Thank you. This objective look at things is a breath of fresh air compared to the RedHat haters and Gentoo zealots in this thread. If anything ever deserved +5 informative, this did.
Funny, I have been updating a Fedora laptop using up2date and a server using yum and I have not had any problems. Mind you, I don't live in fscked up networks that need an HTTP proxy.
Or, you could continue to get updates from both free and paid sources for older RedHat versions if Fedora isn't quite mature enough for you yet.
Or you could sit there and bitch. But I guess that is what these RedHat stories are for - so people can piss and moan about a company that has done more for Linux than pretty much everyone who posted here combined.
You should by able to track any of Stable, Testing, Unstable or Libranet's sources. I run a mix of Libranet (2.8) sources and Unstable, and it works fine for me. A friend did a dist-upgrade to Unstable and that worked ok as well.
Libranet has a pretty good community that should be able to give you a hand with your sources.list as well.
Why don't you try installing a Debian-based distro with a decent installer? Like, say Libranet. The installer and XAdminmenu will get you up and running with a nice Debian desktop system.
My 20 month old loves to play with the Duplo type bricks. They are large and don't represent a choking hazard at all. We have been playing with them for about a year - at first he only liked breaking the stuff I built, but now he likes building things with me too.
Some time after he turns 3 I plan on breaking out my huge stash of Lego I had when I was little and introducing him to those:)
"There is no job that is America's God-given right anymore," Carly Fiorina
That's right Carly, so why don't you move all of HP to India or China? Oh, right, you want to take advantage of American laws, tax breaks, security and live your American executive lifestyle. But it is ok slash R&D spending and screw your engineering talent.
We'll see where HP is after a decade of firing/offshoring R&D and hiring more lawyers and bean counters. Carly sure understands what made HP into HP.
And, in putting in on the Internet, the resource is available for use.
Yes, and if I put an HTTP server on the internet, I am putting up a resource the serves HTML over port 80. I am not provding a resource for you to exploit a buffer overflow, get r00t on by box, dump the database an get CC numbers (insert how stupid it is to have CC numbers on a public server here).
The parent poster is right. Hack your own web server installation. Full disclosure of the exploits. But don't break into my system - it is tresspass and it is not legal or moral.
the world i live in is actually a world that practices this quite well. we do all of our design here in the UK and develop a lot in South Africa. the world i live in, however, is an XP world.
So you don't just write the spec and pass it on then - you work on an iterative process with the developers. This was my point - that is the only way outsourcing works, no matter where you are working. I also am guessing that South Africa is either another office of your company or another established and competent company that you do business with.
the author seems to be under the impression that the success and innovation of a product is purely in the hands of a bunch of software developers. this is rubbish. innovation in the software industry is also about building a product to solve a particular problem - and well
What article did you read? He explicitly states that design is most of the cost for software.
if the functionality is well designed (say with some good interaction design) by a US-based company, the specifications can be written up in the US and sent to the Indian shop for authoring
Sorry, but this is a recipe for disaster and it doesn't save you much, as the author points out. I don't know what world you live in, but the reality is that nobody writes up a perfect spec and then someone codes it. People change, requirements change and no spec is perfect.
Have a look at what is working successfully offshore - either companies are setting up entire divisions there (HP, IBM) or they are working with experienced offshore firms that have some understanding of design as well as implementation. Using offshore programmers as a "body shop" for implementaing a spec works about as well as using onshore programmers for the same task - not very well.
I'll second this. It doesn't have to be big projects either, and it isn't just Anderson/Accenture - I've seen it happen with other large consulting companies on smaller projects that could be done with 2 or 3 people in less than 6 months.
The play is always the same - send in the guys in $2000 suits to close the deal and then dump the specfication on clueless new-grad code monkeys. Not only are the coders terribly inexperienced, but they have not been part of the specification process so they have no information to make good decisions or question anything. A few times in my previous job when I cleaned up from such disasters, looking at the code and documentation produced by these people was almost enough to make you cry.
More clued in clients would often do as you suggest - make it a condition that you have to have at least some of the people involved in the specification actually involved in implementation as well.
it is unbelievably disturbing that people like this are placed in charge of leading efforts to make alternatives to windows for non-technical users
ESR isn't in charge of anything of the sort - this is just his opinion on the matter. If anyone is "in charge" of anything, it might be the people running KDE and Gnome. Did you bother to get their opinions on the matter before making your sweeping generalization?
