Why Ubuntu Should Embrace Wine
Mark Shuttleworth made some interesting comments recently about the role of the Wine compatibility layer in fostering the migration of Windows users to Ubuntu. Shuttleworth views Wine, and binary compatibility with Windows applications in general, as a distraction from the real goal of making the free-software world "thrive on its own rules," not those borrowed from the proprietary ecosystem. I disagree. Here's why.
In an ideal world, software vendors would write native ports of their applications for Linux and release the source under the GPL. But that's never going to happen, even if Linux gains more market share. After all, Apple currently commands about 10% of the desktop market, yet many closed-source developers have yet to port their products, including most modern games, to OS X.
While Ubuntu should stay true to the goal of encouraging the use of free, open-source software wherever possible, it should not close its doors on Wine and winelib as a means of smoothing over the obstacles that prevent many Windows users from switching to Linux–and, more importantly, that discourage developers from writing applications for free platforms.
Porting, maintaining and supporting many applications to run natively on Linux involves a huge amount of work, especially when they're written in languages that don't lend themselves to cross-platform operability. Compiling code against winelib in order to ensure compatibility with Wine, on the other hand, is often much easier. It's also a convenient way of releasing software for Linux without having to make the source public, which many developers are reluctant to do.
I realize that calling for widespread adoption of closed-source Windows applications on Linux is pretty heretical, and I fully expect to face the unmerciful wrath of the GNU stalwarts in the audience. I'd like to stress first, however, that I like freedom and free software a lot, and believe in the ideology upon which open-source code is built. But at the same time, I like having a computer that lets me do as much as possible as easily as possible. Having more Windows applications available via Wine would go a long way towards making the Ubuntu experience the best it can be.
In defense of moderation
Ubuntu should continue to encourage the development and use of free, open-source, native applications for Linux. It should not be reluctant to work with developers who want to make their products available on Ubuntu via Wine.
Dismissing Wine as a distraction from the ultimate objective of a 100% free and open-source computing experience is radical and unrealistic. Traditionally, Shuttleworth and Ubuntu's developers have stood out as reassuring centers of moderation and pragmatism in a software ecosystem dominated by bombastic extremists. Let's hope they have the wisdom to extend their Realpolitik to Wine as well.
Read here;
http://linux.oneandoneis2.org/LNW.htm
The all or nothing style to migration is quite a strange approach, which shuts out a proportion of users that might otherwise migrate to Ubuntu (or other Linux distro). Many companies use F/OSS already they just don’t know it. In-house developed software is the largest base of software used on the planet, which in a sense is open source because I know many people that use solutions contributed to be freely used by others.
I am thinking that no-one wants to slave over re-writing (or pay for) the hodge-podge code that’s built over time in several institutions. Think of all those government systems that still use console apps, built in who knows what language.
Having virtualization layers helps in transitions. Look at Windows 7 with XP mode, or NTVDM to run DOS in XP. The effort put into the various virtualization platforms goes a long way to cementing the confidence that users have for Ubuntu.
After years of enculturation, new Linux users assume that boxed version software runs on anything, while we have better in Ubuntu its a simple and free download. Ubuntu really contrasts with how Windows apps that ask sooooo many cryptic questions like where to save the program, startup items, quick launch, bla bla bla.
The long term solution should be to educate users, perhaps through advertising the fact that users can download equivalent software from a central application. What if there was a project that provided a front end for the kind of information found on http://www.osalt.com/. What if there was an app that you can run on your Windows/OS X system that you can run and it will inventory the software that you have and it will show you the free alternatives available along with user reviews or resources showing how the alternatives work.
For once i would like to disagree with you. Sure wine as its place in on the linux desktop, and couple of times i have had to rely on it to get a window only application to work. But i think Linux Distros should quit get over their obsession with windows (and in some cases mac) and try to do things their own way. I feel distros should really stop trying too hard to bring windows users because we have in too many cases it always back fires, and all we get are people who whine and witch about linux, who don’t understand the free software and open software culture. Let them stick to their windows and when and if they decide to try something different they would seek linux.
