MozillaZine

Weekend Discussion

Friday March 24th, 2000

Time for another weekend discussion. The question this weekend:

What do you feel is an appropriate level of integration between a browser and an Operating System? And at the other end of the spectrum, what are your feelings about Mozilla's cross-platform approach?


#1 My 10 cents

by Kovu <Kovu401@netscape.net>

Friday March 24th, 2000 9:35 PM

Reply to this message

Web browsing has nothing to do with interfacing between the user and the hardware, and that's the SOLE purpose of an OS. End of story. Microsoft once said that if they included a ham sandwich with every copy of Windows 98, the government would have to consider it integrated. I say crap. It slowed down Windows, introduced HUGE security bugs in the OS, made the product bulkier and less stable, and helped drive Netscape to AOL. I say integration is complete crap.

Cross platform is awesome. It will allow every platform, whether major or not, to have a standards-compliant Web browser (assuming anyone from that actually platform ports it -- read: AMIGA). It will certainly allow Netscape to spend FAR more time making the product quality across all platforms since they don't have to spend all their time mired in proprietary code for only the major OSs. I think cross platform rocks, in short.

#19 Re: My 10 cents

by rgelb <nospam@nospam.com>

Saturday March 25th, 2000 1:14 PM

Reply to this message

I disagree. I think integration (in general) is great. I'll give you 2 examples. I had to ditch NS 4.7x because it was crashing too much and couldn't justify putting up with crashes anymore. So I had move to IE/Outlook Express pair. I hate it because there is very little integration between the two. Dragging & dropping, "Send web page" functions between the two are very basic. But at least it doesn't crash. I loved the integration of Communicator. You could email web pages at will, email newsgroup articles at will. Quickly drag links all over the suite. So integration in this case was great. Second example. I just started a new job and they use Lotus Notes 4.6x for email and just about any other intranet function. I hate it. Because it is integrated with nothing. It is a product by itself. Even its windows are not standard (try using window blinds on it). The mouse wheel is not supported. The image attachments that are sent are converted to PCX format. It makes my life very difficult.

So, there is definite value in integration.

#21 Integration WITH THE OS n/t

by Kovu <Kovu401@netscape.net>

Saturday March 25th, 2000 3:52 PM

Reply to this message

n/t

#27 What does n/t mean?

by rgelb <nospam@nospam.com>

Saturday March 25th, 2000 8:53 PM

Reply to this message

Sorry for cluelessness, but what does n/t mean?

#29 no text (in message) n/t

by Kovu <Kovu401@netscape.net>

Saturday March 25th, 2000 10:53 PM

Reply to this message

n/t

#26 Re: My 10 cents

by Russ <ATimeless1@netscape.net>

Saturday March 25th, 2000 8:42 PM

Reply to this message

Agree wholeheartedly with both points. Keep the browser out of the OS ... and go for cross platform. I have been a Netscape user for years and don't intend to give over to IE ... ever. My biggest gripe ... and I'm sure I'm not alone on this one ... with IE is that it slips into your system like a bad date and won't let go. I mean REALLY won't let go ... clogs up the hard drive, sets itself up as the default browser, attaches itself to every HTML file on your drive, etc, etc, etc.

In any case, cross platform is the way to go. Currently, I'm using Netscape 4.72 and Mozilla Build 13 (without crashes of any kind)on a dual-booted Gateway 9100 with Linux and Windoze ... and looking forward to the next release of both.

One problem I DO have, however, is integration between the browser and ISP proprietary software (another topic). There's a lot of that going on out there involving IE and ISP of all persuasions. As with all my other software, I want the freedom to CHOOSE my own browser.

#37 Re: My 10 cents

by macpeep

Sunday March 26th, 2000 9:18 AM

Reply to this message

Web browsing is the process of fetching and viewing files on another computer, over a network. This is conceptually very much the same thing as fetching and viewing local files on your own computer. Unity of interface is a good thing, which is why I don't think it's such a bad idea to have one user interface - one application - for both local and remote documents.

Still, it can be argued that it is not a job to access files, local OR remote.. However I think that integration as such is not a bad thing. It's just a question of WHERE this integration should take place - in the OS or in an application.

#38 Web browsing is browsing THE WEB only.

by Kovu <Kovu401@netscape.net>

Sunday March 26th, 2000 10:39 AM

Reply to this message

Web browsing can actually only be related to the process of browsing file on the World Wide Web, but I see what you mean. Browsing the Internet is what you mean (the Web is a subset of the Internet).

#42 web browsers vs. viewer applications

by macpeep

Sunday March 26th, 2000 11:41 AM

Reply to this message

Web browsing yes of course, is only browsing the web. A web browser application, however, conveniently also allows us to view gopher (not that there are many left) and ftp sites. Why aren't you ready to allow this application to be used to view local files and documents too? Just because the NAME of the application is "web browser"?

I think that the application design in IE - and now Mozilla - is perfect. By allowing the browser canvas to be embedded into other applications, you can plug the net & browsing capabilities into any application if you so desire. You can also use a stand alone browser if you want to.

If you want to use the engine for email, fine.. do so. Telnet? Xterm? IRC? An engine for help? Easy..

This is how IE is "integrated" into the operating system. The browser engine is used where it makes sense - and in a few places where it can be argued that it doesn't make too much sense. I think that in some cases, the integration hasn't been used like it could have. For instance the HTML enhanced folder views in explorer isn't used that much. With a little coding and an embedded web server, you could do just about anything with this. Overriding normal folder views and creating thumbnail views of images, integration to applications like CVS, ZIP, FTP uploads, virus scans.. All using HTML and "CGI". Think about what this could be used for in intranets if applied to shared folders.

Instead of blindly criticizing things because it's Microsoft, I think we should look at what they are doing and learn from it where it's appropriate. IE is a great piece of software.. If it supported standards even better than it does now (and yes, it DOES support standards quite nicely now) it would be almost perfect.

I hope Mozilla will do everything IE does well now, but with the perfect standards support that IE does not currently have. However at this point, I'm not seeing a lot of reason for excitement. The layout engine is much slower than IE even though it's starting to be "fast enough" now so it doesn't matter so much. The user interface (which is still very much work in progress) is absolutely terrible. Slow, ugly, inconsistent.. The mail client is a catastrophy speed wise. While it inherits the same user interface problems that the browser has, the biggest problem is definitely lack of speed.

Modularity of the applications is what we need - in a standard way - so that I can take Mozilla and throw it in where ever I want. I should be allowed to decide just how much integration I want - and with what.

#43 web browsers vs. viewer applications

by macpeep

Sunday March 26th, 2000 11:42 AM

Reply to this message

Web browsing yes of course, is only browsing the web. A web browser application, however, conveniently also allows us to view gopher (not that there are many left) and ftp sites. Why aren't you ready to allow this application to be used to view local files and documents too? Just because the NAME of the application is "web browser"?

I think that the application design in IE - and now Mozilla - is perfect. By allowing the browser canvas to be embedded into other applications, you can plug the net & browsing capabilities into any application if you so desire. You can also use a stand alone browser if you want to.

If you want to use the engine for email, fine.. do so. Telnet? Xterm? IRC? An engine for help? Easy..

This is how IE is "integrated" into the operating system. The browser engine is used where it makes sense - and in a few places where it can be argued that it doesn't make too much sense. I think that in some cases, the integration hasn't been used like it could have. For instance the HTML enhanced folder views in explorer isn't used that much. With a little coding and an embedded web server, you could do just about anything with this. Overriding normal folder views and creating thumbnail views of images, integration to applications like CVS, ZIP, FTP uploads, virus scans.. All using HTML and "CGI". Think about what this could be used for in intranets if applied to shared folders.

Instead of blindly criticizing things because it's Microsoft, I think we should look at what they are doing and learn from it where it's appropriate. IE is a great piece of software.. If it supported standards even better than it does now (and yes, it DOES support standards quite nicely now) it would be almost perfect.

I hope Mozilla will do everything IE does well now, but with the perfect standards support that IE does not currently have. However at this point, I'm not seeing a lot of reason for excitement. The layout engine is much slower than IE even though it's starting to be "fast enough" now so it doesn't matter so much. The user interface (which is still very much work in progress) is absolutely terrible. Slow, ugly, inconsistent.. The mail client is a catastrophy speed wise. While it inherits the same user interface problems that the browser has, the biggest problem is definitely lack of speed.

Modularity of the applications is what we need - in a standard way - so that I can take Mozilla and throw it in where ever I want. I should be allowed to decide just how much integration I want - and with what.

#53 Yes, let's make EVERYTHING BLAND! Up with Vanilla!

by jesusX <jesus_x@mozillanews.org>

Sunday March 26th, 2000 10:51 PM

Reply to this message

Come on. After a while, you forget what app you're in when they all look identical. Viewing a Web page is not the same thing as working in a Word Processor, or editing a photo, etc. With your broad brush, every app is the same.

A Program fetches Data and Views it. Hence, all apps should look the same. But, that doesn't work.

