Slashdot Log In
Microsoft Porting Applications To Linux (Really!)
Posted by
Hemos
on Wed Aug 16, 2000 01:15 PM
from the kicking-down-chairs-and-knocking-down-tables dept.
from the kicking-down-chairs-and-knocking-down-tables dept.
Erbo writes "We've heard the rumors before, more times than we can count, but this time WinInfo claims they're true: Microsoft is working with Mainsoft in Israel and a small French development team to port their apps to Linux, and possibly other Unices. No estimates on availability, of course. Their strategy seems to be to use an "Office for Linux" as a bridge to Windows, similar to Mac Office."
This discussion has been archived.
No new comments can be posted.
Microsoft Porting Applications To Linux (Really!)
|
Log In/Create an Account
| Top
| 616 comments
(Spill at 50!) | Index Only
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Wow, they're... INNOVATING? (Score:3)
It'd be nice to have a modern cross-Win/Mac/Unix office software package (and emacs doesn't count).
I'm looking forward to see what atrocities they try to implement in X11 just to get suchandsuch widget performing in exactly the way they want.
Kevin Fox
Wine and Office (Score:3)
If wine becomes 100% effective for Office, then there may ben no real reason to hold out for a port, which probably won't occur anytime in the near future anyways.
BTW, this was tested with office 97, not 2k. I haven't tried it with 2k yet.
-Restil
An entry point to Windows? I doubt it. (Score:5)
If this report is true, then I think this strategy could really backfire on Microsoft. It's more likely that people will use this as an exit point from Windows to Linux. After all, there's probably quite a few potential Linux users who would switch in a heartbeat, but hesitate because the Office suite on Windows is more productive and more polished. Now, with the option of running Word, Excel, or Outlook on Linux, that objection goes away. (And that's my honest opinion, really: their Office suite is probably the best one out there, but the Windows operating system has plenty of architectural flaws.)
At least it looks like they're operating and planning as two separate companies now. Maybe they think they're going to lose the DOJ case.
Actual quote from my friend when he heard this: (Score:3)
--
This is smart (Score:5)
If it's true, this is a very smart move on Microsoft's part. They've left Linux to its own devices for some time now, and the lack of an office suite has been one of the biggest shortcomings of linux as an office desktop.
But now that viable linux Office suites are coming into their own, and the lack of one won't hold linux back much longer, they can jump in with MSFT Office and claim a big marketshare of office suite installations on Linux.
Hell, if they port DCOM and a bunch of apps that use it, then they can run with the 'it works better on Windows' strategy that they have used with Apple.
Plus, when you've got a few billion in cash, it's not a bad idea to have a few products in your back pocket waiting for hte right time to release.
SteveSo, is WINE porting Microsoft apps to Linux too? (Score:3)
Are you sure their job is to port MS apps? The way I read it, they're basically re-developing WINE, only with non-disclosure agreements... not specifically porting Office/IE/etc.
Wow, that would seem to be WINE's goal as well.
Anyone else read this interpretation?
At this very moment... (Score:4)
At this very moment, a SkiDoo dealer is packing up for his new position as VP of Marketing to Hell.
At this very moment, pigs are growing wings.
At this very moment... Ah, who cares. I don't want that damned paperclip on my Linux box.
Unlikely, here's why (Score:5)
MS doesn't want what little there is of a Linux desktop market share.
Sure, Linux kicks ass on the server side. But aside from loving geeks who devote all their CPU and HD space to Linux, do you know a lot of people who actually think, 'Well, I'd get Windows, but Linux is so much *better* for desktop applications'?
Well, do you?
Linux is still catching up on Windows on the GUI and desktop side. Just look at the Holy Grail of Linux-related GUIs: 'We'll make it as nice-looking as MS'. As long as Linux is running after MS and Windows, they'll never be a threat.
Seriously, who would MS try to convince, here? If people are using Linux as a desktop, then there's something else aside from convenience and wide-ranging applications that they're interested in. Stability? Perhaps. But everybody else still figures stability is a small price to pay for prettiness, especially if autosave is on.
Sorry. MS isn't porting anything to Linux because, let's face it, on the desktop side it's so little of a threat it's laughable.
Freeze the market (Score:5)
This is clearly a response to the Gnome Foundation announcements. The future of non-Microsoft desktops suddenly got a whole lot brighter this week. Microsoft must do everything it can to steer people away from this up-and-coming technology. If they can get people to say to themselves "I'll just wait for MS Office to arrive before I try Linux" then they've succeeded.
