Microsoft Counted As Key Linux Contributor

alphadogg writes "For the first time ever, Microsoft can be counted as a key contributor to Linux. The company, which once portrayed the open-source OS kernel as a form of cancer, has been ranked 17th on a tally of the largest code contributors to Linux. The Linux Foundation's Linux Development Report, released Tuesday, summarizes who has contributed to the Linux kernel, from versions 2.6.36 to 3.2. The 10 largest contributors listed in the report are familiar names: Red Hat, Intel, Novell, IBM, Texas Instruments, Broadcom, Nokia, Samsung, Oracle and Google. But the appearance of Microsoft is a new one for the list, compiled annually."

Comments

whoa

[Anonymous Coward]

Did hell freeze over already??

Re:whoa

[Soilworker]

No. Microsoft just found a way to make money on open source OS.

Re:whoa

[ILongForDarkness]

My understanding is a lot of the stuff they contribute is to get things that should be interoperable there, eg. smb and of course interop helps sell a more hetrogenous environment to corps (so they don't all run and flee to linux, but also linux doesn't break when talking to a Win server).

Re:whoa

[eternaldoctorwho]

Why not? It looks like this will be the year of the linux desktop!

Re:whoa

[realityimpaired]

IIRC from reading the forums and bugreports, samba has accumulated plenty of printing regressions since 3.2 or so, and nothing was ever done about them. It's been quite long since one could use, say, driver for HP LaserJet 8000/8100 directly via samba, without using a local printer port :(

It's also been a long time since I've seen a network-connected printer that didn't have an IPP server built in.

Come to it, it's actually been a long while since I've seen anybody try to use SAMBA to host a print server. Just use CUPS or some other IPP server if you don't have a printer with built-in print capability.

Re:whoa

[ogdenk]

Seems to work fine in OSX. Not sure what you're talking about. IPP printers can be autoinstalled as well as SMB. Can even participate in a Windows Domain and be managed by the domain if you want including scripted mapping of printers and shares. Linux can enjoy most of the same goodies with a little effort.

Quit acting like Microsoft invented LDAP and autoconfiguration. Been around a long time. If it doesn't work in your environment, ditch the retard MCSE and hire a real network admin that knows what he's doing with a broad scope on more than one platform.

Re:whoa

[ozmanjusri]

I heard a talk somewhere about SMB 2.2 features and how the standards were going to get published to help others adapt.

Yes, Microsoft published SMB standards out of the goodness of their hearts, and the threat of continuation of fines of US$2.39 million/day unless they complied.
http://en.wikipedia.org/wiki/European_Union_Microsoft_competition_case

Then they promptly changed their OS so it wouldn't interoperate with the standard...

that's for virtualization.

[Janek Kozicki]

I was wondering "why the hell?" TFA says:

"Much of the work Microsoft did centers around providing drivers for its own Hyper-V virtualization technology. Microsoft's Hyper-V, part of Windows Server, can run Linux as a guest OS."

Why that couldn't be included in the summary?

Re:that's for virtualization.

[Chris Burke]

It's a juicier narrative to portray it in the way the summary did--that even though Microsoft once depicted Linux as a "cancer", Linux must now be so awesome that Microsoft is one of its key contributors. Providing context buffs out some of that luster.

Yeah, instead it's that Linux is so awesome Microsoft can't afford not to ensure it is compatible with their hypervisor. Of course it's no surprise that in the virtualization market being able to virtualize Linux is a key feature.

I find it far more intriguing that the key contributors to Linux are companies and not independent individuals, since the old storyline used to be that devoted hobbyists were gathering on the internet to do a better job than commercial companies, back when the "year of Linux on the desktop" was always right around the corner.

That hasn't been the case since Linux became Linus' job. Though those hobbyists -- including Linus -- did a good enough job that they companies took notice, now didn't they? For over a decade the many contributions of companies to Linux -- not least of which being distros like Red Hat -- have been used as proof that Free Software doesn't mean the death of the paid programmer.

Of course anyone who thought it did in the first place didn't understand the market for programmers. It's always been the case that the majority of programmers are employed solving the specific business needs of specific companies, not creating shrink-wrap per-license software.

I call B.S.

[PCM2]

Last I heard, all of Microsoft's contributions to the Linux kernel have been strictly to improve Linux support for Microsoft products, e.g. to allow Windows Server to be a host for Linux clients. That's fine, but it hardly counts as "key" contributions in my book.