#62 platforms

by macpeep

Monday March 27th, 2000 10:25 AM

Reply to this message

<i>Come on. After a while, you forget what app you're in when they all look identical.</i><br> <br> Yes! That's what I'm hoping for! Why would I have to care about which app I'm in!?

#63 platforms

by macpeep

Monday March 27th, 2000 10:26 AM

Reply to this message

"No HTML"... heh.. oops :)

#65 I should have expected this from a Mac user. (n/t)

by jesusX <jesus_x@mozillanews.org>

Monday March 27th, 2000 12:22 PM

Reply to this message

n/t

#80 Re: I should have expected this from a Mac user. (

by macpeep

Monday March 27th, 2000 9:46 PM

Reply to this message

Heheh, Mac user.. Good one.. Actually I have an Apple IIe at home, a Mac SE, a Toshiba T3200 (PC) laptop, an IBM 390 laptop and a "normal" desktop PC. I use both NT and Linux every day at work and various OS's at home; including, but not limited to, Linux, Win98 and OS/2 Warp 4.

#74 As an Amiga user...

by Kovu <Kovu401@netscape.net>

Monday March 27th, 2000 5:01 PM

Reply to this message

Amiga's last interface makes everything look the same, and it bores me just like Windows does (in fact, Windows borrows quite a bit from the Amiga interface), but we have the excuse that the product stopped growing in 1992 or so when Commodore croaked. I think applications should have exciting, ever-changable UIs, such as Moz skinnable, etc. I certainly don't think they should all look identical.

I do think that programs in app CLASSES, say Word, Excel, Powerpoint being part of an OFFICE class, should have similarly SET UP UIs. As it is, Microsoft puts the Options function here on this program, buries it there on that program, renames it this in that program, etc. Simple stupidity, especially for a company with as much capital as that one, the capital to do it right.

But the point is that apps are apps and OS is OS, and never the twain should meet. Keep your stupid minesweeper program off my OS, it hogs room and I've never even opened it. Keep your stupid IE, keep the rigidly controlled desktop UI and LET ME DO WHAT I WANT TO IT. God I hate Microsoft.

#82 app vs os

by macpeep

Monday March 27th, 2000 9:50 PM

Reply to this message

Yes.. I also think an app is not part of an OS, and never should be. I think one post here said it best: (paraphrased) "we should not integrate with the OS. we should integrate with the user interface. and the question is if that's an improvement." Well, I agree *entirely*, and I think it IS an improvement. IMHO, when I see a file, be it in a file manager or web browser, ftp or whatever, and I want to open it, I just want to (double) click on it and want it to open up. I don't ever want to spend 2 seconds thinking what APP it should be opened in. Possibly, I should be given the option of "open for edit" or "open for view", but "open for view", but still, you get my point I hope.

#2 Cross-platform is great, but...

by luge

Friday March 24th, 2000 9:51 PM

Reply to this message

Two big things: 1) Mozilla must avoid the temptation to replace the OS. It's great that it does everything, and it does it all cross-platform, but I see a huge emphasis on things that aren't browsing, and that is what Mozilla needs to do first and best. The ability to turn Mozilla into a Zope frontend (for example) or to have it be the interface for all things net (IRC? Come on!) is cool, in an abstract sense, but it doesn't help me browse the web- which is what I want a browser for. 2) More importantly- cross-platform-ability shouldn't interfere with usability. I.e., if I were a mac user, I'd be pissed that my cool double scrolling arrows were available on IE but not on Netscape 6. I'm a Linux user, and it pisses me off that every other app on my system does copy and paste the exact same way, and Mozilla refuses to do that. I mean, in the abstract, it's nice that c&p works the same on all platforms- but in the concrete, it pisses me off <i>every single day</i> that I have to use Alt+C instead of just highlighting. If people are used to their OS, and Mozilla sacrifices that comfort in the name of the cross-platform behavior conformance, then people are going to stick to IE, because it works for them in the way that they expect. Just my two cents...

#31 Re: Cross-platform is great, but...

by davidecsdcom <david@ecsd.com>

Saturday March 25th, 2000 11:08 PM

Reply to this message

FWIW, you *aren't* forced to use the ALT+C et al.

I found out the other day that mozilla lets you use CTRL+INS, SHIFT+INS, etc. (Which is what I've used for insert, cut, copy, paste since my OS/2 days.)

#45 I can't select and paste

by luge

Sunday March 26th, 2000 12:54 PM

Reply to this message

I can't select and then paste with the middle button, which is what I can do in every normal X application. It's a serious irritant for me and will be for others, I assure you.

#54 Keyboard shortcuts.

by ppanon

Monday March 27th, 2000 12:01 AM

Reply to this message

Well, in general, there are a whole bunch of keyboard shortcuts in NS 4.x which are not in Mozilla. I'm not sure if it's possible to register the keyboard shortcuts in the XUL GUI definition. If so, then you might be able to just create a new chrome that gives you the mappings you want. I sure hope so because I really miss using n, p , and other keys in Messenger/News Reader.

#68 Actually....

by FrodoB

Monday March 27th, 2000 1:56 PM

Reply to this message

n and t (mark current thread read, go to next message) are now supported in Mailnews. :)

#89 Re: Cross-platform is great, but...

by cplarosa

Tuesday March 28th, 2000 1:01 PM

Reply to this message

I agree with you 100% on both counts. 1) Basic browsing efficiency (which includes application startup time) should not be sacrificed for any additional features; 2) A cross platform application MUST NOT ignore the standards and conventions of any host platform. It should conform to the control styles and conventions of the host platform. Menus, controls, scroll bars, shortcut keys, mouse buttons, etc. must work the way users expect for the platform they are using. This is very important to establishing user acceptance.

#3 Mozilla is a toolkit

by Ben_Goodger

Friday March 24th, 2000 10:46 PM

Reply to this message

...with which people are free to make whatever they wish. File managers, web browsers, mp3 players, spreadsheets.. whatever you wish.

IE is not the OS in windows - it is part of the file management front end. Creating a unified means of dealing with data is IMO a good thing. IE in Win98/2000 does many things right, however it does not go all the way. What if you could have folders on your HD filled with mail messages, saved newsgroup postings and bookmarks, as easily as Word documents and mp3s. What if this was all displayable through a single unified interface that adapted the information it gave to you based on file type? What if you could access and manage remote documents as easily as local ones? What if you could manage files easily across platform boundaries?

These questions are the cornerstone of Aurora - some of this functionality was in place in Mozilla Classic. I think this is the direction we need to be going now. I don't think we need to usurp Explorer or Finder to do it, either. We just create a better system that people want to use.

#4 Skinnable keyboards

by wwrafter

Friday March 24th, 2000 10:47 PM

Reply to this message

What level of integration between an OS and a Browser? About as much as between a word processor or a spreadsheet and the OS. The OS should have APIs to ease the development of a GUI, and the GUI should have APIs to ease the development of applications. The applications should (almost) never talk to the OS directly. Possible Exception: Security validations. As far as cross platform functionality (generically, not specifically Moz), if we were still back in the pre gui days, where every application had it's own keystroke sequence for performing commands, I would say that using the same keystrokes with in an application across all platforms is a good idea. If we all had NCs on our desks, and we only ran one application, I'd say using a consistent keystroke approach would be correct. This is not our reality though. There is no one who can successfully argue that the learning curve on a new application is not greatly reduced when the application uses the platform standard interfaces. In the interest of getting a product out the door, I'd say let's keep going with the cross platform UI. But be prepared. One of the first things that will have to change (based on the feedback you're going to get) is the lack of platform specific compatibility. Take heart however. With the component architecture of the GUI, it shouldn't be too hard to extend XUL to include the keyboard shortcuts and perhaps the mouse-based interaction as well. We might even be able to come up with "skinnable" keyboard layouts (does that sound like an Atari keyboard?).

#8 Shortcuts are skinnable, at least a bit

by cheezycrust <cheezycrust@atmosphere.be>

Saturday March 25th, 2000 12:47 AM

Reply to this message

I'm in the L10n group, so I already saw some (if not all) raw .dtd files, too handle them, so I can tell you that some shortcuts are skinnable. For example, go to chrome/navigator/locale/en-US/navigator.dtd and look at the line that starts with "<!ENTITY openCmd.commandkey". This is where the CTRL-O "Open Web Location" command is defined. If you change the "o" to "%", you can press CTRL-% and open a web location. Thus, some limited keyboard skinnability support is enabled.

#5 not so bad

by tkc

Saturday March 25th, 2000 12:29 AM

Reply to this message

I actually like having IE integrated with Win98...it makes 'exploring' my computer easier. It's nice being 'auto-notified' when there are patches, etc. available.

I just wish I could choose *which* broswer to integrate into Windows. ;)

As for cross-platform: bring it on baby!

#7 Autoupdate Notifier is NOT in the OS!

by jesusX <jesus_x@mozillanews.org>

Saturday March 25th, 2000 12:35 AM

Reply to this message

It's a separate program. I have clipped most of the links between IE and W98, nut left just the basics of IE so that it can still function like a separate browser, and I can Autoupdate just the same as I could out of the box. It takes a bit longer to fire up IE, but now, when the browser crashes, I don't have to reboot.

