When I tested it, it was with the original source on Solaris. I used some of the university's Ultra1 boxes (~200 mhz, 128MB RAM), and spread the source across 1 local and 3 NFS mounted disks (using symlinks).
The compilation took 1+ hour, nearly 1 GB of disk space, and the resulting executable size was ~100MB.
In short, mozilla.org is failing to follow one of free software's most powerful and effective maxims: release early; release often. Just get the browser out there.
The main reason is that the prize for joining is too high. First of all, you need nearly 1 GB of free disk space and waste amounts of time to compile this thing. Second, the development process is too complicated. You just don't use five minutes to fix a small bug in a system with 100 full time developers. When you submit it, they have probably rewritten the entire module. Of course you could try to keep track of what all of them are doing, but that requires too much time.
In emacs I can just find a bug in CC-mode, fix it and submit the fix right away (except for the fact that I can't find a bug in CC-mode (except java-mode, which is crappy)). I don't have to recompile anything. In mozilla, I find a bug, I try to fix it, but I can't figure out where it is, or what will happen if I fix it, or what other things I will break.
The correct approach would be a more kernel oriented design,
And BTW, Javascript is a crappy, and not general enough, extension language.
Netscape did actually create a revolution. The web is largely a result of their ideas back in 1992-1995.
Unfortunately, since then they have done nothing innovative. It seems like they created the web, was pleased and spent the rest of their years parching Navigator.
What they should have done was to keep supporting and extending Navigator, but focus on the design of WWW2.
In addition to your points, you have the following:
Mozilla is just not an innovative product. What you get is basically a Mosaic clone with a couple of thousand of bugs to fix. It is not the way I want the future to be at all.
It requires nearly 1 GB of hard drive space for compilation. Personally, I do most of my development on the university, where resources is limited. The times I compiled it, I had to use several/tmp directories which worked after doing a lot of symlinking.
It is not an easy project to work on. Most of the small parts (like the libs, expat and stuff) works, it's the glue that has to be patched. That requires that you know a lot about all the stuff that should be glued together.
Testing has been hard. The project is not build with a minimal working kernel on which features is connected. It is a mess of features which is patched together.
I guess all of that above is not quite true, but that's the way I see it, and that's why I have made just minor contributions.
You won't. Unless they release just one or two working versions to prevent us from writing our own browser, then destroys the whole open source threat by removing IE, thus removing our access to the common web.
Releasing Mozilla 5.0 is not enough. A couple of weeks after the Mozilla release, Microsoft will release IE 6.0, containing all Mozilla features, and breaking a few new standards in ways Mozilla can't handle. Then we're truly fucked. Mozilla 6.0 will, if it ever is released, be released a couple of years later than that, probably in time for IE 8.0.
This should serve as a wakeup call not for AOL but for the free software community; if we don't contribute, we're never going to have another decent browser to use.
That's right. The whole open source movement is actually in danger.
Unless Mozilla 5.0 get released (or Opera does something drastic), there will be no MSIE competitor on the Windows platform. That means that Microsoft controls the web. And when there's no Windows Netscape users to protest, and the trial stuff is gone, they will have no problem shaking us off and make most of the web directly inaccessible for us.
Who, other than me and other hardcore techies, will use free operating system when that means that they will loose access to the web?
I guess what really happened was that Netscape tried to beat Microsoft in their own game. NS 2.0 to 4.5 has been far from inovative, thei're just patched feature-filled bloatware.
What was really needed (but that's a luxury Netscape can't afford now) was to rethink the whole browser stuff and build another browser built on a better concept, not just a new set of features. That would have been both good for the world and the best way to beat Microsoft. MS have always been concept followers, feature leaders.
Although it's too late for Netscape, it might be time for someone else. That's our only chance. If that does not happen, I truly believe Linux and FreeBSD will be finished as end user desktop systems.
Methinks I'll go convert all my old Office documents into RTF.