Wine can never be perfect and it amount to always chasing and catching up to windows which will always leave us behind Microsoft. I have been using Linux desktop now for close to 2 years half of it exclusively. i have never had the need to install wine, because i try to look for FOSS alternative for all the softwares i use.
to be frank the number one reason an average user will want to run wine is for gaming. and gaming will always suck in wine because of most power games are built to run on Direct X. other than that most of the tools that an average user needs are more than available. ( gimp is enough for most average users, and no average user will ever need corel draw or any of the expensive adobe suits)
security wise wine is not secured. its very prone to windows virus which will also (from experience) run on wine in linux albeit in non root mode but could be just as annoying. a windows worm once claimed 200mb of my ram and slowed the system, reproducing its self every time i deleted it only way i got round the problem was to delete the wine folder. how many noob will not be run to this error?
also putting much effort in wine could potentially kill other FOSS projects imaging if wine was so good that it would run windows apps 100% in linux few new users would see the need to use openoffice, or gimp and even the few commercial app maker that make native binary for linux we not see the need or incentive to do since they know that their windows binary would work just as well. then we would have a situation were the linux desktop space will be swimming with windows applications to the detriment of native applications for linux.
to run up am not against wine, and i believe their doing a great job. But like i said wine as its place that place should never be elevated to be the focal point of linux drive to win market share. The main purpose of linux is not to please windows users
i on the other hand agree with Mark. Here’s why: Wine will never have a stable, usable release, either because Windows keep on changing or the nature of the code itself. The fact is even computer savvy people will find it hard to work with. Why not make a layer in linux itself like Wine, closed or open source doesn’t matter, keep it standard for 4 – 6 years across all linux ecosystem, release the APIs so vendors/developers can make software for it. This could be a business model for Canonical, don’t you think?
budi its a shame when i hear so called linux people say closed or open source dont matter .. dude is it linux we are still talking about? what and how does linux differ from windows or mac .. i could have better hardware support on windows, and have my ipod sync well with my mac. and edit videos clips with my imac, and play the baddest game on windows .. but i will never feel the sense of ownership i get with my ubuntu linux, were 99% of my desktop is free as speech. that is why i use only hardware and in most cases software that dont limit my freedom. when it comes to linux freedom does matter a great deal. it would be suicidal to make “layer in linux itself like Wine” it would be the death to the linux desktop as we know it. because in the end .. we all we would have is a system obsessed with playing catchup to windows and never quite getting there.
I totally disagree with this post, and I am afraid this site is becoming a bit of a troll (yesterday’s post about Intel graphics seem inflamatory, too). It is easy to say that it would be nice to support wine, but it takes a lot of effort, a LOT of effort to support windows compatibility. Just supporting to some extent binary drivers has been a nightmare: X crashes with Nvidia’s or ATI’s and there are lots of angry people, bugs reports everywhere and it’s hard for Canonical or the Ubuntu community to do anything about it. But at least graphics drivers are essential (it’s access to the iron). Committing efforts to wine is distracting, in that it takes a huge effort and we have limited resources. Sure, if we had unlimited resources it would be nice to support it. But this is about prioritizing.
One of the biggest reasons to support Wine are unmaintained Windows applications. First, either no one has the source to these apps, or they refuse to share it. These apps will never be rewritten to run natively on Linux.
Leo: Christopher Tozzi is a fantastic blogger and a great asset to this site. But there’s room for differing opinions and different types of content.
There’s a famous line from Lethal Weapon… “Whatever Leo wants, Leo Gets.” In this case, we continually evolve WorksWithU based on reader feedback. So keep the constructive criticism coming and you’ll see WorksWithU continue to shape its content based on reader feedback and needs.
