of the Internet browser market, although it's a long shot. Nonetheless, I grant this as a possible general benefit. It must be weighed against the costs of cross-platform development, which are high. Whether MS dominates browsers doesn't have much impact on whether they finally decide to compete with my application. By business benefit, I just mean something that makes our products earn more money than we spend making them. That is why the investors give us money, not to be open source evangelists. Investors gave a lot of money to open source evangelists recently and they lost their shirts. For me to hijack investment money for unprofitable purposes would be fraudulent. Life is not made easier for developers by one browser adopting standards if the other browser doesn't. That's not a standard. It's some kind of position paper, perhaps, but not a standard. And strangely enough, just because a group sets itself up as the arbiter of standards doesn't mean that it is. I don't believe open source software is better software. If I did, then I would be part of the crusade. I believe open source software is primarily by, of and for hobbyists, not for the general public. I believe that you have to spend money to create good software. If Mozilla winds up being good software, that will mostly be because businesses have put money into it in hopes of realizing some value from their investment, not out of the goodness of their hearts. None of the open source projects became anywhere near suitable for general use until businesses began to invest in them. That's the thing. It's fair enough saying the Opera and Konqueror don't have a large enough share to justify developing *for* them, but that's the thing - you don't have to. It is enough that you develop to the w3c standards, and then, by the magic of having standards, you website will work in them. And what's even better, if it doesn't, they'll work really hard to make sure it does. I agree there's no point developing for one standard rather than another if they have equal standing (though in fact, in this case, I'd still say designing to the standard that Microsoft, Netscape, and everyone else agreed upon, rather than the one that Microsoft only did would be a more positive community step, but ignore that, that's just me ;-) ), but thing is it doesn't matter how small the share of other browsers there are - there will always be people in niche situations, and the great thing about standards is that all of these small projects are aiming for the same thing. So don't waste money developing for Opera and Konqueror, develop for standards and you get Mozilla, Netscape 6+, Opera, Konqueror, Galeon, K-Meleon, etc.. Thing is, if standards really get their foot in properly, suddenly we won't have to think 'Are they using IE or Netscape?' because people will be able to use their slim, superfast browsers, or their ultra-stable-at-the-expense-of-some-functionality browser, or their 3D browser, or, eventually, XML metabrowser / information finder. That will be a good world, IMHO... I don't think Microsoft dominating browsers has much of an effect on whether they compete with you, but I do think, certainly in a .NET world, it has an effect on whether they succeed. As for your complaint that open source projects aren't good for users, I presume we don't include, say, Apache in that, which seems to work with no trouble out of the box. I know what you're saying, but I seriously think it all comes down to what we define as 'suitable for general use'. I think Code Red / Sir Cam / Nimda prove that products which have had the most money spent on them aren't 'suitable for general use'. If you simply mean are easy to use, that is DEFINITELY something that can be added later a good deal easier than adding stability, security, and simply intelligent architecture. Considering that Linux has only really been thinking about desktops in the last two / three years, I think they're doing remarkably well, and I certainly think they got their priorities right (in fact inadvertantly, since nobody ever made a good software / users choice). When they catch up with Windows 95, say, in terms of usability it'll be better both in terms of stability, security and architecture as well as usability. Just give them a little time (five years grace isn't really much in these terms...). interesting opinion about open source. Just a question: why runs half of the internet on Apache servers if they aren't successfull? And before you say something about "oh they get so much money from business". They were successful way before companies decided that they should invest money in open source software to show people how cool they are. There are many other applications that were useful long ago before the open source hype startet. Just that you weren't interested in them doesn't mean they weren't useful. Maybe development goes faster when companies pay people to help in open source projects but even without them these projects would go beyond any commercial competitor simply because they are made for users and if appropriate for developers using them. They simply have to because this is the only way for being successful and making your name known. To be better and not to do just more marketing like commercial projects can do to get their software to users. Other point about marketshare of browsers. How do you know that either Opera or Konqueror have roughly zero marketshare if both browsers gladly support changing their user agent string? And most users of these browsers change them to something IE like cause of many browser sniffing sites that would not recognize anything but IE. So every statistics about browser usage is just worthless if they did not ask every user wich browser they use but just count on logfiles. Never trust any statistics you didn't fake yourself. (or how this maxim would sound in English) Second you weigh costs of cross-platform development against the general benefit of Mozilla challenging IE. To be more exactly you weigh your costs of developing your product against the benefit of the whole other world. Would be interesting to see a "marketshare" or percentage of developers that have such extreme DOM usage that Mozilla's is not enough like you. I don't think that they are so many that they even deserve a voice to say this. I could say too, that for me the world would be better if only Mozilla existed and no IE. Cause then I would never have had so much problems with the latter. I could do most sites nearly without images, cause CSS2 is sufficient for the designs, but this crap of IE does not support them or displays them plain wrong. I would never have any problems with an IE that doesn't want to set a simple cookie without any warning. As a user I would not get error messages about Javascript problems in sites which are not mine so I'm not interested in them. Or enoying Javascript errors in Getright cause IE could not load the advertising banners (having Javascript turned off in IE). But some of our users decided (or even didn't decide anything, cause Microsoft decided for them) to use IE and so we support it. But I'm more happy with every user that decides that he wants to use Mozilla or even Opera or another browser... > What about Beonex? You see too much fluff there? What about AsaZilla ;) seems like a cool project. I am waiting for IMAP bugs to be fixed, so i'd rather not be a couple milestones back. =P maybe 1.0 is the point where derivative distributions are usable on a daily basis. And yes, I am advocating that mozilla be the default browser here at work, so a version that works on many platforms is needed. =) lots of people use linux, windows, OSX and even one BeOS guy still holding on. but I guess he is outta luck, grin. IMAP really is broken. It doesn't check for email in sub fodlers and it just doesn't work really at all with courier IMAP. I am not complaining though. waiting patiently. I myself respect the drive to 1.0 a great deal. But I would not in the least be surprised to see a Mozilla 0.9.10 before that happened. Don't get me wrong; we've got a very solid piece of equipment here. The 1.0 requirements are extremely high, however. What people like Mike Angelo do not understand is exactly what GSSQ states: "As many as needed." I notice we debate how close we are to 1.0 every time a new milestone comes out. Personally, I have a question about *who makes the decision??