Slashdot Mirror
WineConf 2005 Sets Deadline for Wine 0.9
IamTheRealMike writes "As WineConf 2005 took place last weekend, the Wine developers discussed the project's direction and future. A new deadline of September 30th for the 0.9 release was set by Alexandre Julliard -- the release promises to bring an end to the system of monthly snapshots and provide a new focus on ease of use and stability. A new GPLd application regression testing tool called CXTest was demoed, as was some of the great Direct3D work being done lately. Finally the CEO of Gupta gave a talk. Gupta have ported their 4GL RAD tools to Linux by working with Codeweavers (who sponsored the event), and their experiences were documented in a fascinating presentation. Overall: big thumbs up, but it's not obvious enough that there's enterprise-level support available for Wine. Check out the group photo and the new Wiki!"
Just curious if something like this would be possible, i.e. to have a plugin to VisualStudio.NET that will notify at the time of writing the code whether a certain function exists inside the Wine compatibility layer. The thought behind this would be to get into developers' heads the idea that either not being lazy or rather putting in a bit more effort at the time of writing the code would ensure compatibility down the line and open up more avenues of possibility for their application.
You phrase that is if you think it's wrong that Windows programmers write application without having Wine in mind.
Designing software around a compatibility layer intended to emulate your actual target system is idiotic. You design for a platform, not an emulation of one. If you want your application to run on multiple platforms, then write it for multiple platforms.
Deadlines encourage the development of critical components that are the most needed at the time. While this can lead to bad things if the time constraints are unrealistic, it's better than open-ended all-over-the-map kind of work you'd get otherwise.
I guess this is because it's very important to set goals. While the deadline in itself is kind of annoying, one can afford to wait for the perfect software while it has a (better) running version.
I like wine. I really do, I've been using it for over a year now and have never had many real (unsolveable) issues from the few win games I own or any other win32 app I feel like running under it.
My only true complaint with Wine is how 2/3 of the apps that do work give weird errors and require a lot of tweaking before they actually execute correctly. Its not that I seriously mind messing with the wine config files or spending the time to do so, but it would just be a whole lot easier if it worked by default. Some of the smaller bugs I've found havent been solved over the course of the last year, and I can only hope this new system of deadlines encourages more rapid development to fix the little things those before any new work is done in other areas like directx or 3d support
~You laugh because I'm different, I laugh because I'm insane~
Like debian, wine has been far too long without a stable release. By all means don't rush it, but you do need to actually release a working version once in a while. At the moment any stable version of wine is stable purely through luck - they're just cvs snapshots.
I am trolling
Well, it's just the way you write for multiple platforms in pretty much every language. For example, when writing C code you have to have in mind where it's going to run. If you call mmap you have to know that it doesn't exist on Windows, for instance.
A tool that would allow me to specify a list of systems I want to be compatible with and then warns about things that wouldn't work on some of them would actually be very handy.
The acronym serves to point out that Wine is not a computer emulator. It is an emulation of the Windows platform, though.
If you're writing with a non-complete, fairly incompatible approximation of a platform in mind, then you're a fool. You can draw as many flawed analogies about POSIX C libraries as you like - it just isn't sane thinking when you could target a cross-platform toolkit like wxWidgets.
Wine is extremely useful for running applications which never had non-Windows systems in mind. That's what it's for. Don't mistake it for a target system. I don't even see why you need somebody to tell you why that's a bad idea.
Is the performance target for Wine 1.0 that any (and I mean *any*) Win32 program will run on Linux?
And, if we deleted the word "cold" from the dictionary, we wouldn't have need for central heating anymore. ;-)
It's not deadlines that are toxic to commercial endeavors -- it's deadlines * promises, driven by the imperative to capture customer before the competition does. If you promise more than your competitors, but too far in the future, then you have no competitive advantage. If you promise too much less than your competitors, but sooner, then you have no competitive advantage.
The ideal is to have much more than your competition, much sooner. The only way to do that is to be more productive than average.
