Slashdot Mirror
Mundie Responds
HaiLHaiL writes: "Microsoft's Mundie has a commentary running on ZDNet responding to the responses to his speech. " No real surprises, but it's getting submitted a lot so I figured I'd post it for you. Lots of good points, but I'm sure you can guess the gist of it.
The fact that Microsoft has managed to make a successful thriving business out of software products totally caught everyone off guard. Software in itself is fundamentally worthless, their competitors said. It was such a silly idea, that most big iron vendors didn't even try it. Businesses need custom solutions, not shrink-wrapped packages. So what about the masses with a PC at home? Realistically, that makes up a small percentage of Microsoft's revenue. Most of their killer apps were sold to businesses-- the very same businesses that IBM said wouldn't need them.
Software products are sold to a generic mass market, and as such, they cannot possibly do what every user wants it to do. A single software package will never do what you want, and you will always need to support it, and you will always need to change it to do what you precisely want.
Software products are proprietary by definition. They try to be black boxes. Buy it once and it solves the problem. The business model never takes into account support, for when the product fails, or further development, when the product almost does what you want, but isn't quite there. Your best bet is to hope that the next version, which will cost you to upgrade to, will do what you want, based on your feedback to the vendor.
Amazingly, Microsoft has made billions on a flawed software model. They went out and convinced everyone, (through no monopolistic means of course. Judge Jackson was clearly uninformed), that their bits on a disc are valuable and worth every penny. Since the only value of Microsoft software is the bits printed on a CD, obviously IP rights are extremely important to their livelihood.
The open source way, specifically GPL'd software, suggests a totally different business model. It means that someone can come in, choose from a wealth of open source software utilities, provide you with a custom solution, and you maintain all of the control you need. If anything, it means that a consultant you hire who builds you a point of sale system with open source components can't hide the source from you. You aren't stuck with the mercy of your original vendor. How could this be bad? Sure, maybe you can't resell your custom system, but realistcally, how many people can resell their closed source ones? If anything, you have a much higher chance of reselling a custom open source system.
When I think of system development with Linux, I think 2% custom code, and 98% software integration. When I imagine it with Windows, I imagine the exact opposite. Take a bunch of black boxes and try to glue them together with lots and lots of code. Oh, also, don't forget the software licensing costs!
There are always exceptions here, of course. Closed source works for a lot of business models. But really, people that care about retaining IP rights to their source code as a solutions provider are just looking to keep their clients at their mercy. Typically Microsoft.
And for those of you saying "Software service? Big deal. That's a totally insignificant market", here's a way to prove it to yourself. Look in the want ads for programming positions. I'd wager that 95% of the jobs being offered to programmers are to work on custom systems, rather than working at a company that provides a shrink-wrapped product. There's a reason that COBOL programmers are still in demand, despite almost no new commercial software being written in COBOL in the past 10 years.
Mundie says that Linux can never be used to make one company billions of monopoly dollars. You mean that we don't have to deal with another Microsoft if the world switches to open source? What's the fucking problem?
Oh god no, I'm horrible at drawing ASCII graphics.
But Mundie already addresses your point very early in his response.
I quote:
"As the U.S. Department of Commerce stated in a report titled "International Science and Technology": "Innovation relies on a partnership between the public and private sectors in which the government invests in long-range science and technology and in mechanisms to promote private-sector risk-taking and investment."
The innovations you gave examples of are just that, government investments. The Internet was all part of DARPA, etc.
What Mundie is addressing is the R&D and innovation which is required to take technology A and make it into a marketable product.
I'm a fan of cars, as well as Venn diagrams. So let me use another example.
Honda is a huge proponent of Variable Valve Timing in engines. They call it VTEC. Honda didn't invent this technology, actually I believe it dates back at least 30 some years.
But what Honda has done is transcend it from an interesting idea that can be used to squeeze some power out of high priced racing cars, into a technology which can squeeze some power and fuel economy out of low priced consumer automobiles.
That is, their innovation was making it cheap and efficient to sell.
Honda most certainly has patents on the improvements they made that relate to VTEC which prevents others in the industry from doing the same thing.
But that hasn't stopped other auto manufacturers from also having forms of variable valve timing. Toyota calls theirs VVT, Nissan VVL(variable valve lift), etc.