* It's not exactly something I feel we who read and contribute to MozillaZine can decide. We are close enough to 1.0 where, if people are going to debate it like this, we might as well figure out what the official steps are to declaring a milestone 1.0. "And a Voice from the Heavens cried out, saying '1.0...'" -- the Book of Mozilla Hi, everyone; I've been keeping up with the Mozilla project for a long time, but this is my first post. I downloaded 0.9.4 and was trying it out, and I've noticed that sometimes images don't load. The alt text appears, but the images do not. Sites that I've checked include ESPN (http://espn.go.com/) and Mozillazine itself (where the thumbs up/down build indicators don't load, for example). On ESPN's site, the images that don't load are not consistent; that is, sometimes certain images load while others do not, and upon a reload of the page some of the images that didn't load before do load while some of the ones that did the first time don't. (I'll be handing out maps to that sentence, if anyone needs one. :) ) I'm on a Win98 machine. I didn't know if this was already known, but I figured I'd post it anyway. I'll try to report this to Bugzilla, but I've never done that before, so I wanted to let more experienced people on this board know in case I messed it up. :) Yes, I got same behavior on www.vbrad.com The gif that's supposed to show up on mouse_over didn't show until several tries. Moz worked fine before I have not had any problems with the new build in regards to images. The only issue I have is that I always have to copy a .DLL file for Java to run properly. (I'm on Win2K SP 2) If a number a people keep having an issue with images though, then someone should sumbit it to Bugzilla. With each build it gets better and better. I had the same problem on my linux box. I removed all my old .mozilla directory and reinstalled. This cleared up the problem for me. On my Win2k box I didn't have the missing image problem. Hi, everyone; I've been keeping up with the Mozilla project for a long time, but this is my first post. I downloaded 0.9.4 and was trying it out, and I've noticed that sometimes images don't load. The alt text appears, but the images do not. Sites that I've checked include ESPN (http://espn.go.com/) and Mozillazine itself (where the thumbs up/down build indicators don't load, for example). On ESPN's site, the images that don't load are not consistent; that is, sometimes certain images load while others do not, and upon a reload of the page some of the images that didn't load before do load while some of the ones that did the first time don't. (I'll be handing out maps to that sentence, if anyone needs one. :) ) I'm on a Win98 machine. I didn't know if this was already known, but I figured I'd post it anyway. I'll try to report this to Bugzilla, but I've never done that before, so I wanted to let more experienced people on this board know in case I messed it up. :) why not add a button (like the image button) so that we can turn on/off javascript without opening the preference panel ? http://xulplanet.com/downloads/view.cgi?category=applications&view=prefbar I've used it with 0.9.3 and have just installed it with 0.9.4, works fine. Maybe I should get to work figuring out how to get that on the menubar in the right corner with JS: [x] <br> I was planning on eventually trying something like that myself, but I'm glad someone thought it up before me (saves me the work). Absolutely brilliant! We're going to see a lot of nice stuff like this, including sidebars tabs. I think that stuff like this can be the way to make Mozilla really compete with Explorer. After a while there will be lots of really useful stuff that you can only do with Mozilla. An improved version would allow a user to choose exactly which of all the myriad options should be available on the toolbar. "An improved version would allow a user to choose exactly which of all the myriad options should be available on the toolbar." Actually that version allows for the user to choose from several already and the author is willing to consider adding more. Email him if you have a specific pref that you would like to have added. #63 the bug that led to the pref-toolbar developmentby thomanski Saturday September 15th, 2001 4:01 PM Bug ID 38521: http://bugzilla.mozilla.org/show_bug.cgi?id=38521 There's a huge discussion about what this is supposed to offer and where it belongs (taskbar? sidebar?). Here's a related suggsetion, and apologies if it has already been implemented somewhere. But how about a button right in the toolbar that toggles cookies on and off with one click? I mean the cookie manager is great and all. But being able to click them on to use some e-commerce site and then click them off while surfing without having to use a dialog box would be great. Well, it's actually in the pref-toolbar at http://xulplanet.com/downloads/view.cgi?category=applications&view=prefbar While you may initially only have the choice to switch on/off custom fonts and colors, images and javascript you can add further shortcuts (java, popups, proxies, cookies and XUL cache) by right clicking on that toolbar and selecting each of them. Okay, I've been encountering the same bug/feature/behavior for some time now with 0.9.3, and I just checked that it also occurs in 0.9.4. I'm sure people must know about it (I've sent a bunch of talkback reports), but I've had no luck searching Bugzilla; I'm just not creative/competent enough to come up with the right query. I also posted about this on n.p.m.mac but got no response. That preamble out of the way, here the problem: Mozilla, at least on my mac, seems to have a limit on the number of browser windows you can have open. If you hit that limit and try to open another window, Mozilla will crash without any warning. I've figured out that the limits on my machine seem to be around 10 blank windows (that is, if you hit command-N repeatedly, Mozilla with crash when the 11th window opens) and about 7 non-blank windows (determined by going to a weblog and opening links in new windows via the contextual menu or by opening bookmarks from the bookmark manager). These are the limits I've found while trying to reproduce the crashes - I'm sure different setups and different situation would yield different limits.* Now, I recognize that no browser can open an infinite number of windows at the same time. However, while it would be nice if the limits were a bit higher, the real issue is the way Mozilla just suddenly crashes. It would be nice if there was a warning message that popped up as one approached the limit saying something like "Mozilla is running low on memory/resources/magical window-spawning juice. Please close some windows or restart Mozilla or something." Anyway, does anyone know of any existing bugs in Mozilla that deal with these issues? * (For fun, I decided to see how Navigator 4.08 does with these tests. To my surprise I was able to open 120 blank windows without a warning message or a crash - I could have opened more, but I got bored. As for opening links in new windows, I got to about 35 open windows before I gave up. Is this a native widget vs. XUL issue or what?) I think it's limited by your system's resources. Mozilla is just very resource hungry. Of course, it could be a bug too. Anyone with supercomputers willing to test this? :) Sure, sure. I get that. Getting more memory is on my list of things to do. It isn't so much the limits themselves I have a problem with, as the fact that Mozilla doesn't deal with them particularly gracefully. I seem to remember that Navigator 4.x had warning messages that would pop up telling you that memory was dangerously low or something like that. It seems to me that Mozilla should do something other than just crash without warning. Maybe this is an RFE rather than a bug per se. It must be system resources. I opened 30 Mozilla windown and watched my memory usage go from 115MB to 150MB. The odd thing is when I closed all those windows my memory usage stayes at 150MB. 35MB is a lot of RAM to leak. I guess we need two RFEs -- a system resources running low message and a don't leak memory option :) Those 35MB is probably stuff being cached, not leaked. Such a leak would've long been found, and there are actually some bugs about this issue (sorry, don't know the bug number right now) File a bug. I just tested this (actually testing something else but similar steps). I was able to open 100 windows (about:blank) under os9.1 without any problems. I was testing new window creation time and I was happy to note that the 100th window only took 2 seconds to open, only about 1 second longer than the first window. I experienced no crashes or memory warnings. PowerMac G4 256 MB RAM OS 9.1. --Asa I'm wondering if we ought to have a collection to help Mike Angelo get some therapy - he seems to have some strange ideas about Mozilla 0.9.4: * It's "very buggy" (this claim is purely based on [incorrect] bug counts and not actually on user experience of the browser. In fact, Angelo actual admits later that he hasn't even done some simple testing of 0.9.4 !). It is also later contradicted by saying that "the changes...are behind-the-scenes improvements in performance and stability" (er, doesn't that make Mozilla *less* buggy ?). * Because 0.9.4 was delayed a week (which, to me, is a minute amount of time in such a large project as Mozilla), he accuses Mozilla.org of "poor program management" and making "inappropriate concessions to AOL/Netscape" (what were those concessions ? Making -turbo the default ? Adding in disabling of window.open() ?). * He's been claiming that Mozilla.org has been keeping the identity of who would be using the branch to create a commercial Mozilla-based release "a closely guarded secret". Er, what ?! AOL/Netscape are obviously those people (although clearly, other vendors could step in and use 0.9.4 as a base for a commercial browser as well), but Angelo now claims Mozilla isn't a "true open source project" because of this non-existent "secret" ! * In a show of paranoid delusion, he is convinced that -turbo is the one of the biggest evils in the universe. He's saying "don't get rid of pre-0.9.4 releases, install 0.9.4 in a separate dir, don't enable -turbo". Never mind the fact that the user can disable -turbo in *both* the Installer (where it is ultra-clear - it's got a big dialogue box of its own explaining about it) and the Preferences dialogue [oh, AND the status tray pop-up menu !]