WineConf 2004 Wrapup

IamTheRealMike writes "Well, the attendants are back home and the writeups have been written - WineConf 2004 is over, and Brian Vincent of Wine Weekly News fame has written a comprehensive account of the conference. Wine hackers the world over congregated in snow-covered Minneapolis to talk shop and try and locate the magic bullet to make Wine better, faster. Cheers!"

Believe me, I love it as much as anybody, but. . .

[Speeshus Ness]

We get yet another conference in without dealing with some of the fundamental flaws in the Wine implementation, that if they continue to go unaddressed will get locked into core code and we'll end up with as much of a dog as the system it's trying to replace. I'm speaking of (at least - this is the short list)

-- Thread resolution and collection.
-- Incredibly sloppy video management and screen management code.
-- Conflicts with the kernel in the management of I/O and wait states for unsupported hardware.
-- TCP/IP stack is circa 1993, even Microsquid could do a better job.

I applaud this ra-ra attitude, but unless we're willing to take an *unflinching* look at where we are, Wine is really going to stay as a novelty.

CrossOver

[cozziewozzie]

I wonder how many of the improvements can be attributed to the Crossover code. IIRC, the Crossover people release all their changes back to the WINE tree after a time. IMHO, this is a good example of a company staying alive while helping out the community.

Anyway, running Office smoothly is a great thing. This and Photoshop are two very important steps to getting Linux on more desktops (last time I tried Photoshop, it crashed after a while and Office complained about some access violation).

Re:I got confused...

[MarvinMouse]

...and I thought they were talking about the drink! I was thinking "wine hackers"? Shouldn't that be "sommeliers"? Man, it's still way too early in the day...

Nah, Sommeliers are closer to quality assurance workers, than wine hackers. Wine hackers would be more like people who make their own wine at home, and try to get the alcohol content as high as possible.

Re:Wine and DirectX

[SydShamino]

And to keep people centered on Linux when it comes to gaming, latest DirectX support needs to be a top priority.

Of course, I mean for the Wine folks. For the rest of the Linux community, getting developers to release native Linux games is more important.

Re:Wine and DirectX

[kfg]

And me still playing Grand Prix Legends and Red Baron 3D would be another exception. There may be more of us out that than you realize.

And while implimenting the latest DirectX might well be of some high priority it is inherently impossible to achieve in a timely manner, with regards to people who will only run the latest and "greatest" games. Wine will always be at least a generation behind.

So why not start from the beginning and work up, getting games people already have to run?

My Windows partition exists solely for these games. If DX8 were fully supported I could ditch the thing.

KFG

Re:Wine and DirectX

[Anonymous Coward]

"Centered on linux" amounts to hacking the kernel, or doing cool open source work, or just getting _NOVEL_ shit done using an alternative OS (I would like to say "free software" instead here, but that's not necessarily the case, for better or worse). Honestly, what's the point of keeping people centered on linux by emulating Window$?

I haven't kept up with WinE for a long time, but I do recall it used to be necessary to have a few proprietary binaries for it to work, i.e., it's not even free-as-in-beer-window$-emulation. (This may have since changed, which would be cool; in which case, disregard this)

Now, notice that I'm not attacking the WinE project - it's a great technical exercise and it may be useful (though I really doubt that the manhours saved from applying WinE in productive situations will ever amount to those put into WinE's development, that's cool). It's a natural thing to develop as much cross-platform support as possible; it's one of those geek imperatives.

However, I'm attacking this "we need to emulate windows for the good of linux" mentality. It's worse than the various Gnome/KDE suite apps which apparently try to clone Win-apps down to the annoying "features" (Right now, I'm thinking about gnumeric's limitation to 256x65536 sheets... I don't know why anyone would want something to behave EXACTLY like Excel.) It illustrates an appalling herd-mentality: "We have an alternative OS, but man, it's just not going to be good enough for the world until it copies what everyone's familiar with."

Re:I hate to whine

[caino59]

i'll bite...

this just isn't for joe consumer, but for converting business over to a linux desktop.

most system admins that would love to have linux on their place of employment's desktop will have no trouble setting up wine. all the users have to do is use it, not set it up...

Re:CrossOver ? with MS Blessings...?

[da5idnetlimit.com]

Well, the point had always been that the Office division of MS is a milk cow, System brings quite some money in but is attacked by Linux and XBOX Division (as well as the media division) is loosing money like hell...

Wine continuation just means that when (rofl) Linux has dethroned Windows on the desktop(/rofl) Microsoft can with no problem continue pushing it's Office suite everywhere...

Maybe putting all this devellopment time and brains on OpenOffice / MSOffice compatibility and TheGimp Tools/Dev/Filters would allow us what we really need..a really free, top to bottom OS...with all the goodies softs available for free...

"This and Photoshop are two very important steps to getting Linux on more desktops"

I might be wrong, but I think I'm closer to the mark than you are...

Get Oppenoffice working for cheap AND MS doc compatible (almost totally done), push for the SMEs and Big Companies to get cheaper hardware by getting them Microsoft free and then you will see that Photoshop is announcing a native Linux version by it's nice, userfriendly editor...