But they aren't quite like Honda's solution, and that is what makes cars like the S2000 unique. By pulling 240 hp out of a normally aspirated 2.0 liter engine.
So I guess instead of attacking a strawman argument, why don't you contemplate Microsoft's true position.
Instead why don't you envision a world in which all government funded research projects are licensed with something akin to the GPL. Imagine this world and how it will impact our economy?
Would it be a good thing?
Where has all the great intellectual rhetoric of the past gone?
But actually, that isn't what the last 50 years have shown us. In exact opposition, the last 50 years have shown us that open systems are the one s that exhibit massive, uncontrolled growth and contribute the most surprising things to society. The "PC era" that Mundie invokes in this article was possible not because of IBM's lightning wit, but because Compaq and the rest pried IBMs intellectual property away to make PC clones. The Internet was the result of public sector research. HTTP, SMTP, DNS, POP, IMAP, SSL, ICP, HTML/SGML/XML, and every other enabling technology of the Internet was given away freely by its creator. The web was created, and given away. BIND, Sendmail, NCSA httpd, Apache, and free operating systems are examples of key technologies that enjoy wide, free distribution unconstrained by their licenses.
There is only one example of an underlying enabling technology that fell under strong intellectual property protection. RSA encryption was patented and required licensing until last year. This "protection" literally crippled encryption innovation for some time. People were forced to either invent their own encryption schemes that weren't covered by RSA's patents, license RSA's patents for large sums of money, or ignore their patents. If you have set up an Apache HTTPS server before this year, you know what a pain in the ass it was to do so legally in the United States. The intellectual property protection afforded to RSA was a huge blow that slowed the growth of encryption for years.
There are so many more examples of technology that was freely distributed to the benefit of society. The C and C++ languages upon which Windows is built are an example. Think of where Microsoft would be if they had to pay a recurring licensing fee for every C++ object they compiled. Consider also how damn hard it would be to debug a C++ program if the format of the object file were protected under intellectual property laws. Think of what Windows would be if the inventors of TCP/IP had refused to license the protocols to Microsoft. Windows would of course be worthless with TCP/IP networking. What would Windows 2000 be if LDAP and Kerberos had not been available to the developers? Microsoft is standing on the shoulders of a giant so big, that they don't even realize it.
Mundie is flat wrong in his argument: almost all of the software technology that we take for granted today was the direct result of research and development performed in the open and given away.
This is true, and I think it's am important point. Microsoft (and any other software company for that matter) has the write to craft their licenses in any way they see fit, as long as the consumer has a choice to accept or reject that model Of course, monopolies run into other problems, since people are no longer free to make decisions in their own best interest... but that's another topic.
What bugs the heck out of me though is the continual insunation that Mundie and others are making now that the GPL is threatening to take away their IP out from under them! That's why they keep repeating this "choice" thing. The lie is: "If everyone starts using GPL software, then we will be forced to give up our own investments in IP." It makes no sense, as anyone who has ever bothered to read the GPL knows, but it falls into the old saying:
--
Your Servant, B. Baggins
I think this boils down to Mundie stating that you can't get rich selling GPL'ed Software. That might be true but the flip side is that you can save a lot of money using it. Isn't that what RH's Bob Young has been saying all along. He has staed that Redhat will reduce the OS market by 80% in dollar terms.
Help fight continental drift.
Legendary inventors such as Alexander Graham Bell, Thomas Edison and Henry Ford (who held thousands of patents between them) succeeded precisely because they were able to use funding, management and market insight to deliver their innovations as unique, practical and useful products.
Actually wasn't Bell successful because he got to the patent office first, and beat some other guy to the punch?
Summarized and dissected:
1) Helping customers and partners to be successful through source access programs.
Their philosophy is exclusive, and therefore limited in how effective it can be. Students and other poor people are NOT allowed to participate in their philosophy.
2)Building the development community and offering the tools to produce great software.
A community is a spiderweb network arrangement of people, with free associations. Shared Source is a star topology network, with Microsoft strictly arbitrating all associations between clients. They don't fit my definition of "community" very well.
3) Improving feedback processes in order to create better products for Microsoft's customers and partners.