Re:I call B.S.

[TWX]

If it increases interoperability, that is a rather high priority for any operating system whose proponents wish it to remain viable. I am no fan of Redmond, but I have managed to make a lot of money supporting their products.

Re:I call B.S.

[John Mister]

Last I heard, all of Microsoft's contributions to the Linux kernel have been strictly to improve Linux support for Microsoft products, e.g. to allow Windows Server to be a host for Linux clients. That's fine, but it hardly counts as "key" contributions in my book.

Why wouldn't it count as key contributions? Windows has market share of 95% on desktops and almost 50% on servers (used more on internal servers like exchange than your typical apache+centos cheapo host). Still, MS works to maintain some compatibility when they really have no reason to. I think that deserves some appreciation.

Re:I call B.S.

[PCM2]

If you make a closed, proprietary system that is not interoperable, then work to change everybody else's system so that it can work with yours, do you really deserve a pat on the back for that? Every action from beginning to end was wholly self-serving.

Re:I call B.S.

[maroberts]

Last I heard, all of Microsoft's contributions to the Linux kernel have been strictly to improve Linux support for Microsoft products, e.g. to allow Windows Server to be a host for Linux clients. That's fine, but it hardly counts as "key" contributions in my book.

A large number of contributors put in source code which is "relevant to their interests". e.g. graphics card manufacturers contribute towards open source drivers and improvements to X.

Personally I see nothing wrong with this, and quite frankly makes a good change from when Microsoft did everything possible to hide how their stuff works e.g. *cough*Samba*cough*

Re:I call B.S.

[PCM2]

Still, MS works to maintain some compatibility when they really have no reason to.

No reason at all? Are you sure you've thought this through?

Re:I call B.S.

[PCM2]

Every single contribution to the linux kernel (or any open source project) is inherently self serving. Every one of the companies listed benefits from the contributions they provide. That's the entire point of open source, you modify it to suit your needs. So what if you don't like Microsoft, too fucking bad.

Yes, but unlike Microsoft, most of the companies who contribute hardware drivers to the Linux kernel (such as Broadcom, for example), don't have a history of trying to destroy Linux. In this case, the fact that Slashdot is claiming Microsoft is suddenly "a key Linux contributor" is even more valuable to Microsoft than the actual kernel contributions it has made. Framing the story in this way helps Microsoft craft messaging that subverts Linux.

Hyper-V

[Enderandrew]

I do believe they've basically only added support for running Linux as a guest OS within their VM solution, Hyper-V. They haven't contributed to the betterment of Linux on the whole.

Re:Hyper-V

[Trepidity]

I do think that's a legitimate contribution, even if it's obviously self-interested. To the extent that people use Hyper-V, it's good for Linux to have support for running under it, and it's good that Microsoft contributed the resources to make that happen instead of leaving it for other contributors to try to get it working. Similar to how Sun/Oracle employees contributed a considerable amount of the kernel's Xen support.

It is fair to be aware that that's the entirety of their contribution, so it doesn't signal some more general engagement with kernel development.

Re:Hyper-V

[TheNinjaroach]

They haven't contributed to the betterment of Linux on the whole.

I was with you on this for the past decade. Then on November of 2011, they went and did this. Real Linux drivers for SQL Server? Yeah!

And if you don't think that counts towards the betterment of Linux, then we're just going to have to disagree!

Is this news to anyone?

[John Mister]

Microsoft has always been one of the best innovators about new technology. Against the popular belief on Slashdot, they have contributed a lot to computer technology, innovations, and of course, Linux too. Stop the hate and accept that Microsoft also has many technically knowledgeable persons who also contribute to Linux. When reading this hate about MS I can't but think that YOU are who is having problems with dealing with it.

Re:Is this news to anyone?

[lattyware]

I didn't see any hate there, just surprise that a company was contributing to what is essentially a rival product. That's pretty reasonable, and doesn't portray Microsoft in a bad light at all.

Re:Is this news to anyone?

[Anonymous Coward]

I hope this was meant to be funny.

Of course not. Microsoft has contributed more to the knowledge base of predatory marketing, monopoly abuse, and price manipulation than any other company in recent memory. You would have to go all the way back to Standard Oil to find a company that has given us more in these important fields. Many companies look up to Microsoft and use their history as an template for themselves, and like Microsoft, they are much richer for it.

