Microsoft goes Linux

Microsoft has several divisions and groups. One of these makes a rather popular and profitable database server, MS SQL Server. They expect to become more popular and more profitable if their server software runs on the most popular server OS, rather than being restricted to a single OS. That sounds like a perfectly reasonable business strategy, seen from the viewpoint of the group that handles the SQL Server software.

It is a brave move by the MS management, however, as it undermines their Windows server marketing. Is a license of MS SQL Server on Linux an extra sale for SQL Server - or is it a lost sale of Windows server?

No.

No.

If SQL Server on Linux goes well, you can expect to see Exchange Server or other backend server software supporting Linux. We may even see MS Office for Linux desktop at some point. But Windows is not going to become Linux-based.

On Tue, 08 Mar 2016 08:08:31 GMT, Jan Panteltje Gave us:

Bullshit. They are not worth that much... at all.

And MS would make their own Linux based offering. They do not need Ubuntu.

On a sunny day (Tue, 08 Mar 2016 10:05:42 +0100) it happened David Brown wrote in :

Microsoft Linux!

It is said, that those who criticize Unix are bound to re-invent it. Microsoft Linux would be an addition.. Of course it will be incompatible with everything else out there ;-) Just like rathead.

Ah, well, what more proof do we need for our fortune-telling? /Somebody/ said it, so surely it is merely a matter of time before it comes true.

Obviously it would be an additional choice, if such a thing were to exist.

In the unlikely event that MS releases their own version of Linux, or the even more unlikely event that they do so for the desktop rather than the server, their prime motivation would have to be compatibility with both Windows programs and Linux software. Anything else would be pointless. And while Microsoft may be many things, they are not stupid.

?

It is sometimes forgotten that the most commercially successful Unix of the 1980, i.e. Xenix, was made and sold by..... Microsoft.

You seem clueless, I have been with Unix since when was it 1979? The people who know it made that statement. And they are / were right.

Their prime motive would be profit.

Profit works, anything else is obama crowd.

That is the most stupid statement I have read in well, when was the big bang? Anyways, take one look at any MS windows version, it is obvious they are stupid, even stupid to the point they do not understand selling something that really works is a better strategy,

It is (MS windows) for idiots by idiots. Just like those who cannot read and write and only click on pictures.

Your question mark has been noted.

4 u to figure it out.

On Tue, 08 Mar 2016 14:02:07 GMT, Jan Panteltje Gave us:

What a retarded crack, from the SED utter retard.

I know that (though I never used it). But the computing world today is somewhat different from the 1980's, so I don't see Xenix as particularly relevant except historically.

Eh, no.

As a general point, it seems that many of the design decisions of early Unix were remarkably successful, and have scaled well across a wide range of uses. Other systems, such as Windows, have copied a good number of features from the *nix world over the years (but note also that the *nix world has copied from others too).

The old saying is - like all old sayings - an exaggeration compressed into a pithy remark. It is not supposed to be taken literally, or considered as some sort of holy prophecy.

And in particular, it does /not/ mean that Microsoft will build Windows

11 on Linux!

OK, but they make profit by selling systems - a MS Linux would only sell if it were compatible with existing software.

Yes, isn't it amazing how big and profitable MS is despite being idiots?

Oracle made that decision 20 years ago. Probably Microsoft watched how that went en decided that it was better to do this as well. Of course, it remains to be seen if they launch a "Microsoft Linux" as the preferred base to run it on, or support some existing distribution.

I expect it will only be supported on an "enterprise class" Linux. (which is the same thing as a normal version, but with a pricey support contract attached)

On a sunny day (Tue, 08 Mar 2016 15:59:35 +0100) it happened David Brown wrote in :

From the OS POV windows was based on MSDOS and that filesystem was a cheap hack of CP/M, with features removed. It got much worse when they integrated that shit into the GUI and browser (win 98 and up). And that mess was a dead end by design. That they keep their customers hostage with 'upgrades' automatic as well and needs a sticker on their PC, what an annoying sad bunch of crap programmers, what a bloat, what a stupidity, Microsoft.