This is an unequal flow of information, which makes me wonder how Microsoft thinks of their partners. Imagine what would happen if our relationships with wives and girlfriends (ideally a partnership) worked like this. The Man (Microsoft) would do what he wanted. The woman would give everything she earned to the Man. The Man would provide everything that the Woman needed. Occaisionally, he would sit down and listen to the various ways he could improve the quality of what he provided to her, to make her happy. If he decided not to implement suggestions, that would be touch luck for the Woman. How long would it take for the Woman to tell the Man to screw himself and his "partnership"?
4) Maintaining the integrity of our customers' environments.
Integrity simply means that words and actions are aligned. Microsoft doesn't seem to understand what partnership and community actually mean, so how can we expect them to have integrity? Integrity is easy if you are the author of the dictionary.
5)Increasing educational access to get technology into the hands of universities worldwide, and to seed the future of a strong technology industry.
This is called indoctrination. It's not a philosophy, it's a strategy.
6)Protecting software intellectual property rights based on the firm belief that software offers value as the basis of a successful business.
Software is the basis of Microsoft's business, but other businesses base themselves on things like financial services, building houses, making industrial machinery, etc. Reminds me of a guy at American Express that I used to work with. He actually told me "if Amex were to adopt open source, how could we make money if we gave all our software away?" I had to remind him that Amex made money off charge cards (not software), and they weren't required to distribute source if they didn't distribute the binary.
If tits were wings it'd be flying around.
I believe this is quite literally the best response that Microsoft has to the threat of the GPL: if you can't beat it on technical merits, strangle the money supply instead.
Microsoft knows what would happen if Red Hat and VA Linux Systems went under: whole segments of the open source community, including Slashdot and Sourceforge, would suddenly find themselves quite strapped for cash. Linux and OSS development would be permanently crippled, at least relative to today's heady pace. Eventually, Microsoft would once again beat Linux on technical merits.
The best solution to this problem is for companies like Red Hat and VA Linux to turn a profit, and soon. This is realistic for Red Hat; I'm really really hoping that it will also be realistic for VA soon.
ObJectBridge (GPL'd Java ODMG) needs volunteers.
Finding God in a Dog
I like how Mundie casts it solely as a money issue, and how he cites a few notable, successful, capitalist inventors (Alexander Graham Bell, Thomas Edison and Henry Ford) to make it seem like all innovation is about wealth.
He left out a lot of inventors who weren't in it for the money, or who got cheated by big capital: Tesla died penniless, as did Baron Karl von Drais de Sauerbrun, Jan Matzeliger, Mandee Daguerre, Walter Shaw, Samuel Morse, William Friese-Greene, Lee de Forest, Johann Gutenberg, Henry Trengrouse, and on and on....
Then there are all the inventors/researchers who did what they did not for money, but for the love of it. Let's look at computer scientists. Does anyone think that Nicholas Wirth, Edsger Dijkstra, Grace Hopper, Steve Wozniak, Don Knuth, John von Neumann, Alan Turing, Brian Kernighan, Dennis Ritchie, Kenneth Thompson, Linus, etc, etc, were doing it all for the money?
There's doing it for money, which is the world Mundie understands, and then there's doing it for love, which he finds very threatening.
bukra fil mish mish
-
Monitor the Web, or Track your site!
Eloi, Eloi, lema sabachtani?
www.fogbound.net
Companies have the choice of protecting or relinquishing the intellectual property resulting from their research and development consistent with their particular customer and business needs.
We know what choice Microsoft has made. As much as we want to flame Microsoft for making buggy, expensive software, it's their business model, and it's obvious that Mundie is advocating something more than shared-source here. He's rubbing it in the face of the Linux industry when he says it: companies have the choice whether to hang on to their source or not, and the success of the company is often indicative of the choices they make.
Whether you love 'em or hate 'em, you just can't point to any other company and say they've had the same results. It's easy for Mundie to say that shared-source (rather than open source) has played some role in that growth, because there's no way any of us can refute it. But at the same time, he could just as well have been saying that the success of Microsoft is due to Gates having a bad haircut, and that every CEO/founder/President should have a bad haircut.
In their defense, though, Mundie is saying that it's a choice, and it's a choice Microsoft has made. It's not like they aren't aware of the choice: they're making it to satisfy their business needs, like stockholders, and I sincerely doubt the stockholders would jump for joy if Microsoft gave up the source code tomorrow.