The concept of freely distributable and modifiable code must be seperated from the concept of The Unix Way
Go check up on the number of successful open source Windows on sourceforge and get back to me.
My response: "But Eric, most usability experts recommend you design the interface first and then write the code".
Ever heard of model-view-controller? Those mysterious un-named "experts" are wrong. You can and should design the model before the view/controller. Otherwise your GUI will be built on top of junk.
I was wondering the same thing. I made a trip this summer and I drove the ETR from end to end (and back again later in the weekend). We were doing 140 kph, and since they were taking pictures of our plates so they could mail us the bill it would be pretty easy to deduce how fast we were going by timestamping the pictures.
I was never able to find anything online about restrictions for this data.
At least when I go down and drive the I-90 in New York State from Syracuse to Buffalo I don't have an EZPass so I pay the tolls anonymously in cash:)
that decided to suddenly drop support for their product not even a year later?
I am trying hard to resist flaming, but you are an idiot. For 50th time since this story broke, RedHat announced there would only be one year of support for 9 when it was released. That's right, when it was released.
I don't care how much you and every other "RedHat sucks, RedHat has no integrity" moron weren't paying attention. They clearly announced their intentions, and started promoting the RHEL line at that time.
You are free to disagree with RedHat's move, but all of this anti-RedHat FUD is really disgusting.
This is one of the Evolution testing releases that go along with Gnome 2.5. The goal is a stable Evolution 2.0 and Gnome 2.6 later in the spring. Check out he roadmap.
So by all means, pick up 1.5 if you want to help with bug fixing, but this is not a "stable" release.
I think that most people ship binary modules with the kernel interface distributed as source for this reason. You compile the part that links directly with the kernel and that loads the binary-only portion at runtime.
For example, NVidia and VMware do this, and I have never had problems with any kernel.org or vendor (RedHat, Debian) kernel using these modules.
Of course, typically the binary portion is shipped for an i386 architecture only, which could be another problem.
It's not like they would lose money by giving away the source to the driver -- they'll sell the hardware anyways, won't they!?
Typically the source either contains "intellectual property" that the company does not want to GPL (sometimes this is silly, but a PHB makes the decision anyway) or it contains licensed 3rd party "intellectualy property" that legally cannot be GPLed.
Personally I prefer GPLed modules, but given a choice between binary-only and nothing, I choose binary-only. However, I understand and respect the opposite opinion.
Redhat wants us to develop and test fedora for free, turn around sell it to enteprise for big bucks.
RedHat gives you Fedora for free. They sink lots of resources into working on Fedora, not to mention bandwidth for hosting, mailing lists, etc. If you want you can help by creating packages or filing bug reports. How exactly is this a bad thing?
I guess what I'm saying is that I don't want to develop for free.
Then don't. Nobody is asking you to do anything. Many of us write open-source software on our spare time because we have an itch to scratch or we want to give something back. Other people write docs, file bugs reports and do packaging for the same reasons.
What about a model where OSS developers get paid for the quality of code they check in?... that would be better than the current state of affairs.
WTF? It would be better than the current state of affairs? I'll give you a clue: there is nothing wrong with the current state of affairs. People are paid to work full time on Open-Source software by companies selling support for it. Other people do it on a volonteer basis for free. What is the problem here exactly?
As a Canadian who apparently knows more about this than you, let me rebut...
the National Post is noted for having a right-wing bias in its editorial section
LOL. Let me guess what kind of bias you have? The National Post might have a more conservative view than the Globe and Mail, but the Globe is so far left of center it is easy to notice.
Second, the piece in the National Post was written by an Earth Sciences professor of a University famously known as "last chance U": Carleton has one of the lowest admission standards in the province of Ontario, which tends to be correlated with the quality of faculty they tend to attract.
This is straight up bullshit. Carleton might have been called "last chance U" by certain other Ontario universities over its arts program admission standards in the 1990s. However, the admission standards for programs like Science, Engineering and Journalisim were never in question. Slandering Carleton's faculty like this shows how truly ignorant you are.