That's not a good idea. Currently, RTF is yet another undocumented format. Try PDF, ps, text, some open SGML/XML DTD or another open format instead. Personally, I would go for PDF.
Does mormons have several gods?
on
Gaming on Linux
·
· Score: 1
It is normally 4 or 5 guys in white. They have name signs with "dr." in front of their names. They act strangely too. When I let them in, one of them runs straight towards the window, blocking my entire view and refuses to move away. He says something like "for my own safety" or something. Guess he means that it's a sin to watch the world from the 4th. floor or something. I don't really understand that one, but the guys are completely nuts after all.
They are drug addicts to. And violent. Normally, all but one of them attack me, and try to hold me while the last one tries to force me to take narcotics, saying that "it will make me feel better".
Fortunately, the "coffee" will work by then. They just run out, throwing up all the way. Some guy even jumped out of the window once.
cat core | h_analyze -verbatim | ffi -english
on
Gaming on Linux
·
· Score: 1
I think your HUM2x14 coprosessor instruction detector is defect. Usually, it can be fixed by connecting both EAR1x units to a power outlet, and keep the connection until you start to smell burned meat. If that doesn't help, I think your only hope is that "lobotomy" thing. That's pretty dangerous, but a life without a working HUM2x14 is simply not worth living.
Actually, when I think about it, it could be that the whole thing is caused by you FFI beeing low on power. I would check that out before running any inreversible bugfix.
What is this "family" stuff anyway? A lot of people saying that they are my "family" wants to meet me for no apparent reason. I just assume that they're from Jehovas Vitnesses, give them coffee mixed with Thinner for Correction Fluid (that's what the bottle says), and tell them to try next door instead. It's strange, but they actually keeps coming back, sometimes bringing men dressed in white clothing. Those times, I have usually have to apply my collection of LARTs on them to make them go away.
I didn't quite understand the phrase "nothing better to do than sit in front of a computer all day". Isn't that kind of like "compile a compiled language program with a compiler"
Forced absence from the computer && linux games
on
Gaming on Linux
·
· Score: 2
About the "getting away from the computer" problem
People keeps telling me that I need to get away from my computers sometimes. Unfortunately, they are completely right. I need to get away from the computer to eat, go to the bathroom, and sometimes when I'm tired and don't want to wake up with a face looking even more like a chessboard than it did this morning. It could be worse, though. If the keyboard had been designed so I actually woke up with QUERTY on my face, it would be far more difficult to lie to people, telling them that my face looks so weird because I had a shaving accident.
All this is of course due to major design flaws:
First of all, "tired" is implemented in such a way that it is sometimes impossible to compute the correct time to consume the complete collection of energy drinks. What I want is a hose from the computer right into my weins, so that it can fill me up with caffeine when I stop typing for a long period.
Second, the complete approach of eating is wrong. We go away from the computer, find some food which we insert through a large opening. Some totally inpredictable time later, we just have to move to a unit called the bathroom to get rid of all the unneccesary stuff we accidently pushed in through the hole in our face in the first place. Why do we push unneccesary stuff through that hole? Why do we at all let the body, which is our network connection and firewall, do such work as separating useful stuff from garbage? What we need is a separate external unit, which processes food collected in ways we, as end users, don't want to know about, then inserting the resulting pure fuel into us through some secure channel.
Taking care of that, we don't really need to get away from the computer. Now that that's sorted out, we can work further on the sleeping problem. The problem is that, while the computer can detect sleep and wake us up, it will still not prevent us from spending the time required for this process pressing our faces into the keyboard.
The solution to this problem is quite obvious. To prevent us from pressing our faces into the keyboards, we just need to put something soft between the face and the keyboard. Normally, objects matching the "soft enough" criteria already is there already; Our hands. If we just could prevent them from slipping away when we fall asleep, all our problems would be solved. Those of you that didn't instantly thought "use a binary key system requiring only ten keys, and stapple the fingers to they keys" need to upgrade your brains to at least version 2.14.122.