#79 Re: not so bad

by shk41

Monday March 27th, 2000 9:35 PM

Reply to this message

Amen! I love IE because I can open My Computer so easily! I think that it would be completely stupid to release the product using separate copy&paste, keyboard sequences, and so on, than the OS it is on, because no one would feel comfortable using the browser! I don;t care too much if the UI is changed (just as long as it is for the better which I am yet to see), but the functionality is essential. Face it! People are used to things like Windows and they won't feel comfortable with something completely different. Personally, I have not seen any benefit in using Mozilla. It is slow, the address bar SUCKS, copy&paste doesn't work, text boxes are messed up, and the UI is pretty ugly to use. Plus, the hand symbol is so stupid! The IE one is way better! And sometimes when you move the mouse over a submit button after a large text box, you see an image of the text box breifly about 100 pixels to the right.

PS: Mozilla is getting all worked up on this "standards-compliance thing, but half the webpages I view with it don't function right! They should support ALL the popular code features of IE, Netscape, etc.

#6 Clearly Defined Boundaries.

by jesusX <jesus_x@mozillanews.org>

Saturday March 25th, 2000 12:32 AM

Reply to this message

History shows us, that in the real world, clearly defined boundaries are the key. Computers are not just machines anymore. They truly are extensions of the human mind. Just like every other machine, they are designed to serve people, and thus, should model the users it is made for.

Countries have clearly defined borders, and yet many interact very smoothly, while still being different and independent. States have clearly defined borders the same way. Even people have clearly defined borders. The OS and Browser model is no different.

I feel that a browser and OS should interact well, but NOT integrate. Too many problems are created when you try this. Security, performance, flexibility, etc. Plus, when you tie a browser to one OS, then you wind up losing strict adherance to standards, plus opening up the OS to the network at large. IE has shown this over and over. IE does not offer great standard compliance, but it's horribly insecure, and in some areas, broken. Flat out. While Communicator has even worse standards adherance at times, there is rarely a security compromise, due to it's clearly defined borders.

Mozilla has the most clearly defined borders, and sticks to those borders VERY well. Standards compliance is second to none, speed is fantastic, cross-platform suport is also second to none. But it could work a little better in the interoperability department. Since some code HAS to be platform specific, there should be a little better support for each platforms' APIs. Cut and paste for example, is a pain in the ass. Like the Linux guy said above, I want my normal shortcut. I want to just right click and then COPY. Also, I'd love to be able to browse my local computer like I can in Communicator, especially when I'm testing web pages locally.

I don't THINK it would take much to add these features, because these would have to go into the platform specific code of each port, so it SHOULDN'T break any XP compatibility. But over all, Mozilla is DEFINITELY on the right track. Total XP compatibility, standards out the wazoo, almost pixel perfect redering across multiple platforms, open and standards-based extensibility of the browser (which is ALSO XP by it's very nature!), it's a web designer's dream, and a web users answer.

Yes, it has some not-so-shiny aspects, but for all reality, it IS a version 1 product. It will grow and mature like everything else, but Mozilla has a leg up on the competition already: clearly defined borders.

#9 Basic Integration needed badly

by btbernie01

Saturday March 25th, 2000 1:34 AM

Reply to this message

I agree with jesusX re "Clearly Defined Boundaries". The OS has to provide services that allow the applications to interoperate as seamlessly as possible. Unfortunately, DOS, Win 95, Win NT and Win2000 try to dominate and dictate how to integrate to an extent that basic integration in a cross-platform way appears to be difficult to achieve sometimes. My question: Is it possible for a skilled end-user to configure the right mouse-click context menu to run an operating system command with the URL as parameter ?. In Communicator and Internet Explorer, there is not even the most basic option "Copy URL to clipboard" when I right-click on a page. What has been done in this area recently? I am curious because as a JavaScript developer, I like to create browser based semi-integrated development environments.

#13 Kind of: "Copy Link Location"

by Kovu <Kovu401@netscape.net>

Saturday March 25th, 2000 9:51 AM

Reply to this message

If you're referring to Communicator 4.x, there is kind of a "Copy URL to clipboard" command, it's called "Copy Link Location" But these are for right-clicking LINKS on the page you're on, not the page you're on. You can choose "Add Bookmark" or "Create Shortcut" but those aren't the same thing.

#10 Cross-Platform Uniformity is a godsend IMHO :P

by Hendy99 <gbhendy1@bigpond.net.au>

Saturday March 25th, 2000 1:34 AM

Reply to this message

I like being able to fire up Moz on my platform of choice (Win32, Linux or Mac, which I use on a regular basis) and know that no matter what i'm using, moz will be virtually the same =)

Just downloaded a Linux build for my own machine and it's flying using XFree86 4.0 :) I had trouble even getting it to start on 3.3.5 (Turbolinux 6, Celeron 400, TNT2) - Ideas? =)

#41 Re: Cross-Platform Uniformity is a godsend IMHO :P

by gregstoll

Sunday March 26th, 2000 11:18 AM

Reply to this message

The latest daily build seems to be broken (it's about a meg smaller than it usually is). Try getting an older one...

#11 Well, *I* think...

by Waldo

Saturday March 25th, 2000 2:28 AM

Reply to this message

Pete Collins has said it a few times in the newsgroups, and I agree...

Mozilla represents MUCH more than just a browser, and I don't think a lot of people yet "get" the potential.

If Linux is an cross-platform operating system, Mozilla is a cross-platform Operating Environment...for the Internet.

Something like that anyway. It's going to be really cool seeing what happens in the next year or so. I really hope it doesn't bomb because people don't like the widgets or something stupid like that.

W

#23 Umm...

by luge

Saturday March 25th, 2000 4:50 PM

Reply to this message

I don't need an "Operating Environment." I have one of those, and I'm quite happy with it. In fact, lots of people have "Operating Environments," and most of them are either a) quite happy with their "environments" or b) too lazy to learn a new "environment." <BR>What I need (and what Mozilla needs to provide) is a web browser, because my current one stinks. If Mozilla insists on forcing people to learn new things (or throw away perfectly good old ones) just so that they can browse, it is going to fail and fail miserably, no matter how good a piece of software it is. Worse, if it throws away time and effort on becoming an "environment," it's not only going to fail commercially- it's going to be failed software, because it'll be a cool "environment" but a bad browser-unstable and unsupported. <BR>Look, I understand the need to be revolutionary. But if this organization is going to be so fixated on revolution that it tries to force people to use software they don't want, guess what? They aren't going to use it. And that would be sad... <BR>luge

#33 It's a browser too, don't worry.

by Waldo

Sunday March 26th, 2000 3:33 AM

Reply to this message

Hmm. I think the title says it all. W

#12 It all depends on who programs it :)

by WillyWonka

Saturday March 25th, 2000 8:39 AM

Reply to this message

I would like to see more programs use the mozilla code to interact with the operating system.

I wouldn't mind having a cross platform shell which uses mozilla, but if I want to use Browser X as my web browser, I should be able to without having to worry about Mozilla taking over items which should be left to the web browser - like browsing the net!

I think we're actually already on our way there with XMLTerm. Its not totally cross platform, yet :), but its getting close. It takes the underlying os and creates a graphical command line using the mozilla XP engine.

If you can launch a browser window from xmlterm I hope it uses the system defined browser as opposed to defaulting to Mozilla or Nav6.

In windows explorer, when I type in a url, it assumes I want to open it in IE. Which 99% of the time is NOT the case. Thats what I find really annoying about the browser integration.

The other thing that mozilla has to clear up before this is possible are the crashes. If my email crashes, I don't want the browser to crash. If my new mozilla shell crashes, I don't want to loose the web pages I have up.

I know alphanumerica is working on a crash rebound program, but I don't know if it will work for only the web browser, or if it will launch all moz windows to their previous state. I suspect the former.

#24 Re: It all depends on who programs it :)

by petejc <pete@mozdev.org>

Saturday March 25th, 2000 5:11 PM

Reply to this message

* I know alphanumerica is working on a crash rebound program, but I don't know if it will work for only the web browser, or if it will launch all moz windows to their previous state. I suspect the former. *

Nope the plan is, it will save all of your open windows. So you will be able to totally recover from the crash.

pete

#25 Re: It all depends on who programs it :)

by svn <svn@xmlterm.org>

Saturday March 25th, 2000 8:35 PM

Reply to this message

"If you can launch a browser window from xmlterm I hope it uses the system defined browser as opposed to defaulting to Mozilla or Nav6."

I agree that XMLTerm should respect the user's overall desktop preferences. In fact, the plan is that the icon mode of XMLTerm should use Gnome icons under Gnome, KDE icons under KDE, Windows icons under Windows and so on.

As for the the general topic of integration, I think integration is good, provided it doesn't interfere with freedom of choice. The XPCOM technology of Mozilla makes it fairly easy to do that. Once Chrome overlays become available, you will be able to swap components at will and even have them change the menus!