. * He complains that there isn't a non-Talkback installer. Er, does this matter at all anyway - you don't notice any diff unless Mozilla crashes, in which case a dialogue box appears, which you do NOT have to send off to Mozilla.org if you don't want to ! "Whether you participate in Mozilla talkback should be your choice, not Mozilla's". That's why you don't have to send the Talkback when there's crash - it's called choice <sigh>. * More paranoia - "we recommend your disconnect your computer from the Internet when installing 0.9.4". If that isn't the biggest load of drivel, I don't know what is ! And what reason does he give for this suggestion ? Answer: none whatsoever (it's not even tied into his crazy Talkback rant). In conclusion, if Angelo was talking about a commercial product like this, the company probably would have threatened him with a lawsuit by now - I believe the term is "libel". Now, I know Mozilla is a free, Open Source project, but I must say I'm a bit surprised guys like Angelo can get away with this rubbish unchallenged (yes, I know about www.mozillaquestquest.com, but it hasn't changed the situation :-( ). Isn't it about time one of the big IT news sites exposed MozillaQuest for what it is (rather than now seemingly linking to it each time a Mozilla comes out) ? the Register guy removed the link to mozillaquest which _actually_ UNCOVERED the story by help of some netscape inc. bigheads sending mails... Slashdot reports 0.9.4 without giving the mozillaquest link... Aren't you still satisfied? Why do you make us support Mangelo just because he is been treated like an asshole?... Don't bash him cluelessly on every damn instance. THAT IS THE THING MADE JWZ, CREATOR OF MOZILLA *HATE* FROM MOZILLA and EVEN PROVIDING IE ICON FOR HIS GRUNTLE PAGE... Forget that Club guy yes? Okay... 0.9.4 released. Netscape -if anymore released- (I bet) won't have Popup window filtering since AOL is the absolute abuser of popup windows nearing crack sites... Thats all OK... As I said above... Don't mess with poor guy, so we can have OBJECTIVE thinking about their FAULTS. Also I sure wonder how many of you mailed that Matrix corp. rumoured to be his boss begging to get him fired. Based on such much hostility, majority expects that it must been done... So, congrats on new release (really!) but consider taking time to make mozilla/netscape 6 at least at 20% share again so web can care of it. Just a question: is Mozilla is trademark of someone? If so, Mike Angelo should not have the legal right to call his web site MozillaQuest and use all this FUD against Mozilla. Because having the name Mozilla in the website make some people think it's official or semi-official. Yup. I too read the latest copy-paste-modify "Mozilla 0.9.x is released. Behind schedule and very buggy." story on MozillaQuest and laughed all the paranoid warnings he had there. While it's obviously a pile of crap, e.g. that he calls for people not to try any later versions of Mozilla than 0.9.3 / Netscape 6.1, is totally paranoid about turbo mode etc., he has some valid points. "He's been claiming that Mozilla.org has been keeping the identity of who would be using the branch to create a commercial Mozilla-based release "a closely guarded secret"." Well, it is. There was some discussion here that started from a statement that "a major vendor will take over the 0.9.4 branch" and nobody ever confirmed who that major vendor is, even though lots of people asked directly. It could be Nokia for example.. Of course, it's just crap that the secrecy would make it less open source or open in general. That's just a commercial company that doesn't want to pre-announce its plans. " Because 0.9.4 was delayed a week (which, to me, is a minute amount of time in such a large project as Mozilla), he accuses Mozilla.org of "poor program management" and making "inappropriate concessions to AOL/Netscape" (what were those concessions ? Making -turbo the default ? Adding in disabling of window.open() ?)." Yeah, those are the things he was talking about - the turbo mode in particular. I thought he was being quite clear about that point... "In a show of paranoid delusion, he is convinced that -turbo is the one of the biggest evils in the universe. " That was kinda funny, wasn't it? :) But he has a point about the hypocricy about it. Everyone was complaining about how Microsoft loads IE at boot time (which is, by the way, only indirectly true - it's not actually truly "pre-loaded") and now that Mozilla does the same thing, it's the greatest feature on earth. Personally, the load time is fast enough even without turbo mode, so I'll probably disable it myself. "I'm a bit surprised guys like Angelo can get away with this rubbish unchallenged" Why does so many people on MozillaZine seem so incredibly keen on shutting up everyone that has anything bad to say about Mozilla? You'd think that in an OPEN source project, people would appreciate such ideas as free speech and freedom of forming an opinion about things. Mike Angelo has some good things to say about Mozilla too. For example he praises the IRC client.. Personally I believe that there's something wrong with him (seriously) because of the strange fixations he has on bug counting and all that, but I don't see him writing anything that is straight out wrong. Yeah, he exaggerates and makes bizarre conclusions from misunderstandings, rumors and speculation ("Netscape denies using Mozilla code!" comes to mind) but he doesn't really make up stuff nor lie about things. Talking about threatening him with a lawsuit or writing to his employer to get him fired is, IMHO, really silly and sad. Just ignore the guy and his site! When Slashdot has used him as a source recently, half of the user comments have been along the lines of "don't use him as a source!", exposing him for what he is. I think everyone has gotten the message already. There's no need to sink to his level. >Personally, the load time is fast enough even without turbo mode, so I'll probably disable it myself. Say what? You *disabled* it? Hypocrisy averted, methinks! "Say what? You *disabled* it? Hypocrisy averted, methinks!" Yes, because like I said, I think Mozilla loads fast enough without it and I rather not have the overhead of having Mozilla loaded when I'm not actually using it. #40 Re: Re: Re: Re: I'd laugh if it wasn't so sad...by SubtleRebel Saturday September 15th, 2001 9:06 AM I think SmileyBen's point was that with Mozilla -turbo can be disabled, but it is virtually impossible to disable the preloading that occurs with IE. Since the primary criticism of IE preloading is that it wastes resources for those who do not use IE, it would only be hypocrisy if Mozilla did the same thing, but since Mozilla makes preloading an OPTION then Mozilla is not being hypocritical by touting this feature for those who want it. #48 Re: Re: Re: Re: Re: I'd laugh if it wasn't so sad.by asa Saturday September 15th, 2001 11:52 AM Not only can it be disabled (Edit|Preferences|Advanced or right click on the tray icon) but it can be averted alltogether by unchecking a checkbox in the installer or using a zip build that doesn't even have an installer. We have introduced a fairly big chunk of code into the codebase to handle this turbo functionality. In the very near future it will be a great feature for those who chose to enable it. We decided that it would be good to get some crash data about this feature. The best way to determine if a new feature is causing instability (and a feature like this is a prime candidate for causing instability) is to get talkback data and compare that to talkback crash data without the feature being used. This is what Milestones are for. We have not filled in the turbo checkbox in the installer on the trunk builds and there is no loss of choice here even in the 0.9.4 win32 Milestone builds. Users are free to enable and disable a handful of features in our testing binaries. If they aren't happy with the choices we provide in these testing binaries then they are free to pull the source and build whatever they want from that. --Asa maybe instead of bashing him, we could try to help him make Mozillaquest a real news site worth reading it. If he posts something wrong just try to explain it to him in a friendly way. And if there's something new he might post himself (just like a new milestone has arrived) send him some infos, even about things that are just normal to most readers but may lead to misunderstandings. If he's really interested in having a good newssite he will accept this form of help as log as it is not agressive but just friendly. If he just wants to bash Mozilla he wont but I think it would be worth a try :) Lots of people (myself included. he used to send me articles/editorials for preview/input/editing) did this for about the first year of the site. In those days his articles/editorials were a lot more interesting and well informed (not that the writing was any better). Then he discovered that by writing negative stories he could get picked up by larger and more respected sites (expecially slashdot) and all his articles started emphasizing the negative until they became pure bashing editorials. He obviously has an axe to grind and his beautifully designed homepage is the place he's doing it. It is likely that any offers of help would be a waste of time. --Asa This Mike Angelo seems to have something personal against the Mozilla project. Or maybe Microsoft pays his bills... ;-) Is there anything to be done about it? Does anyone here know Angelo? I'm curious what's wrong with him. I don't think MozillaQuest is _good_ by any means, and in the last month it's become increasingly crankier, perhaps because of the increasing tide of personal attacks against Angelo here and on /. -- which according to a message downthread, have included email to his boss attempting to get him fired from his unrelated day job. (Nice work, guys. Jeez.) In any