OK maybe win12 then :-) They have no choice. Remember, the reality of it: Every embedded system I have hare runs : Linux. Robot vacuum cleaner, Linksys WiFi access points, Samsung TV, Humax cable receiver etc etc, Linux is everywhere. Whole continents have moved to Linux, and then there is Android (I do not like that) but it is a fierce competition to Win10 crap. The cheap Pipo boxes are dual boot Android Linux.

They are not good at Profit anymore, Billy The Gates was, but now with Balmer send away it is a the mercy of investors who will likely rip it apart. Look at the stupidity with Nokia, they destroyed that company, now look at what they did with that money and knowledge: No body buys windows phones. It is either Samsung or Apple, and Apple is insecure and supports terrorists, that leaves Samsung with AMOLED displays and Android, and some Chinese makes (Note Jobs is no more).

So MS is dead, I was reading today a big ad by the Billy the Gates foundation aimed at high school pupils to 'change the world', he (Billy) has the money.

He probably sold those MS shares ;-)

He DID change the world, do computahs save time? NO, they waste time. Instead of doing real tronix people play with slimulations, do not know what end a of a soldering iron to hold on to, or not to burn their fingers on hot coils. What we really need now is screens with components that smoke and get hot! THERE is a market.

So Billy no, he is not gonna change the world.

I think I mentioned a few things that show it is. MS windows being the real proof and evidence.

What was that energy company, it was big too, so was the Roman empire. Dinos were big too. Mosquitos won.

Until the last month, would have disagreed slightly, since it was sold to SCO - and we know what boon that has been to the lawyers!

ARRRGGhhhh! NOOoooooo, please!

Jon

Just a way of undermining competition. The don't have to make money on this, just make it harder for anyone else to do so.

Standard operating proceedure as MS

RL

n

ist.

an

bang?

e stupid,

really works is a better strategy,

Natural monopolies get bigger and more powerful once they've made themselve s natural monopolies. IBM didn't foresee how big the market for personal co mputers was going to be, otherwise they would never have sub-contracted the operating system to Bill Gates. When the penny dropped they got to work on OS/2, but it was too little, too late.

Bill Gates made some clever decisions early in the process, and IBM made so me dumb ones, but since then all Microsoft has had to do is to hang onto it 's natural monopoly. It's starting to look as if Linux is powerful enough, and sufficiently easy to use, to eventually drive Microsoft out of the mark et, but it's going to take quite a while.

--
Bill Sloman, Sydney

The Unix v7 code was licensed from AT&T by Microsoft. In turn, Microsoft sold Xenix licenses to SCO, IBM, Tandy, Altos, and a bunch of others. Microsoft also sold a version of Xenix. The tangled history is at: In about 1987, Microsoft sold ownership of Xenix to SCO, in trade for

25% ownership of SCO. SCO tried desperately to kill Xenix but failed because it was the ideal text based applications platform of the time. SCO finally managed to mostly kill it in about 1995. However, until about 2010, I was still servicing Xenix customers. I also know of some rather important applications currently running on Xenix. It's really difficult to kill something that works well.

At various time, Microsoft has declared war on Unix and it's mutations. The strategy is always embrace, extend, extinguish: There have been emulators, VM's, developers tools, guest operating systems, and courseware. I just found "Essentials of Windows for Unix Developers" by Microsoft. It includes a 180 day trial of Windoze Server 2003 Enterprise Edition and a CD full of powerpoint slides. I think they were expecting Unix developers to jump ship to Windoze after running Server 2003 for a while. It didn't happen.

Plenty more experiments which tried to bring Unix and Windoze together, such as Java based Tarantella: We've come a long way since then.

Drivel: Unix timeline:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

The public story is incomplete - I'm not sure whether what I'm about to say has ever been publicly acknowledged, but I knew the people involved.