Mozilla does not depend upon extensions like Chatzilla (the IRC client) and XMLTerm. The user can choose not to download and/or use these extensions without affecting any of the core browser functionality. Most Mozilla developers at Netscape are working on these core functions at the moment, as they should be. It's mostly non-Netscapers like me who are dabbling with "non-traditional" uses of Mozilla at the moment. Again, it is about freedom of choice.

Saravanan

#14 cross-platform?

by ywwg

Saturday March 25th, 2000 10:59 AM

Reply to this message

Cross-platform is great and all, but I would have preferred mozilla "look and feel" like my GUI of choice (GNOME), and not like some unique interface unto its own. The theme is nice, but I was hoping it would blend with the rest of my desktop better.

#15 Browser and OS integration

by vulturex <jonesev@home.com>

Saturday March 25th, 2000 11:12 AM

Reply to this message

Being a Unix user (Linux at home, Solaris and AIX at work) I don't see the browser integrating with the OS ever. The browser integrates with the user interface. The question is if that is an improvement.

In my opinion, the general computer user interface will need to be "smarter" in the future, not neccisarily more integrated. If I am reading an email message about a change that one of my coworkers implemented in some source or configuration file, my UI should offer me an instant way to open that configuration file, preferably with the new changes highlighted.

It is that level of intelligent "related information" concepts which will take user interfaces to the next level in the near future. I can see that browser integration can be a part of that, but I don't see it being there yet.

For now: I want my web browser to be a web browser, and not my entire user interface.

#52 Re: Browser and OS integration

by johnlar <johnlar@tfn.net>

Sunday March 26th, 2000 8:27 PM

Reply to this message

I have a friend who was able to test OS X and he hints at the fact that this kinda thing will be possible with their design (I don't know if you exact situation would be possible, as it is an extreme, but the general idea)

#16 OS/browser integration, xplatform

by _jon

Saturday March 25th, 2000 11:19 AM

Reply to this message

kovu, you hit the nail on the nose! =)

"Web browsing has nothing to do with interfacing between the user and the hardware, and that's the SOLE purpose of an OS. "

exactly. integration is a necessary evil though, but as long as i maintain control over what an app does to my system, including what goes on behind my back like sending information about my system to retrieve an update, i'm happy. i'd say most computer users today don't have a clue about using their computer, save clicking things. there's also the trend of improving efficiency, however such an improvement may be perceived. if it takes dragging an icon instead of typing an address, many will say that's good and constitutes an improvement. same for automatically obtaining an update- system files, virus defs, whatever. again, i'd rather go fetch it myself. let those who want to use the feature use it, just don't force feed me.

regarding mozilla's cross platform approach, it's the only way to go. i've yet to go through all the code to see how deep this goes, but i see at least a few areas that would realize benefits from it. since there's no effectively captive audience with the internet, it's prudent to implement designs and functionality that will prove usable by the broadest possible group.

for you w98 users, IE can be 'dis-integrated' with few side effects if you know your way around the OS. i have an installation that offers netscape when the windows icon in explorer is clicked as well as when URLs are typed in an addressbar (got tired of the behavior resulting from the integration one day).

for the mozilla developers out there, nice job! let's hope the industry learns from your example =)

#17 OS/browser integration, xplatform

by _jon

Saturday March 25th, 2000 11:29 AM

Reply to this message

so there's no confusion about my sentiments on integration, i run w98 (an installation that hasn't been 'dis-integrated') for testing only. w95 has not that horrible (==not optional) shell integration, so that's the primary OS on my home pc. there's red hat linux and NT4 partitions, too. the box at the office runs nt4 and has a 98 part. if i had a choice with the integration, i might be running 98. same goes for w2k, but i'll wait for the first bunch of patches before checking it out.

#18 Re: OS/browser integration, xplatform

by pepper <jonas.utterstrom@vittran.norrnod.se>

Saturday March 25th, 2000 12:20 PM

Reply to this message

There are actually parts of IE that are good. The taskbar is greatly improved by allowing you to drop icons on it. This way you don't need to bring down all applications to access the desktop. The new draggable menus are also cool, wonder how hard it would be to add a similar feature in Mozilla? The other thing is HTML Help which really rules over WinHelp.

Unfortunately, you have to install IE to get this features on NT and Win95. But you still don't have to use the Browser. Unfortunately, you also get the worthless Active Desktop and NT Explorer integration. But both can be disabled with a little work.

What I would like from Mozilla is a similar HTML Help application written in XUL. That would be cool.

#20 Re: Help files

by WillyWonka

Saturday March 25th, 2000 2:25 PM

Reply to this message

At work we use RoboHelp to generate the help files for our program. We will be releasing a mozilla version (Just as soon as the activex control gets built again. Come on web shell guys!). Anyways, Robohelp allows you to export a help file similar to the MS chm files. But instead of compiling it like the chm's it places a Java applet on the left side - which has the exact same functionality.

The html pages loaded fine in mozilla but I don't have the beta jvm so I can't test the index, etc.

#36 Re: Help Files

by pepper <jonas.utterstrom@vittran.norrnod.se>

Sunday March 26th, 2000 7:59 AM

Reply to this message

Hmm,an java applet that doesn't sound to nice. I'm still waiting for the first usable Java application. But I wil not turn it down until I have had a first look on it.

I still would like to have a XUL version of HTML Help so that I can use it both in Windows and Linux.

#69 Off-topic, but....

by FrodoB

Monday March 27th, 2000 2:05 PM

Reply to this message

There now exist dozens of usable Java applications. Now, usable Java client applications are still not frequent (the speed still lags in this department; something that the next two versions of Java are directly targeting), but server apps in Java kick butt. :)

#73 Re: Off-topic, Java Accelerator

by WillyWonka

Monday March 27th, 2000 3:40 PM

Reply to this message

I've heard of a Java Accelerator which suposably speeds them up 6 fold. Wonder if we'll be seeing more of these in the coming months?

#90 Compiling Java to executables...?

by Silverthorn <shawn.fumo@the-spa.com>

Tuesday March 28th, 2000 2:57 PM

Reply to this message

Yeah, there still might not be tons of huge programs running around as non-applet programs, but I certainly wouldn't say it is dead either...

As was said, once the speed can get going a bit more...

But while in this theme, what about the programs that can compile to regular non-interpreted machine code? I had d/led a microsoft thing that could do that a long time ago, but I haven't heard anything about programs like this for a long time, on any platforms...

At first this might seem to defeat the purpose, but think about it in terms of creating programs for multiple platforms.

Once you write the code, all you have to do is have one of these programs on each platform, and quickly compile it without having to re-write anything.

That should make it around as fast as a C/C++ program, without a lot of the hassles. And you could also distribute the regular java version for the people that want to run it interpreted, or with some of the JIT compiling...

It seems like this sort of thing would get more people using java for regular applications, until some of the speed issues are worked out better.

Is this being done? Are there any big programs that being distributed as binaries but are written in Java?

Just curious, Shawn

#22 integration

by Kovu <Kovu401@netscape.net>

Saturday March 25th, 2000 4:15 PM

Reply to this message

Some of us seem to be getting confused with this whole "integration" concept. Integrating e-mail, newsgroups, and Web page development is not the type of integration being discussed here. It's perfectly fine for applications to integrate with other applications, that's not the issue.

As far as IE's integration, that's with unrelated .dll files for the most part, just to make it so deleting IE "breaks" the OS and thus proves MS's argument in court. Period. IE3 and before were a completely separate program and didn't botch up your OS like win98. If Windows 95 wasn't missing so many UI features I would run it instead. And if someone would finally release a program that automatically removed IE (like Win98 lite, but easier) from the OS I would use it in a heartbeat. I think that should be part of any settlement in the trial -- make that program fully available.

And any cool "features" that IE has are UI-related Windows-ONLY (or Mac-only) features that have nothing to do with how good the application IE, which amounts to a broken, proprietarily jaded version of Mosaic, is as a standards-compliant Web browser. Netscape can add UI features at any time, what matters now is making the Web browser fully functional so that there's a user market for these great new features.

As far as Windows Update goes, that is an application that checks the Internet to see if what's on it matches the OS. It should be considered an application, not part of the OS. This integration thing is crap. Integrate applications with other applications, fine, but keep your applications out of my OS. Anyone who has used an Amiga and seen the thing reboot in 10 seconds knows what I mean. Windows is an ugly, bloated nightmare, and was BEFORE IE came along. Now it's only gotten worse.

#28 integration

by socbyset <socbyset@hotmail.com>

Saturday March 25th, 2000 10:16 PM

Reply to this message

It's important for a browser to "know its place" as just another application sitting on the OS. Both IE-Windows and Netscape/Mozilla violate this rule in opposite ways.

IE Windows' integration is annoying and unnecessary. As was mentioned before, it just spreads dll's around in an attempt to make Windows blow up if you try to remove it-- Although I must admit that it is convenient to turn any folder's window into a browser by typing a url in the title bar.

Mozilla's lack of integration with OS's user interfaces is even more annoying. They seem to have this idea that since the Internet will create an uber-platform for networked computing, they should make Mozilla look and behave exactly alike on every platform. Then, in theory, people would just learn the Mozilla environment, it would become the internet "platform" and particular OS's would be downgraded in importance to plumbing, just commodities. But this is awfully hubristic, people want their browsers to behave just like all their other applications (especially Mac users- they are fanatical about the Mac OS user experience). If every app decided to reinvent the UI wheel then the whole computing experience would be a maddening hodgepodge, you would have to relearn the most basic tasks for every app you use.

