Actually, I was planning to buy (yes, purchase, pay money for) CodeWarrior from the instant I heard that they were porting it to linux. I have used CodeWarrior on the Mac for years (it's NOT unstable, I don't know where the hell you got that from) and I'd say it's the best, most well rounded IDE I have ever used.
but I cannot tolerate either running a Red Hat system (way too unstable, in my experience with 5.2), or the idea that CW will only work on Red Hat linux. Does this mean that you can only compile apps with it that run on red hat linux?
Distribution exclusivity is unnacceptable. I wish I could get CodeWarrior for linux, but it doesn't exist.
I can only think of two reasons that would definitely force me to switch from RedHat: (a) if they pull some technical crap that makes their distro clearly inferior to others (b) if they pull some marketing crap that tries to dominate the distro market not on technical merit but incompatible lockins.
Well, then it's time to switch because 1. they have pulled some technical crap making their distro clearly inferior to others (see linuxconf), and 2. they have been pulling massive marketing crap to dominate the market. I tried Red Hat, found it too unstable for my needs, and switched to (eventually) SUSE. SuSE is what Red Hat wishes it was - I hope Red Hat copies everything they do, so at least the market leader can also be the technological leader.
Actually, I used slack for a long time before I decided to try Red Hat, and loved it. Slack is great, but the times I tried using the RPM retrofitted onto it, things didn't work (probably due to some sort of libc/glibc conflict, but at the time I wasn't aware enough to realize that). I was using a libc based system, and had a significant amount of trouble getting various packages to compile and run properly on my system (all too often,./configure; make; make test; make install didn't work without changing scripts or actual code).
So I was, after a while, sick of having to spend hours finding and downloading an obscure library slackware didn't include that some other distro did, and sick of fixing things in programs that didn't work with my system for some reason or another (when the make process for a package I really wanted came up with a bunch of compile time errors (not warnings), I could often fix them with 2 to 7 hours of effort (average)). I heard about the holy grail of typing "rpm -i package.rpm" and having it just work immediately sounded pretty damn cool. And I knew that my experience of using rpm on slackware was less than perfect, so I wanted to get a real Red Hat system - they were, after all the market leader, and I wanted them to be the best. So I got a copy from a friend, and installed it.
I later regretted this move. For one thing, Red Hat uses the init scripts with the folder for every runlevel, and S09SERVICE links and such. I know some people like it, and I even know that my SuSE system I use now uses it by default, but one of the things I really didn't like about red hat at first was the switch from slackware's simple single level rc.d directory with a series of text files to initialize different parts of the system.
The next big problem was linuxconf. I don't know how it's version number got to the 0.9x's, because the version I used was not, in my opinion, suitable for an alpha quality release. Practically _NONE_ of the functions I tried to use to administrate my system worked. I'm talking important, basic things like adding users and groups, changing network settings, etc... I couldn't believe Red Hat would put such a useless tool in the center of it's admin interface. I wanted to use shell scripts - and I'm sure there probably were some I could use - but the names I knew from slackware didn't work, and I didn't know where else to look (All my linux knowledge came from slackware experience at that point). Then RPM came crashing down - sure, it was nice to be able to type "rpm -i gimp-0.91.rpm", but that lost it's appeal when the latest development versions failed to appear as RPMs. Sure, I could do a plain old source compile and install, but that would break my system's rpm-purity. It was an annoying feeling to have to worry if rpm's database was up to date in addition to worrying if my actual system was up to date. I couldn't believe rpm didn't have some graceful way to cover this situation (typing --nodeps for the rest of your system's life is NOT graceful), at least not a well documented one. I read everything I could find, even a large portion of Maximum RPM, but no solution presented itself. RPMs were not so cool after all, I had to start doing source compiles. I am not the type of linux user who is interested in only using old stuff.
SRPMS didn't work for me at ALL, by the way.