Feel free to argue this paper based on its merits. Trying to besmirch the reputation of the author and the newspaper that this story was from just makes you look like a fool to anyone who knows better.
No, but when 9 was released Red Hat said that they were going to release the "consumer" distro every 6 to 12 months and only support back one release. They also said that they were focusing on the RHEL product and that the "consumer" distro would be a proving ground for RHEL. More recently they announced that they would not be selling retail any more and then that RH10 would morph into Fedora.
If you didn't notice the writing on the wall with the release of RH9 then you weren't paying attention.
Fedora is what RH10 would have been only better - more packges, better repositories, more community oriented. It might not be what you want if you are the original poster in this thread, but you can hardly say that RedHat sprung this on anyone out of the blue.
The nice part is that this is Linux - if you don't like the choice between Fedora and RHEL then choose another distro.
IMHO this is a bad move for RedHat only because of the no advance notice. Had they said this 6 months ago, everyone would be in a better position to deal with it.
You and/or your company are morons. This was announced months ago - back when RH9 shipped in fact. Hell - here is the Slashdot story on the very topic from January 2003!
You don't have to like RH's business model, but don't bad-mouth them for your own incompetence.
So they are going to rely on Volunteers to do the work for them? It seems a little dirty that RH has decided that they want to use the community to provide Fedora -- to maintain this 'farmiliarity' -- but not do it in house.
Have a look at the mailing lists and who is doing the work. RedHat is hosting Fedora and their developers are working on it as well. They are doing it in house, but out in the open and allowing the community to participate in the process.
When RH 9 was released, the details of this were clear: they were going to EOL 7 and 8 in December and support for 9 would only last a year. All that happened more recently is that RH 10 morphed into Fedora. If you bought 9 expecting something else then you are a moron.
If you want 5 years of support then you can buy RH Enterprise, or you can switch to another distro.
Slashdot Mirror
This is different how?
Apple is not a convicted monopolist.
Thank you. This objective look at things is a breath of fresh air compared to the RedHat haters and Gentoo zealots in this thread. If anything ever deserved +5 informative, this did.
Funny, I have been updating a Fedora laptop using up2date and a server using yum and I have not had any problems. Mind you, I don't live in fscked up networks that need an HTTP proxy.
Or, you could continue to get updates from both free and paid sources for older RedHat versions if Fedora isn't quite mature enough for you yet.
Or you could sit there and bitch. But I guess that is what these RedHat stories are for - so people can piss and moan about a company that has done more for Linux than pretty much everyone who posted here combined.
You should by able to track any of Stable, Testing, Unstable or Libranet's sources. I run a mix of Libranet (2.8) sources and Unstable, and it works fine for me. A friend did a dist-upgrade to Unstable and that worked ok as well.
Libranet has a pretty good community that should be able to give you a hand with your sources.list as well.
Why don't you try installing a Debian-based distro with a decent installer? Like, say Libranet. The installer and XAdminmenu will get you up and running with a nice Debian desktop system.
My 20 month old loves to play with the Duplo type bricks. They are large and don't represent a choking hazard at all. We have been playing with them for about a year - at first he only liked breaking the stuff I built, but now he likes building things with me too.
Some time after he turns 3 I plan on breaking out my huge stash of Lego I had when I was little and introducing him to those :)
"There is no job that is America's God-given right anymore," Carly Fiorina
That's right Carly, so why don't you move all of HP to India or China? Oh, right, you want to take advantage of American laws, tax breaks, security and live your American executive lifestyle. But it is ok slash R&D spending and screw your engineering talent.
We'll see where HP is after a decade of firing/offshoring R&D and hiring more lawyers and bean counters. Carly sure understands what made HP into HP.
And, in putting in on the Internet, the resource is available for use.
Yes, and if I put an HTTP server on the internet, I am putting up a resource the serves HTML over port 80. I am not provding a resource for you to exploit a buffer overflow, get r00t on by box, dump the database an get CC numbers (insert how stupid it is to have CC numbers on a public server here).
The parent poster is right. Hack your own web server installation. Full disclosure of the exploits. But don't break into my system - it is tresspass and it is not legal or moral.
the world i live in is actually a world that practices this quite well. we do all of our design here in the UK and develop a lot in South Africa. the world i live in, however, is an XP world.