case, while not exactly defending MozillaQuest, again I am finding that most of the response to him here seems even more mistaken than he is. >> * It's "very buggy" (this claim is purely based on [incorrect] bug counts and not actually on user experience of the browser. << The bug counts appear to be correct, as always. As for users, significant bug reports started rolling into both MozillaZine and SlashDot as soon as the stories were posted. Response to this build has not been enthusiastic. >> Because 0.9.4 was delayed a week... he accuses Mozilla.org of "poor program management" and making "inappropriate concessions to AOL/Netscape". << He says bugs and those two reasons are his main suspects for reasons for the delay. I'd have to agree on all three fronts. Saying you don't have a release schedule, failing to evaluate incoming bug reports, forcing releases out on an accelerated schedule when their targeted bugs aren't fixed -- these are all issues relating to "bugs, poor program management, and inappropriate concessions to AOL". >> He's been claiming that Mozilla.org has been keeping the identity of who would be using the branch to create a commercial Mozilla-based release "a closely guarded secret". Er, what ?! AOL/Netscape are obviously those people..., but Angelo now claims Mozilla isn't a "true open source project" because of this non-existent "secret" ! << It's a valid issue. Asa has also refused to answer the question, and it's not obvious that AOL is who he was talking about. There is a conflict between having secret arrangements with commercial vendors and being "open." It's one of the more interesting issues facing the open source community today and is coming up on a number of projects. When is open not open? >> In a show of paranoid delusion, he is convinced that -turbo is the one of the biggest evils in the universe. << No, he's not. He's saying that enabling turbo by default was not a good idea, and he gives specific reasons -- it slows down Window launch, it eats resources even when you're not using it, and it keeps buggy code running in the background, which is a risk. We've seen half-a-dozen people say the same things about turbo on MozillaZine without them being accused of "paranoid delusion." >> He complains that there isn't a non-Talkback installer. << That is a legitimate privacy concern. >> More paranoia - "we recommend your disconnect your computer from the Internet when installing 0.9.4". If that isn't the biggest load of drivel, I don't know what is ! And what reason does he give for this suggestion ? Answer: none whatsoever (it's not even tied into his crazy Talkback rant). << Yes, he does give a reason. "The installer version of Mozilla presents some potential privacy breaches, which we shall address shortly in another article. In the meantime, we recommend that you disconnect your computer from the Internet when installing mozilla-win32-0.9.4-installer.exe so that you can adjust your security and privacy preferences before taking Mozilla 0.9.2 [sic] online for the first time." While he says he will go into more detail on the privacy issues later, he does state them as the specific reason -- and given the removal of non-talkback builds, this seems like a legitimate concern. Go ahead and call me an insane idiot now, but it continues to appear to me that Angelo is far more correct than most of his critics. He is being flamed without apparent regard for the facts of the case, and pranks like trying to get him fired from his day job do his critics no credit. ">> He complains that there isn't a non-Talkback installer. << That is a legitimate privacy concern." Whether or not he thinks it is a privacy concern, he has no reason to complain. After all, why should Mozilla provide a non-Talkback installer? Especially for this build where Talkback data is even more important than usual. If Angelo or anyone else wants to test out Mozilla without providing Talkback feedback then that option is always there, but Mozilla personell should not be expected to spend time building it. >> He's been claiming that Mozilla.org has been keeping the identity of who would be using the branch to create a commercial Mozilla-based release "a closely guarded secret". Er, what ?! AOL/Netscape are obviously those people..., but Angelo now claims Mozilla isn't a "true open source project" because of this non-existent "secret" ! << "It's a valid issue. Asa has also refused to answer the question, and it's not obvious that AOL is who he was talking about. There is a conflict between having secret arrangements with commercial vendors and being "open." It's one of the more interesting issues facing the open source community today and is coming up on a number of projects. When is open not open?" Huh? Regardless of how many commercial vendors are utilizing Mozilla code, regardless of whether or not everyone knows who they are, the project is still open source. Anyone who wants to can download the code. Anyone who wants to can potentially contribute code. Anyone who wants to can create a project based on Mozilla code. When these things are true, then it is open source. If a specific company wants the project to go a certain direction and provides resources to help it move their way, then that does not stop the project from being open source. It may help their most hated bugs get squashed sooner or it may help their favorite features to be implemented faster than others, but anyone making similar contributions can have similar influence. That is because it is open source. I do not see how that can be questioned. In fact, this is the very essence of open source. A licence which prevents a company's contribution being viewed the same as an individual's is most certainly (rightly or wrongly) not an open source one. Open source is there to get contributions for as many directions as possible - it is totally absurd to suggest that a company paying for contributions to an open source project is against the spirit, which couldn't be further from the truth. >The bug counts appear to be correct, as always. As for users, significant bug reports started rolling into both MozillaZine and SlashDot as soon as the stories were posted. Response to this build has not been enthusiastic. We always get a few more bug reports from a Milestone than a nightly builds. I think it probably has to do with the fact that we get about 2500 downloads of nightly builds and about 100000 downloads of a Milestone build. I've read every comment at slashdot and here at mozillazine and don't see a single report of a real issue that isn't already in Bugzilla. I evaluate the slashdot and mozillazine comments as overall quite positive. I realize that others will have a different evaluation but I've been reading responses to Milestones since about M6 and I have seen them get more positive with every Milestone (with the possible exception of 0.9 for Mac). >It's a valid issue. Asa has also refused to answer the question, and it's not obvious that AOL is who he was talking about. There is a conflict between having secret arrangements with commercial vendors and being "open." It's one of the more interesting issues facing the open source community today and is coming up on a number of projects. When is open not open? What question have I "refused to answer" and please point me to my statement of refusal. There is no secret arrangement and there is no conflict. If you came to me and said that you wanted to check in a few fixes to the 0.9.5 branch for a commercial StraussZilla release you were doing I'd say "great!" and offer any help I could in improving your release. There's nothing that isn't "open" about you're desire to work on a branch after we're done with it. There's also nothing wrong with my commenting on mozillazine that I heard that you were interested in doing a commercial release based on the Mozilla 0.9.5 branch plus a few fixes. If you didn't want me to mention that you were doing this release I wouldn't but there's nothing secret when the day after we release 0.9.5 there are checkins to that branch from your cvs account. You really don't understand how this works or have much insight into what's going on in this project or you would have seen that for the last 3 months in Bugzilla and the newsgroups Netscape employees have been talking about their next release and the fixes that will be landing on the 0.9.4 branch after Mozilla 0.9.4 is released. Were you watching when Mozilla did an 0.9.2 release and Netscape continued to develop on that branch (mostly pulling high value low risk fixes from the trunk to the branch) for a while longer and released a product called Netscape6.1? There's no secret arrangement here. If you think there is then you're paranoid and out of touch or incapable of deciphering pretty blatant cues about what's happening with checkins and bugfixes. >No, he's not. He's saying that enabling turbo by default was not a good idea, and he gives specific reasons -- it slows down Window launch, it eats resources even when you're not using it, and it keeps buggy code running in the background, which is a risk. We've seen half-a-dozen people say the same things about turbo on MozillaZine without them being accused of "paranoid delusion." That's not what he's focused on. We've had turbo for some time and he never complained. He's complaining that we're taking away user choice. His arguement is broken from the start. There is just as much choice in unchecking a box as there is in checking it and there's an entire build made available that doesn't even have an installer if you're paranoid about installers and