Re:Is this news to anyone?

[s.petry]

Microsoft has always been one of the best innovators about new technology. Against the popular belief on Slashdot, they have contributed a lot to computer technology, innovations, and of course, Linux too.

First, I really hope you do some investigation on the history of Microsoft and the products that they claim to have innovated. You will find that many of us have very legitimate bias against Microsoft and their so called "innovation".

Example: Microsoft during Windows 95 release was adamant about not producing a TCP/IP stack for Windows claiming that the Internet was a waste of time and there is nobody in their right mind that would use it. Microsoft released and poured cash into their own proprietary network protocol (NetBUI).

When the hopes of crushing the Internet were dashed, Microsoft started releasing a TCP stack which broke communication with non Windows hosts. The original TCP/IP specification was to respond to an ACK once. Microsoft released a stack which sent and expected 2, and invented the term "Crippled Network" for anything that did not respond that way. Throttling bandwidth to any non-Microsoft host to make it appear that anything was slower than Microsoft. (An interesting piece of trivia is that most *NIX was still faster than Windows at networking even with the throttled bandwidth.).

Sun found the (to be kind) quirk that Microsoft had build in to their TCP/IP stack. This was reported everywhere, and most vendors started releasing similar code because Microsoft refused to follow the specification. As vendors migrated their stacks, Microsoft increased the ACK count again. At least they stopped reporting any non windows host as "crippled" which stopped many of the complaints to other vendors about "Why does windows show your OS as crippled?"

This is a company that has done the same with any open specification that they adopt. Kerberos, NFS, LDAP, and the list can go on and on and on.

When it comes to "innovation", Microsoft does do a good job of watching the market and buying up things that appear to be good. Often times, this puts many other good companies out of business. Example here is that in WIndows 98 time, there were several web rating companies. NetNanny, Cybersitter, and more. Windows liked their ideas so much, they put a very limited and broken version of that service in to Windows and put all of those companies out of business. Not so much innovation here, but rather a predatory method of dealing with competition which people dislike.

Stop the hate and accept that Microsoft also has many technically knowledgeable persons who also contribute to Linux. When reading this hate about MS I can't but think that YOU are who is having problems with dealing with it.

Honestly, I think Microsoft has done a good job at giving people a consistent look and feel on a computer. For some odd reason, they do away with in Windows 7, and Office 2010 and the "Ribbons" which is why there is such a low adoption rate and Microsoft started losing more market share than they should.

Outside of the look and feel, Microsoft has not innovated anything in the market. I wish that was a troll statement, but nothing they have done has been "new" or innovative. That's not to say that they have no patents, but every patent I have seen could be invalidated in court. Look at the 7 they are suing B&N for as an example. All 7 of those are either obvious or have prior art. Groklaw has lots of information

When you see all the hate for Microsoft, do you ever wonder if it's warranted?

Re:Is this news to anyone?

[Anonymous Coward]

Apple is gaining on them though. Lately, MS has been one of the good guys (shame I'm not that found of their software solutions)

Re:Is this news to anyone?

[Anthony Mouse]

Lately, MS has been one of the good guys

So they claim. But it seems to me more that they're on the back foot and therefore incapable of acting too overtly malicious without causing excessively many customer defections. I mean they're still doing this, and patent trolling, and pushing automatic updates to Internet Explorer that default to making Bing your search engine even though nobody likes it, etc.

They've still got a ways to go before anyone ever trusts them again. Like years. That's what happens when you ruin your own reputation.

Re:Is this news to anyone?

[jo_ham]

The iMac hard drive thing has been blown out of proportion. They used a custom firmware to repurpose the LED activity light pin (they they don't use) to carry temperature information to cut down on parts and part variability. However, their own documentation has instructions for what to do if installing a non-special firmware drive in that bay (eg, one of Apple's own SSDs if you specify that as a BTO option, or a third party replacement drive); you install a jumper to short two pins together and it carries on as normal, and knows not to attempt to monitor the internal temperature of a non-special drive.

If you get an iMac from the factory with an SSD in that bay, the pins come pre-shorted with a little jumper installed at the factory. They just didn't tell anyone about it, since they don't consider the internals on an iMac to be user serviceable.

Re:Is this news to anyone?