Joe Panettieri
Editorial Director
WorksWithU
Thanks Joe, and sorry, I think I was a bit over the top. I like this site and it’s one of my few rss feeds, together with . I personally like the original content (news about system 76 new offerings, and such). What I think would be a shame is if this would become something along the lines of OS News, which was initially promising but quickly became something else. Mostly, a collection of rants and troll posts aimed at attracting heated discussions. On the other hand, you are right that there is (and should be room) for disagreement. I clearly mixed both things (my disagreement with the contents of the post with my concerns about the direction of this site), so sorry about that.
Please keep up the good work!
(Errata: I meant to say “together with LWN”)
I think that WINE has helped many people switch to Linux. Sure it may not be perfect, but Linux software isn’t perfect. And until we get the usability of a video editing software on par with commercial applications, until games for Linux don’t suck the big one, until adobe releases native photoshop or gimp becomes an international standard, wine is sill going to exist.
When I tell others about Linux, a frequent question is if they can use the software THEY bought will work. I tell them that WINE is able to do run their software, generally without issue. I’m glad I can tell them that, because my word about Linux would instantly be discarded if they learned that they could no longer use software they know and love with this great operating system. I am in full support of the wine project and I don’t consider it an obsession with windows.
Think of it this way. Have you ever had to edit video? What options do you have? Kino? Everything else is incomplete or requires a steep learning curve. Yet windows movie maker seems to be what everyone loves and uses.
Any games? WoW is the most popular and the most worked on in wine.
Users expect their software to just work. Plain and simple. Wine makes that possible, and those who simply state that we need to stop being Windows are right. Just not right now while we are still not piquing the interests of the big software makers. People know software by name, and if you can’t run big name software all of a sudden they are disinterested. This might not be the case for all, but it’s a major issue.
Also note that I wrote my last post on my iPhone, so sorry if there are grammatical or spelling errors.
@travist120
Points well taken, still, keep in mind that supporting wine is extremely difficult and that’s why I think it would be too distracting to support it officially (see my arguments above)
Maybe a Canonical/CodeWeaver partnership would make a lot more sense. Offer a discounted version, customized for Ubuntu, and get them lots of costumers. Everyone wins 🙂
Ubuntu shouldn’t embrace Wine, I agree with Mark, ISVs should support Linux 100%. Apps like Opera, Firefox, Open Oficce are third party apps that are available for every platform and work well on every platform. ISVs should start developing for Linux and pre-packaging their apps for major distros so users can install them without even adding repos. That would be good for noobs too.
Gdebi should be more widely used.
I have to take Mark’s side on this issue as well. I don’t think it’s the place of a Linux distributor to worry about running Windows apps, especially not when there is still so much low hanging fruit in the Linux software ecosystem itself that needs attention. Perhaps when the team working on the next release of Ubuntu looks around and says, “Well, geez, there’s really not much left to do…”, then maybe it will be time to start putting energy into something like Wine. Until then though, it would be diluting the available energy being put towards the “real” goal. At that point though, if the meta-goal is to improve cross-platform capabilities, it would seem to be more sensible to be putting time into Mono, but I digress…
Now, all that is not to say that Wine is unimportant. I think Wine is very important, in fact, but only to relatively small subset of Linux users. Those are the people who should be “embracing Wine”, and really, they are. Transgaming and Codeweavers have both decided that making certain Windows applications work well in Linux is important, and have built what seem to be sustainable businesses around that. Assuming they keep going strong, and Wine project itself keeps rolling along, we get the best of both worlds. The people who are focused on building the best Linux platform out there keep doing that, and the people who want to run Wine on Linux have a great platform to layer their technologies on.
I agree with Quentin (#15). There are lots of people out there working on Wine, doing phenomenal work at that. After last year’s giveaway, I went out out and bought a license for Crossover, not because it’s the end all be all or it works perfectly (still can’t play Civ IV without a CD hack), but because I feel I should support that work.
