MozillaZine Readers Sound Off
Sunday April 4th, 1999
MozillaZine readers Ho-Sheng Hsiao and Oliver White (aka cyanide) have written essays regarding recent events in the Mozilla community. The first, on the benefits of downloadable chrome, can be read here (or, click Full Article below). The second, musings on Mozilla and Open Source, can be found here. You can also get to the second essay by clicking the "Next" link at the end of the first essay.
The opinions stated in these essays do not necessarily reflect the opinions of MozillaZine.
If you have written an essay regarding Mozilla's first anniversary, feel free to submit it to MozillaZine. We can't promise we'll print all submissions, but we'll try.
#1 Re:MozillaZine Readers Sound Off
by Orlando Medina <firstname.lastname@example.org>
Sunday April 4th, 1999 11:43 PM
Just get the lizard done! It really doesn't matter whether people have lost faith or not, whether the recent spate of setbacks have caused us to falter or to strengthen our resolves. One thing remains true: WE'RE ALL EAGERLY AWAITING THE RELEASE OF MOZILLA.
With regards to helping development, I would glady join in if I could figure out easilly how each component works. Please teach us how to use the "interesting" componenents "as if I was a 4-year old". Contributions to Open-Source development starts with an "idea" one person would like to contribute. How can anyone come up with an idea if s/he can't even begin to make sense of all your docs!?
#2 Re:MozillaZine Readers Sound Off
by Niels L
Monday April 5th, 1999 12:18 AM
Hundreds would have worked on it, if GPL licensed. So obvious, do it now please.
#3 Re:MozillaZine Readers Sound Off
by Niels L
Monday April 5th, 1999 12:19 AM
Hundreds would have worked on it, if GPL licensed. So obvious, do it now please.
what's so holy about the GPL?
I highly doubt that making Mozilla GPL-compliant at this time would change anything, nor would it have if it were released GPL-compliant to begin with.
In order to attract developers you need to give them a reason to come aboard--as JWZ said in his sign-off message.
Methinks the GPL-happy fanatics of Slashdot need a dose of reality..to me the NPL is just fine. I'm sure it's fine to many others out there as well. The GPL is NOT the magic cure-all for everything, nor is it the open source beacon certain types pretend it is. The QPL hasn't scared developers away from KDE (again, contrary to Slashdot opinions), and I'm sure that if Mozilla 1.0 were out today, millions would be using it happily the NPL notwithstanding.
There's another point to consider: there are already a lot of developers in the open source world that are already quite busy working on other things...like KDE or Gnome. Not everyone out there is inclined to be a developer, and those that are are more likely to be involved in other things at the moment.
If it was under GPL the people working in KDE or GNOME or other GPLed software could use it, thus improving it where they feel. There is currently an engine for kfm and another for the gnome-help browser. These desktop environments could benefit from a fully standard compliant browser engine. The mozilla project would attract more developers, and possibly ship sooner than later. I see no need for KDE or GNOME to write their own engines if an exellent one is available to them.
But if mozilla were to be GPL'ed it would turn off all the other developers. NPL takes the best of both worlds and we are seeing Windows, Mac and Unix developers working on it. The GPL only attracts linux users (what about the bsd, irix, solaris, hp-ux, sco, tru64 and all the other unices let alone windows, mac, BeOS, os/2, amiga and who knows what else)
#8 Re:MozillaZine Readers Sound Off
Monday April 5th, 1999 9:47 AM
The Chrome idea is a good one, but unlikely to happen. I cannot see anybody building a browser handing over the control of the interface to whoever has a website and wants to create a better visual and navigational effect.
I think it would only work if there were strict rules for what your code can do, and if your code does not meet those rules, then nothing will happen with the chrome when the website is loaded. The last thing a user needs is a fu**ed up, buggy, GUI to pop up out of the blue when they are browsing.
Strict rules, and an all or nothing approach to actually displaying the chrome whould be the way to go. I personally would take advantage of it.
think of it like this: Webmaster: "I want people to think my site is cool so they'll come often" Visitor: "hmm this site really is cool. I'll go there often" Webmaster: "thanks, here's a skin for my site that you can try if you like" Visitor: "thanks-now this site is even cooler" So you see there has to be that give and take in order for it to work.
#10 Re:MozillaZine Readers Sound Off
Monday April 5th, 1999 10:11 AM
Also, do not forget the ability to create web-based applications with fully-functional UIs (that don't require HTML, DOM, and JS coding). We hope to implement our site administration using XUL when it is completed.
#11 Re:MozillaZine Readers Sound Off
by Robin Colgrove <email@example.com>
Monday April 5th, 1999 11:34 AM
Oliver White asks us to help by finding and complaining about bugs. For non-professional programmers, there is a threshhold effect the software must cross in order not to take to much time away from other tasks. I tried to take mozilla's advice and use the m3 release of apprunner as my usual browser but had to get rid of it after it crashed my powerbook twice requiring a re-install of the system before it would even boot. I am willing to put some effort into hammering out the bugs in new software but I draw the line at programs that trash my system folder. Theres' too little time in the day already. I will try again with m4.
#12 Re:MozillaZine Readers Sound Off
by Chris Knoll <firstname.lastname@example.org>
Monday April 5th, 1999 11:34 AM
The configurable UI through XUL is great, even if you can't download it off the web automatically, make it like a skin for winAmp. Also, I think once a product is released that eveyrone agrees is an excellent one, developers will be more inclined to take a look at the code and offer enhancements. I think JWZ was a bit off saying that there should have been more outside development support, I think after everyone sees the product, and sees that they can make it even better, people will be all for that, only time will tell!