Wow. When someone fixes that food processing unit, then the most urgent problem in the world is actually solved! Now I just have to convince the university not to force me to move down there and sit in a room in 8 hours several times each year, and convince my cand scient cooperators to meet using IRC or news instead of RL.
About Linux games
What we need is not linux games. What we need is Unix/X11/GLX games. Fixing hardware and GLX support in XFree86 would be the most important step towards linux gaming.
And to you Playstation junkies; I have a computer capable of running games. Why should I use a lot of money getting another unit doing that work, when I could buy some MIT Press books and an Communications of the ACM subscribtion for those money instead?
From CSvax:pur-ee:inuxc!ixn5c!ihnp4!houxm!mhuxi!eagle!m it-vax!mit-eddie!RMS@MIT-O
From: RMS%MIT-OZ@mit-eddie Newsgroups: net.unix-wizards,net.usoft Subject: new UNIX implementation Date: Tue, 27-Sep-83 12:35:59 EST Organization: MIT AI Lab, Cambridge, MA
Free Unix!
Starting this Thanksgiving I am going to write a complete Unix-compatible software system called GNU (for Gnu's Not Unix), and give it away free to everyone who can use it. Contributions of time, money, programs and equipment are greatly needed.
...the rest of the message cut away to save space
I guess this proves that GNU was an operating system project. Unfortunately, the GNU kernel (named HURD) never got finished. That's the reason why GNU currently only is known as the "support software" for Linux. While the thruth is, as RMS says, that Linux just replaces the missing kernel in the GNU operating system.
The naming is equivalent to what would happen if I wrote a new libpng, and then linked it with GIMP and called the complete thing "MacroPaint".
Why should we get more people to use Linux? And are the reasons important enough that we should lie to people and pretend they're not clueless?
I think linux has enough users for a while. Currently Linux has a lot of potential for incompability. Take a look at GGI and stuff. Does that compile on FreeBSD, Solaris, Irix?
Compability and following standards is a far more important issue than open source. We must not sacrify the first one for the second.
I don't think fame is the reason Richard Stallman keeps telling people to call Linux GNU/Linux.
First of all, the FSF is the initiator of the GNU/Linux project, so they have the right to name it whatever they want.
Secondary, I think the reason for repeating the GNU/Linux phrase is that RMS don't want the GNU project to drown in the current Open Source Hype Wave. GNU is near equivalent with the FSF, the political organization. It is the political message, not the GNU software, which is FSFs and RMSs preferred product. The GNU project was, AFAIK, not only an Operating System project, but also the horse in front of the FSF wagon.
Now, the Open Source people are disconnecting the wagon, and keeps selling the Open Source idea. The GNU/Linux phrase is RMSs way of connecting the wagon to the horse.
Personally, I don't understand the Open Source people. They are selling Open Source big, but it doesn't look to me like they care about product quality, Right Ideas[tm], political messages or anything. So why do they sell Open Source? Do they just want to be part of something big? The "Hey, ma, I'm in the history books" attitude?
I think that might be the case. If you ask a standard Linux luser what kind of application to write, one that makes you work more effective and helps software evolution, or one who will kill all resistance and double the number of Linux users, he would go for the latter.
I like Richard Stallmans attitude; "It's better to go in the right direction than to go somewhere fast".
Slashdot Mirror
When I tested it, it was with the original source on Solaris. I used some of the university's Ultra1 boxes (~200 mhz, 128MB RAM), and spread the source across 1 local and 3 NFS mounted disks (using symlinks).
The compilation took 1+ hour, nearly 1 GB of disk space, and the resulting executable size was ~100MB.
The main reason is that the prize for joining is too high. First of all, you need nearly 1 GB of free disk space and waste amounts of time to compile this thing. Second, the development process is too complicated. You just don't use five minutes to fix a small bug in a system with 100 full time developers. When you submit it, they have probably rewritten the entire module. Of course you could try to keep track of what all of them are doing, but that requires too much time.
