| |||||||||||||
|
| |||||||||||||
Too short!Too short!Posted Dec 31, 2002 3:51 UTC (Tue) by dskoll (subscriber, #1630)Parent article: Red Hat errata policy updates and product end of life 12 months is way too short. I have clients still running RH6.2 quite happily. It must be at least 24 months, preferably 36 months. And yes, I do actually buy official Red Hat CD's.
(Log in to post comments)
Too short! Posted Dec 31, 2002 5:10 UTC (Tue) by aturner (guest, #4037) [Link] I think what they're basically trying to do is get people like your clients to switch to advanced server which has longer life times. Honestly, considering what they're asking for non-advanced server, I can't blame them. Keeping all those packages up-to-date on all those different versions must of been a real PITA.
Too short! Posted Dec 31, 2002 10:18 UTC (Tue) by AlanCox (guest, #4858) [Link] Nothing stops other people maintaining old releases for longer. Unlike proprietary software you have all the pieces. My guess is that if sufficient people want 6.2 after all that time, you'll find people making a business out of supporting it.Alan
Too short! Posted Dec 31, 2002 12:41 UTC (Tue) by RobSeace (subscriber, #4435) [Link] True... We have at least one machine still happily running RH5.2, whichhas been unsupported for ages, now... But, did that mean we're screwed, and forced to upgrade? No, absolutely not... It just meant we couldn't just look to RH to provide us with neatly packaged RPMs to keep stuff up to date... We can still do the work ourselves, if necessary... At some point, the effort of doing so may outweigh the effort of upgrading the system, at which point we may do so... But, until then, there's nothing wrong with happily chugging along with an ancient, unsupported system...
Too short! Posted Dec 31, 2002 14:22 UTC (Tue) by jwharmanny (guest, #971) [Link] The only real problem I see is security updates. Old distro's come with old packages. All major software (kernel, Apache, SSH, etc.) have several vulnerabilities which are only resolvable by upgrading the respective applications and libraries to newer versions. Upgrading the whole distro would be much easier then upgrading those major apps by hand.Besides, would it be fair to expect a distributor to support software which is deprecated for a long time by their original authors?
Too short! Posted Dec 31, 2002 14:33 UTC (Tue) by dskoll (subscriber, #1630) [Link] > Nothing stops other people maintaining old releases for longer.You are absolutely correct. So tell me again why I should buy official Red Hat CD's instead of just downloading ISO's? It's a dumb business decision.
Too short! Posted Dec 31, 2002 14:39 UTC (Tue) by tjc (guest, #137) [Link] So tell me again why I should buy official Red Hat CD's instead of just downloading ISO's?No one is telling you why you should do anything. The reason I buy Red Hat CDs is to support the contributions they make to the free software community. Buying an RHN account is another good way to do this.
Too short! Posted Dec 31, 2002 16:54 UTC (Tue) by dskoll (subscriber, #1630) [Link] No one is telling you why you should do anything.Nevertheless, from a business point of view, one reason I do buy Red Hat CD's is to make sure Red Hat is viable in order that it can continue to provide support. It's not the only reason, but it's an important one, and if Red Hat lessens its support, it means that reason is less of a reason now. As for saying "find someone else to support the software", I think that's not very realistic. In spite of widely-available source, Red Hat is still in the best position to support its own products. Inevitably, a company builds up internal knowledge about its systems, and Red Hat could probably fix a problem with RHL much more cheaply and more quickly than an outside person. I'm the author of a few free software projects, and I know that I can fix things in my software much more quickly than anyone else, because I'm past the learning curve. I urge Red Hat to reconsider this decision, and to provide support for 24 months.
Too short! Posted Dec 31, 2002 18:01 UTC (Tue) by tjc (guest, #137) [Link] As for saying "find someone else to support the software", I think that's not very realistic. In spite of widely-available source, Red Hat is still in the best position to support its own products. Inevitably, a company builds up internal knowledge about its systems, and Red Hat could probably fix a problem with RHL much more cheaply and more quickly than an outside person.I'm the author of a few free software projects, and I know that I can fix things in my software much more quickly than anyone else, because I'm past the learning curve. Yes, exactly. A lot of people don't realize what goes into maintaining software, especially someone elses software. It's in some ways a lot more work than developing it in the first place. There are a couple of things I don't like about sawfish, but after looking at the code for a few hours I realized that it would probably be easier to start from scratch than to try to figure out how it works. That, and my lisp skills are pitiful. I urge Red Hat to reconsider this decision, and to provide support for 24 months. It might be in their best interest not to provide extended support. Support takes resources, but it doesn't generate any revenue other than good will. It's likely that the drop in good will be less than the savings in support costs. I think most of the people who post messages like "that's it, I'm not using Red Hat any more!" probably weren't using it in the first place. Trolls, for the most part. People like you and me will probably continue to use it despite any misgivings (although I have recently developed a rather acute addiction to apt-get).
Too short! Get real Posted Jan 2, 2003 23:19 UTC (Thu) by smoogen (subscriber, #97) [Link] 6.2 was released in the spring of 2000. That means the product was supported for 3 years when it is 'discontinued'. While that seems like a short time, it is 12 development cycles in Red Hat time. Most of the items on it are not maintained at that level upstream and there isnt a yearly revenue stream of at least $400,000 to pay for the engineers that would be needed to keep the product up. That 400,000 would pay for about 2-3 engineers, 1-2 qa people at $60,000 on average salary. The rest would get soaked up in insurance for both company and people... plus all the other overhead costs...
|
|
||||||||||||