Around 1986-7, Hewlett-Packard and SCO approached Microsoft to license the Windows source code, intending to translate the assembler code parts to make the whole thing portable, and to provide back-end rendering to the X Window System. Windows would become a graphical windowing environment for Unix systems. An *application*, not reliant on MS-DOS or its successors.

The cost of rewriting was predicated on an evaluation of the code quality - conducted by inspecting samples and computing various metrics like comment density - and on Microsoft's own measures of the number of lines of code. The code examples inspected were not too bad, and the size was quoted as 200,000 lines of code excluding comments. On that basis, HP and SCO went ahead with the deal, and the code was delivered.

It didn't take very long before it became obvious that the code samples were not typical, and the quality - especially the comment density - was seriously worse. But what really killed the deal was the size. There was 480,000 lines of this crappy assembler code, and HP baulked at the cost of rewriting it. The whole deal was off.

If they had decided to renegotiate and to bite the bullet, there would probably never have been an OS/2 nor a Windows NT, XP, and their successors. Microsoft would not have an operating system of their own, but would be a desktop and applications company on an open platform (assuming that they could have survived that).

Which seems to be where they're going now... that they're having trouble even *giving away* their operating system.

The Hewlett Packard division involved was the one producing X Windows software including the HP Widget toolkit (and later, OSF/Motif) at the same site where the calculators were made, in Corvallis, Oregon. They also made the inkjet printers there.

Perhaps someone who worked there at the time would like to chime in and correct or augment details of this story? I heard all this from Bob Miller, project manager for OSF/Motif, in 1989.

Clifford Heath.

This is the first time I've heard this story.

That's odd. Windoze 1.0 was release in late 1985. During the next few years, MS introduce later versions of Windoze, OS/2, various productivity apps, and did an IPO. I find it very difficult to believe that MS would even consider licensing source code in the middle of a spectacular period of growth, to what could easily be recognized as potentially a competing product. However, it is consistent with the way SCO operated. Almost everything in the SCO products was licensed from and not written by SCO. Are you sure about the 1986-87 dates?

I was on the borderline of SCO and HP during this period. I don't see how SCO and HP could collaborate on anything. For example, the HPNP printer driver in ODT and OSR5 was written by someone at HP. When bugs were found, SCO demanded that HP fix them. HP's attitude was that HP wrote it for SCO, so now it's SCO's problem. I had to write some workarounds. For example: I don't see how these two companies could have collaborated on anything. However, the Microsoft part might be possible. The pres of Microsoft at the time was Jon Shirley, who would probably have sold the company crown jewels if he thought it would generate revenue.

That was probably the first time that happened. The repeat performance happened again in 1998 with SCO, IBM, and Sequent in Project Monterey: Everyone had a different idea of what they were trying to accomplish, what they had to work with, and what they were expected to do.

Maybe. If true, then MS would have not had a reason to sell Xenix to SCO in 1987 and would have probably built Windoze on top of Xenix to compete with their former collaborators.

It's not too difficult to see which way MS is heading. Nadella came from the Azure cloud (software as a service) division of Microsoft. He's pushing the company and customers (kicking and screaming) in a direction where we rent all our software from Microsoft (pay by the month or whatever). MS is intentionally not giving away the OS because we'll soon not be able to own our own OS but instead pay monthly fees for its use. One area where this has already happened is in corporate server farms. These were formerly owned by the corporations and maintained by their IT staffs. Today, the entire network is virtualized in the cloud, and the IT function is provided by an offshore service company. To the corporation, this is better because it's cheaper.

Thanks for the info and detail. I'll send this to some of the former SCO people and see what they can add.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

That amount of assembly lines seems to be quite high, since Win NT

3.51 was available, for Alpha, MIPS, PowerPC and x86. When supporting multiple platforms, you really try to minimize the amount of assembly code needed.