accidentally enabling turbo which can be turned off with great ease at any time. >> He complains that there isn't a non-Talkback installer. << >That is a legitimate privacy concern. It's not a legitimate privacy concern. If you're concerned about privacy then don't use our testing builds. We make no guarantees that it won't melt your harddrive and send all your tax evasion schemes to the IRS. I'm pretty sure that we even warn you on our download page that these things are likely to happen. THESE BUILDS ARE PROVIDED FOR TESTING AND DEVELOPMENT PURPOSES ONLY! Try Beonex (a custom Mozilla based browser distribution with a focus on security, privacy and support) or build it yourself without talkback or turbo or SSL or any other option you're interested in. If this isn't enough and you really want to use a mozilla based browser then hack the code until you're sure that nothing is infringing on your privacy. >> More paranoia - "we recommend your disconnect your computer from the Internet when installing 0.9.4". If that isn't the biggest load of drivel, I don't know what is ! And what reason does he give for this suggestion ? Answer: none whatsoever (it's not even tied into his crazy Talkback rant). << >Yes, he does give a reason. "The installer version of Mozilla presents some potential privacy breaches, The installer presents no privacy breaches. THis is silly. There are plain zipped builds that don't have an installer if you're really that paranoid, but mangelo consistently fails to mention that. There are no connections made to the net during the install of the full installer (where it's all downloaded in one blob) and the net installer has to connect to retrieve the packages you choose (connecting to the net is the whole point there). If you don't like talkback in your installerless talkback builds then before you start it (after you've unzipped it) delete the damn thing. But this kind of defeats the purpose of these builds and if you're not interested in helping to test Mozilla then find an end user distribution to use or find a different browser alltogether. To suggest that there are privacy concerns here is plain silly. >Go ahead and call me an insane idiot now, but it continues to appear to me that Angelo is far more correct than most of his critics. He is being flamed without apparent regard for the facts of the case, and pranks like trying to get him fired from his day job do his critics no credit. You have some proof that someone tried to get him fired from his job? A posting from some anonomous person suggesting that some other unnamed person(s) in this discussion (or was it just some person in the mozilla community?) tried to get mangelo fired is suffiecient evidence for you to repeat it multiple times? You're just repeating rumors. There's no real value in that. I won't call you any names. It does appear that you're willing to make a lot of leaps from speculation and partial information that most people aren't. Mangelo has consistently misrepresented issues by giving half the story like his last turbo rant where he neglected to tell readers that there was a non-installer build available in every Milestone release and 2 non-installer nightly builds and that turbo was still optional in the installer. Much of his poor reporting is because he simply doesn't familiarize himself with the actual mozilla builds before writing about them. He reads a post to the newsgroup or here and writes an entire article (a paragraph or two pasted on top of cuts from his last 4 or 5 Mozilla stories) with little or no investigation. That's fine if you call it editorial but it's not reporting and it's definitely not respectable journalism. --Asa > What question have I "refused to answer" and please point me to my > statement of refusal. I think it was more an "implicit" refusal than anything else. You had mentioned that there as a major commercial vendor who was going to base a browser release on 0.9.4 and then didn't answer anybody's questions about the statement. > If you didn't want me to mention that you were doing this release I > wouldn't but there's nothing secret when the day after we release > 0.9.5 there are checkins to that branch from your cvs account. So, to paraphrase (infer), there was a major vendor who approached you to say that they would base their browser on 0.9.4 but that they wished you to keep it secret. Which is why you've never mentioned their name. (That's fair - but it would have cleared up a lot of confusion if you'd simply made the statement that you weren't allowed to comment any further and/or that you'd been told to keep it secret. Or, it would have made even *more* sense if you'd never said anything at all.) Further, if we all keep an eagle eye on checkins to the 0.9.4 branch to CVS we might be able to figure out who it was ourselves. Jason. #82 Re: Re: Re: Re: I'd laugh if it wasn't so sad...by JBassford Sunday September 16th, 2001 8:04 AM > if we all keep an eagle eye on checkins to the 0.9.4 branch Make that the "0.9.5" branch. Jason. Actually you got it right the first time. mozilla.org has finished work on the 0.9.4 branch and is working on the tip of the trunk (working toward a 0.9.5 branch and 0.9.5 release). Some Netscape developers continue to check in to the branch from which Mozilla 0.9.4 was released (you can watch this branch progress by querying bonsai for checkins to the MOZILLA_0_9_4_BRANCH). The 0.9.5 branch hasn't happened yet and won't until October 5th at the earliest (see http://www.mozilla.org/roadmap.html ). #85 Re: Re: Re: Re: I'd laugh if it wasn't so sad...by AlexBishop Sunday September 16th, 2001 11:21 AM Here's a quote from the Mozilla Roadmap: "Drivers propose these changes to the last roadmap's plan in order to use 'more obvious' milestone numbers, to pick up the pace slightly (5 weeks instead of 6), and to align better with several known commercial contributors' tentative release plans. If you are planning a Mozilla-based product release whose schedule does not jibe well with the above milestones, we welcome your feedback (we will keep confidential information to ourselves, and will sign NDAs if necessary)." http://www.mozilla.org/roadmap.html So, the reason Asa didn't reveal that the 'major vendor' is Netscape was probably because he's under an NDA. Saying that a commercial release is going to be made from 0.9.4 without saying by whom is still useful as it shows that there's commercial interest in the current builds. A vendor could just download the 0.9.4 source code, work on it a bit and then make a commercial release from it without ever informing The Mozilla Organization. They could do it entirely in secret. If they choose to tell The Mozilla Organization, it's their prerogative to ask that it be kept secret. Even so, it wasn't too difficult to work out that Netscape are planning a 0.9.4-based release. You don't even have to watch the checkins: just the amount of Bugzilla activity by Netscape engineers and the filing of meta bugs to track 'must fix' issues for eMojo should make it easy enough to work out. Personally, I'm interested in what's going to be in the next (post-eMojo) release. So far I've figured out it's probably codenamed MachV and a quick search on mozilla.org brought up a load of specs for features that Netscape presumably want in their next release. Alex Yeah.. In short, it's no big secret that Netscape is going to use 0.9.4 for a Netscape-branded release. However Asa posts here as a member of the Mozilla organization, but since he is also a paid employee of Netscape he probably feels that it is inappropriate to comment on Netscape's plans and/or is obligated by contract not to comment, as Alex said. At least that's my understanding of what's going on. So all this conspiratorial talk about secrets is silly. And even if there was some kind of secret I would echo those who say that this wouldn't corrupt Mozilla as an open-source project. Since when does "open-source" mean that everyone who uses the code has to share all their plans with the world? Where in the MPL does it say that? >> Since when does "open-source" mean that everyone who uses the code has to share all their plans with the world? Where in the MPL does it say that? << The open source and free software movements are not defined by their licenses, but by various position papers, mostly by RMS and ESR, as well as by consensus of the community. One of the most persistent self-definitions in the community is "free as in free speech, not as in free beer." A broad appeal to social freedoms and the open society (a la Karl Popper) is often part of this trend. There is a plain conflict between this "open society" definition and secret arrangements with closed entities. How serious this conflict is, and whether in fact the "open society" definition is the best one could hope for, are reasonable questions, and not ones I intend to try to solve here. My point is that it is counterproductive to jeer at someone for simply raising these questions. ---- The open source and free software movements are not defined by their licenses, but by various position papers, mostly by RMS and ESR, as well as by consensus of the community. One of the most persistent self-definitions in the community is "free as in free speech, not as in free beer." A broad appeal to social freedoms and the open society (a la Karl Popper) is often part of this trend. There is a plain conflict between this "open society" definition and secret arrangements with closed entities. --- Freedom of speech is also the freedom *not* to speak. You'll find that many advocates of open source and free speech are also advocates for privacy rights (see: the EFF). They go hand in hand (most of the time--they do conflict on occasion). #134 Re: Re: Re: Re: I'd laugh if it wasn't so sad...by JBassford Tuesday September 18th, 2001 4:47 AM > You'll find that many advocates of open source and free speech are also advocates for privacy rights I think that, for the most part, these people are advocates for *personal* privacy rights - not corporate privacy rights. People who argue for personal privacy and liberty also, normally, argue for disclosure of corporate and government activity. Jason. <quoth strauss>The open source and free software movements are not defined by their licenses, but by various position papers, mostly by RMS and ESR, as well as by consensus of the community. One of the most persistent self-definitions in the community is "free as in free speech, not as in free beer."