It reminds me of the early days of Communicator 4.x, when you were forced to download mail, news, and Composer just to use the damn browser- Netscape sometimes thinks about what users want second, and what it wants first.

#30 That's what skins are for, see the Chromezone n/t

by Kovu <Kovu401@netscape.net>

Saturday March 25th, 2000 10:56 PM

Reply to this message

n/t

#32 Re: That's what skins are for, see the Chromezone

by socbyset <socbyset@hotmail.com>

Saturday March 25th, 2000 11:35 PM

Reply to this message

But still there's the issues of inheriting ui preferences, inheriting changes in the ui when the os is updated, standard keystrokes per os (does xul handle this?), etc.. (Mac users: does mac Mozilla support macos Navigation Services for open/save?) Granted, customization via XUL can solve some problems but isn't that approach a bit kludgey? And in Netscape branded mozilla, will OS-specific skins come standard?

#34 Well...

by Waldo

Sunday March 26th, 2000 3:37 AM

Reply to this message

To answer your last question-- I'll bet after the public totally rejects the current widget set, we'll be seeing OS-specific skins.

(that's just my prediction. I don't want to start another discussion about widgets)

As far as inheriting preferences goes-- there's only so much you can inherit if you want to support web standards. Native OS FEs are not capable of doing the stuff that needs to be done to comply with those standards.... so what do you do? Either chuck the standards or cheat a little on the interface. There are pros and cons to both but I guess they decided to do it this way...

W

#46 So you are saying...

by luge

Sunday March 26th, 2000 1:09 PM

Reply to this message

that in order to support web standards I can't use standard MacOS scrollbars or cut and paste? Obviously, some XP widgets were very necessary for certain types of web display. But scrollbars and text entry (for example) are just unecessary...

#48 Yeah but..

by Waldo

Sunday March 26th, 2000 3:35 PM

Reply to this message

Where do you draw the line? First off, I'm all in favor of native OS scrollbars, at least in the outside window of the browser..

But let's say the web standards call for super-wide (or super-narrow) scrollbars (I don't know if they do, just saying what-if) or scrollbars with semi-transparency or with a colored background or whatever...

This is stuff that the native scrollbars don't do. To support the specs and maintain control over them xplatform, they included the widget stuff in Mozilla itself.

As some of the developers have said, the alternative to doing this would have been to have NO browser support for mac or linux. I guess it was a resource thing-- to integrate all the different types of OS's and widgets and keep them up-to-date would have been unfeasable with the manpower they had.

Still, you can do lookalike widgets if you want, thanks to XBL, so hopefully all the native widgets will be invisibly recreated fore ach platform.. I think we'll see this happen pretty fast after teh beta phase begins.

W

#84 Re: integration

by shk41

Monday March 27th, 2000 10:01 PM

Reply to this message

I agree. I think it's stupid for them to just go and rebel against what people are used to, because what people are used to in an apllication is not particularly bad. I think their making a mistake by being so different, but they might be able to make it work IF Mozilla's interface was MORE useful than the standard. Why would you come out with a program that uses a different interface and different functionality than anyone is used to, and that is really pretty annoying, irritating, and useless, and expect for people to want to use it? The only people who are going to use Mozilla are the immature teenagers that think the interface LOOKS kind of cool. It does not make sense. Mozilla needs to take into consideration what it's REALLY got on the other browsers, not just focus on one thing, cross-platform functionality. That's only ONE step.

#35 I don't want integration

by jeromekwok

Sunday March 26th, 2000 7:43 AM

Reply to this message

I don't want to use a 30MB+ browser to "dir" and copy files. If you can make the browser thin, fast, efficient, easy to use, welcome!

#40 Re: I don't want integration

by jbailey999

Sunday March 26th, 2000 10:45 AM

Reply to this message

But I *like* Emacs! =)

#39 Nothing wrong with using a browser to surf files

by Kovu <Kovu401@netscape.net>

Sunday March 26th, 2000 10:43 AM

Reply to this message

There really isn't anything wrong with using a browser as part of a file managament system. But there is also no good reason at all to make this "integrated" with the OS. 4.x allows you to FTP -- that's file management. But it's also a clearly defined application that doesn't do bad things to my OS.

#44 XPCOM and XUL are the future

by ttfkam

Sunday March 26th, 2000 12:23 PM

Reply to this message

#1 The main reason most MS-Windows developers remain MS-Windows developers is COM.

You are never going to tear them away from a language-independant, reasonably fast, component architecture no matter how bad any other part of the operating system becomes or how good any part of an alternate OS becomes. Don't believe me? Try talking to a VB coder for more than five minutes. And you thought Linux zealots were rabid...

Any part that is unless that part is an XPCOM service running alongside the OS on any arbitrary platform -- independant of the browser. You do this, and you cripple the number one argument for MS-Windows.

#2 XUL makes sense as a general UI framework. It's component-based, skinnable/themeable, scriptable, has a uniform API on multiple platforms, etc.

For all of the naysayers regarding XUL becoming "integrated" into the OS, let me remind you that GTK also started as a widget toolkit for a specific, specialized application: GIMP.

XUL is the future: Structured markup on top of a compiled framework to produce a complete MVC architecture.

Whether the Mozilla developers intended this result or not, they are going to see their baby leap out of a browser-only role with both feet and become the tie that binds for all platforms.

The funniest part about it all is that almost no one will see it coming.

#47 Re: XPCOM and XUL are the future

by petejc <pete@mozdev.org>

Sunday March 26th, 2000 3:03 PM

Reply to this message

*The funniest part about it all is that almost no one will see it coming. *

Here, here . . .

Well said.

#50 Re: Re: XPCOM and XUL are the future

by Waldo

Sunday March 26th, 2000 3:37 PM

Reply to this message

I agree :)

I think the GTK/Gimp comparison is pretty interesting. Wish I could moderate it up ;)

W

#51 Re: XPCOM and XUL are the future

by ttielkes <t.r.tiekes@zap.a2000.nl>

Sunday March 26th, 2000 5:01 PM

Reply to this message

I agree with you, however, XPCOM and XUL are pretty hard to use for, say, an average VB programmer. Microsoft put years of development into the Visual Studio tools, and it shows.

OTOH, I see most VB/Delphi apps migrating to browser-based webapps in the next years. This will probably show with the release of Visual Studio 7 with the Web Forms and ASP+.

If Mozilla wants to be able to run similar apps (DHTML UI and XML data), it better get some good XML support. IE4 had *TWO* XMLDOM parsers (C++ and Java) and an XML DSO (databinding). This was back in 1997.(Check out this sample from Drew Marsh: <http://bowery.nyc.proxicom.com/JavaCOM/dhtml.htm> )

To me, it seems the Mozilla vision is: "We're going to build a nice HTML renderer". Microsoft says: "We're going to build the future application platform"

#58 Re: Re: XPCOM and XUL are the future

by basic <_basic@yahoo.com>

Monday March 27th, 2000 1:59 AM

Reply to this message

> I agree with you, however, XPCOM and XUL are pretty hard to use for, say, an average VB programmer.

I don't know about XPCOM, but the XUL being hard problem can be solved easily with a XUL editor.

> OTOH, I see most VB/Delphi apps migrating to browser-based webapps in the next years.

I see moz at the center of it...

> If Mozilla wants to be able to run similar apps (DHTML UI and XML data), it better get some good XML support.

Moz runs on XML, Moz itself uses DOM for it's UI. As for DHTML UI? Well, if it can do XUL, DHTML is nothing...

> IE4 had *TWO* XMLDOM parsers (C++ and Java) and an XML DSO (databinding). This was back in 1997.(Check out this sample from Drew Marsh: (LINK) )

Expat is not good enough? As for Java, there are many XML parsers out there. I don't see a connection between a Java-based XML parser and Moz though. Jazilla OTOH...

> To me, it seems the Mozilla vision is: "We're going to build a nice HTML renderer".

Where did you get that? Moz has been accused of trying to build the ultimate app platform. In fact it is the NS Comm users that want "just a nice HTML renderer", but Moz will deliver much more. If Moz wanted to just build a nice HTML renderer, Moz 5.0 would have been released half a year ago. Switching to nglayout is not the only reason that it is not released today...

Basic

#60 Re: XPCOM and XUL are the future

by ttielkes <t.r.tiekes@zap.a2000.nl>

Monday March 27th, 2000 8:23 AM

Reply to this message

>I don't know about XPCOM, but the XUL being hard problem can be solved easily with a XUL editor.

A GUI layout tool is worth nothing if it isn't connected to the rest of the IDE. These things take -years- to build and perfect. What tools are there *today* to use?

>> If Mozilla wants to be able to run similar apps (DHTML UI and XML data), it better get some good XML support.

>Moz runs on XML, Moz itself uses DOM for it's UI. As for DHTML UI? Well, if it can do XUL, DHTML is nothing...

