GordonGR asked a good question which I split into its own thread.
]]>ConnorBehan wrote:Which makes sense because a lot of people use binary blobs that would depend on all the lib32 stuff if they used 64 bit Arch. @levir3: My guess (based on how friggin easy it is to support i686 if you already support x86_64) is that i686 support will go on for another 10 years.
Ok, so it ended up being more like 4.5 years.
Cue long sessions pondering about the temporal and fleeting nature of software and life itself....
And this is a pretty deep necrobump, really.
]]>Which makes sense because a lot of people use binary blobs that would depend on all the lib32 stuff if they used 64 bit Arch. @levir3: My guess (based on how friggin easy it is to support i686 if you already support x86_64) is that i686 support will go on for another 10 years.
Ok, so it ended up being more like 4.5 years.
]]>I Not believe in anythink of the dev say in this thread, I believe in Fact and mailisting the fact say: 'Maintain i686 for a maximum of 25 years more'
You need to look up the word "fact". Connecting the concept of "fact" with something somebody said on the internet in the same sentence is like saying: "I never had problems with suspend after a kernel update."
]]>ngoonee wrote:nomorewindows wrote:The design philosophy of Linux is to make it work on just about any hardware. If this distro doesn't have maybe another has or can be easily ported. Just because another distro is dropping i686 support makes it a good idea.
Have you read the discussion or are you simply responding to a thread title?
For one thing, the last line is supposed to read Just because another distro is dropping i686 support doesn't mean it is a good idea. I kept rewriting it, and still couldn't get it right.
probably /sarcasm is not used here /sarcasm-fin
I Not believe in anythink of the dev say in this thread, I believe in Fact and mailisting the fact say: 'Maintain i686 for a maximum of 25 years more'
]]>nomorewindows wrote:The design philosophy of Linux is to make it work on just about any hardware. If this distro doesn't have maybe another has or can be easily ported. Just because another distro is dropping i686 support makes it a good idea.
Have you read the discussion or are you simply responding to a thread title?
For one thing, the last line is supposed to read Just because another distro is dropping i686 support doesn't mean it is a good idea. I kept rewriting it, and still couldn't get it right.
]]>The design philosophy of Linux is to make it work on just about any hardware. If this distro doesn't have maybe another has or can be easily ported. Just because another distro is dropping i686 support makes it a good idea.
Have you read the discussion or are you simply responding to a thread title?
]]>ngoonee wrote:SanskritFritz wrote:After all one of the key points in using arch is that it is so lean that it is feasible to run on old hardware.
I must have missed that memo. Old/slow hardware was never really the target (check Puppy Linux out, that's more like what you're referring to).
Oh come on. I didn't say it was a target, just that it is a nice side effect of the simplicity arch provides. Don't be a nitpick.
Not trying to be a nitpick, sorry, but 'key points' is stretching it (and misleading to the less well-informed).
It IS a very nice side-effect (and I take advantage of that myself on some older machines) but that's as far as it goes. I'd be more than happy to see i686 support properly maintained, but with most of the devs (not TUs though, I think I remember quite a few TUs mentioning 32-bit much more frequently) on 64-bit the days are likely numbered.
Maybe Arch should totally drop 32-bit support by the beginning of next month. Or haven't we done that before?
]]>Gusar wrote:I don't think there's really a need for a special "32bit maintainer". I'd have no problem if the devs do away with the need to sign off on 32bit packages, and move the package from testing to core at the same time as the 64bit package. If a bug crops up, that's what the bugtracker is for. Yeah, this would mean that 32bit folks are perpetually beta testers and not "users", but I'd be fine with that. Much better than needing to search for a different distro. I like Arch very much and only one of my machines is 64bit capable.
Allan wrote, that there were some 32 bit related bugs already. Ok, you report, and test, but how would the problem be fixed?
...
You're mistaking Gusar for someone who doesn't know how to fix bugs .
]]>Let me reiterate - there are no plans to drop i686... But we will likely reduce the requirement for package signoffs for i686 (officially... unofficially we already have). When that happens, it will be clear that x86_64 is our primary architecture and i686 will be secondary. That brings potential for other secondary architectures...
]]>SanskritFritz wrote:After all one of the key points in using arch is that it is so lean that it is feasible to run on old hardware.
I must have missed that memo. Old/slow hardware was never really the target (check Puppy Linux out, that's more like what you're referring to).
Oh come on. I didn't say it was a target, just that it is a nice side effect of the simplicity arch provides. Don't be a nitpick.
]]>After all one of the key points in using arch is that it is so lean that it is feasible to run on old hardware.
I must have missed that memo. Old/slow hardware was never really the target (check Puppy Linux out, that's more like what you're referring to).
]]>I don't think there's really a need for a special "32bit maintainer". I'd have no problem if the devs do away with the need to sign off on 32bit packages, and move the package from testing to core at the same time as the 64bit package. If a bug crops up, that's what the bugtracker is for. Yeah, this would mean that 32bit folks are perpetually beta testers and not "users", but I'd be fine with that. Much better than needing to search for a different distro. I like Arch very much and only one of my machines is 64bit capable.
Allan wrote, that there were some 32 bit related bugs already. Ok, you report, and test, but how would the problem be fixed? Otherwise, I agree, dropping 32 bit would be undesirable with so many old machines still working. After all one of the key points in using arch is that it is so lean that it is feasible to run on old hardware.
]]>