In emacs I can just find a bug in CC-mode, fix it and submit the fix right away (except for the fact that I can't find a bug in CC-mode (except java-mode, which is crappy)). I don't have to recompile anything. In mozilla, I find a bug, I try to fix it, but I can't figure out where it is, or what will happen if I fix it, or what other things I will break.
The correct approach would be a more kernel oriented design,
And BTW, Javascript is a crappy, and not general enough, extension language.
Netscape did actually create a revolution. The web is largely a result of their ideas back in 1992-1995.
Unfortunately, since then they have done nothing innovative. It seems like they created the web, was pleased and spent the rest of their years parching Navigator.
What they should have done was to keep supporting and extending Navigator, but focus on the design of WWW2.
In addition to your points, you have the following:
I guess all of that above is not quite true, but that's the way I see it, and that's why I have made just minor contributions.
You won't. Unless they release just one or two working versions to prevent us from writing our own browser, then destroys the whole open source threat by removing IE, thus removing our access to the common web.
Releasing Mozilla 5.0 is not enough. A couple of weeks after the Mozilla release, Microsoft will release IE 6.0, containing all Mozilla features, and breaking a few new standards in ways Mozilla can't handle. Then we're truly fucked. Mozilla 6.0 will, if it ever is released, be released a couple of years later than that, probably in time for IE 8.0.
That's right. The whole open source movement is actually in danger.
Unless Mozilla 5.0 get released (or Opera does something drastic), there will be no MSIE competitor on the Windows platform. That means that Microsoft controls the web. And when there's no Windows Netscape users to protest, and the trial stuff is gone, they will have no problem shaking us off and make most of the web directly inaccessible for us.
Who, other than me and other hardcore techies, will use free operating system when that means that they will loose access to the web?
I guess what really happened was that Netscape tried to beat Microsoft in their own game. NS 2.0 to 4.5 has been far from inovative, thei're just patched feature-filled bloatware.
What was really needed (but that's a luxury Netscape can't afford now) was to rethink the whole browser stuff and build another browser built on a better concept, not just a new set of features. That would have been both good for the world and the best way to beat Microsoft. MS have always been concept followers, feature leaders.
Although it's too late for Netscape, it might be time for someone else. That's our only chance. If that does not happen, I truly believe Linux and FreeBSD will be finished as end user desktop systems.
It is normally 4 or 5 guys in white. They have name signs with "dr." in front of their names. They act strangely too. When I let them in, one of them runs straight towards the window, blocking my entire view and refuses to move away. He says something like "for my own safety" or something. Guess he means that it's a sin to watch the world from the 4th. floor or something. I don't really understand that one, but the guys are completely nuts after all.
They are drug addicts to. And violent. Normally, all but one of them attack me, and try to hold me while the last one tries to force me to take narcotics, saying that "it will make me feel better".
Fortunately, the "coffee" will work by then. They just run out, throwing up all the way. Some guy even jumped out of the window once.
I think your HUM2x14 coprosessor instruction detector is defect. Usually, it can be fixed by connecting both EAR1x units to a power outlet, and keep the connection until you start to smell burned meat. If that doesn't help, I think your only hope is that "lobotomy" thing. That's pretty dangerous, but a life without a working HUM2x14 is simply not worth living.
Actually, when I think about it, it could be that the whole thing is caused by you FFI beeing low on power. I would check that out before running any inreversible bugfix.
People keeps telling me that I need to get away from my computers sometimes. Unfortunately, they are completely right. I need to get away from the computer to eat, go to the bathroom, and sometimes when I'm tired and don't want to wake up with a face looking even more like a chessboard than it did this morning. It could be worse, though. If the keyboard had been designed so I actually woke up with QUERTY on my face, it would be far more difficult to lie to people, telling them that my face looks so weird because I had a shaving accident.
All this is of course due to major design flaws:
First of all, "tired" is implemented in such a way that it is sometimes impossible to compute the correct time to consume the complete collection of energy drinks. What I want is a hose from the computer right into my weins, so that it can fill me up with caffeine when I stop typing for a long period.
