Posted by
ryuzaki0
on from the kinda-scary-stuff dept.
Adam Cody wrote in
to note that this job
on Microsoft's web site. It's a marketing position and
the responsibilities include a competitive analysis of Linux.
Should I apply?
Poor Vinod, we knew him...
by
overshoot
·
· Score: 3
Inquiring minds want to know:
Did Vinod V. meet with an untimely end? Did he succumb to the 'seductions' of Open Source and turn to the Light Side? Or has he been banished for his part in an embarrasment to M$?
-- Lacking <sarcasm> tags,/. substitutes moderation as "Troll."
Know your enemy, and know yourself
by
Benjamin+Shniper
·
· Score: 3
And you will come through a thousand battles unscathed. -Sun Tzu
The job post seems harmless enough, but it is, alas, the middle steps in their embrace and extend policy for shutting Linux out of the operating system market.
In a business environment where reading Sun Tzu is required, this shows how deeply the warrior philosophy is engrained. They are seeking knowlege of the strengths and weaknesses of Linux. They wish to find these so they can integrate the strengths into the next version of NT, while using the weaknesses to destroy Linux vendors. They are not afraid of lengthenning the NT code base to get what they want - if there are features in Linux that aren't in NT, then they will publicly praise these features. If there are weaknesses in Linux that aren't in NT, then they will publicly praise NT. They do not praise Linux, or Apache (as Ballmer was accused of doing) but instead the features.
My friends, features can be stolen. They have all the money in the world to steal features with.
They will tell the poor sap who takes this job that they are trying to improve NT by adding Linux-like features. They will tell themselves that. But their vision of a single OS for everything is more important than the methods used to achieve it.
This person they wish to hire is the middle step in the chain to deflating the Linux user base. They will use this person to take all the features in Linux that can be taken and put them into NT. And then they will use this person to say "Look, Linux is now not as good as this thing me and Bill Gates have made: Windows 2002; Linux + Windows" This is exactly how they attacked Java, OS/2, Macintosh, and others.
-Ben Shniper
Re:I have a question however...
by
IntlHarvester
·
· Score: 3
Not that I disbelieve you, but when someone posts "I see BSOD's with NT all the time" without posting specific STOP messages or some indication of what is going on, I'm inclined to right it off as Linux advocacy at it's worst, or maybe if I'm charitable, just someone whose confused NT with 95.
Think if somone posted "Linux Kernel PANICS for me all the time!" You wouldn't give it much thought unless they had some specific debugging info. --
OK, they want someone to help them tell the public why to use NT instead of Linux. Don't we have enough people telling everyone why to use Linux instead of NT?
This is not to say that I approve of all of the various unfair Microsoft strategies, including writing FUD. But doesn't every company say why you shouldn't use their competition?
Re:Why not an MS Linux distribution?
by
Grenamier
·
· Score: 3
Thanks for the reply, maynard. I feel kind of honoured that you spent so much time on my comment.:)
Grenamier makes the point that Microsoft could create their own distribution of Linux with a proprietary Win32 kicker...How different is this from Corel using Linux as a vector to market their proprietary Corel Office software?
Good question. On fourth thought, there isn't much difference, especially since Corel has already composed its own distribution. OTOH, I do think there's a slight difference in that it's in Corel Office's best interests to have Linux stay free and open and carrying along as it has. Microsoft would be interested in having Linux development slow down or be influenced to allow Microsoft more control over its direction.
That said, I don't want people to get the idea that I'm really anti-Microsoft. At home I run Windows 98, Windows NT and Linux (on a separate computer). I'd prefer to have NT and Linux both succeed. Linux really is something special though, and I don't to see anything badly interfering with it.
Even though the premise is good, there are some fundamental mistakes of fact which ought to be resolved:
Thanks for clearing up these things for me. I should've known better on a lot of them.
Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that. [snip] But they can't "Steal" something out in the open like a GPL'd package, so I don't know what you mean by that.
Some of the previous replies speculated that Microsoft might be looking for features of Linux to steal for NT. I was deciding that it wouldn't be a good enough reason to build a distribution. As far as "stealing" goes, I hadn't thought much about it, I suppose stealing from free-software could be burying GPL'ed code in a closed-source product and ignoring the GPL's terms.
No. I've got a couple problems with this. First, USB support is already in active development, as is NTFS support in the kernel [...] that point is moot.
I should've remembered this, but I was only tossing out ideas on how Microsoft might have wanted to candy-coat the distribution for new users coming from Windows.
[...] What this means for Microsoft is that they will never control the momentum of kernel development unless they're willing to attempt a legal attack against the GPL; though that may well be in their interest.
It would be unlikely for Microsoft to attempt something like that for a while until people have gotten used to the notion of a Microsoft distribution and some of their lawyers get some vacation time.:)
OK. More problems of minor ignorance. First of all, if MS were to release a proprietary Win32 widget library, this would be no different from The OpenGroup with their Motif product line, or Troll Tech with their previous QT-1.x product line. In both cases each company has (and had) every legal right to sell a proprietary binary only software package which could not only cost to develop for, but could even cost just to use in runtime licensing fees!