"Runs on XML"? Come on..

Please show me how to load XML into a DOM in a Mozilla ECMAScript. Please show me how I can POST this XML doc over the web in an easy way.

If Mozilla "runs on XML" so great, show me how to include XML into my web pages in an easy way, how to bind UI elements to XML data, how to persist XML cross-session, et cetera..

And XUL? Is't very nice, but that's not the think I'm talking about. If I can build W3C DOM conformant DHTML UIs in IE5, why bother with some non-standard layout format?

>> IE4 had *TWO* XMLDOM parsers (C++ and Java) and an XML DSO (databinding). This was back in 1997.

>Expat is not good enough?

Expat has no DOM, no validation. Don't start about Schemas, XSL(T), XPath or ...

If you think it *is* good enough, give me an example of how I can parse an XML file from my server inside an HTML page from my server.

>> To me, it seems the Mozilla vision is: "We're going to build a nice HTML renderer".

>Where did you get that?

I got that from trying to build something with it.

As for vision, take 3 minutes to read this: <http://davenet.userland.c…999/11/06/theAncientGeeks>

#70 IE errata

by ttfkam

Monday March 27th, 2000 2:33 PM

Reply to this message

The comparison of XUL with DHTML behaviors gave me a nice chuckle. Thanks. Work kinda sucked this morning. No seriously, DHTML behaviors is a good implementation of scripted HTML. This brings up some of my gripes with HTML: presentation and content in one piece.

An HTML "div" tag means symantically a page division, but it also has a layout meaning with states that it should leave an open carriage return after the closing tag. On the other hand, XUL defines the component only. Yes, you can take away that layout behavior from the "div" tag, but then you're left with a open XML tag container... just like XUL only you had to work to get it there. There is a strict demarcation between the logic (usually found in a separate, linked script), presentation (CSS) and the component itself (XUL).

This separation is commonly referred to as an MVC (Model-View-Controller) architecture. In other words, modification of one component does not usually affect the functionality/presence of the other two. DHTML behaviors tend to blur this distinction. MVC is kinda like object-oriented programming. It starts with folks bemoaning the apparent loss of their preferred mode of development/use, follows with an excited "Oh! I get it!", and finishes with a resigned "I can't believe I used to do it the other way."

I agree with you about Visual Studio and its depth, but a GUI development platform based on DOMs and stylesheets is not that terrible to create. XPCOM development tools will probably lag as you say, but using/reusing existing objects will be a relative breeze.

IE4's XML parsers and DOM support was a shoddy exercise to say the least. After all, why do you think that Microsoft changed the API so dramatically from v4 to v5? The Java parser sucked arse then and still sucks arse now.

With regards to instantiating a DOM via JavaScript, you are correct up to a point. Instantiation of an external DOM is a DOM level 2 item which the coders have not tackled en masse yet. The funny thing about the Mozilla group is that they seem to be trying to finish one spec and implement it fully before working in earnest on its successor. But you are correct, Mozilla is behind in this area.

Schemas, XSLT, XPath? Have you even glanced in the general direction of the W3C draft for Schemas? Check out the MS implementation and compare with the spec. Yes, I know, the Microsoft.XMLDOM came out early on in the spec, but do you see any work being done on it in the new technology previews?

XSLT is in fact in development and folks are coding away on a workable implementation to be included into the main branch. It isn't there yet, but it's not like they're ignoring it either.

Honestly I don't know the status of XPath support. I'll concede this one unless anyone steps up to correct me.

#77 Give me something to work with!

by ttielkes <t.r.tiekes@zap.a2000.nl>

Monday March 27th, 2000 6:34 PM

Reply to this message

The IE4 XML parser sure is shoddy. It was released some time before the W3C XML 1.0 and DOM1 specs, IIRC.

And yes, the (dataChannel) Java parser sucks arse. It always has.

What I want to say is: IE4 offered pretty strong (pre-W3C) XML support back in 1997. I am the the last person to say that it was perfect. But it *was there*, for people to play and work with, and learn about XML and it's uses!

I don't see how loading/saving/POSTing XML isn't part of *a workable browser implementation* of DOM1. The DOM1 spec explicitly states that some (crucial) functionality in not included in the interfaces. Perhaps I should say: "Why is there work being done on DOM2 if there are no ways to use the DOM1 functionality?"

I know jack about XUL.

I want to build simple UIs working through the HTML DOM1. I want to use XML as a data format using XML DOM1. IE4 did all this, be it in a time when XML and DOM specs weren't out yet.

IE5 does all this *now*, using both W3C HTML DOM1 and W3C XML DOM1.

I want my (little) apps to work on Mozilla. I don't mind to write wrapper methods in ECMAScript because IE5 wants to hear "oDom.load(url)" and Mozilla likes "oMozDom.loadXMLformURL(url)".

I want to work with Mozilla, and I want Mozilla to work. The Microsoft IE team (through documentation, articles, samples) shows to me a clear vision of XML as the *de facto* data container/format of the web. That, and HTML DOM1 is all I need.

If Mozilla can give that to me I'll follow it to the end of the world ;-)

#56 Re: XPCOM and XUL are the future

by socbyset <socbyset@hotmail.com>

Monday March 27th, 2000 1:04 AM

Reply to this message

The idea of something like XUL/Mozilla taking UI out of the hands of the OS and turning it into a open source and Web standards-based application architecture is compelling. It would make NC's more viable and fulfill a lot of the cross-platform hype that used to surround Java. It would also be a great leap forward for open standards-based computing. It's something that I would love to see. But I still think that in practice it may have a hard time gathering steam. Say I am a graphic artist working on the MacOS. I spend maybe 60% of my time in Photoshop and Illustrator and 10% doing miscellaneous tasks like copying files, organizing my stuff, etc; and 20% doing tasks on the Web. Well, if my browser is Mozilla then I will be frustrated by its ignorance of the way my Mac and my Mac apps do things. Not just widgets but all the Mac interface guidelines, look and feel, dialogs, copy/paste, etc. I will choose IE in a second. The point is, I get good standards compliance and a browser that genuinely tries to be "mac". What will make me switch to Mozilla?

Maybe I'm not thinking big enough, maybe the landscape will have so many different wierd devices and set top boxes and NC's in a few years that XUL/Mozilla etc. will multiply and dominate the universe, and traditional desktop computing will be a much smaller share of internet computing. But I don't think it's a foregone conclusion that we will get from here to there.

#57 Re: XPCOM and XUL are the future

by basic <_basic@yahoo.com>

Monday March 27th, 2000 1:28 AM

Reply to this message

> Any part that is unless that part is an XPCOM service running alongside the OS on any arbitrary platform -- independant of the browser.

XPCOM is nice, but if it does not grow into DXPCOM, I'd say that it will exist only in Moz for the most part. So let see some action on that front!!

> You do this, and you cripple the number one argument for MS-Windows.

NOT!!! The number of machines with MS-Windows sitting out there is the number one argument for MS-Windows. If this is done, the argument for DCOM is crippled!! Then it doesn't matter how many machines are sitting out there with MS-Windows as one can write for more than one platform.

> It's component-based, skinnable/themeable, scriptable, has a uniform API on multiple platforms, etc.

> For all of the naysayers regarding XUL becoming "integrated" into the OS, let me remind you that GTK also started as a widget toolkit for a specific, specialized application: GIMP.

> XUL is the future: Structured markup on top of a compiled framework to produce a complete MVC architecture.

XUL is going through growing pains. As it is now, I don't want it integrated into my OS, but I would love to see it grow into a XP complete UI description language. And don't forget XBL. As for GTK, is there a chance where GTK might adopt XUL in some way (when it is more matured of course)?

> Whether the Mozilla developers intended this result or not, they are going to see their baby leap out of a browser-only role with both feet and become the tie that binds for all platforms.

Interesting, you don't think this is a conspiracy do you? ;-)

> The funniest part about it all is that almost no one will see it coming.

Some will... and the rest will follow.

#71 Re: XPCOM and XUL are the future

by ttfkam

Monday March 27th, 2000 2:59 PM

Reply to this message

XPCOM doesn't necessarily have to grow into DXPCOM. DCOM is basically trying to redo CORBA. Drawback of DCOM - trying to implement it through a firewall without leaving a gaping security hole. I'd prefer XPCOM to grow into CORBA.

As far as the "number of machines" argument, for the desktop, I can sorta see the premise.

For the server arena, UNIX still reigns.

However as more "Internet Appliances" hit the market in force, and some are starting to hit, the OS means nothing. Most "Mom & Pop" spots are the type to get the appliance. These are the folks that got AOL "cause it's easy". They don't care what it runs on. They care if it works or not and helps them get done what they want to get done.

Developers, on the other hand, do care about COM. If you took away COM from MS-Windows, I'd warn you to try not to get trampled by the ensuing stampede. Without COM, VB ends up little more than the BASIC from DOS days.

XPCOM is working toward COM capatibility. If you have the choice between making a COM component or an XPCOM component with all other things being equal, why on earth would you choose the technology that limits you to just MS-Windows?