[ogdenk]

Bullshit. MacOS existed LONG before Windows was anything more than a glorified DOS shell. The internet was in full swing on UNIX boxes and Macs before crappy Windows machines flooded the net via AOL. Most PC users were late to the party and saddled with a shitty IP stack (remember Trumpet Winsock?) and buggy software. And DOS wasn't even REALLY "their" innovation. They bought a half-assed CP/M clone from some guy named Tim Patterson and claimed they had an OS the whole time to IBM.

We existed just fine before Win95. We had slick GUI-based machines like the Atari ST, the Amiga, and the Macintosh that were ALL superior to MS's offerings. And not in a small way.....we're talking LEAPS AND BOUNDS more advanced. And all were capable of online connectivity. All had more capable graphics and sound than PC's of the same time period.

The ONLY thing MS brought to the table was "Good Ol' Boy" predatory business tactics, manipulation and extortion. They INVENTED absolutely NOTHING and forced an industry into shoving unreliable cheap PC clones down our throats with their software bundled. I hope Gates and Ballmer choke on their breakfast. F**k them.

Re:Is this news to anyone?

[aiht]

In all seriousness, there's some cool stuff coming out from Microsoft Research. Everything else, if it can be considered innovative, is half-baked. The dimwits who specified and documented winapi had no clue how to formally specify stuff. Thus all the undocumented behavior that applications exploit in light of no documentation and no clear direction as to the rationale and intended uses behind various APIs. Thus we have stuff that MS had to work around over and over to maintain compatibility with broken applications; stuff that wine people have to deal with as well. As far as MS complaining that app writers are getting things wrong: well duh idiots, you can't write the docs, you'll pay for it. Yeah, I've been consistently pissed about that, even back in the times of 16 bit winapi -- even as a kid back then I realized that they were not saying things that should have been said.

Of course with various non-standard Linux APIs, you're entirely on your own. But at least there's no pretense of documentation, and you can look at the code.

Wow, you've really managed to piss off an AC today, huh?

As a fun example of winapi, I happen to have an MSDN page open right now on the GetDIBits function. It copys bitmap pixels around, and it returns:
on success: nonzero or "the number of scan lines copied from the bitmap." (because that's very helpful)
on failure: 0
Then it also says "This function can return the following value: ERROR_INVALID_PARAMETER"
And what is the value of ERROR_INVALID_PARAMETER? 87
How does someone sit down a design an API that returns a positive value on succes, and also on failure?

Re:Is this news to anyone?

[LordLimecat]

Example: Microsoft during Windows 95 release was adamant about not producing a TCP/IP stack for Windows claiming that the Internet was a waste of time and there is nobody in their right mind that would use it. Microsoft released and poured cash into their own proprietary network protocol (NetBUI).

Funfact: When I was working on my A+ back around 2004, the course material we used indicated that "one" of the communications protocols was TCPIP, but it was esoteric and of course everyone used NetBEUI.

I think I still have the book talking about how the future is NetBEUI and how TCP/IP is some backwoods protocol that noone uses.

Re:Is this news to anyone?

[ogdenk]

You missed my underlying point. I'm not arguing who had the best product. I agree, it wasn't Microsoft. We got to this point in consumer technological advancement primarily because of Microsofts marketing.

Wrong. You missed my point. We advanced a LOT QUICKER before they arrived and dominated the industry through force.

We already had great inexpensive machines that were advancing quite fast. The difference is your great grandmother didn't and we didn't give a shit. To be fair to the elderly though, my WW2-vet granddad had an ST and then a Mac. Long before Win95.

The best products and services won't amount to a hill of beans without marketing. That alone changed the world. Forever.

That's is why people hate MS. It's the truth that never should have been. To the parent, it pains them too much to admit it. It's where idealism and reality clash head on.

Most of the groundwork for the "technological advancement" you see today existed in the early 80's. Marketing only made the Walmart crowd care and drove x86 PC prices and quality down. We were better off without them. It changed the world for the worse, not the better and has held TRUE advancement back a decade or two.

3D accelerated graphics existed before Windows. Web browsers existed on machines more capable than the PC in 1992. Gopher before that. Hi-rez displays and 24-bit color existed before Windows. Broadcast quality hardware-accelerated video playback existed before Win95. Advanced sound chips existed before Windows. Touch screens existed before Windows. Pen-based input has existed since 1952 on mainframes. Preemptive multitasking existed before Windows. GUI's have existed in various forms since the 70's.