If these guys get to play that game, then certainly Microsoft can. The GPL only covers software which can be thought to derive from a previously GPL'd source base. For example, if I create a source tree completely by scratch the GPL has nothing to say about what kind of license I distribute under. But if I derive an offshoot from a previously GPL'd program, say my own special version of GNU ls, and I distribute this code in binary form for either free or for charge, then I must also offer to distribute source, and I can't prevent the recipient from also distributing either the binary or the source.
I made a bad assumption here. I thought Microsoft would need to modify the kernel and a lot of the system software in order to create a UI with the same infrastructure and integration with the system as the Explorer UI on Windows. Of course, WINE, GNOME and KDE show that such changes wouldn't really be necessary. Duh on me.
[...] But baring such a casastrophic -- and unlikely -- outcome, we should expect that even if Microsoft released the most popular distribution of Linux, with all sorts of proprietary and expensive add ons to run commercial MS software, plenty of people will still just tinker around and maintain their own special Linux flavors. Do you really think Debian would go away? Not even if Caldera, Suse, and RedHat were flattened by the MS steamroller could they stop the Debian project without a Legal or Congressional victory (heh, maybe a bullshit Free Software Tax Act, that might make for good assault legislation against free software and is right up this congresses alley). Just like MS can't prevent end developers from creating a quality stable kernel and giving that away for free, they can't prevent a bunch of developers and end users from collecting a bunch of free software, creating a usable Linux distribution, and giving that away for free.
Quite true. But I'm not really concerned that other distributions could become extinct, more that Microsoft could influence Linux development in ways that don't really benefit Linux. For instance, I wouldn't want to see Linux catch a case of creeping featuritis that somehow spreads into the kernel. In theory, Linus and the people who work on the code would be able to avoid such a thing simply by saying no, but what if in the future a large number of users demand feature after feature after feature to be added to parts of Linux that have a wide impact? A compromise here, a lapse there and suddenly the kernel is growing or becoming buggier, perhaps...certainly we can live without MS Office (actually, I really like Word and Outlook on NT), but I hope most of the future's changes to Linux can be of the take-it-or-leave-it variety.
I get paranoid about good things that become too popular. Like this great seafood restaurant I know that's kind of gotten crappy since it suddenly started becoming "cool". With all the masses of people expecting things now, now now, there just isn't as much care taken in the food as there used to be. Now it tastes the same as all the other restaurants and it's hard to find a reason to pick this one over the others. I don't have to eat the new dishes that they serve now, but I want my old fried lobster back the way it was before.
I've snipped a lot of what you wrote, most of which I haven't marked. Not to ignore it, but because I agreed and I didn't have anything to add. I probably didn't add much here anyway.:) Even so, thanks for setting me straight on a few things...I should be more careful.
Did anyone notice that no actual Linux experience was required for this job ? And that excellent writing qualities were sought ? Sounds like they are looking for a FUD-farmer...
The simple explination is "know thy enemy". Doing a competitive analysys is pretty common, and I'm sure that the marketing people over at RedHat, VAR, and so on have done these. And I'd venture to guess they do this not only against MS products, but against SCO, IBM, probably even other Linux vendors.
MS is looking for what to say when a customer says "but what about that Linux thing? why is Win2000 better?", they'll know what to say besides "it's free software".
If anything, I take this as one of the best compliments the Linux community can get. We're really competition. Now if only we can get a hold of the complete analysis when complete....
Looks like "Winux" is becoming a reality.
by
Grenamier
·
· Score: 5
To me, it looks like Microsoft is creating its own distribution of Linux. For one clue, the title of the job is "Product Manager". No company has Product Managers for products they don't sell. Some of the duties include "product development for the development team" and "technical assistance for the sales force."
The position also advertises "high visibility" outside of the company. People who just watch competitors and make reports about them don't normally get much exposure (VinodV didn't get his 15 minutes on purpose). But if Microsoft is selling a "new" product, its Product Manager would be expected to flog it to the press.
So what's Microsoft up to? Scrapping NT for Linux? Of course not, they'd be throwing away all of their leverage. Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that.
This is my wild speculation:
Free-sourced software is the Internet happening all over again for Microsoft...everyone in the industry is jumping on the bandwagon and stories about the success of Linux, Apache and Perl are everywhere. Microsoft missed the populization of yet another decades-old "phenomenon" and they've known it since Hallowe'en at the least.
What did Microsoft do about the Internet? Embraced and tried to extend it. Now everything they sell has some Internet features, but they didn't manage to spread Active Server Pages everywhere, and Apache still tromps MS Internet Information Server.
I believe Microsoft will start by bundling Apache like Apple did, all "cleaned-up" with a Windows interface. Perhaps in the next Service Pack and in Windows 2000. Ballmer was saying the other day that Apache was simply better, in fact. They'll make a big deal about Microsoft supporting Open Source, and how they support innovations like that.