So you don't just write the spec and pass it on then - you work on an iterative process with the developers. This was my point - that is the only way outsourcing works, no matter where you are working. I also am guessing that South Africa is either another office of your company or another established and competent company that you do business with.
the author seems to be under the impression that the success and innovation of a product is purely in the hands of a bunch of software developers. this is rubbish. innovation in the software industry is also about building a product to solve a particular problem - and well
What article did you read? He explicitly states that design is most of the cost for software.
if the functionality is well designed (say with some good interaction design) by a US-based company, the specifications can be written up in the US and sent to the Indian shop for authoring
Sorry, but this is a recipe for disaster and it doesn't save you much, as the author points out. I don't know what world you live in, but the reality is that nobody writes up a perfect spec and then someone codes it. People change, requirements change and no spec is perfect.
Have a look at what is working successfully offshore - either companies are setting up entire divisions there (HP, IBM) or they are working with experienced offshore firms that have some understanding of design as well as implementation. Using offshore programmers as a "body shop" for implementaing a spec works about as well as using onshore programmers for the same task - not very well.I'll second this. It doesn't have to be big projects either, and it isn't just Anderson/Accenture - I've seen it happen with other large consulting companies on smaller projects that could be done with 2 or 3 people in less than 6 months.
The play is always the same - send in the guys in $2000 suits to close the deal and then dump the specfication on clueless new-grad code monkeys. Not only are the coders terribly inexperienced, but they have not been part of the specification process so they have no information to make good decisions or question anything. A few times in my previous job when I cleaned up from such disasters, looking at the code and documentation produced by these people was almost enough to make you cry.
More clued in clients would often do as you suggest - make it a condition that you have to have at least some of the people involved in the specification actually involved in implementation as well.
Nice strawman.
it is unbelievably disturbing that people like this are placed in charge of leading efforts to make alternatives to windows for non-technical users
ESR isn't in charge of anything of the sort - this is just his opinion on the matter. If anyone is "in charge" of anything, it might be the people running KDE and Gnome. Did you bother to get their opinions on the matter before making your sweeping generalization?
The concept of freely distributable and modifiable code must be seperated from the concept of The Unix Way
Go check up on the number of successful open source Windows on sourceforge and get back to me.
My response: "But Eric, most usability experts recommend you design the interface first and then write the code".
Ever heard of model-view-controller? Those mysterious un-named "experts" are wrong. You can and should design the model before the view/controller. Otherwise your GUI will be built on top of junk.
I was wondering the same thing. I made a trip this summer and I drove the ETR from end to end (and back again later in the weekend). We were doing 140 kph, and since they were taking pictures of our plates so they could mail us the bill it would be pretty easy to deduce how fast we were going by timestamping the pictures.
I was never able to find anything online about restrictions for this data.
At least when I go down and drive the I-90 in New York State from Syracuse to Buffalo I don't have an EZPass so I pay the tolls anonymously in cash :)
eschasi asks: "eschasi recently got a new job..."
George is getting upset!
that decided to suddenly drop support for their product not even a year later?
I am trying hard to resist flaming, but you are an idiot. For 50th time since this story broke, RedHat announced there would only be one year of support for 9 when it was released. That's right, when it was released.
I don't care how much you and every other "RedHat sucks, RedHat has no integrity" moron weren't paying attention. They clearly announced their intentions, and started promoting the RHEL line at that time.
You are free to disagree with RedHat's move, but all of this anti-RedHat FUD is really disgusting.
So is Evolution 1.5 a development release? Are they following the same numbering scheme as the Linux kernel?
Yes and yes.
If you don't want to be testing 1.5 then you should be waiting for a stable 2.0. Of course, if you can, testing 1.5 is a good thing.
This is one of the Evolution testing releases that go along with Gnome 2.5. The goal is a stable Evolution 2.0 and Gnome 2.6 later in the spring. Check out he roadmap.
So by all means, pick up 1.5 if you want to help with bug fixing, but this is not a "stable" release.
I think that most people ship binary modules with the kernel interface distributed as source for this reason. You compile the part that links directly with the kernel and that loads the binary-only portion at runtime.
For example, NVidia and VMware do this, and I have never had problems with any kernel.org or vendor (RedHat, Debian) kernel using these modules.