That said, I don’t even know what you really mean by “Ubuntu should embrace Wine.” How? What would they do that they’re not? I just don’t see what Ubuntu’s place is in that equation. It’s not like they’re impeding Wine development or telling people not to use it or anything like that. They’re just letting the Wine developers develop Wine. I’m fine with that. What else specifically would you like to see happen here?
And finally. Yes, I’m one of those GNU/Linux puritans you decry so much (it’s cool, I make fun of “pragmatists” on my blog, I still like ya). But rather than turn on my Stallmanesque firebreath here, let me ask you a pragmatic question. Who the heck is going to switch operating systems just to run the same application set? I mean, I hear the “pragmatists” say all the time that “nobody but us geeks care about their operating system.” If that’s true (and I’m willing to admit it probably is most of the time), the only way we win in the long run is if our applications are better, our environment is better, our offerings are more compelling than the other guy’s. So let’s get to work on that.
@Leo and p.daniels: what I was trying to suggest in the post (but which maybe didn’t come across as clearly as I hoped) was that Ubuntu should actively encourage software developers who currently refuse to publish for Linux to explore Wine as an easy way of making their software run on Ubuntu. In most cases, it wouldn’t involve much work to compile code against winelib (as opposed to the native win32 API) and release it for Linux.
A real-life example of this approach which has just come to mind (too late for inclusion in the post) is Google’s Picasa photo-editing software. The .deb package for Picasa is basically just the Windows build bundled with Wine and some Linux scripts to make Picasa integrate nicely into Gnome and KDE, and look prettier than most Windows applications running in Wine. This is a great example of a software developer using Wine to distribute a Windows application on Linux with minimal extra effort. Ubuntu should be encouraging similar offerings from other Windows-oriented developers.
I see Wine and winelib as an easy way to reach out to developers who so far have ignored Linux because they 1) don’t want to release source code and 2) don’t have the resources to write and support native Linux applications alongside Windows ones. Unfortunately, almost no developers have explored this route, because the Linux community has generally been hostile to the idea of closed-source Windows applications integrated into Linux via Wine. Instead, Linux users (at least the most vocal among them) have demanded that all developers work on the free-software community’s terms, which is never going to happen.
Although creating a 100% open-source platform is a noble goal, Ubuntu users need to think pragmatically and realize that the larger software world is never going to acquiesce wholly to the free-software ideology. Wine can be a good way for the free-software community and proprietary developers to meet one another half way.
Wine is the lazy option to solving the Linux software availability problem.
Canonical could make more of an effort towards developing the applications businesses and consumers need to make Linux a viable option. We need to find a way of funding the work that needs to be done instead of looking for stopgap measures.
When there is money to be made, the software houses will start paying attention. If we simply opt for Wine, the software houses will code to the lowest common denominator and the quality of software will diminish.
To get the applications we want we need to fund their development. Remember free software means free to view and edit the source code. It doesn’t and never was really meant to mean free as in beer.
In case anybody missed my point. Funding software development is the answer!
“sudo apt-get install wine” and problem solved. There really is no reason to include WINE in the default install. Too many people see Ubuntu as a Windows replacement when it’s nothing of the sort. Ubuntu is Linux plain and simple. And if a user wants to run Windows applications inside of Linux and not pretend that Ubuntu is Windows then he or she is free to do so.
[…] in response to Canonical#8217;s support of WINE and Microsoft Windows compatability). The community response to Shuttleworth#8217;s comments were of mixed results. I must admit, that I agree with […]
Apple doesn’t have even nearly a 10% share of the desktop market. It may have this share where it’s most successful in, the USA, but for the whole world I’d guess the right figure is around 3-4%.
It seems to me that the poster is trying to make Mark’s not-anti-Wine-comment to look anti-Wine. Some excerpts from the post:
“It [Ubuntu] should not be reluctant to work with developers who want to make their products available on Ubuntu via Wine.”
“Dismissing Wine as a distraction from the ultimate objective of a 100% free and open-source computing experience is radical and unrealistic.”