Then Microsoft ActiveLinux 3.1 will come out (I thought that was the most probable name and no version number less than 2.2 - 3.1 is good luck for Microsoft.) They'll make it easily installable, perhaps add USB and NTFS support, and then use Active X and IE for UNIX to put a Windows interface on Linux that has some Win32 APIs, the way OS/2 had some Win32 API's with DAX.
I'd be willing to bet Microsoft keeps that interface closed. They'll argue using a creative interpretation of GPL or LGPL so that they won't have to reveal the source and perhaps any undocumented Win32 calls that they port over. What they do will probably be in violation of the (L)GPL, but what legal force will be able to challenge them? ESR doesn't have that kind of cash.
MS shops and ISV's won't care about that though. Those dyed-in-the-wool Microsoft groupies, PHBs and non-technical CEOs who've heard so much about Linux can try it without leaving the MS fold or the Windows interface.
And they'll be impressed. Instead of creating a bad distribution to discredit Linux, they'll create a really good one, with all the goodies and easy installation (that you can start from Windows like Caldera, probably). Microsoft users who need hand holding will use it, Windows users who want to get into Linux (but don't want to get their hands dirty) will use it, and ISVs will write to that closed interface, because Microsoft will give them tools to make an easy port that lets Windows program keep Win32 calls for the UI. Microsoft will make ActiveLinux become the path of least-resistance for users and vendors from the Windows world who want to expand into Linux but have no interest in free-source issues.
If Microsoft succeeds in keeping the interface closed, and porting Office and BackOffice to their distribution, then it'll mean trouble because other distros won't look very competitive to the new users and the industry press. Microsoft could end up gathering more users than the other distributions, and then I don't know what...Microsoft was never able to get most of the web servers to run on MS-ware, so they didn't succeed there, but what if they managed to amass the biggest share of Linux users? Would it give them enough power to influence Linux's direction?
I'm not trying to be alarmist, I'm just trying to figure out their strategy...I could have it all wrong, but I do believe Microsoft will create its own distribution, and that it will try to create some reason to use Microsoft's distro instead of anyone else's. And Rob shouldn't work for them...what would happen to Slashdot??
Sorry for the length, I got off on a tear there...
-- --
John Truong
Why not an MS Linux distribution?
by
maynard
·
· Score: 5
This is the best comment of the bunch, I think. Grenamier makes the point that Microsoft could create their own distribution of Linux with a proprietary Win32 kicker. They are in their rights doing this, and if I were MS this is exactly the tactic I would use. How different is this from Corel using Linux as a vector to market their proprietary Corel Office software?
Even though the premise is good, there are some fundamental mistakes of fact which ought to be resolved:
Grenamier (jtruong@idirect.com) Wrote: To me, it looks like Microsoft is creating its own distribution of Linux. For one clue, the title of the job is "Product Manager". No company has Product Managers for products they don't sell. Some of the duties include "product development for the development team" and "technical assistance for the sales force."[Emphases mine]
The position also advertises "high visibility" outside of the company. People who just watch competitors and make reports about them don't normally get much exposure (VinodV didn't get his 15 minutes on purpose). But if Microsoft is selling a "new" product, its Product Manager would be expected to flog it to the press.
So what's Microsoft up to? Scrapping NT for Linux? Of course not, they'd be throwing away all of their leverage. Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that.[emphases mine]
Yes, I agree completely. They wouldn't bother with a project unless they were planning to either make something for sale, or were planning a "grassroots" marketing effort. But they can't "Steal" something out in the open like a GPL'd package, so I don't know what you mean by that.
This is my wild speculation:
Free-sourced software is the Internet happening all over again for Microsoft...everyone in the industry is jumping on the bandwagon and stories about the success of Linux, Apache and Perl are everywhere. Microsoft missed the populization of yet another decades-old "phenomenon" and they've known it since Hallowe'en at the least.[Emphases Mine]
I really like that quote. It well describes the mentality which must be fermenting in MS upper management strategy meetings.
What did Microsoft do about the Internet? Embraced and tried to extend it. Now everything they sell has some Internet features, but they didn't manage to spread Active Server Pages everywhere, and Apache still tromps MS Internet Information Server.
I believe Microsoft will start by bundling Apache like Apple did, all "cleaned-up" with a Windows interface. Perhaps in the next Service Pack and in Windows 2000. Ballmer was saying the other day that Apache was simply better, in fact. They'll make a big deal about Microsoft supporting Open Source, and how they support innovations like that. [Emphases Mine]
This follows well.
Then Microsoft ActiveLinux 3.1 will come out (I thought that was the most probable name and no version number less than 2.2 - 3.1 is good luck for Microsoft.) They'll make it easily installable, perhaps add USB and NTFS support, and then use Active X and IE for UNIX to put a Windows interface on Linux that has some Win32 APIs, the way OS/2 had some Win32 API's with DAX.
No. I've got a couple problems with this. First, USB support is already in active development, as is NTFS support in the kernel... that point is moot. Of course they could always release a binary only driver, but like with VMware, I think they're likely to find kernel developers unwilling to maintain backward compatability to support binary only drivers. As well they should! What this means for Microsoft is that they will never control the momentum of kernel development unless they're willing to attempt a legal attack against the GPL; though that may well be in their interest.