I don't expect, or truth be told necessarily want, Microsoft to go away. What I'm hoping for is a loosening of the chains to my MS-Windows box. I would like to "want" to work on MS-Windows because it may be the better tool for a job. I would also "want" to work on Linux for some jobs. I would "want" to work on a BSD. If they all have a common component toolkit, I'll be less likely to have to use a screwdriver to cut a piece of wood... so to speak.

#49 Browser Technology as GUI?

by Stromgol <rochette@autray.net>

Sunday March 26th, 2000 3:37 PM

Reply to this message

Maybe integrating a complete browser into an OS is not such a great idea but using the renderer/parser (Nglayout!?!) as a low-level GUI lib. could be something we should consider.

I don't want to "browse" through my hard Drive... but ,if the integration of browser technology in an OS would mean using XML (..XUL?!?) to build a gui instead of a high level language, I'm all for it. It would bring an incredible amount of portability in the gui-related code of programs and it would be a breeze to modify, beign non-compiled. Well, at least that would be the case if the implementations on different platform would follow the standards... In other words, we should make XUL an gui standard on every platform ;)

#55 Integration? What about...

by basic <_basic@yahoo.com>

Monday March 27th, 2000 12:56 AM

Reply to this message

Browser integration into almost anything is inevitable in this web/net crazy age. How it is integrated is the question. IE/Win98 was a good experiment to show how not to integrate a browser. Ooops...

I take that back. Browsers should not be integrated into anything!!!! But a WWWeb/XML based GUI/shell will be. Win98 was another "nice try", but IE is hardly what I'd call a XML based shell. I believe that Moz is the best candidate for such a GUI/shell... so far. There will be better ones later, but Moz proves that it can be done. And speaking of IE, just saw this on Web standards mailing list.

Subj: [stds]Mac IE5 available now

Date: 3/26/00 9:57:24 PM Pacific Standard Time

From: <paulv@transmitmedia.com> (transmit)

Sender: <standards-admin@webstandards.org>

CC: <standards@webstandards.org>

Available now:

<<http://mschus.http>://www.conxion.com/download/ie5mac/install/5/macos/en-us/ie5en. bin>

-----------------------------------------

Paul Vachier <paulv@transmitmedia.com>

-----------------------------------------

base: San Francisco, CA

_______________________________________________

Standards mailing list

<Standards@webstandards.org>

<http://www.webstandards.o…ailman/listinfo/standards>

Anybody checked it out yet? How is it's standard compliants?

#64 IE5 for Mac (lots of stuff)

by Silverthorn <shawn.fumo@the-spa.com>

Monday March 27th, 2000 10:47 AM

Reply to this message

Well, I just installed it on a computer in the Mac lab at school. As much as I hate to admit it, it does seem to be a pretty impressive program.. ;)

The new interface is fairly flashy, going on that whole candy theme again (how much candy can we take with MacOSX coming out? hehe).

The ability to change the color on the interface, configure the icons on the toolbars, collapse all toolbars and status bar at once. It all works well, and the throbber is really nice...

But anyway, it seems good in the standards department. I just went through almost the whole list of top 10 msie4.5 mac bugs and everything seemed to be fixed. I also went to a bunch of pages with CSS on them, and I didn't see anything out of place.

Also on their website, they mention it can show the acid box test ok. They actually claim to be the only browser able to show it correctly, but I think Mozilla can show it fine? They're probably going on the technicality that Mozilla is still alpha?

They claimed ping support on the site, but when I tried to view one, it didn't know what to view it with. And on the site it opens up to, there is a demo page with flash, but flash isn't installed? Perhaps because this is still an early version?

But I think Mozilla should be really careful right now. This version seems to work really well, and the interface is really flashy and Mac-like. And it also keeps the various old features like the url completion (now the box is transluscent).

If someone were to compare Mozilla and this side by side, IE would look a lot better, both in the interface and speed and features, at least from a regular user's point of view...

So, I guess we'll see how all of this goes. I'm still not sure that I'd be ready to install it on my home computer if they made a PC version that was just as good, but it looks good, and I'm glad they finally seem to be getting their act together in terms of the standards.

I viewed a page I put together as an experiment before (using Mozilla as the preview browser), and IE4.5 on mac showed it pretty well, but the right edges of some divs were ragged. But now it looks fine. So Mozilla and IE5Mac are the only ones that can view it right.

Perhaps Netscape 6 and a final IE5 will finally be more where we want things to be...

Shawn

#66 IE Mac doesn't really count as an IE, I'm afraid

by Kovu <Kovu401@netscape.net>

Monday March 27th, 2000 12:34 PM

Reply to this message

IE Mac falls outside the Windows-dominant machine that is Windows IE. Mac IE is a completely separate program from Windows IE, with different goals in mind, etc. And MS's Mac division is just a cute little pet of the main company to appease the evil deals MS made with Apple to try and show some semblance of competition in the market. Since the Mac team isn't part of trying to rule the world, in short, it isn't surprising that the Mac software MS puts out is actually pretty good software (and it also isn't surprising that software comes out at least a year AFTER the Windows versions, either, though that's not the Mac team's fault).

But all those interfaces should be possible, and easier accomplished, with Mozilla. And I'd like to see someone REALLY take that Mac IE through the W3C wringer and THEN see where it comes out against Mozilla. I sincerely doubt it gets a 95% plus on CSS1, much less an 80% plus on CSS2. Anyone?

In any case, Mac IE will never be a real threat to Mozilla. Mac people may not want to admit it, but if every single Mac user used IE it wouldn't matter a hill of beans in the main market. There just aren't that many Mac users out there in comparison to Windows/Linux. Judge Jackson discounted Mac and Linux as insufficient to show any real competition to Windows and I fully agree. Not that Amiga isn't less of a market, don't get me wrong, but IE Mac will still never be a threat to Mozilla.

#72 True... but...

by Silverthorn <shawn.fumo@the-spa.com>

Monday March 27th, 2000 3:26 PM

Reply to this message

What you say makes sense. Perhaps I'm being too optomistic, and of course I very much doubt the candy-colored interface part of things would go in a windows version, etc.

But still, I have some hope. They seem to be raving about the new Tasman rendering engine. That seems to be a direct attempt to compete with the Gecko engine in Mozilla.

I can see creating the interface and everything, but I can't see them expending all that effort on a new rendering engine just for the Mac. As you said it isn't THAT big of a market.

I'm hoping that they'll stick Tasman into the windows version eventually. If they already have it written, seems like it'd be the sensible thing for them to do. They could put a lot of hoopla into it...

They yell about Tasman over here:

<http://www.microsoft.com/…features2.asp?Page=tasman>

But as you said, we'll have to wait to see exactly how compliant it really is before even speculating about if it'll reach over into the windows platform...

Shawn

#76 Mac IE vs. Netscape 6 here

by Kovu <Kovu401@netscape.net>

Monday March 27th, 2000 6:19 PM

Reply to this message

<http://developer.netscape.com>

Sorry, Mac IE still don't get it against Mozilla.

#78 Re: Thanks...

by Silverthorn <shawn.fumo@the-spa.com>

Monday March 27th, 2000 7:46 PM

Reply to this message

Hmmm... good information. Of course somewhat slanted the way it is presented, but the facts speak mainly for themselves. Plus IE tends to be more guilty of weird stuff anyway...

So, my conclusion is that if MS can get its act together and put Tasman into both platforms, then at least we'll have a decent base of HTML 4 and CSS1.

DOM will still be shot, but at least it'd be something...

They had BETTER even out the platforms and do it soon. Having a platform lag in comparison to another is sort of understandable, but whoever heard of totally different support for html across platforms. As it said, that could cause a major mess with regards to the sniffing.

I should go download the opera beta...

Anyway, sorry for being so off-topic here. Ok, my bad attempt to bring it on-topic: IE is too integrated in that it is totally different on different platforms. Mozilla is relatively the same across, so that's a good thing. ;)

Shawn

#75 Re: IE5 for Mac (lots of stuff)

by stephan <stephan@micropop.com>

Monday March 27th, 2000 5:27 PM

Reply to this message

FYI: Opera 4 (beta) passes the box acid test.

#59 Cross-Platform

by dzirpel <dzirpel@ford.com>

Monday March 27th, 2000 2:37 AM

Reply to this message

Great idea! When and where can I get an version for the (classic) Amiga-platform (with support for PowerPC-CPU's)?

regards

Detlef Zirpel C.H.I.P. - Analyst

Vehicle Homologation and Conformity (VHC) John-Andrews Research & Engineering Centre D-MD / PS-1 Ford-Werke A.G. 50725 Köln Germany

Tel.: +49 221 903-7786 (office) +49 2291 3722 (privat) e-mail: <dzirpel@ford.com> (office) <dzirpel@t-online.de> (privat)

#67 3 percent platform dependent...

by Kovu <Kovu401@netscape.net>

Monday March 27th, 2000 12:37 PM

Reply to this message

There is a site (my site actually) trying to organize an Amiga Classic port here: <http://home.att.net/~Amiga401/amigamozilla.htm> but there doesn't seem to be much progress at this point. Anyway, the 68x80 port would probably precede the PPC port.