“…it should not close its doors on Wine and winelib…”
When I compare these comments to what Mark said, it seems like the poster isn’t commenting Mark’s words at all. In his comment Mark didn’t say one word against Wine, but emphasized the strengths that Ubuntu needs to thrive on. On the other hand, he did say this about Wine:
“[Wine and native Linux ports] both play an important role [in the success of Ubuntu]”.
I don’t like when people’s words or thoughts are twisted. In this post, the latter seems to be done. It’s as if the writer is trying to protect something he sees as important, Wine, by making a false or exaggerated alarm. That kind of manipulation of discussion and thoughts of readership is all too common nowadays.
Ari: I stand corrected on Apple’s market share–the figure I used was based on the United States, not the world. The point I was trying to make, however, is still valid: attaining larger market share does not necessarily mean vendors will start porting their software to Linux. Apple controls a significant portion of the American market–currently the most lucrative in the world–yet many vendors, especially game developers, still ignore OS X.
As for twisting Shuttleworth’s words: I agree that he didn’t express any explicit animosity towards Wine. What he did imply, however, was that Wine’s role should be limited to temporarily filling the gaps in the Linux application lineup until more commercial vendors port their software to Linux. My argument is that, since realistically most commercial vendors don’t have the interest or the resources to port code to Linux, Ubuntu should instead work with proprietary developers to encourage them to build their applications against winelib, an easy way to ensure Linux compatibility without demanding much extra work on the part of developers. Shuttleworth doesn’t believe that Wine should be used for this purpose, which is where I disagree with him.
I guess sometimes I have to stand on the shoulders of a giant to see a little further. I was looking at some of the documentation on the samba project and I was intrigued by something. There are a whole lot of people out there – project maintainers – that put a great deal of effort into project that help interoperability between systems like wine, samba, and so on. In spite of this effort the likes of Microsoft will maneuver changes to break their code or outright sabotage their efforts through all kinds of means in for the sake of Windows market share. Maybe its this uncertainty, combined with the fact (echoed by the samba team) that its a follower strategy that makes laying a foundation on wine risky business wise.
Microsoft is not making money in its non-core businesses, selling Xboxes and search, so it wants to protect its turf and it likely will not play nice to open source even when it could make office and IE for Linux. Wine, Samba, Mono, etc., are great projects that have the backing of individual users and developers but what happens when Microsoft begins to change things as usual setting project back into the dinosaur era.
The VFAT lawsuits are an indicator of how low Microsoft is willing to go. Next time they may roll an update to change the way SMB/CIFS or Windows UI libraries behave. Oh, darn they already did with Windows 7.
Wine is not a critical make or break feature. If left out there is a guarantee it will be installed by the user – at least I know I will even though I don’t even use it. Its always just in case I want to use free software utilities written for Windows. I can self support, but what about companies that now demand support, especially when their built for vista software breaks.
[…] Read more at Works With U […]
no, not wine directly
but a frontend with a list of working program/games like playonlinux or wine-doors
playonlinux ran so badly on my system. I wouldn’t touch it again with a bargepole.
Shuttleworth himself said Wine has its place. Users should be made aware that it exists if no native app meets thier needs. Look at how Ubuntu handles codecs or the flash plug-in, or proprietary drivers. They’re easily visible but not set up by default, and attempts are made to let users know why they are somewhat of a “last resort” option, for the user’s own good.
WINE should support up to Windows XP. That is a fixed target.
They don’t have to play catch up with Microsoft forever.
Right now, I don’t know of any (non-Microsoft) software that requires a version past Windows XP. Given that Vista was such a shit, why would any software maker do this?
If Ubuntu can run all WinXP compatible code, it can run basically all important applications made up to 2009, and probably well beyond. That would be a very powerful advantage.
I personally use one WinXP app in WINE (it’s a pdf reader/editor that lets you hightlight, markup, etc). If I couldn’t have used that passably, I wouldn’t have switched because there is no open source equivalent and I don’t have time to develop my own.