Since mathematically everyone can't be above average, the typical experience is that you end up with an unfinished pile of shit when your deadline rolls up. Deleting some of the pile simply leaves you with a smaller pile of shit. So the only solution is to let the deadlines slip.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
The critical difference is who sets the deadline. A bunch of developers saying "we want to have it done by dd/mm/yy" may be good motivation. A manager or salesman coming in and saying "it will be done by now()" is not - most of us have probably had to deal with people like that at some point. It's not good.
The deadlines we set ourselves are hopefully sensible ones.
Aide-toi, le Ciel t'aidera - Jeanne D'Arc.
With all of the changes coming in Longhorn it is gonna be interesting to see how long it will take for WINE to gain parity.
The have a lot of API to implement.
So the only solution is to let the deadlines slip.
Or ship the product as is. At least that's what Microsoft does.
This is not a troll or flamebait. I really wonder how much longer the concept of running Windows applications on Linux will be relevant? WINE has been a slow moving product. I remember playing around with it ten years ago and to this day most applications fail to run properly under WINE. This isn't a statement about the quality of the development team but rather the difficulty in porting the libraries from a closed source OS that is continuously changing.
It just seems to me that Linux is coming into it's own. How much longer will it be before many programs are released for Windows and Linux? As more and more applications are released for both platforms WINE becomes less and less important.
I haven't purchased a new Windows program for some time now as Linux comes with everything I need for my day to day work. It would be nice to be able to run some of my older stuff under Linux but that's mostly games. I've supported WineX or whatever they are calling their product now days for years with monthly donations in the hopes that it will become very useful but still there are only a handful of games that run perfectly under WineX. Most don't run at all.
I'm thinking that rather that rewrite the Windows libraries we should concentrate on libraries like SDL that can be installed on both Windows and Linux thus giving the developer a uniform API that can targeted both OSes.
The race isn't always to the swift... but that's the way to bet!
BTW, Lame Ain't Mp3 Encoder, but it is, my head explodes!-)
Escher was the first MC and Giger invented the HR department.
...so ironically enough, people will move to Linux+wine instead as the path of least resistance :).
Which would be the case if the new APIs mattered at all to existing applications, which they don't -- they will continue working just as well as they always have with the existing ones. Microsoft is quite anal about this. So no need to emulate Longhorn until apps actually depend on it, which won't be till some time after it's released, I think.
Work is punishment for failing to procrastinate effectively.
I think there's three reasons, two of which are undeniable. First, there's a lot of legacy apps out there that are unsupported and won't be updated because the vendor went out of business. Personally I support 3 applications like that. So it becomes a matter of finding a native replacement which may not exist or be worth switching too if works with Wine. Second, there's a phenomenal amount of software that's been created for Windows. All kinds of odd little apps for doing things like interior design and such that don't exist on Linux. Finally, and this is the item a lot of people will disagree with, many commercial applications are just plain better on Windows. More effort has been put into designing UI's and such. (For the record, I personally use free software always as my first choice based on the principle of free, as in libre, software.)
----- obSig
I'm with you, man. I work at a library with an aging WinNT population (Thanks to the B&M "First Hit Is Always Free" Foundation) that would probably go catatonic if they were "upgraded" to XP. From what I understand, they're trying to raise funds to get newer machines to replace the current ones (PIII 600Mhz, 256MB RAM).
;-)
Basically, what I was wondering was what the arguments were to push for Linux where you are. I'm sure Linux would do wonders with our aging hardware here and possibly save the place some money in hardware upgrades as well as maintenance costs. However, the MS mentality is prevalent and probably difficult to sway.
I guess the questions go:
1) How'd you get Linux in past the bureaucracy?
2) What distro and what do you do for maintenance?
3) How do the patrons react? (that whole "learning curve" argument)
4) Why not cold-turkey to OO.org? Public demand for the popular and familiar hasn't stopped the government from making unpopular decisions before.