When I needed to install a newer gimp, turned out I also had to install a newer GTK. And this newer gtk conflicted with the rpm gtk I had installed (no rpm available for the new one, of course), so I had to -erase the old one. Installing the compiled gtk did not stop the warnings when I tried to install any package that depended on gtk. This pattern occured with many a package, and I got accostumed to typing --nodeps. What the hell was the point of having rpm if I didn't use it's dependencies? My system's combination of rpm-based programs and compiled programs resisted eachother and I eventually couldn't deal with their fighting, the linuxconf fiasco, my printing problems, and my general feeling that Red Hat had just downloaded a bunch of crap off the net and put it on a cd without testing it even once from a new user's perspective. I missed slackware. But I felt like trying other distros.
To make a long story short (I'm late for class), SuSE had the best combination of actually working as advertised (try YaST! it is excellent compared to every other (linux) admin tool I've tried) and having a higher level of control than slackware.
Oh shut up. This is a damn fine way to do research for a paper. I can't think of a more technically literate community he could have tapped into. Obviously he has to form his own opinion to write a paper at all. There is more honesty and unbiased (read: un-bought) opinion in one day's worth of slashdot than six months of news.com
Infospinner has done the research for you. On the main page of their website they reference this article from pcweek that talks about 20 other companies doing clustering solutions before infospinner is even mentioned. I can't believe they are this stupid. I wish there were criminal penalties for abusing the patent system like this (and I wish patent officials could be held responsible for their incompetency - even notary publics are held to be responsible)
>As for IIS being faster than Apache, of course it >is. We all know the performance differences which >happen when using fork/exec as opposed to >spawning a new thread. There are two remedies for >this performance difference. The first is to >rewrite the Apache code to run threaded (not a >particularly trivial task).
Hey - here's a third solution: READ THE APACHE MANUAL, understand that there are supposed to be x servers lying around at any given time (See the Min/MaxSpareServers config option), and if you are going to have a high traffic site, make sure that there are enough spare apache instances to handle the amount of connections you expect. If there were 60 MinSpareServers in their "fair test" on the linux machine, all the fork/exec'ing would be done at the beginning and there would be NO OVERHEAD WHATSOEVER from spawning new processes, because they would be spawned before testing (to have this work effectively, you also have to disable the limit of n requests per child, which is not a bad thing on linux).
The people who did these tests are A. incompetent at apache administration, or B. intentionally distorting their results. If anyone disagrees, please explain why.
Don't worry, public code is still The Answer. Simply because it evolves quickest. There is a slowing and maybe even a stopping of development as every project nears maturity - and for the wider category of computer users, their needs are going to be satisfied in time. Time and the free software movement will eventually limit programming to specialization of existing software to fit a particular set of needs - it's already hard to think of a software project that hasn't been done, or had significant work put into it. It is usually easier to use what's already available for free than to write it yourself, unless you are doing research or learning. In any case, I don't know about you, but all this free power that the Open Source Founders (tm) gave to me leaves me with an overwhelming sense of gratitude, and I fully intend to give back as much as I possibly can, in advocacy and in documentation and coding. There are lots of people like me out there, and as long as that is true, it is ok that a lot of computer users are people like you, because there is plenty of time, and the nature of open source is to only allow positive progression. GPL is a coercive license, but every program that's GPL'd is that way because the guy who put the original work into the program wanted it that way, and it's up to him. If you don't like that, too bad. Write a clone of it your damn self, or go to the store and pay money for an inferior binary-only commercial program.
Taxing Internet commerce will be fairer overall, be extremely efficient (since it is, by necessity, automated!), and allow lowering of other taxes. I mean, the government has to collect as much money as it needs anyway, and it is not as if the Internet needs tax breaks in order to grow.
The government is already getting far more than it needs. and I am not seeing any lowering of other taxes. Don't hold your breath, it ain't gonna happen. The government is a glutonous monolith absorbing rights and power from citizens of all types, one group at a time. It can't afford to ever lower taxes long term at it's current rate.
the only question is - where is critical mass - how much are americans going to put up with ?
It is wrong for a government to hold a gun to my head (or a jail sentence, same thing) and say "pay up". If I don't want to pay for a federal service, I should lose my right to benefit from it, but I should not have to.
I believe governments should be profitable. If government provides a necessary service, people will be willing to pay for those services on a more granulated scale, eh? If not, fuck it - we don't need it anyway.
Slashdot Mirror
It should be illegal to make more than one law at a time. Can I sue them for wasting my money?