Still, even if it's true, I can't see how it'd be very good if they're using MainWin (basically the equivalent of WineLib) to do the port. While the entire Gnome Foundation initiative is centering around CORBA and the Bonobo framework, a ported MS Office will still be using a ported DCOM. Furthermore, it'll look and feel like a Windows app, right around the same time that Linux apps are starting to take on a more unified look and feel. It'll only talk to itself. In other words, MS Office will feel as isolationist and foreign in the future standardized Gnome desktop as the current version of StarOffice feels in the current Linux desktop. Who wants that? More importantly, who wants that and at a cost of $500?
--
More proof of M$ Monopolism. (Score:3)
Well, I love the admission that the article makes. I mean, my last time programming was assembly language on an Amiga 1000. But even with my woefully outdated programming skills, it's painfully obvious that M$ apps have had an interface to the OS advantage over everyone else.
Why should Windows source code be required to port Office (let alone write it)? Corel certainly didn't get Windows source code to write WordPerfect 8.
But the process is even more complex than it sounds, since most Microsoft applications--especially those in the Office suite--use a number of proprietary interfaces, and each application requires specific workarounds.
Mainsoft has access to the Windows NT 4.0 and Windows 2000 source code, a necessity for the work it is doing.
(If M$ wasn't using their market dominance unfairly, wouldn't porting Office require only Office source code, not Windows source code?)
If this article is true, it's just a far more blatant piece of proof that Microsoft is corrupt, and really has to be broken into an OS division, distinct and different from their internet and applications businesses.
Re:Source? (Score:5)
-B
Oooo. (Score:5)
For those who don't recognise that reference- first of all, furrfu- second of all, this is the real leverage that comes with having Office on a platform. Given the expectation that 'oh, Office is on platform X, therefore legitimising it', Microsoft can and does use this as a weapon. For example, they literally told Apple to kill off Quicktime or they'd kill Office for MacOS- the quote comes from an exchange like 'We think it would be better if Quicktime, uh, wasn't.' 'Let's get this straight, are you asking us to knife Quicktime for you- to knife our baby?' 'Yes, we're talking about knifing the baby'.
Should MS apps be established on Linux it'd be like that only instead of dealing with a single point of development and control, MS would be dealing with little groups and individuals, threatening them that if they didn't stop work on their projects, MS would kill Office for Linux (and presumably blame said developer). This degree of blackmail might not work on RMS types but there is a level where it is frightening. Basically it's a sort of extortion, and the point is to engender a climate of fear and obedience. Some of us (mac people into development) have been able to watch this sort of thing going on in the real world for longer than you linux people have... and yes it seems to be illegal, the antitrust case nailed them for JUST this sort of behavior. Now we've got to see if that sticks, or if they get to ignore that as well.
At any rate- there is no benefit from having Office available for your platform. None. There's no significant compatibility between versions, ports are never in synch, it takes large amounts of motivation for them to produce software even half good (i.e. IE for mac) and even if they do they take pains to use it to cut off your other options and change the 'territory' right out from under you so your choices are dead.
The people screaming 'nooooo!' are, ironically, a lot closer to the mark than the people screaming 'yay' here. You've got to look at the business practices that inevitably go along with this sort of 'beachhead'. These guys kill markets- that is their whole schtick. Why would you want them coming over and killing your market too, even if your market is largely mindshare instead of commercial? All it will do is kill your choices without giving you the supposed benefits you think you'd be getting. And that's because, as was repeatedly found by the judge in the antitrust case, they really make a special EFFORT to kill your choices and kill your market- we're not talking about 'network effects', we're talking 'knife the baby'. An MS guy actually accepted those words, mid-negotiation, as descriptive of what they were trying to do. How can that be right? How can that be a market?
Re:Easy because MS will just require root privs. (Score:3)
Most software has to be installed as root, especially if it's going to be of any use to its participents.
Where did you get this idea? Virtually all software I have used in unix-like systems can be installed in home directories just fine. If it's autoconf/automaked, then ./configure --prefix=/home/user. RPM also provides relocation, and I assume Debian has a way, too.
With a file disk quota of 5 megs standard on most machines, do you think most users will be able to install Office (or StarOffice, or anything for that matter?)
5 meg standard? We're talking desktop systems here. If an entire organisation is using it, then individual users won't have to install it themselves. Home users probably won't be using quotas.
What I think the OP was trying to say was that it would require root privs to run, i.e. it would have to be installed suid root. Isn't this the case with their Frontpage extentions for unix? There is absolutely no need for either to be suid root, assuming good design. However, MS aren't exactly consistent when it comes to good design :)
MS Thought Process (Score:4)
Imagine This. (Score:3)
MS makes a Linux distro. They make a closed-source, propriotary DE that supports most of the Win32 API. They include support for DirectX, COM, and OLE. Now, all these changes require major changes to the kernel. While these changes are GPL'ed, MS has now effectivly forked the kernel.
People start writing apps for this new MS/Linux. Because these apps rely on the modified kernel, and propriotary DE, they will not run on regular Linux.