Second, the complete approach of eating is wrong. We go away from the computer, find some food which we insert through a large opening. Some totally inpredictable time later, we just have to move to a unit called the bathroom to get rid of all the unneccesary stuff we accidently pushed in through the hole in our face in the first place. Why do we push unneccesary stuff through that hole? Why do we at all let the body, which is our network connection and firewall, do such work as separating useful stuff from garbage? What we need is a separate external unit, which processes food collected in ways we, as end users, don't want to know about, then inserting the resulting pure fuel into us through some secure channel.
Taking care of that, we don't really need to get away from the computer. Now that that's sorted out, we can work further on the sleeping problem. The problem is that, while the computer can detect sleep and wake us up, it will still not prevent us from spending the time required for this process pressing our faces into the keyboard.
The solution to this problem is quite obvious. To prevent us from pressing our faces into the keyboards, we just need to put something soft between the face and the keyboard. Normally, objects matching the "soft enough" criteria already is there already; Our hands. If we just could prevent them from slipping away when we fall asleep, all our problems would be solved. Those of you that didn't instantly thought "use a binary key system requiring only ten keys, and stapple the fingers to they keys" need to upgrade your brains to at least version 2.14.122.
Wow. When someone fixes that food processing unit, then the most urgent problem in the world is actually solved! Now I just have to convince the university not to force me to move down there and sit in a room in 8 hours several times each year, and convince my cand scient cooperators to meet using IRC or news instead of RL.
About Linux gamesWhat we need is not linux games. What we need is Unix/X11/GLX games. Fixing hardware and GLX support in XFree86 would be the most important step towards linux gaming.
And to you Playstation junkies; I have a computer capable of running games. Why should I use a lot of money getting another unit doing that work, when I could buy some MIT Press books and an Communications of the ACM subscribtion for those money instead?
Somehow I don't think RMS cares what the press likes. He may make himself unpopular, but a lot of people will know a lot more about FSF/GNU.
I hope RMS won't kill me for citing him :)
I guess this proves that GNU was an operating system project. Unfortunately, the GNU kernel (named HURD) never got finished. That's the reason why GNU currently only is known as the "support software" for Linux. While the thruth is, as RMS says, that Linux just replaces the missing kernel in the GNU operating system.
The naming is equivalent to what would happen if I wrote a new libpng, and then linked it with GIMP and called the complete thing "MacroPaint".
Why should we get more people to use Linux? And are the reasons important enough that we should lie to people and pretend they're not clueless?
I think linux has enough users for a while. Currently Linux has a lot of potential for incompability. Take a look at GGI and stuff. Does that compile on FreeBSD, Solaris, Irix?
Compability and following standards is a far more important issue than open source. We must not sacrify the first one for the second.
I don't think fame is the reason Richard Stallman keeps telling people to call Linux GNU/Linux.
First of all, the FSF is the initiator of the GNU/Linux project, so they have the right to name it whatever they want.
Secondary, I think the reason for repeating the GNU/Linux phrase is that RMS don't want the GNU project to drown in the current Open Source Hype Wave. GNU is near equivalent with the FSF, the political organization. It is the political message, not the GNU software, which is FSFs and RMSs preferred product. The GNU project was, AFAIK, not only an Operating System project, but also the horse in front of the FSF wagon.
Now, the Open Source people are disconnecting the wagon, and keeps selling the Open Source idea. The GNU/Linux phrase is RMSs way of connecting the wagon to the horse.
Personally, I don't understand the Open Source people. They are selling Open Source big, but it doesn't look to me like they care about product quality, Right Ideas[tm], political messages or anything. So why do they sell Open Source? Do they just want to be part of something big? The "Hey, ma, I'm in the history books" attitude?
I think that might be the case. If you ask a standard Linux luser what kind of application to write, one that makes you work more effective and helps software evolution, or one who will kill all resistance and double the number of Linux users, he would go for the latter.
I like Richard Stallmans attitude; "It's better to go in the right direction than to go somewhere fast".