Actually, yes, that _was_ censorship.
I'd like to see a politician-burning trend start...
Hell, I wish Sun could have hired OJ himself to take care of Mircosoft.. I'm sure the lawers would be useful later, though.
Try SuSE: get rpms AND stability!
Actually, I was planning to buy (yes, purchase, pay money for) CodeWarrior from the instant I heard that they were porting it to linux. I have used CodeWarrior on the Mac for years (it's NOT unstable, I don't know where the hell you got that from) and I'd say it's the best, most well rounded IDE I have ever used.
but I cannot tolerate either running a Red Hat system (way too unstable, in my experience with 5.2), or the idea that CW will only work on Red Hat linux. Does this mean that you can only compile apps with it that run on red hat linux?
Distribution exclusivity is unnacceptable. I wish I could get CodeWarrior for linux, but it doesn't exist.
I can only think of two reasons that would definitely force me to switch from RedHat: (a) if
they pull some technical crap that makes their distro clearly inferior to others (b) if they pull some marketing crap that tries to dominate the distro market not on
technical merit but incompatible lockins.
Well, then it's time to switch because 1. they have pulled some technical crap making their distro clearly inferior to others (see linuxconf), and 2. they have been pulling massive marketing crap to dominate the market. I tried Red Hat, found it too unstable for my needs, and switched to (eventually) SUSE. SuSE is what Red Hat wishes it was - I hope Red Hat copies everything they do, so at least the market leader can also be the technological leader.
So I was, after a while, sick of having to spend hours finding and downloading an obscure library slackware didn't include that some other distro did, and sick of fixing things in programs that didn't work with my system for some reason or another (when the make process for a package I really wanted came up with a bunch of compile time errors (not warnings), I could often fix them with 2 to 7 hours of effort (average)). I heard about the holy grail of typing "rpm -i package.rpm" and having it just work immediately sounded pretty damn cool. And I knew that my experience of using rpm on slackware was less than perfect, so I wanted to get a real Red Hat system - they were, after all the market leader, and I wanted them to be the best. So I got a copy from a friend, and installed it.
I later regretted this move. For one thing, Red Hat uses the init scripts with the folder for every runlevel, and S09SERVICE links and such. I know some people like it, and I even know that my SuSE system I use now uses it by default, but one of the things I really didn't like about red hat at first was the switch from slackware's simple single level rc.d directory with a series of text files to initialize different parts of the system.
The next big problem was linuxconf. I don't know how it's version number got to the 0.9x's, because the version I used was not, in my opinion, suitable for an alpha quality release. Practically _NONE_ of the functions I tried to use to administrate my system worked. I'm talking important, basic things like adding users and groups, changing network settings, etc... I couldn't believe Red Hat would put such a useless tool in the center of it's admin interface. I wanted to use shell scripts - and I'm sure there probably were some I could use - but the names I knew from slackware didn't work, and I didn't know where else to look (All my linux knowledge came from slackware experience at that point). Then RPM came crashing down - sure, it was nice to be able to type "rpm -i gimp-0.91.rpm", but that lost it's appeal when the latest development versions failed to appear as RPMs. Sure, I could do a plain old source compile and install, but that would break my system's rpm-purity. It was an annoying feeling to have to worry if rpm's database was up to date in addition to worrying if my actual system was up to date. I couldn't believe rpm didn't have some graceful way to cover this situation (typing --nodeps for the rest of your system's life is NOT graceful), at least not a well documented one. I read everything I could find, even a large portion of Maximum RPM, but no solution presented itself. RPMs were not so cool after all, I had to start doing source compiles. I am not the type of linux user who is interested in only using old stuff.
SRPMS didn't work for me at ALL, by the way.
When I needed to install a newer gimp, turned out I also had to install a newer GTK. And this newer gtk conflicted with the rpm gtk I had installed (no rpm available for the new one, of course), so I had to -erase the old one. Installing the compiled gtk did not stop the warnings when I tried to install any package that depended on gtk. This pattern occured with many a package, and I got accostumed to typing --nodeps. What the hell was the point of having rpm if I didn't use it's dependencies? My system's combination of rpm-based programs and compiled programs resisted eachother and I eventually couldn't deal with their fighting, the linuxconf fiasco, my printing problems, and my general feeling that Red Hat had just downloaded a bunch of crap off the net and put it on a cd without testing it even once from a new user's perspective. I missed slackware. But I felt like trying other distros.