I'd be willing to bet Microsoft keeps that interface closed. They'll argue using a creative interpretation of GPL or LGPL so that they won't have to reveal the source and perhaps any documented Win32 calls that they port over. What they do will probably be in violation of the (L)GPL, but what legal force will be able to challenge them? ESR doesn't have that kind of cash.
OK. More problems of minor ignorance. First of all, if MS were to release a proprietary Win32 widget library, this would be no different from The OpenGroup with their Motif product line, or Troll Tech with their previous QT-1.x product line. In both cases each company has (and had) every legal right to sell a proprietary binary only software package which could not only cost to develop for, but could even cost just to use in runtime licensing fees!
If these guys get to play that game, then certainly Microsoft can. The GPL only covers software which can be thought to derive from a previously GPL'd source base. For example, if I create a source tree completely by scratch the GPL has nothing to say about what kind of license I distribute under. But if I derive an offshoot from a previously GPL'd program, say my own special version of GNU ls, and I distribute this code in binary form for either free or for charge, then I must also offer to distribute source, and I can't prevent the recipient from also distributing either the binary or the source.
Harsh terms if you want to take something someone else wrote, change a line here and there (or not even that if you release under a BSD license), and then call it yours. But a Win32 emulation library, which has already been created for NT, is something wholly authored by Microsoft, and therefore completely their's to distribute under whatever license they choose. They don't even have to sell development libraries to third party developers; meaning they could, if they wanted to, completely monopolize the office software for Linux market given their current monoploy on office software and the immediate demand from the business community should they ever offer such a product. They could even use their old standard tactic of the Interface Shuffle by porting Win32 and MS Office to Linux, and then changing the Win32 layer regularly enough to prevent WINE from ever maintaining compatability.
This is what they mean by embrase and extend.
The LGPL license is another animal, designed to allow developers to offer a library of functions for other developers to include in their code without demanding that said developers using that library release their source upon releasing their software.
MS shops and ISV's won't care about that though. Those dyed-in-the-wool Microsoft groupies, PHBs and non-technical CEOs who've heard so much about Linux can try it without leaving the MS fold or the Windows interface.
And they'll be impressed. Instead of creating a bad distribution to discredit Linux, they'll create a really good one, with all the goodies and easy installation (that you can start from Windows like Caldera, probably). Microsoft users who need hand holding will use it, Windows users who want to get into Linux (but don't want to get their hands dirty) will use it, and ISVs will write to that closed interface, because Microsoft will give them tools to make an easy port that lets Windows program keep Win32 calls for the UI. Microsoft will make ActiveLinux become the path of least-resistance for users and vendors from the Windows world who want to expand into Linux but have no interest in free-source issues.
Microsoft won't do it unless they think it will significantly extend their MS Office monopoly. This is where the serious bucks are for them anyway. If MS ever begins to think Corel's Office for Linux, or StarOffice, might become a serious threat to MS Office marketshare (+15% of the market, say) then this is exactly the tactic I'd expect them to take.
If Microsoft succeeds in keeping the interface closed, and porting Office and BackOffice to their distribution, then it'll mean trouble because other distros won't look very competitive to the new users and the industry press. Microsoft could end up gathering more users than the other distributions, and then I don't know what...Microsoft was never able to get most of the web servers to run on MS-ware, so they didn't succeed there, but what if they managed to amass the biggest share of Linux users? Would it give them enough power to influence Linux's direction? [Emphases Mine]
You can't stop Microsoft from releasing their own commercial and costly Linux distribution with a proprietary Win32/MS Office kicker, they're in their legal rights to do so. This is no different from Apple releasing the core MACH and BSD components from MacOS-X under the APSL while keeping the crown jewels (Carbon and the OpenStep libraries) propritary. But like I said, the Linux core components like it's kernel and basic tools will never be under Microsoft's control unless they can pull a legal victory out of their ass which undermines the GPL. If they pull that stinker off the entire Open Source community is fucked. And it might well be in their interest to attempt this move if they ever think the dollar losses from OS/Office sales to Linux ever exceeds the potential PR disaster from such a move.
But baring such a casastrophic -- and unlikely -- outcome, we should expect that even if Microsoft released the most popular distribution of Linux, with all sorts of proprietary and expensive add ons to run commercial MS software, plenty of people will still just tinker around and maintain their own special Linux flavors. Do you really think Debian would go away? Not even if Caldera, Suse, and RedHat were flattened by the MS steamroller could they stop the Debian project without a Legal or Congressional victory (heh, maybe a bullshit Free Software Tax Act, that might make for good assault legislation against free software and is right up this congresses alley). Just like MS can't prevent end developers from creating a quality stable kernel and giving that away for free, they can't prevent a bunch of developers and end users from collecting a bunch of free software, creating a usable Linux distribution, and giving that away for free.
Just like how Microsoft can't control the kernel source tree, the Debian project is exactly why Microsoft could take over the lion's share of the Linux user market as a vector to generate revenue from a Win32 and MS Office port, yet still not control all the Linux distributions out there. I don't know about you, but I can live without MS Office for Linux. I won't be paying astronomical fees just to run bad software because Frank down the street is too stupid to convert his email to ascii. But hey, that's me.