MS also wasn't the first to combine this functionality but when everyone else did it, their machines were written off as scientific workstations, gaming toys or "just for creative types". You don't seem to realize the Amiga didn't die in 1990. Neither did the Atari ST/TT/Falcon. They were just forced out of the general US market because people were content buying a more expensive and less capable Packard Bell.

The only advancement that happened was the PC sucking up everyone else's hard work and research as they languished with MS claiming they invented something. You're right that MS won, you're wrong thinking we gained from it. We lost. Big time.

Re:Is this news to anyone?

[ogdenk]

The trash-80 was doomed from the start. The Atari 8-bit and C64 languished until the mid-90's due to a devoted user base. The ST and later TT/Falcon had quite a few professional applications in DTP and MIDI work. They found niches for a while and persisted until they got axed. The Amiga was popular for 3D modeling and video editing work. In pure numbers the PC outnumbered them but they remained dominant in quite a few niches for years afterward because the PC well......sucked. Even with more raw clock speed the PC......sucked. Only very recently has the PC sucked a lot less due to absorbing a lot of the features that made machines like the DEC, SGI and Amiga machines cool.

I don't care about growth. Retards will buy anything you can convince them they need. I care about real, honest-to-god innovation and engineering. MS has brought very little new to the table. Without MS the PC still would have had a dominant business foothold thanks to Novell, IBM OS/2 and various UNIX versions and wait......*GASP*..... LINUX which predates WinNT even. MacOS has been network-capable since the Mac Plus IIRC. Berkeley and Sun on the UNIX side contributed a lot to small-scale IT as well. The IT boom was already in full swing and growing fast before MS even got on the boat. Even infant Linux was around for the party on a small scale.

MS was the axe-wielding disruptive psycho latecomer in server-side business IT that seemed really good at sweet-talking execs.

Ethernet's been with us since the 70's thanks to DEC (RIP), Xerox and Intel. Localtalk's been around since I was a little kid. Small offices have been hooking computers together for ages. Ever even SEEN coax ethernet? You really think that crap came from the post-Win95 era? I helped run a bit of it though twisted pair wasn't far around the corner. Believe me, people saw value in personal computers before Windows existed. Especially in cubicle farms. Get off my lawn. Ever even HEARD of Digital Research and CP/M, DR/DOS and GEM (also used on ST)? They could have easily carried MS's torch had Gates never been born.

In other news...

[Lohrno]

Several backdoors were found in the latest versions of the code...

Remember - they were threatened

[Anonymous Coward]

Remember - they were threatened with having their HyperV drivers removed due to lack of support.

And that could easily have spelled disaster for their cloud capability.

compiled annually?

[ILongForDarkness]

This is linux, there list needs to go to daily builds :-)

Canonical

[gtirloni]

Nice to see Canonical finally contributing something to the Linux kernel.

Wait... Nevermind, my bad.

Top 20 varies quite a bit

[pavon]

About 24% of changes are the result of people who have not declared an association with any company, and there is a very long tail of companies that have small changes, so while the top 5 corporate contributers are fairly consistent, the top 20 varies significantly from release to release.

In this case, these drivers have been 2.5 years in the making. They had been held out of the kernel for that time because their quality wasn't up-to-par before finally being approved. The metric used in this report basically comes down to git commits, and includes all the commits that were made in private git branches before being folded into the mainline kernel. So Microsoft has 2.5 years worth of work on Hyper-V credited to them during the 6 months in question, which amounts to 1% of the changes in that time period. It is a one-time blip, and not indicative of a trend.

MS a key contributor?

[CosaNostra Pizza Inc]

And yet, they haven't updated the linux version of Skype since they acquired the company. I have to wonder what their motivations are.

Very old news: lwn.net had this in July 13, 2011

[nick_urbanik]

I'm surprised to see this as news; it was discussed about nine months ago in Jon Corbet's article in LWN.net.

K. Y. Srinivasan topped the list of changeset contributors with a massive set of cleanups to the Microsoft HV driver in the staging tree; it's impressive to see how much cleanup less than 15,000 lines of code can require.

It appears that Microsoft's contribution needed a lot of cleaning up to bring it up to scratch.

What happened to the borg graphic

[greenreaper]

. . . used for Microsoft topics? I miss that!