</quoth> Actually, that's not really true. Certainly the free software movement is about freedom, but the whole creation of the open source movement was that it isn't about freedom, or anything of the sort, simply pragmatics - that distributed, wide development produces better software. Open source advocates would certainly not have ay problem with companies being involved and not stating every last decision they make. Free software advocates won't mind so long as the freedoms are respected. But let me say this again: the sort of thing we're discussing - Netscape paying people to contribute to the community things they want is *absolutely* in the spirit of open source, embracing all contributions, so long as they give back while they take. >>> Because 0.9.4 was delayed a week... he accuses Mozilla.org of "poor program management" and making "inappropriate concessions to AOL/Netscape". <<< >> He says bugs and those two reasons are his main suspects for reasons for the delay. I'd have to agree on all three fronts. Saying you don't have a release schedule, failing to evaluate incoming bug reports, forcing releases out on an accelerated schedule when their targeted bugs aren't fixed -- these are all issues relating to "bugs, poor program management, and inappropriate concessions to AOL". << There *is* a release schedule: every 5 weeks mozilla.org stabilizes the tree as best they can and calls it a milestone. mozilla.org can't control how much work developers do for each release, nor can mozilla.org keep developers from overcommiting to a release in the bug system. And other than the in-progress Mozilla 1.0 definition, mozilla.org doesn't define what goes into each milestone. They stick a stake in the ground so people can see where we are, and they hold the release for a week give or take while the most egregious of the rough spots are sanded down. That's all a milestone release is. >> He complains that there isn't a non-Talkback installer. << > That is a legitimate privacy concern. < Talkback is an *option* in all the installers. It is on by default because the reason mozilla.org provides binaries is to get feedback, but it can turned off just like any other part of the install users decide they don't need or want. If a user happens to install it accidentally they would find that it announces itself the first time it is activated, giving the user an option to turn it off, and thereafter it asks before sending anything each and every time it is activated. It can also be turned off completely at any time. There are no legitimate privacy concerns with the Mozilla use of Talkback. Some people may wish not to use it, but it's all completely above board, optional, and opt-in when sending data. When I mentioned the lack of a release schedule, I meant for Mozilla 1.0. Yes, there is a rough schedule for build cycles, but none for the overall product. After some more poking at the installer, I see that there is a button to turn off TalkBack, although it is buried fairly deep on the Mac installer. Still, it's there, and this is no more obscure than having a separate installer for non-TalkBack builds, so I agree that the concern expressed on MozillaQuest about lack of a TalkBack-less installer is not a substantial criticism, and is even misleading. As he has not specified what other privacy concerns he has with the installer, I am not able to express an informed opinion on whether they are valid or not. My initial point was in response to the statement made by his critic that he hadn't given any reason for recommending disconnection during the install process. In fact he did give a reason, though he did not do much to substantiate it. Hi, forget about him. He's just replacing the version-ID in his text when a new release is available. The sentence "Mozilla 0.X.X is the buggiest Mozilla Milestone ever." and his "Bug-counts" are included in every "review" made by him. Mike Angelo is not real..... he's a script. Cheers, Alex The release notes mentioned that the speedups for tubo mode had been reduced, apparently so it can work for different profiles. Unfortunately, it's true. For me, 0.9.4 takes six to seven seconds to start up, wherease 0.9.3 took about a second. Aaargh! :-( On the bright side, 0.9.4 is very stable and responsive, which 0.9.3 was definately lacking in. I suppose responsiveness is more important than startup time, but I wish it didn't have to be a tradeoff... There are already patches being worked on and reviewed that return much of the performance inprovements. I still see a major improvement on first launch after a reboot or a logout. On my machine without turbo Mozilla takes 15 seconds on first launch after reboot and with turbo it takes about 2 seconds. All subsequent launches without turbo are about 5 seconds and with turbo about 4 seconds. The old turbo was about 1 second on all subsequent launches and I think we'll see much of that performance return on the trunk very soon (next week or so). Thanks for testing turbo, and be sure to report any crashes you experience. We're trying very hard to root out any instability that it might cause. --Asa Toy factory does not appear right in 0.9.4. Is this a know problem? The scroll bar appears messed up but does still work. If thats the only problem your seeing then I'll send you an updated jar that will fix it I downloaded 0.9.4 tar.gz on linux, but I can't get Java to work. It did work in 0.9.3, though. Creating links to the NS6.1 installation's plugins stuff doesn't seem to work anymore. I also tried to manually install the Java plugin from inside a running mozilla. It reported a successful install, but the results are the same: On startup, I first get this message: LoadPlugin: failed to initialize shared library /mozilla/0.9.4/plugins/java2 [/mozilla/0.9.4/plugins/java2: cannot read file data: Is a directory] Later on, the browser crashes when trying to load a java applet: # # An unexpected exception has been detected in native code outside the VM.# Program counter=0x483d99a1 # # Problematic Thread: prio=1 tid=0x80d02b0 nid=0x4bb9 runnable # INTERNAL ERROR on Browser End: Pipe closed during read? State may be corrupt System error?:: Success Has anyone else encountered this problem? Is this a known bug? What's the bug number? First of all, I would like to say that I am fairly impressed with 0.9.4 Recently I had to go from a 1.4Ghz with 512 of DDR ram to a 300 Mhz with 128Mb of RAM, and I couldn't believe how much Mozilla seemed to crawl (not surprising though, heh). Still, the difference between last friday's nightly build, and 0.94 is very good! Now then, did anyone check the mail client before 0.9.4 went out? I just tried composing a mail today and the thing is going totally bezerk on me. Some text wont display until I highlight it, the cursor is randomly jumping around the text box. Quoting inserts blank quoted lines for no reason... I just find it hard to believe that no one checked that stuff before 0.9.4 was released. Sometimes I face similar problems when I install over a previous build (i.e. without uninstall it). Another workaround is to try a new profile. Well, the mail client has been a sorry sight for as long as I have been testing Mozilla. I have been using the Mozilla browser as my main browser for a long time now, and it's super, but the mail client leaves me weeping each time I try it. I just can't write mail without being connected. So I have to stick to Outlook Express 6 as a mail and news program. It's really a pity. This reminds me of some comments in earlier days when it looked like there would be never a NS 6 or Mozilla 1.0. This was when people said "forget the mailnews and composer just get the browser ready". Nowadays the browser is really cool and hightly usable and complaints go to the mail client. I think this is just a sign that Mozilla is on it's way, at least the browser :) I agree that Mail is not on par with NS4 (and there aren't new exciting standards here to support, as in browser). It's still slow (although folder and compose window performance has been greatly improved) and it has numerous ui and connection (2 of them recently fixed in nightly builds) bugs. But you can't say it isn't usable. Actually, for a non natively english speaking user is much more usable than NS4, because of it's better support/customisation of codepage settings. What do you mean ? Offline mode in mail has a few bugs but it works. Could you explain ? #138 Re: "I just can't write mail without being connectby bertilow Tuesday September 18th, 2001 11:00 AM Here's the bug that keeps biting me: http://bugzilla.mozilla.org/show_bug.cgi?id=75434 I've been using previous builds off and on. Now I'm ready to use 0.9.4 as my default browser, although the mail client