Of course, typically the binary portion is shipped for an i386 architecture only, which could be another problem.
It's not like they would lose money by giving away the source to the driver -- they'll sell the hardware anyways, won't they!?
Typically the source either contains "intellectual property" that the company does not want to GPL (sometimes this is silly, but a PHB makes the decision anyway) or it contains licensed 3rd party "intellectualy property" that legally cannot be GPLed.
Personally I prefer GPLed modules, but given a choice between binary-only and nothing, I choose binary-only. However, I understand and respect the opposite opinion.
Redhat wants us to develop and test fedora for free, turn around sell it to enteprise for big bucks.
RedHat gives you Fedora for free. They sink lots of resources into working on Fedora, not to mention bandwidth for hosting, mailing lists, etc. If you want you can help by creating packages or filing bug reports. How exactly is this a bad thing?
I guess what I'm saying is that I don't want to develop for free.
Then don't. Nobody is asking you to do anything. Many of us write open-source software on our spare time because we have an itch to scratch or we want to give something back. Other people write docs, file bugs reports and do packaging for the same reasons.
What about a model where OSS developers get paid for the quality of code they check in? ... that would be better than the current state of affairs.
WTF? It would be better than the current state of affairs? I'll give you a clue: there is nothing wrong with the current state of affairs. People are paid to work full time on Open-Source software by companies selling support for it. Other people do it on a volonteer basis for free. What is the problem here exactly?
As a Canadian who apparently knows more about this than you, let me rebut...
the National Post is noted for having a right-wing bias in its editorial section
LOL. Let me guess what kind of bias you have? The National Post might have a more conservative view than the Globe and Mail, but the Globe is so far left of center it is easy to notice.
Second, the piece in the National Post was written by an Earth Sciences professor of a University famously known as "last chance U": Carleton has one of the lowest admission standards in the province of Ontario, which tends to be correlated with the quality of faculty they tend to attract.
This is straight up bullshit. Carleton might have been called "last chance U" by certain other Ontario universities over its arts program admission standards in the 1990s. However, the admission standards for programs like Science, Engineering and Journalisim were never in question. Slandering Carleton's faculty like this shows how truly ignorant you are.
Feel free to argue this paper based on its merits. Trying to besmirch the reputation of the author and the newspaper that this story was from just makes you look like a fool to anyone who knows better.
No, but when 9 was released Red Hat said that they were going to release the "consumer" distro every 6 to 12 months and only support back one release. They also said that they were focusing on the RHEL product and that the "consumer" distro would be a proving ground for RHEL. More recently they announced that they would not be selling retail any more and then that RH10 would morph into Fedora.
If you didn't notice the writing on the wall with the release of RH9 then you weren't paying attention.
Fedora is what RH10 would have been only better - more packges, better repositories, more community oriented. It might not be what you want if you are the original poster in this thread, but you can hardly say that RedHat sprung this on anyone out of the blue.
The nice part is that this is Linux - if you don't like the choice between Fedora and RHEL then choose another distro.
IMHO this is a bad move for RedHat only because of the no advance notice. Had they said this 6 months ago, everyone would be in a better position to deal with it.
You and/or your company are morons. This was announced months ago - back when RH9 shipped in fact. Hell - here is the Slashdot story on the very topic from January 2003!
You don't have to like RH's business model, but don't bad-mouth them for your own incompetence.
I wouldn't consider Fedora to be any more unstable then RH 10 would have been.
Furthermore, if you can't get the $350 together to buy the enterprise version for a work-related project then you have other problems.
So they are going to rely on Volunteers to do the work for them? It seems a little dirty that RH has decided that they want to use the community to provide Fedora -- to maintain this 'farmiliarity' -- but not do it in house.
Have a look at the mailing lists and who is doing the work. RedHat is hosting Fedora and their developers are working on it as well. They are doing it in house, but out in the open and allowing the community to participate in the process.
When RH 9 was released, the details of this were clear: they were going to EOL 7 and 8 in December and support for 9 would only last a year. All that happened more recently is that RH 10 morphed into Fedora. If you bought 9 expecting something else then you are a moron.
If you want 5 years of support then you can buy RH Enterprise, or you can switch to another distro.