Why in the world is a slashdotter browsing jobs on the Microsoft web site? Hmmmmmmm?
--
Save the whales. Feed the hungry. Free the mallocs.
Re:Just a question
by
Anonymous Coward
·
· Score: 3
hehe.. I made a bet with a guy at work that a linux search at M$ would turn up a job description. I was guessing but after all the new jobs with keyword llinux in them compared to six months ago I assumed I had a good shoot. Can't wait to see what the "press" does with it.
Slashdot Mirror
Inquiring minds want to know:
Did Vinod V. meet with an untimely end?
Did he succumb to the 'seductions' of Open Source and turn to the Light Side?
Or has he been banished for his part in an embarrasment to M$?
Lacking <sarcasm> tags,
And you will come through a thousand battles unscathed. -Sun Tzu
The job post seems harmless enough, but it is, alas, the middle steps in their embrace and extend policy for shutting Linux out of the operating system market.
In a business environment where reading Sun Tzu is required, this shows how deeply the warrior philosophy is engrained. They are seeking knowlege of the strengths and weaknesses of Linux. They wish to find these so they can integrate the strengths into the next version of NT, while using the weaknesses to destroy Linux vendors. They are not afraid of lengthenning the NT code base to get what they want - if there are features in Linux that aren't in NT, then they will publicly praise these features. If there are weaknesses in Linux that aren't in NT, then they will publicly praise
NT. They do not praise Linux, or Apache (as Ballmer was accused of doing) but instead the features.
My friends, features can be stolen. They have all the money in the world to steal features with.
They will tell the poor sap who takes this job that they are trying to improve NT by adding Linux-like features. They will tell themselves that. But their vision of a single OS for everything is more important than the methods used to achieve it.
This person they wish to hire is the middle step in the chain to deflating the Linux user base. They will use this person to take all the features in Linux that can be taken and put them into NT. And then they will use this person to say "Look, Linux is now not as good as this thing me and Bill Gates have made: Windows 2002; Linux + Windows" This is exactly how they attacked Java, OS/2, Macintosh, and others.
-Ben Shniper
Not that I disbelieve you, but when someone posts "I see BSOD's with NT all the time" without posting specific STOP messages or some indication of what is going on, I'm inclined to right it off as Linux advocacy at it's worst, or maybe if I'm charitable, just someone whose confused NT with 95.
Think if somone posted "Linux Kernel PANICS for me all the time!" You wouldn't give it much thought unless they had some specific debugging info.
--
Business. Numbers. Money. People. Computer World.
This is not to say that I approve of all of the various unfair Microsoft strategies, including writing FUD. But doesn't every company say why you shouldn't use their competition?
Bruce
Bruce Perens.
Thanks for the reply, maynard. I feel kind of honoured that you spent so much time on my comment. :)
:)
:) Even so, thanks for setting me straight on a few things...I should be more careful.
Grenamier makes the point that Microsoft could create their own distribution of Linux with a proprietary Win32 kicker...How different is this from Corel using Linux as a vector to market their proprietary Corel Office software?
Good question. On fourth thought, there isn't much difference, especially since Corel has already composed its own distribution. OTOH, I do think there's a slight difference in that it's in Corel Office's best interests to have Linux stay free and open and carrying along as it has. Microsoft would be interested in having Linux development slow down or be influenced to allow Microsoft more control over its direction.
That said, I don't want people to get the idea that I'm really anti-Microsoft. At home I run Windows 98, Windows NT and Linux (on a separate computer). I'd prefer to have NT and Linux both succeed. Linux really is something special though, and I don't to see anything badly interfering with it.
Even though the premise is good, there are some fundamental mistakes of fact which ought to be resolved:
Thanks for clearing up these things for me. I should've known better on a lot of them.
Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that.
[snip] But they can't "Steal" something out in the open like a GPL'd package, so I don't know what you mean by that.
Some of the previous replies speculated that Microsoft might be looking for features of Linux to steal for NT. I was deciding that it wouldn't be a good enough reason to build a distribution. As far as "stealing" goes, I hadn't thought much about it, I suppose stealing from free-software could be burying GPL'ed code in a closed-source product and ignoring the GPL's terms.
No. I've got a couple problems with this. First, USB support is already in active development, as is NTFS support in the kernel [...] that point is moot.
I should've remembered this, but I was only tossing out ideas on how Microsoft might have wanted to candy-coat the distribution for new users coming from Windows.
[...] What this means for Microsoft is that they will never control the momentum of kernel development unless they're willing to attempt a legal attack against the GPL; though that may well be in their interest.
It would be unlikely for Microsoft to attempt something like that for a while until people have gotten used to the notion of a Microsoft distribution and some of their lawyers get some vacation time.
OK. More problems of minor ignorance. First of all, if MS were to release a proprietary Win32 widget library, this would be no different from The OpenGroup with their Motif product line, or Troll Tech with their previous QT-1.x product line. In both cases each company has (and had) every legal right to sell a proprietary binary only software package which could not only cost to develop for, but could even cost just to use in runtime licensing fees!