What's your damage, Heather?
You're right. It's inexcusable for the Slashdot editors to pay attention to what the readers find interesting!
They should decide what we will read, and when we will read it! AND WE SHOULD LIKE IT!
I mean, letting readers decide what's covered by the media is as silly as... as silly as letting users decide what features the OS and applications should have! That way lies anarchy! Madness! Declining Microsoft stock prices!
For every complex problem there is an answer that is clear, simple, and wrong. -- H L Mencken
His argument is that hoarding ("protecting") IP is the only way to economic success. And he may be correct. There's no doubt that keeping your code proprietary and out of the eyes of others will make you more money when your product is the code itself. But who cares? Who is arguing against this? This is not an argument. This is a truism.
The discussion is at a more fundamental level - should people be allowed to monopolize ideas indefinately? He casually skirts this larger undercurrent by preemptively Fearing us with some blather about IP resulting in economic growth, IP making us rich and happy. *You* don't want to be the one to ruin the economy, *do you* hippy free code slacker? What's good for Microsoft is good for America.
Seeing as he was talking to a business school, it does make sense that he was saying that Open Source is not the way to make money (so, and helping old ladies accross the street all day isn't either). His arguments seem to stem from the assumption that making money is the ultimate test of human endeavor. Whereas the Free Software community has different values.
yes this is rantish, i don't care
It's 10 PM. Do you know if you're un-American?
At least now he can distinguish between Open Source and the GPL, although I believe the title of the article is mis-leading.
There is nothing in an open source model that can keep someone from competeting against it. If you can build a significantly better mousetrap, then people will buy it anyway. DEC VAX/VMS was a completely open source operating system that was a SIGNIFICANT player in the late 80's and early 90's. Their OS source code was available for a nominal fee (to pay for the Microfiche it came on).
What Mr. Mundie and Microsoft in general still seems to be wresting with is competing against the GPL. The GPL is a software house that produces code that's free, is of good quality, and can't be bought, incorporated, dismantled, or undersold. All their tried and true techniques of competition don't work.
The only way to compete with the GPL is to be more customer focused, have better quality, and respond to changes quickly. MS's customer base is too big and too divserse to do that, and they lack real cross-platform development abilities.
Perhaps Microsoft is starting to feel a similar pain to what Netscape felt when Microsoft released IE and IIS for free? Netscape couldn't buy it, they couldn't dismantled it, and they couldn't undersell it, and it was good quality (Esp. for Windows platforms), and their last resort was to open-source the browser.
-- You can't idiot-proof anything, because they're always coming out with better idiots.
But the problem that no one really mentioned is that GPL is protecting my intelectual property. Its protecting it in the exact way I want it to be protected and its protecting it so well, that Microsoft cannot steal it! And thats why they cry out loud. They see all the revenues lost, they could gain if they would be able to steal my intelectual property. And thats makes them mad.
If programs would be read like poetry, most programmers would be Vogons.
this line was funny and painful at the same time:
...
The issue at hand is choice; companies and individuals should be able to choose either model, and we support this right.
yet from the beginning it seem MS has wanted to make this choice for us
_f
Whether this is a bad thing or not is open to debate.
--
TO BUY A NEW CAR WOULD MAKE YOU SEXUALLY ATTRACTIVE.
If I use a patented washer in some mechanism that I sell, well, I can buy that washer from a supplier that holds patent rights. If I use a patented software hack, only a component in my system, I can get shut down if I get caught. The big guys sue each other over patents but end up cross-licensing each other into a big guys club.
Stallman's concern was that in the software land-rush patent-grab where the most obvious stuff gets patented but it is tres expensive to challenge any of those patents once they get issued, one wouldn't be able to write any software at all.
What the GPL does is stake out territory, not only in high-level stuff like OS's and compilers, but also low-level stuff like algorithms. GPL code lying around makes a strong case for prior art that someone cannot patent, say, a compression algorithm found in GPL's source somewhere.
The extension of this, is that if Microsoft needs a compression algorithm found in a popular piece of GPL code, they will have to prove that they did not look at the popular code and hence have to open-source all their stuff, which they are not eager to do.
IMO the GPL is a fair way of fighting the software patent land grab, and software gets written on account of government funding at universities and gets GPL's, well, Microsoft needs to suck it in.