does present a bit of a problem. I'll do a bit of experimenting with a copy of my ~/nsmail directory safely out of reach. Does anyone know of a way to eliminate the small icons appearing in the personal taskbar entries? I'm unable to find an entry in the Prefs dialogue or in prefs.js. I decided to give the latest milestone a try to see where Netscape is headed. I sure hope this isn't it. Delays in loading pages are so long that I thought my computer had crashed! Pages with complex backgrounds (including those on tables) redraw these backgrounds pixel by pixel when a new URL is loaded. I thought the problem was that I had installed Mozilla without the Mail / News and Chatzilla components, so I tried a new install with everything. Same problem. I was using a fresh profile with the latest version of the classic Mac OS, 9.2.1, and I even created a second test profile to see if the first one was just corrupt. Nope -- apparently there's something wrong here. Netscape 6.1 is a fine product. Except for the occasional site that can't seem to decide whether I have Flash installed or not (I always do), it's a pleasure to use. I hope that future versions will not destroy that pleasure. FYI, these are the URLs I tried: http://www.cbs.com/primetime/bigbrother2/ http://www.macsurfer.com/ http://news.cnet.com/ even on windows 2k, big brother loads a ton of external files onto that page if you view the page source: cbs has a slow page thats all.. news.cnet.com is not ever slow.. and latestly nightlies are better than the 0.9.4 milestone on this. the first thing you can do to fix this slow down is in the advanced system preferences up the memory cache to 8192 or 16384.. it helps tremendously with page loading.. I think 8192 should be the memory default not 4096.. it takes more ram to load mozilla itself than that.. some pages have a lot of content: you can check this by download: saving a whole web page with IE's save feature, you open a few windows or downloading in the background and its going to slow mozilla down. So up your memory cache size will help! try today's build 2001091508 with 8192 for the cache size.. damn this is faster than the 094 milestone.. give the nightly a try. I have quick launch enabled, and I would substitute this build for the milestone!!!!!!!! go mozilla! whooo hooo! Here's a hypothetical question: Pretend that you haven't tested Mozilla for awhile, but you know how these things work, and that you have all the buildbar comments logged. Given a choice, would you try out Milestone 0.9.3, or go to a nightly given the "thumb-up plus !" mark? What I'm trying to say is that for some reason, the milestone invariably gets complaints about speed, bugs, and other issues. Isn't the intention of a milestone to eliminate high-priority bugs and improve performance, and shouldn't the nightlies be more adverturous? It is generally not the case that we do anything about performance on the Milestone branch unless it is very low risk (rarely the case). The Milestone almost never has any bug fixes that nightly trunk builds don't have. A Milestone build is hopefully better than the trunk build the day the Milestone branch was cut because we take a few fixes on the branch before we release it. The trunk nightly builds have all those fixes and more but may also have more problems. We start to say no to major checkins near the end of the cycle to make sure the Milestone doesn't get trashed by a risky checkin. Those fixes land on the trunk and not the branch so the trunk might have some really great bug fixes that the branch doesn't have. This particular cycle the trunk had several key performance fixes, a few major layout fixes and some UI cleanup right after the branch was cut. These fixes were deemed too risky for the branch. If you're wondering how you can be most helpful the answer is definitely to test nightly builds. The most important thing we get from the Milestone is the broad range of configurations and the sheer volume of test boxes out there. If you are involved enough in Mozilla to have an active Bugzilla account I recommend you skip Milestone builds alltogether. Leave those to the people that don't want to get anything more than their feet wet. Hope this answers your question. If it doesn't please follow up and I'll try to answer anything more specific. --Asa Mozilla should use todays nightly 2001091508 with quick launch and 8192 mem cache size for the milestone 094: a noticeable feeling of a speed difference over the milestone 094 2001091303 build.. wow, cool. Then look at this http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=SeaMonkeyAll&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=day&mindate=&maxdate=&cvsroot=%2Fcvsroot every day. These are the daily checkins. From there you can decide whether a build worths a try or there aren't significant improvements for your kind of Mozilla usage. Checkins you see there affect next builds. Sometimes we have to wait for many hours before a checkin into the cvs will be built into a nightly. This is the case, particularly for win32 builds. I'm not sure why win32 builds are so late, maybe development and testing tools are easier to use in other plattforms. Asa's build comments here in Mozillazine is another source of build information (warnings about stability issues and regressions are invaluable). In numerous occasions, I see a very very important fix in the trunk, shortly after a milestone release. E.g. a recent fix in pop3 login that didn't make it into the branch afaik. Until now, message download was a frustrating experience for me, just because of this bug. Wouldn't it be silly to wait until the next milestone ? :-) It'd be nice if, on the main page, the date (maybe even the time?) of the last comment posted to a story was given as well as the total number of comments. It would make it easier to keep track of ongoing discussions, I think. Just an idea. #84 cannot access https://internet-banking.dbs.com.sg/by riconeo Sunday September 16th, 2001 11:04 AM i'm wondering how come i cannot access it using mozilla 0.9.4 in Winxp. #86 Re: cannot access https://internet-banking.dbs.comby riconeo Sunday September 16th, 2001 11:33 AM i'm able to get it to work after installing personal security manager. Sorry for posting this bug. sorry abt the bug question. :) keep up the good work! long live mozilla! #104 Re: cannot access https://internet-banking.dbs.comby gssq Monday September 17th, 2001 12:14 AM Ooo, another Singapore user? :) Seems to work for me 2001081403 Win98 #87 since when is windows Cross Platform (XP)? ;-) n/tby eiseli Sunday September 16th, 2001 1:11 PM
#89 Re: since when is windows Cross Platform (XP)?by AlexBishop Sunday September 16th, 2001 1:57 PM Didn't you hear? Windows is going to be cross-platform because Microsoft is going to eliminate all non-Windows platforms. ;-) Alex It's not really on-topic (apart from the fact that 0.9.4 doesn't have this feature), but Netscape releases for a long time (maybe as far back as 1.X ?) have been able to save a page as plain text, HTML source or PostScript. Bug 42441 ( http://bugzilla.mozilla.org/show_bug.cgi?id=42441 ) details this problem [I'd argue that bug 64705 ( http://bugzilla.mozilla.org/show_bug.cgi?id=64705 ) is a duplicate, but it's been re-opened after someone marked it as such for no obvious reason I can see]. The reason I mention this one is two-fold: 1) it hits me most weeks - it's nice to have a plain text version of a Web page for filing away (or incorporating into another text document) and having to resort to "lynx -dump" is annoying, particularly with tables and 2) progress on sorting this one out seems slow considering it's a feature in many previous Netscape versions and has been a bug for some 8 months now (I'd help out if I knew C++, but I'm barely a beginner at it). I'm off to vote for it ( http://bugzilla.mozilla.org/showvotes.cgi?voteon=64705 ), which will, er, increase the votes from 1 to 2, but let's hope someone considers this a bit of a higher priority in the future - I can't be the only person missing this Netscape feature, surely ? When reading about this issue and the coresponding bugs I had just one Question. Mozilla is an open source project controlled by the people of mozilla.org. Netscape is just the largest contributor and a customer of mozilla.org. Mozilla is a plain new project that didn't reach V 1.0 yet. So why is functionality that was in the old Netscape but missing in Mozilla not an enhancment but a missing functionality with even a own keyword for it? If mozilla.org is completely independent of Netscape than Netscape should be treated as every other browser. Just a question ;) > So why is functionality that was in the old Netscape but missing in Mozilla not an enhancment but a missing functionality with even a own keyword for it? What is the distinction you make between an "enhancement" and "missing functionality"? Keywords are for tracking classes of bugs that matter to a significant number of contributors. There are many Mozilla contributors that want Mozilla to achieve parity with Netscape 4.x products. They asked for a mechanism to track those bugs. If you do not find that tracking bugs that relate to feature parity with Netscape 4.x products is useful then you don't have to use the keyword. >If mozilla.org is completely independent of Netscape than Netscape should be treated as every other browser. Just a question ;) mozilla.org exists to provide a central point of contact and community for those interested in using or improving the source code. It is chartered to act as the virtual meeting place for the Mozilla