#13 Re:MozillaZine Readers Sound Off
by Mark Treiber
Monday April 5th, 1999 11:40 AM
The only reason that the GPL is often equated with linux is because the GNU system in linux based. But if you look at GLIB which is part of the basis for GTK, it is designed to be cross-platform across all unixes. Granted the most popular systems for GPL tend to be either linux or *BSD based.
#14 Re:MozillaZine Readers Sound Off
Monday April 5th, 1999 12:03 PM
oh btw it would be very nice if we had alot more "skins" even right now before the beta comes out.
#15 Re:nice if we had alot more "skins"
by basic <email@example.com>
Monday April 5th, 1999 2:37 PM
I would like it too, but XUL is still subject to changes. And the XML parser has not been fully tested yet. Most of these problems (and more) are targeted for M4, but there doesn't seem to be any indication that they can meet it. Anyone want to prove me wrong?
#16 Re:MozillaZine Readers Sound Off
Monday April 5th, 1999 6:15 PM
Bah, we don't need the Greedy Public License. The NPL and MPL are far better.
#17 Re:MozillaZine Readers Sound Off
by paul <firstname.lastname@example.org>
Monday April 5th, 1999 8:38 PM
After spending some time browsing Mozilla's code, trying to do some hacking on different levels I'm not happy with the results.
It's not quite correct to call the Mozilla project "Open Source".
The level of _internal_ BS is much higher comparing with any other Open Source project.
If you want people to help you:
Publish the dataflow. Yes, Mozilla's dataflow is not perfect yet, Cut & Paste code resides in ... hmm... critical places of the system.
But it'l be honest and helpful to publish it. It'l be *much* more efficient than asking people to spend their free time hacking yet another heap of undocumented code full of hacks and temporary solultions.
"'Hacking Mozilla' tutorial" is not the tutorial. It's XPCOM propaganda. BTW - some places are handy, even Doczilla failes on it's own samples. It's not 'pre-Alpha' stage. Stop BS.
What to do to get more help from people who are not @netscape.com:
Finish good and scalable design ... not finished yet... You are a bit late with XPCOM... scalability was the priority - not the option ...
2. Some day ( say when )
Publish API's, marking hacks and temporary solutions with some special color.
3. Always. Keep key-developers...
4. Don't know when.
Assign one person who will answer to the particular design requests from the outher space and publish the e-mail of that person. It should be his *responcibility* to answer ( and publish answers widely ) questions like ( simplified ):
"In the existing design parser is registered as a stream listener. what is the best way to register some custom filter for pre-processing the input stream before invoking parser on it?"
There still is some chance that Mozilla would not die, but Netscape should make some real and serious steps in helping developers from the 'outher space'.
<insert parser image here>
#18 Re:MozillaZine Readers Sound Off
Monday April 5th, 1999 10:55 PM
I think the web-based aps with fully functional UIs is genius. Are we talking in addition to the browser UI, replacing it, or merely using the webpage for a UI kind of like Netcenter and Webmail? Did you guys abandon the Webtop thing or is that the basis for all this?
Mozilla won't DIE. Good God, AOL may be many things but poor they are NOT and dependent on this browser they most definitely ARE.
#19 Re: MozillaZine Readers Sound Off
by Nicka <email@example.com>
Monday April 5th, 1999 11:27 PM
OK, So here goes, I read these pages using my nightly build of Mozilla (Post M3) and thats how I am posting it too. I started working on the XPCOM stuff recently. and there *is* stuff that people can work on without understaning the entire code-base (I sure as hell don't yet) - Start off working on things internal to one project and slowly branch out. OK it's not perfect, but it's always better to do something than whine on.
That's a little more like it.
As for me, since I'm not a C++ wiz, I'm grateful for the nightly builds and the fact that most of them run and are stable.
#21 Re:MozillaZine Readers Sound Off
by Chris Knoll <firstname.lastname@example.org>
Tuesday April 6th, 1999 9:06 AM
Robin Colgrove, I'm not sure how putting M3 on your system could thrash your os (I assume you are running windows of some sort), no files go into the system directory, and there is nothing stopping you from just deleting the directory that you unzipped the M3 into to remove it COMPLETELY from your systems (I don't think any registry settings are even set). I think if you had problems with your system crashing, I think your problem lies elsewhere.
#22 Re:MozillaZine Readers Sound Off
by Keith Hankin <email@example.com>
Tuesday April 6th, 1999 11:48 AM
For the past two weeks I have been downloading the nightly build for Linux, and get the same problem. It apparently requires a libunicharutil.so library. Does anyone know how I can get this? Has anyone gotten Mozilla to work on Linux?
#23 Re:MozillaZine Readers Sound Off
by Robin Colgrove <firstname.lastname@example.org>
Thursday April 8th, 1999 11:30 AM
Thanks to Chris Knoll for replying to my message. Let me provide a little more detail. In my lab I run both MacOS8.5 and ppcLinux,2.2.1 kernel. I was running M3 on my G3 Powerbook. I had not been able to crash this machine no matter what I did until trying to use apprunner. Then, I got a repeated crash when trying to access various pages. These crashes required a hard reboot and when restarting, the Powerbook could not read the system file on the internal HD. I had to boot from a CD, replace the system folder and then everything was OK again. Foolishly, I tried using apprunner again and got the same result. That's when I gave up on M3. My more general point is that developmental software has to reach a threshold of stability before general testing so that it does not take unacceptable amounts of time away from our Real Jobs or Real Lives. I do not know how this crash was caused but I did want to alert people to the potential problem. I am hoping M4 will be more stable.