Wine is chasing a moving target with insufficient resources to ever catch it. This it is always in a state of almost, but not quite, running a lot of things yet never getting over the hump before the thing in question is obsoleted by new thing that almost but not quite runs.
Wine started out to clone Win16. Assuming anyone cared, is there a non-trivial Win16 app that runs on it now? Photoshop, Quicken, Office, Wordperfect? No, because the action moved to Win95/NT's Win32 before Win16 ever stabilized. Wash, rinse, repeat as the action moved to supporting WinXP applications. Because of the unusual length of that product's lifespan things got much closer and several large applications were actually usable, especially if you sprung for Crossover. But Microsoft got back in the rewriting the API game pretty big time with Vista & 7 so XP support will never quite get finished because the action has shifted back to taillight chasing.
I remember this story.... DOSEmu reached 1.0 just in time to ship in exactly ONE RedHat major release cycle before being dumped for lack of interest. Now you can't even run it anymore since modern kernels don't even include the support for running 8086 code. And once you go 64bit it is a hardware limitation.
FreeDOS released a 1.0 version. Yea, nobody except a few industrial control people cared by the time it happened.
ReactOS is trying to make a clone of NT/XP. And have been for well over a decade without producing anything approaching something useful. Pretty safe bet the same will remain true as they celebrate their twentieth anniversary.
Haiku is closing in on a decade of effort to clone BEOS to a stable release. But hey, they are self hosting and might actually get a 1.0 in time to celebrate their tenth anniversary. BEOS was great tech.... for the 1990's. But how much multimedia are you doing restricted to VESA video modes or running in an emulator.
Linux could do what it did because of the long toil of the GNU Project to have a full userland and development environment ready for it to complete.
Think the lesson to be drawn is that unless you have a plan to have massive resources to bring to bear at a big cloning project, don't bother as you will just get a lot of folks worked up, many will spend a lot of effort and in the end it ends in fail.
If a major player were to throw a thousand third world developers at it, Wine would succeed in a year or two. Success defined as having a top ten list of major current Windows apps running as reliable as they do on Windows and 90% of random apps installing and running even if with imperfections. Problem is the ongoing effort to keep up with something like Windows is probably beyond what volunteers or a small company like CodeWeavers are likely to muster so unless a fairly large revenue stream was found it would only be a temporary success.
Posted Nov 5, 2010 9:29 UTC (Fri) by mlankhorst (subscriber, #52260)
[Link]
Yes wine will not be able to run all programs perfectly, it's still useful though. 16-bits support is still functional last time I checked, and even some dos applications might still be supported..
Posted Nov 8, 2010 8:04 UTC (Mon) by nix (subscriber, #2304)
[Link]
DOSEmu reached 1.0 just in time to ship in exactly ONE RedHat major release cycle before being dumped for lack of interest.
And now we have DOSBox, which can run most things anyone cares about on DOS anymore (old games) under emulation accurate enough that even a lot of 1990s-era demos work.