code, to provide technical and architectural direction for the project, to collect changes, help authors synchronize their work, and periodically make new source releases which incorporate the best work of the net as a whole http://www.mozilla.org/mission.html. We http://www.mozilla.org/about.html operate discussion forums (mailing lists, newsgroups, or whatever seems most appropriate), coordinate bug lists, keep track of and publicize works in progress, and generally attempt to provide ``roadmaps'' to the code, and projects based on the code. http://www.mozilla.org/mission.html. The Mozilla community is a collection of contributors working to make Mozilla better. Netscape is a big part of this community. The community is not independent of Netscape, it is inclusive of Netscape. Many issues important to Netscape are also important to the greater community. It is important to Netscape to have make a product that is sufficiently featureful to replace their 4.x product. The features and bug fixes contributed by Netscape to mozilla.org are valued by the community. Do you think that there is something wrong with mozilla.org providing a member of the Mozilla community with a keyword that helps them track their conributions to mozilla.org? --Asa >What is the distinction you make between an "enhancement" and "missing functionality"? The distinction I've seen in bugzilla is that normal enhancments have a lower priority (enhancment is lower that normal, at least in the description of serverity) than these 4xp bugs. >Do you think that there is something wrong with mozilla.org providing a member of the Mozilla community with a keyword that helps them track their conributions to mozilla.org? No, of course not. It was not even the keyword that confused me but more the severity. It was just that I thought putting these bugs on a higher priority would be more....."natural". Cause the severity description tells more about a not working feature than something that's completely missing. And at least some of the 4xps are really enhancements for Mozilla even if they are missing for Netscape. So thanks for your time to clear me up a bit :) I didn't want to upset someone or say something against Netscape, I know without their help there just would not be a Mozilla. Have just been a bit confused about this :) http://bugzilla.mozilla.org/show_bug.cgi?id=9412 #121 Re: See bug 9412 about the severity "enh" problemby SubtleRebel Monday September 17th, 2001 11:25 AM Cool. I think that this change in Bugzilla would help a lot. Seriously. On a lighter side. Just imagine what will happen to MozillaQuest's "defect curve" when a large number bugs get appropriately reclassified as enhancements. #120 Re: Slightly off-topic: Saving a page as plain text...by damian Monday September 17th, 2001 11:15 AM You got my vote. I marked 64705 as a duplicate again. I remember in the earlier builds that there was an FTP client that was similar to the "graphical FTP client" type applications, with twisties for folders, loading the directories that you "twist", and keeping the whole as a windows explorer-like app. It seems to be back to the Netscape 4.78-like FTP client. Where did the other app/version go? I think you can get it back by editing your prefs. I remember that when we had the XUL tree view I used to test the HTML view with something like user_pref("network.ftp.generate_html", true) so perhaps removing that pref or setting it to false will get the XUL tree back. Maybe helpful, maybe not. --Asa Is there a "bug" on this that outlines the decision to drop the XUL version? I did rather prefer it myself. I did a preliminary search on Bugzilla but didn't find any mention of it. Thanks in advance for any info you can provide. There is a GUI already: Preferences/Debug/Networking, there is a checkbox "Enable html directory listing". See bug 77715: http://bugzilla.mozilla.org/show_bug.cgi?id=77715 Unchecking this pref in a 0.9.4 mozilla build brings back the directory viewer for me. Directory viewer has been disabled for ftp by default on Apr 26 in version 3.225 of all.js, see http://bonsai.mozilla.org/cvsview2.cgi?diff_mode=context&whitespace_mode=show&file=all.js&root=/cvsroot&subdir=mozilla/modules/libpref/src/init&command=DIFF_FRAMESET&rev1=3.224&rev2=3.225 #122 Re: GUI exists: Edit Preferences/Debug/Networkingby SubtleRebel Monday September 17th, 2001 11:31 AM Thanks for pointing that out. I would have never guessed that that pref would do that. Afaik this bug : http://bugzilla.mozilla.org/show_bug.cgi?id=92170 isn't fixed in either 0.9.4 or current nightlies. It haunts my browsing experience, since I open multiple windows and have them minimized all but the one I read. The other windows pop up when they start to retrieve data (you have to use really fast internet connection to avoid this) and this is quite annoying. I think that bug has been annoying a lot of people. I know that some people just use one browser window, but many such as myself have this tendancy to open up like 8 at a time - and having them all fight to be on top is extremely annoying. I know the developers for k-meleon took it one step farther and put an option in the mouse menu for "open in background". Bugzilla bug 100000 http://bugzilla.mozilla.org/show_bug.cgi?id=100000 has been filed. Slashdot have a story http://slashdot.org/articles/01/09/17/1231204.shtml about it (and they managed to avoided Slashdotting Bugzilla this time!). So who won the sweepstakes? Alex OK, I found the results of the sweepstakes. They were posted in netscape.public.mozilla.general and in bug 100000 itself: ======================================================== The date of filing of Bug 100,000 http://bugzilla.mozilla.org/show_bug.cgi?id=100000 was: 2001-09-16 23:24 And the winner is... Niko Pavlicek! 2001-09-20 11:11:11 <niko_pavlicek@web.de> Niko Pavlicek Error: 3 days, 11 hours and 47 mins Runners up: 2001-09-20 13:15:37 <andersma@luther.edu> Mark Anderson Error: 3 days, 13 hours and 51 mins 2001-09-21 10:10:10 <jonasj@jonasj.dk> Jonas Jorgensen Error: 4 days, 10 hours and 46 mins Honourable mentions: 2001-09-10 13:37:42 <alexbishopuk@yahoo.com> 2001-09-23 13:07:20 <jesusx@who.net> 2001-09-23 17:08:31 <pgp1@cornell.edu> 2001-09-24 10:10:10 <psolanki@netscape.net> 2001-09-24 20:15:30 <greg@tcp.com> Gerv ======================================================== I came fourth! As you can tell, I'm thrilled at doing so well in a trivial contest. If you've no idea what I'm talking about you'll want to look here: http://groups.google.com/groups?hl=en&selm=3B79E8B7.ECBA88DD%40mozilla.org Alex I recognized a weird behaviour of Mozilla 0.9.4 and the preceeding nightlies on my machine. I don't know how to use bugzilla, so I try posting in this forum: At work (1.5 GHz Pentium IV, 512 MB RAM, 10 MBit LAN, WIN2K), Mozilla 0.9.4 is a joy to use, compared e.g. with Netscape 4.77. At home (300 Mhz K6-2, 256 MB RAM, 56K Modem, WIN98), the gui etc. works at a resonable speed. But if I try to surf the net from home, Mozilla really crawls: Their is only seldom activity in the Dial-up network - everything else seems to be fine. Mozilla seemingly doesn't want to download information. If I try to access mozillazine and enter the URL, Mozilla connects etc. But from this time on, I get the information faster if in the meantime I start Netscape and type in the URL. There doesn't seem to be a problem with the dial-up-network as Netscape and IE work fine. I've already deinstalled Mozilla several times and cleaned the regestry and all the whole system, removed the profiles etc. and reinstalled. Does anyone have an idea? Disconnect and log back on usually fixes this problem right away. 0.9.4 I think has some networking download bug that sucks. This is why I had suggested for the 2001091508 build to replace the milestone because it is much better in that respect. Go to the nightlies directory and backup the parent and look for september 15th and try that one. Is it just me, or does the progress bar move at about half the speed that it did before? Is it just me or is "Only accept images from the requested server" missing. It was never included in the Netscape build but I'm sure it was in mozilla 0.9.3. To bad they disabled it, that was a useful feature. Try the build 2001091508 - its the best ever mozilla build I've ever tested for rendering and performance. Better than the milestone 094 and newer nightlies. - I have not been able to run mozilla since 9.1 when I tried to install a theme unsuccessfully. (this is on Win2k) I get the following core dump message: "The instruction at '0x60392c6e referenced memory at '0x00000000'..." So apparently we are trying to reference a null pointer. When VC++ comes up for debug it says that there was an unhandled exception in gklayout.dll (Access Violation) I have done the feedback agent thing on every release since then but still no fix seems forthcoming. I have tried re-installing and de-installing with extreme predjudice: -- uninstall -- kill the install directory, -- remove any mozilla keys in the registry but nope, no joy so far. Any further ideas to restore mozilla to working order short of an OS re-install? maybe deleting (but backup ;)) your profile helps more. Additional there are the files mozreg.dat and mozver.dat in the Windows directory, at least on 98. I'm just wondering if there's a keyboard shortcut that expands or hides the Sidebar. I know [F9] does it but I often want the bar that lets the user click on it to expand or close it. If it's there, I can either use the shortcut or click on it, depending on what my current task is. Can you help me? Thanks. |