If these guys get to play that game, then certainly Microsoft can. The GPL only covers software which can be thought to derive from a previously GPL'd source base. For example, if I create a source tree completely by scratch the GPL has nothing to say about what kind of license I distribute under. But if I derive an offshoot from a previously GPL'd program, say my own special version of GNU ls, and I distribute this code in binary form for either free or for charge, then I must also offer to distribute source, and I can't prevent the recipient from also distributing either the binary or the source.
I made a bad assumption here. I thought Microsoft would need to modify the kernel and a lot of the system software in order to create a UI with the same infrastructure and integration with the system as the Explorer UI on Windows. Of course, WINE, GNOME and KDE show that such changes wouldn't really be necessary. Duh on me.
[...] But baring such a casastrophic -- and unlikely -- outcome, we should expect that even if Microsoft released the most popular distribution of Linux, with all sorts of proprietary and expensive add ons to run commercial MS software, plenty of people will still just tinker around and maintain their own special Linux flavors. Do you really think Debian would go away? Not even if Caldera, Suse, and RedHat were flattened by the MS steamroller could they stop the Debian project without a Legal or Congressional victory (heh, maybe a bullshit Free Software Tax Act, that might make for good assault legislation against free software and is right up this congresses alley). Just like MS can't prevent end developers from creating a quality stable kernel and giving that away for free, they can't prevent a bunch of developers and end users from collecting a bunch of free software, creating a usable Linux distribution, and giving that away for free.
Quite true. But I'm not really concerned that other distributions could become extinct, more that Microsoft could influence Linux development in ways that don't really benefit Linux. For instance, I wouldn't want to see Linux catch a case of creeping featuritis that somehow spreads into the kernel. In theory, Linus and the people who work on the code would be able to avoid such a thing simply by saying no, but what if in the future a large number of users demand feature after feature after feature to be added to parts of Linux that have a wide impact? A compromise here, a lapse there and suddenly the kernel is growing or becoming buggier, perhaps...certainly we can live without MS Office (actually, I really like Word and Outlook on NT), but I hope most of the future's changes to Linux can be of the take-it-or-leave-it variety.
I get paranoid about good things that become too popular. Like this great seafood restaurant I know that's kind of gotten crappy since it suddenly started becoming "cool". With all the masses of people expecting things now, now now, there just isn't as much care taken in the food as there used to be. Now it tastes the same as all the other restaurants and it's hard to find a reason to pick this one over the others. I don't have to eat the new dishes that they serve now, but I want my old fried lobster back the way it was before.
I've snipped a lot of what you wrote, most of which I haven't marked. Not to ignore it, but because I agreed and I didn't have anything to add. I probably didn't add much here anyway.
-- John Truong
Did anyone notice that no actual Linux experience was required for this job ? And that excellent writing qualities were sought ? Sounds like they are looking for a FUD-farmer...
superblog.org: all your favourite blogs on o
The simple explination is "know thy enemy". Doing a competitive analysys is pretty common, and I'm sure that the marketing people over at RedHat, VAR, and so on have done these. And I'd venture to guess they do this not only against MS products, but against SCO, IBM, probably even other Linux vendors.
MS is looking for what to say when a customer says "but what about that Linux thing? why is Win2000 better?", they'll know what to say besides "it's free software".
If anything, I take this as one of the best compliments the Linux community can get. We're really competition. Now if only we can get a hold of the complete analysis when complete....
To me, it looks like Microsoft is creating its own distribution of Linux. For one clue, the title of the job is "Product Manager". No company has Product Managers for products they don't sell. Some of the duties include "product development for the development team" and "technical assistance for the sales force."
The position also advertises "high visibility" outside of the company. People who just watch competitors and make reports about them don't normally get much exposure (VinodV didn't get his 15 minutes on purpose). But if Microsoft is selling a "new" product, its Product Manager would be expected to flog it to the press.
So what's Microsoft up to? Scrapping NT for Linux? Of course not, they'd be throwing away all of their leverage. Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that.
This is my wild speculation:
Free-sourced software is the Internet happening all over again for Microsoft...everyone in the industry is jumping on the bandwagon and stories about the success of Linux, Apache and Perl are everywhere. Microsoft missed the populization of yet another decades-old "phenomenon" and they've known it since Hallowe'en at the least.
What did Microsoft do about the Internet? Embraced and tried to extend it. Now everything they sell has some Internet features, but they didn't manage to spread Active Server Pages everywhere, and Apache still tromps MS Internet Information Server.
I believe Microsoft will start by bundling Apache like Apple did, all "cleaned-up" with a Windows interface. Perhaps in the next Service Pack and in Windows 2000. Ballmer was saying the other day that Apache was simply better, in fact. They'll make a big deal about Microsoft supporting Open Source, and how they support innovations like that.
Then Microsoft ActiveLinux 3.1 will come out (I thought that was the most probable name and no version number less than 2.2 - 3.1 is good luck for Microsoft.) They'll make it easily installable, perhaps add USB and NTFS support, and then use Active X and IE for UNIX to put a Windows interface on Linux that has some Win32 APIs, the way OS/2 had some Win32 API's with DAX.
