I too was very much turned off by the marketing "features" creeping into both netscape and mozilla releases.
Which ``marketing "features"'' crept into Mozilla releases? The only thing I can think of was the keywords stuff, and it was in for about half an hour before the developer in question did The Right Thing and backed it out (pending pref control defaulted to off, etc.)
Well yeah sure. I wish their interface defaulted to a less color oriented theme. Wish that fonts were more happening and consistent with the other browsers as well.
It's important to note that the XPIDL interfaces to which Brendan refers are programming interfaces, not the user interface.
How would it ever cease to be open source? That would be a Neat Trick(tm).
Nobody -- not Netscape, not AOL, not I, not RMS, not the Governor of Rhode Island -- can take away the rights granted to you under the MPL and NPL regarding the Mozilla code.
Seriously, though, we build the system continuously and close the tree daily for verification and detection of regressions. (We publish the builds that are used to verify.
(The Extreme Programming stuff is pretty cool, but it uses the acronym XP, which causes no end of confusion around the Mozilla camp.)
I must correct myself: while OpenVMS uses 64-bit addressing, the compiler is run in a mode that uses sizeof(int) == sizeof(long) == sizeof(void *) == 4.
My offer stands, though: find code that assumes sizeof(void *) == 4 in Mozilla, and I'll fix it for you.
That's a pretty interesting accusation, Dr. Spong. I've not heard any reports of such assumptions in Mozilla code in many months, though there are some problems on Linux/Alpha: lack of -mieee in CFLAGS on systems which aren't correctly detected as Alphas, and some issues which might be related to glibc/pthreads stuff (``CAN'T HAPPEN'' things in pthread_mutex_lock, etc.). These aren't universal problems with the architecture, though, as evidenced by the fact that we have a working M10 build for OpenVMS/Alpha, which is also a 64-bit platform. (The M10 build for that, as well as Linux/SPARC, will be hitting the FTP site shortly.)
If you can find a case of code depending on 32-bit pointer width, please file a bug and Cc: shaver@mozilla.org on it. I will _personally_ repair it, if you don't get rapid response for the owner of the code in question.
You certainly can write XPCOM components in JavaScript, sir. See nsSample.js for the quick sample I provided when I finished enabling them. -- due for an update and better comments shortly. (And you've been able to implement XPCOM interfaces in JavaScript for some months now.)
Slashdot Mirror
Since when is Mozilla not free software?
Nobody -- not Netscape, not AOL, not I, not RMS, not the Governor of Rhode Island -- can take away the rights granted to you under the MPL and NPL regarding the Mozilla code.
Seriously, though, we build the system continuously and close the tree daily for verification and detection of regressions. (We publish the builds that are used to verify.
(The Extreme Programming stuff is pretty cool, but it uses the acronym XP, which causes no end of confusion around the Mozilla camp.)
My offer stands, though: find code that assumes sizeof(void *) == 4 in Mozilla, and I'll fix it for you.
If you can find a case of code depending on 32-bit pointer width, please file a bug and Cc: shaver@mozilla.org on it. I will _personally_ repair it, if you don't get rapid response for the owner of the code in question.
You certainly can write XPCOM components in JavaScript, sir. See nsSample.js for the quick sample I provided when I finished enabling them. -- due for an update and better comments shortly. (And you've been able to implement XPCOM interfaces in JavaScript for some months now.)
Crap, I see this on my Win98 box, too. I'll post more if I figure out where it's failing.
I think that's related to GTK themes, but I'm not 100% certain. I'll check with our GTK wizard (pavlov).