#61 The browser engine is a service...

by Quelish

Monday March 27th, 2000 10:02 AM

Reply to this message

From experience with n-tier application development, I believe that there should be clear and clean distinctions between the OS and the browser.

The OS has the sole reponsibility of managing the system's resources and making them available in an easy to use manner. The GUI is simply an interface to the OS that, if one ever desired, could be swapped for another GUI (eg. LiteStep). The OS should have a standard interface that any GUI can use therefore swapping GUI's isn't going to break anything.

The browser engine provides a service: rendering pages. The browser GUI uses this service much in the same way that the OS's GUI uses the OS. And in just the same way you should be able to swap about the browser's GUI without breaking anything.

Now, if the OS's GUI happens to use both the OS and the browser engine to display files, folders, web pages, or whatever, then that's perfectly fine. But the OS and the browser are two different services. The browser's engine might even use the OS's services to do it's job -- But they are not the same!! And just like you should be able to swap out the OS's GUI, you should be able to swap out the browser engine and still maintain compatability.

The idea here is that each level has a standard interface. So long as each level above it uses that interface you should be able to swap things out without breaking applications.

From an architectural standpoint, this makes sense. You separate the layers, standardize the interface, and then you're home free. Bit now back to the REAL world of how applications are being written today...

#81 "Standards-Compliance"

by shk41

Monday March 27th, 2000 9:46 PM

Reply to this message

I don't get all this standards-compliance crap! Mozilla needs to support all the features of every popular browser, that way, people will enjoy it compatibility with all pages. "Standards-compliance" in the sense they are using it is like not caring that English and Spanish are the the two most popular languages in thw world, and speaking Latin, even though nobody else is. Face it, people can't be popular on the web only making "standards-compliant" web pages. "EVERYBODY has to use it or we won't! We don't care that IE is more popular than Netscape and supports the div and span tags because Netscape doesn't yet!"

They ought to wait for every SINGLE browser to support something before they do!

#83 Re: "Standards-Compliance"

by ttielkes <t.r.tiekes@zap.a2000.nl>

Monday March 27th, 2000 10:00 PM

Reply to this message

What standards dictate what standards Mozilla should support?

Where did this standard-mania come from?

Why not just copy the IE5 features that make the most sense, and make 99.99% of all web users very happy? (Wishfull thinking :-)

#87 Re: Re: "Standards-Compliance"

by Tanyel <tanyel@straightblack.com>

Tuesday March 28th, 2000 1:48 AM

Reply to this message

"Why not just copy the IE5 features that make the most sense, and make 99.99% of all web users very happy?"

Probably because this is a matter of pride rather than what is best for general Internet users.

#93 Please no.

by Stromgol <rochette@autray.net>

Thursday March 30th, 2000 6:57 PM

Reply to this message

Why should we leave the future development of the web to some company... if they become the only HTML definer in the industry we will end up having a browser (IE) up to date and the rest lagging behind because there is no way MS will reveal its "enhancements" prior to the release of the version of IE that incorporates them. Oh, and by the way, maybe I'm just weird... but I think it would be to Ms' advantage to release these "enhancements" with windows-only features, where we could see things like

"<link type='WindowsUpdate' Active-server='windowsupdate.com/checkforIE'>"

#94 Re: Please no.

by ttielkes <t.r.tiekes@zap.a2000.nl>

Friday March 31st, 2000 8:20 AM

Reply to this message

Most useful MS stuff is right there in the W3C archives to download. Data islands, behaviours, go check it out before you post.

#95 well...

by Stromgol <rochette@autray.net>

Friday March 31st, 2000 1:18 PM

Reply to this message

Ok, it is true that IE5 support a lot of W3C standards as of right now.(I never said otherwise) But, if we rely on MS to actually MAKE the standards... I bet they will slowly be getting away from the W3C.

By the way, IE5 is not all clean though, there is some things like the <link> to .ico (wich is a proprietary format).

Anyway, I understand your point. I too would like to see identical rendering of websites between the browsers, but, I also think that we should not encourage MS for adding some non-compliant HTML by making a standard out of them.

I am only stating my opinion. I'm not saying yours is not as valid as mine though.

#96 Re: well...

by ttielkes <t.r.tiekes@zap.a2000.nl>

Friday March 31st, 2000 7:18 PM

Reply to this message

Ok.

I don't disagree with you.

However, W3C "legislation" follows practice, at least..good practice ;-)

#85 Bull#$&! n/t

by Kovu <Kovu401@netscape.net>

Monday March 27th, 2000 11:26 PM

Reply to this message

n/t

#86 Re: "Standards-Compliance"

by Ben_Goodger

Tuesday March 28th, 2000 12:47 AM

Reply to this message

if we all we wanted to do was build services and features, we'd probably just embed Trident like other apps.

however, we happen to think we can do better than Trident. So far we're the only organisation that's been able to produce something to contend with it, so I think we'd be foolish not to.

#88 Next Step: web like applications

by jmarranz <jmarranz@eresmas.com>

Tuesday March 28th, 2000 2:58 AM

Reply to this message

I think that XPToolkit and XPCom can be the killer developer frameworks of the Microsoft Windows dominance.

Many developers must decide to develop GUI applications over Windows plattform or X-Windows (XT,GTK or Qt), but develop on both systems costs very much, because they are very different.

Mozilla Group would may document, and maintain releases of the XPToolkit and XPCom and promote the development of web like applications, not only web based. This could hurt the MFC brutal dominance in desktop based applications.

#91 Part Integration is Important

by alik

Thursday March 30th, 2000 7:35 AM

Reply to this message

Hi, I am not a technical guy, but just someone who really wants Mozilla to kick off because of all the had work people have put into making something so wonderful.

However, we do not live in a perfect world and we have seen many good applications go down the tube, just because they did not do the popular things.

To Sell mozilla idea to people it has to do a lot more than just browse webpages or read e-mails. I work in an ISP and Most browser users do not even know the difference between IE and Netscape. They want ease of use, and better performance. I know this forum is for technical people, but most of internet users do not have the know-how and preferences we have. Therefore, the one of the ideas should be to make thier lives easy and make Mozilla their choice.

Here are a few important points..

1. Stability - It has to be rock hard! If it must crash.. it should do that gracefully.. it should be able to bring back the e-mail i was typing (like Rebecca) and recover the webpages i had open.

2. The Address book - to be honest (and do not kill me on that) I think it should take over the windows address book. It is really annoying to open the browser just to copy an address into my word document. (Also, why can we not have an address book for Mozilla, ICQ, AIM??) This feature should be a Option in Installation.

3. Auto Update - Although I have not used IE a lot, it offers plug-ins much better than Netscape. Instead of telling you to visit a website and download the required plugin, it just asks you if it should download the plug-in. This is much easier for the novice.

4. Non standard HTML tags - I know standards are very importnat to us. But we all know that millions of pages are not made on standards. They have many IE tags which look nice. people like justified text (althought it is more difficult to read), they like Page Transitions.. and these tags are not standards. I think mozilla should be able to show all these pages!! without sacrificing its own quality. I mean if mozilla supports HTML 4.0 to 100%.. why not suppliment this with a few extra tags?? However, if you guys do not thinks that should be the case, make the non standrad HTML tags an option. If the browser comes across a page which has IE specific tags, it should tell the user that the page has these Non standard tags and if the user wants, mozilla can download a small upgarde to show these. Or a little red light in the cornor telling the user, that the page is displayed correctly, but has Non standard HTML.

Guys mozilla is a a great achievement. But it is only the first step. More and more individuals and comapnies should be encouraged to develop things that work with mozilla to make it more powerful. A few of ideas i have are:

a. An Organizer that works with mozilla address book, and maybe shows calander and schedular using Geko.

b. Bookmark manager that can integrate with bookmarks file but can also store my passwords and other information (but this should make sure that no website can upload my bookmarks with password ;-) unless the company that made the software offered an online bookmarks manager with sync facility)

c. etc.....

So integration may not be imporant at OS level (except maybe for address book), but it is importnat at application level. There should be synergy in these. IE may appear to have won the browser battle. But that was only A battle. This war has many fronts.. I hope mozilla wins the WAR!

And last but not least. Many thanks to all of you who made this happen!

Ali K

#92 Very interesting point of view

by Stromgol <rochette@autray.net>

Thursday March 30th, 2000 6:47 PM

Reply to this message

You really have a point. A lot of computer users don't even know what is HTML, now, try to explain to these users what is W3C compliant ...

Browsers, and software in general, should be made in a way that it would work in and with the environment in wich it will be used. Therefore, the interactive "Tag Upgrade" is a reallllyyyy good idea. I wonder how flexible and "upgradable" Nglayout is in that sense... does anybody know if, with its the current architechture, it is easy to add/correct HTML? This would be usefull not only in the sense that it would provide IE compatibility but that it could make it easy to write patches and upgrades to make the current browser compliant to the future revisions of HTML or XHTML.

Perhaps a plugin-like architechture to let third-party write these would be desirable?

anyway, very inspiring post alik :)