To make a long story short (I'm late for class), SuSE had the best combination of actually working as advertised (try YaST! it is excellent compared to every other (linux) admin tool I've tried) and having a higher level of control than slackware.
How do you make that a hotlink? Check out http://www.w3.org/MarkUp/
Oh shut up. This is a damn fine way to do research for a paper. I can't think of a more technically literate community he could have tapped into. Obviously he has to form his own opinion to write a paper at all. There is more honesty and unbiased (read: un-bought) opinion in one day's worth of slashdot than six months of news.com
Infospinner has done the research for you. On the main page of their website they reference this article from pcweek that talks about 20 other companies doing clustering solutions before infospinner is even mentioned. I can't believe they are this stupid. I wish there were criminal penalties for abusing the patent system like this (and I wish patent officials could be held responsible for their incompetency - even notary publics are held to be responsible)
>As for IIS being faster than Apache, of course it >is. We all know the performance differences which >happen when using fork/exec as opposed to
>spawning a new thread. There are two remedies for >this performance difference. The first is to >rewrite the Apache code to run threaded (not a >particularly trivial task).
Hey - here's a third solution: READ THE APACHE MANUAL, understand that there are supposed to be x servers lying around at any given time (See the Min/MaxSpareServers config option), and if you are going to have a high traffic site, make sure that there are enough spare apache instances to handle the amount of connections you expect. If there were 60 MinSpareServers in their "fair test" on the linux machine, all the fork/exec'ing would be done at the beginning and there would be NO OVERHEAD WHATSOEVER from spawning new processes, because they would be spawned before testing (to have this work effectively, you also have to disable the limit of n requests per child, which is not a bad thing on linux).
The people who did these tests are A. incompetent at apache administration, or B. intentionally distorting their results. If anyone disagrees, please explain why.
Oh come on, when has MICROS~1 ever do anything to promote their products that __DIDN'T__ insult your intelligence as a professional?
Don't worry, public code is still The Answer. Simply because it evolves quickest. There is a slowing and maybe even a stopping of development as every project nears maturity - and for the wider category of computer users, their needs are going to be satisfied in time. Time and the free software movement will eventually limit programming to specialization of existing software to fit a particular set of needs - it's already hard to think of a software project that hasn't been done, or had significant work put into it. It is usually easier to use what's already available for free than to write it yourself, unless you are doing research or learning. In any case, I don't know about you, but all this free power that the Open Source Founders (tm) gave to me leaves me with an overwhelming sense of gratitude, and I fully intend to give back as much as I possibly can, in advocacy and in documentation and coding. There are lots of people like me out there, and as long as that is true, it is ok that a lot of computer users are people like you, because there is plenty of time, and the nature of open source is to only allow positive progression. GPL is a coercive license, but every program that's GPL'd is that way because the guy who put the original work into the program wanted it that way, and it's up to him. If you don't like that, too bad. Write a clone of it your damn self, or go to the store and pay money for an inferior binary-only commercial program.
Taxing Internet commerce will be fairer overall, be extremely efficient (since it is, by necessity, automated!), and allow lowering of other taxes. I mean, the
government has to collect as much money as it needs anyway, and it is not as if the Internet needs tax breaks in order to grow.
The government is already getting far more than it needs. and I am not seeing any lowering of other taxes. Don't hold your breath, it ain't gonna happen. The government is a glutonous monolith absorbing rights and power from citizens of all types, one group at a time. It can't afford to ever lower taxes long term at it's current rate.
the only question is - where is critical mass - how much are americans going to put up with ?
Actually, taxes ARE evil.
It is wrong for a government to hold a gun to my head (or a jail sentence, same thing) and say "pay up". If I don't want to pay for a federal service, I should lose my right to benefit from it, but I should not have to.
I believe governments should be profitable. If government provides a necessary service, people will be willing to pay for those services on a more granulated scale, eh? If not, fuck it - we don't need it anyway.