I'd be willing to bet Microsoft keeps that interface closed. They'll argue using a creative interpretation of GPL or LGPL so that they won't have to reveal the source and perhaps any undocumented Win32 calls that they port over. What they do will probably be in violation of the (L)GPL, but what legal force will be able to challenge them? ESR doesn't have that kind of cash.
MS shops and ISV's won't care about that though. Those dyed-in-the-wool Microsoft groupies, PHBs and non-technical CEOs who've heard so much about Linux can try it without leaving the MS fold or the Windows interface.
And they'll be impressed. Instead of creating a bad distribution to discredit Linux, they'll create a really good one, with all the goodies and easy installation (that you can start from Windows like Caldera, probably). Microsoft users who need hand holding will use it, Windows users who want to get into Linux (but don't want to get their hands dirty) will use it, and ISVs will write to that closed interface, because Microsoft will give them tools to make an easy port that lets Windows program keep Win32 calls for the UI. Microsoft will make ActiveLinux become the path of least-resistance for users and vendors from the Windows world who want to expand into Linux but have no interest in free-source issues.
If Microsoft succeeds in keeping the interface closed, and porting Office and BackOffice to their distribution, then it'll mean trouble because other distros won't look very competitive to the new users and the industry press. Microsoft could end up gathering more users than the other distributions, and then I don't know what...Microsoft was never able to get most of the web servers to run on MS-ware, so they didn't succeed there, but what if they managed to amass the biggest share of Linux users? Would it give them enough power to influence Linux's direction?
I'm not trying to be alarmist, I'm just trying to figure out their strategy...I could have it all wrong, but I do believe Microsoft will create its own distribution, and that it will try to create some reason to use Microsoft's distro instead of anyone else's. And Rob shouldn't work for them...what would happen to Slashdot??
Sorry for the length, I got off on a tear there...
-- John Truong
Even though the premise is good, there are some fundamental mistakes of fact which ought to be resolved:
Grenamier (jtruong@idirect.com) Wrote:
To me, it looks like Microsoft is creating its own distribution of Linux. For one clue, the title of the job is "Product Manager". No company has Product Managers for products they don't sell. Some of the duties include "product development for the development team" and "technical assistance for the sales force."[Emphases mine]
The position also advertises "high visibility" outside of the company. People who just watch competitors and make reports about them don't normally get much exposure (VinodV didn't get his 15 minutes on purpose). But if Microsoft is selling a "new" product, its Product Manager would be expected to flog it to the press.
So what's Microsoft up to? Scrapping NT for Linux? Of course not, they'd be throwing away all of their leverage. Trying to steal Linux technology for NT? Perhaps, but they wouldn't need to sell a product of their own for that.[emphases mine]
Yes, I agree completely. They wouldn't bother with a project unless they were planning to either make something for sale, or were planning a "grassroots" marketing effort. But they can't "Steal" something out in the open like a GPL'd package, so I don't know what you mean by that.
This is my wild speculation:
Free-sourced software is the Internet happening all over again for Microsoft...everyone in the industry is jumping on the bandwagon and stories about the success of Linux, Apache and Perl are everywhere. Microsoft missed the populization of yet another decades-old "phenomenon" and they've known it since Hallowe'en at the least.[Emphases Mine]
I really like that quote. It well describes the mentality which must be fermenting in MS upper management strategy meetings.
What did Microsoft do about the Internet? Embraced and tried to extend it. Now everything they sell has some Internet features, but they didn't manage to spread Active Server Pages everywhere, and Apache still tromps MS Internet Information Server.
I believe Microsoft will start by bundling Apache like Apple did, all "cleaned-up" with a Windows interface. Perhaps in the next Service Pack and in Windows 2000. Ballmer was saying the other day that Apache was simply better, in fact. They'll make a big deal about Microsoft supporting Open Source, and how they support innovations like that. [Emphases Mine]
This follows well.
Then Microsoft ActiveLinux 3.1 will come out (I thought that was the most probable name and no version number less than 2.2 - 3.1 is good luck for Microsoft.) They'll make it easily installable, perhaps add USB and NTFS support, and then use Active X and IE for UNIX to put a Windows interface on Linux that has some Win32 APIs, the way OS/2 had some Win32 API's with DAX.
No. I've got a couple problems with this. First, USB support is already in active development, as is NTFS support in the kernel... that point is moot. Of course they could always release a binary only driver, but like with VMware, I think they're likely to find kernel developers unwilling to maintain backward compatability to support binary only drivers. As well they should! What this means for Microsoft is that they will never control the momentum of kernel development unless they're willing to attempt a legal attack against the GPL; though that may well be in their interest.
I'd be willing to bet Microsoft keeps that interface closed. They'll argue using a creative interpretation of GPL or LGPL so that they won't have to reveal the source and perhaps any documented Win32 calls that they port over. What they do will probably be in violation of the (L)GPL, but what legal force will be able to challenge them? ESR doesn't have that kind of cash.