When you are at that point, most editors will come and shell out Linux versions... Binaries only, maybe, but Linux versions anyhow...

Wine was all right and fine an idea 2-3 years ago, when Linux didn't fully have the basic apps.

Now that we have them...

There is only two position in IT Market, the best or the cheapest... If we get all for free, editors will try and provide the best for a fee... Or so it was to work ...

Re:Wine still a pain

[stratjakt]

The second and most obvious thing is that because wine exists then less software will be made for linux in the meantime.... this is the biggest barrier to linux taking more market share.

Why? Who cares?

The average user wouldnt care if the app their running was compiled for linux natively, or is being emulated, so long as it works seamlessly.

A WINE that worked, was effortless to install, had a compatibility with XP in the high 90%'s (including the latest DirectX - big issue, games are probably the most cited reason home users stay on windows), would be a huge reason to run linux on the desktop.

Users could simply be told "look, here's this free product, that's more stable, more secure, AND runs all Windows applications to boot". People start using linux, MSFT responds by breaking compatibilities in their next OS, breaking compatibility with XP in the process, pissing people off, driving more to linux..

There are plenty of apps I use that only have Windows versions, and I frankly prefer to any OSS counterparts I've seen. DVD/CD burning software, some dev tools (yeah, I much prefer Visual Studio to anything in the OSS world I've seen)..

And as far as games go, I know there's a handful of linux titles, but by and large there's no comparison.

As I see it, at that point, hardware support would be the biggest (only?) thing holding linux back from the desktops of the world.

That's why it's 0.9

[Weaselmancer]

Plus, Wine is not a product, it's a project. Codeweavers makes a product based on Wine, and so does Transgaming.

Codeweavers product is aimed at people who want to use Linux, but communicate 100% with MS Office people. And use MS plugins in their Linux browsers.

Transgamings product is aimed at the hacker/enthusiast who wants to be on the cutting edge running DirectX games on their Linux install.

Eventually, Wine will be a near 100% replacement for the MS API. Buy a MS piece of software at CompUSA, drop it in your Linux distro, and it works perfectly.

And once that happens, you will see Linux begin to take over the desktop. And that's why Wine developers are heroes. Keep up the good work!

Weaselmancer

PS: The submitter is hoping for the "magic bullet" that'll speed up wine, but may have missed just such a magic bullet in the article he posted. It's a shared memory wineserver, currently experimental. I'll quote from the WineHQ page:

Gav showed a dramatic demo of American McGee's Alice running under both WineX and WineX with shared memory. In that particular game the sound and graphics threads needed to sync with each other at an astounding rate. Typical WineX performance produced about 50 frames per second. By moving to shared memory the framerate nearly doubled to about 95 a second.

same old wine story

[Anonymous Coward]

I've been using Linux for 10 years. Back in the day there was dosemu and wine. The story was that wine was getting close, wine was getting better, wine was getting usable.

Every once in a while I would give wine another try and find that wine was still not working. I don't think very much has changed in 10 years with the wine project.

At this point I have no interest anymore in using wine, I just use linux and native linux apps.

IMHO the wine developers should focus their efforts on linux and native linux applications, and users should request native linux applications from the software vendors.

WINE is not just for "basic apps."

[autechre]

Sometimes you have this old, closed-source 16-bit program (possibly for an old, out-of-date piece of external hardware) that you really can't replace because all of this other stuff is built up around it. The company isn't around anymore, but the version of the application won't run on modern, supported versions of Windows for some reason. This is where WINE can really help you out.

It can help in other ways, too. My Playstation 2 was having a problem reading discs. In searching for a local repair place on the Web, I found out that several people sell cheap "self-repair" guides, but these are in some wacky Windows hypertext browser format (probably to prevent copying). Worked fine in WINE, and I had repaired my own PS2 for $10 in less than an hour.

Re:Wine and DirectX

[Anonymous Coward]

Really, the whole Wine/DirectX thing is only interesting to 19 year old college kids where games are a vital application. You get one game working, and it doesn't matter because they're all off playing the next one (which doesn't work on Wine) -- it's a losing battle.

If one REALLY wanted to move users, they'd get perfect Visual Basic, Access, and Delphi app support. Then a ton of companies could ditch Windows and use Linux on the desktop, something that is impossible right now without expensive rewrites.

However, the irony is that a "Hello World" VB app is much more closely tied to Windows than a 500MB DirectX game, and therefore harder to emulate.

Re:Looks like he "proved" what he wanted to.

[Frizzle Fry]

Since he ended at -1, I don't think that proved it. His post is modded up, which causes people to react by pointing out what's wrong with it, and it immediately dives to -1. So the system works okay in my opinion, provided that the post doesn't spend too long at 5 before getting struck down to -1. I think that he did show that some poeple react to the tone of authority (by the fact that anyone modded him up at all), but not that the moderation system is broken. More of a statement on human nature than on slashdot.

Re:Legality question..

[Anonymous Coward]

If you don't have XP, then you can't legally use the ntfs.sys driver at all.

But then, if you don't have a licensed copy of ntfs.sys, you're pretty unlikely to have any NTFS partitions. It's not really a big problem, shall we say.