Everybody switches to this new distro, because it has more apps and better technology.
MS has won, despite the GPL.
Stuff to Watch Out For (Score:3)
The conspiracy-theorist in me believes Micros~1 is trying to steal thunder from LinuxWorld Expo without actually making the announcement themselves. However, such a move doesn't surprise me. Bill Gates has never demonstrated loyalty to anything other than making a buck. If he thinks he can make money selling Office to Linux users, he'll do it without the slightest hesitation.
Those who remember the port of IE to Solaris, however, know that Micros~1 will not adapt their apps to the host OS, but will instead try to graft on enough of Windoze's "architecture" to make it work, just barely.
For example, it wouldn't surprise me in the least for LinuxOffice to require a Windoze-style registry, needed to support per-user preferences. This despite the fact that .*rc files have been around forever. Also, expect a butt-load of COM/DCOM components to get installed requiring root privileges (though the need for these prvileges will never be adequately explained).
But I suspect the biggest delays will come from trying to implement Micros~1's demands for copy protection and to prevent "unauthorized" use of "their" software. So, assuming they work out how to translate NT's methods of "license" administration, expect to be required to install a "license" manager that's completely incompatible with any other "license" managers you may have (and may even interact badly with them). Oh, and the CD key will have 96 digits :-).
In short, this is FUD of the highest order, even higher than normal, since it doesn't come directly from Micros~1, thereby giving them plausible deniability. Even if it's true, I wouldn't expect anything to ship for at least a year and, like the initial release of all Micros~1 products, will probably work quite poorly. Thus, even if you're working on an office suite, this story may effectively be ignored.
Schwab
Finally!!!! (Score:5)
Re:A good idea but... (Score:3)
NT itself is stable
If NT itself allows applications, after they have crashed, to leave leaked memory or leve the system in a strange state or leave the system in a slowed state then it is not a stable OS. It may not fall/crash but it is not stable. Kinda like a ship in a storm, your footing is not stable but, hopefully, the ship won't sink. You may be flung overboard but, hopefully, the ship won't sink!
From the makers of IE for Unix... (Score:3)
So from what I can gather, these are the geniuses that brought us Unix IE. Yup. Just look at the explosion of IE users on Unix. If they do just as good a job on Office, then MS better watch out or their OS market will disappear overnight!
My favourite part is how they say their software isn't a Windows emulator. They just build every single DLL that you'll ever need into a Unix library. And probably a registry too, for good measure.
And you thought administrating Windows was bad enough on a Windows box...
--
Re:A good idea but... (Score:4)
Re:interesting? (Score:5)
Weather report (Score:4)
Re:Hubris (Score:3)
What the hell? (Score:5)
Porting DCOM to UNIX is one thing, but Office is another beast entirely. Microsoft nearly destroyed their market for Mac applications when they tried to offer a weak port of the Windows version of Office -- people simply started refusing to upgrade. These days, the two have pretty much completely different code -- you can't really port the Windows version of Office to any other OS, because they're joined at the hip with DLL-Hell, private system calls, etc.
Re:A bridge to Windows? (was Re:Hmmm) (Score:3)
I think (one of) the other thing(s) M$ is trying to accomplish with this move is to shore up the lifespan of their proprietary data formats. The targets are coorporations who have licensed Office. One of the main things that seems to slow the move off of Windoze at a corporate level is the tremendous investment in M$ Office and legacy data that is already in the Office formats.
The Office data formats are, after all, the key to keeping their current captive customer base captive. The office applications themselves simply aren't good enough on their own to do that. A *nix Office suite is a way to keep those formats alive, which is important to M$, since if the formats are dropped in favor or, say, an open standard, then there is really no incentive for corporate customers to stay with M$.
M$, I believe, fears that as work goes forward to make format conversions easier, it will be more cost effective for a corp to invest in data conversion and move off of the Windoze platform than it will be to keep putting up with (paying for) the spew of M$ corp.
Other office suite software vendors are targetting the office data formats and application functionality. If one of those outfits gets there (usable in a corp environment) first, M$ would be left with their collective asses uncovered. So, for M$ to produce a Linux Office suite is also a CYA strategy...
Heh. Disingenuous is M$'s middle name. Their contempt for users -- and everybody else, for that matter -- is huge. They really, really don't expect anyone to figure it out, and when someone does, they apply spin. I don't think you misunderstand.Also, note that this effort is going to be incorporated into the spin machine. 'See, we develop apps for Linux, too!' It's a good PR move for M$, since they will co-opt a certain percentage of *nix advocates with this move. Since Linux advocates in particular seem to be the most vocal and persuasive of the M$ detractors (often because they are former M$ users, I think), anything M$ can do to cut down on their number is a plus.
All in all, it's about Leverage.
Usual attack, beware (Score:5)