OK. More problems of minor ignorance. First of all, if MS were to release a proprietary Win32 widget library, this would be no different from The OpenGroup with their Motif product line, or Troll Tech with their previous QT-1.x product line. In both cases each company has (and had) every legal right to sell a proprietary binary only software package which could not only cost to develop for, but could even cost just to use in runtime licensing fees!
If these guys get to play that game, then certainly Microsoft can. The GPL only covers software which can be thought to derive from a previously GPL'd source base. For example, if I create a source tree completely by scratch the GPL has nothing to say about what kind of license I distribute under. But if I derive an offshoot from a previously GPL'd program, say my own special version of GNU ls, and I distribute this code in binary form for either free or for charge, then I must also offer to distribute source, and I can't prevent the recipient from also distributing either the binary or the source.
Harsh terms if you want to take something someone else wrote, change a line here and there (or not even that if you release under a BSD license), and then call it yours. But a Win32 emulation library, which has already been created for NT, is something wholly authored by Microsoft, and therefore completely their's to distribute under whatever license they choose. They don't even have to sell development libraries to third party developers; meaning they could, if they wanted to, completely monopolize the office software for Linux market given their current monoploy on office software and the immediate demand from the business community should they ever offer such a product. They could even use their old standard tactic of the Interface Shuffle by porting Win32 and MS Office to Linux, and then changing the Win32 layer regularly enough to prevent WINE from ever maintaining compatability.
This is what they mean by embrase and extend.
The LGPL license is another animal, designed to allow developers to offer a library of functions for other developers to include in their code without demanding that said developers using that library release their source upon releasing their software.
MS shops and ISV's won't care about that though. Those dyed-in-the-wool Microsoft groupies, PHBs and non-technical CEOs who've heard so much about Linux can try it without leaving the MS fold or the Windows interface.
And they'll be impressed. Instead of creating a bad distribution to discredit Linux, they'll create a really good one, with all the goodies and easy installation (that you can start from Windows like Caldera, probably). Microsoft users who need hand holding will use it, Windows users who want to get into Linux (but don't want to get their hands dirty) will use it, and ISVs will write to that closed interface, because Microsoft will give them tools to make an easy port that lets Windows program keep Win32 calls for the UI. Microsoft will make ActiveLinux become the path of least-resistance for users and vendors from the Windows world who want to expand into Linux but have no interest in free-source issues.
Microsoft won't do it unless they think it will significantly extend their MS Office monopoly. This is where the serious bucks are for them anyway. If MS ever begins to think Corel's Office for Linux, or StarOffice, might become a serious threat to MS Office marketshare (+15% of the market, say) then this is exactly the tactic I'd expect them to take.
If Microsoft succeeds in keeping the interface closed, and porting Office and BackOffice to their distribution, then it'll mean trouble because other distros won't look very competitive to the new users and the industry press. Microsoft could end up gathering more users than the other distributions, and then I don't know what...Microsoft was never able to get most of the web servers to run on MS-ware, so they didn't succeed there, but what if they managed to amass the biggest share of Linux users? Would it give them enough power to influence Linux's direction? [Emphases Mine]
You can't stop Microsoft from releasing their own commercial and costly Linux distribution with a proprietary Win32/MS Office kicker, they're in their legal rights to do so. This is no different from Apple releasing the core MACH and BSD components from MacOS-X under the APSL while keeping the crown jewels (Carbon and the OpenStep libraries) propritary. But like I said, the Linux core components like it's kernel and basic tools will never be under Microsoft's control unless they can pull a legal victory out of their ass which undermines the GPL. If they pull that stinker off the entire Open Source community is fucked. And it might well be in their interest to attempt this move if they ever think the dollar losses from OS/Office sales to Linux ever exceeds the potential PR disaster from such a move.
But baring such a casastrophic -- and unlikely -- outcome, we should expect that even if Microsoft released the most popular distribution of Linux, with all sorts of proprietary and expensive add ons to run commercial MS software, plenty of people will still just tinker around and maintain their own special Linux flavors. Do you really think Debian would go away? Not even if Caldera, Suse, and RedHat were flattened by the MS steamroller could they stop the Debian project without a Legal or Congressional victory (heh, maybe a bullshit Free Software Tax Act, that might make for good assault legislation against free software and is right up this congresses alley). Just like MS can't prevent end developers from creating a quality stable kernel and giving that away for free, they can't prevent a bunch of developers and end users from collecting a bunch of free software, creating a usable Linux distribution, and giving that away for free.
Just like how Microsoft can't control the kernel source tree, the Debian project is exactly why Microsoft could take over the lion's share of the Linux user market as a vector to generate revenue from a Win32 and MS Office port, yet still not control all the Linux distributions out there. I don't know about you, but I can live without MS Office for Linux. I won't be paying astronomical fees just to run bad software because Frank down the street is too stupid to convert his email to ascii. But hey, that's me.
[snip]
#define COMMENT_TONE HUMOROUS
Why in the world is a slashdotter browsing jobs on the Microsoft web site? Hmmmmmmm?
Save the whales. Feed the hungry. Free the mallocs.