Slashdot Mirror
Microsoft is the Industry's Most Innovative Company?
mjasay writes "According to a recent analysis by IEEE, Microsoft's patent portfolio tops the industry in terms of overall quality of its patents. And while Microsoft came in second to IBM in The Patent Board's 2006 survey, its upcoming 2007 report has Microsoft besting IBM (and even its 2006 report had Microsoft #1 in terms of the "scientific strength" of its patent portfolio). All of which begs the question: Just where is all this innovation going? To Clippy? Consumers and business users don't buy patents. They buy products that make their lives easier or more productive, yet Microsoft doesn't seem to be able to turn its patent portfolio into much more than life support for its existing Office and Windows monopolies. In sum, if Microsoft is so innovative, why can't we get something better than the Zune?"
I wonder if they included Microsoft patents such as their Virtual Desktop Pager patent? (http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(Microsoft.ASNM.+AND+%22Virtual+desktop+manager%22)&OS=AN/Microsoft+and+) Honestly, a vast portion of Microsoft's patents are complete bullshit that should NEVER have been awarded. Remove cases of OBVIOUS prior art (Linux has had virtual desktop pagers as described in that patent forever, and when they received this patent Microsoft had never used such a thing), and Microsoft's patent portfolio is shit. ~nog_lorp
the innovation is going to vista techs that no one seems to want like there crappy DRM system that mess up networking when you are playing a .mp3
Just because someone comes up with a patentable idea, doesn't mean it's a GOOD idea.
Similarly there may well be plenty of good ideas which arn't patentable.
Microsoft Research is really cool. They crank out cool stuff all the time! Take a look! The problem is that most of their stuff never sees the light of day. MS just gets the patent then bury it and move on. WinFS and other neat things came out of there. They hire a lot of PhDs, too... James Larus, the guy that wrote SPIM (MIPS simulator) works there now...
Gotta get me one of these!
What the hell is wrong with the Zune 2? The reviews have been overwhelmingly positive and it beats the hell out of the iPod classic.
Things that I have either heard of or seen coming from Redmond:
Any of which could have had multiple patents. A lot of what they do is impractical as a product now (the wall for instance), but is an investment in the future. Like in the early 90's when they purchased tons of digital rights. And some, like the Network LOD, are designed for developers to tie them into MS products.
But Microsoft, like AT&T when it had too much money, take a bunch of academics, give them money, and tell them to do cool things. After all, the whole deparment will pay for itself with a couple of nifty inventions.
Your ad here. Ask me how!
Yeah,the Zune is innovative... two way encrypted handshakes to ensure only MS's software can do anything with it.
Had Microsoft made it driverless, or using the generic MTP that MS had as a standard spec, it would be different.
At least you can find software to use ipods on Linux.
That's because innovation isn't measurable by the number of patents you produce. Let me tell you my patent story.
I used to work at a company that made a widget. Details left out because of possible NDA/lawsuit goodness.
There were 3 or 4 other players in this widget space. There are about 3 or 4 useful functions any of these widgets can do.
One of the other players decides to patent "feature A from this widget, combined with feature B from this other widget". A multi function widget, merely taking two functions from two widgets and combining them. In other words, peanut butter is ok, and jelly is ok, but putting peanut butter with jelly is *hugely innovative* and deserves a patent.
We held meetings and began to file patents too. They were all equally insane.
There was NO INNOVATION going on in these meetings. Just carving up the widget patent space - that has existed for years - with each of these little companies nit-picking each other to death with patent suits and royalty fees.
Patents do not equal innovation.
Weaselmancer
rediculous.
http://www.research.ibm.com/areas.shtml
http://research.microsoft.com/research/default.aspx
There's no real contest though. If they were course listings, one reads like MIT and the other like a community college.
"Gold still represents the ultimate form of payment in the world." - Alan Greenspan, 1999
Well, they got people to pay hundreds for a box with a 300 page book that nobody read and a CD.
They practically invented the EULA for the masses.
They entered new markets by simply buying companies and their portfolios.
They probably weren't the first in any of these, but they perfected integrating these into a government-proof business strategy.
So yeah, they're pretty innovative.
I mostly pay attention to theoretical areas like programming languages and automated reasoning, and MS has made significant contributions in those fields over the last few years.
Yeah, and not only that, Microsoft seems to have understood that the first company to crack the parallel programming nut will be at the forefront of computing in this century. Lately, they have hired a few world-renowned experts in parallel programming and supercomjputing. Dan Reed (formerly of the Rennaissance Computing Institute) comes to mind. However, I doubt that this is going to be enough to solve the parallel computing conundrum. Sadly, computer science is dominated by a bunch of aging computer geeks who still think like Charles Babbage when it comes to computer programming and CPU design. Solving the parallel computing problem will take a strong willingness to break away from the orthodox fold. In my opinion, it is time to declare the algorithm dead and embrace a non-algorithmic computing model. We must reinvent the computer, especially now that the industry is taking its first painful step away from sequential computing to massive parallelism. We made a mistake fifty years ago when we chose Babbage's model but, it wasn't so bad because most of our computers had single-core CPUs. Unless we choose the correct path now, we will pay a heavy price later. Eventually, we will be forced to change. Better now than later. Is there anybody at Microsoft who can see the writing on the wall? Who know?
Lol, this isn't interesting. I'm sorry, which part of that Apple rant has anything to do with fonts? If I develop a paper plane, does that mean I can sue Boeing for developing fighter jets?
The thing is, where is this alleged research going? We don't see it in MS's products; this was stated in the article summary. This is always the answer trotted out when anyone questions MS's patents and MS Research.
When IBM comes up with some great new technology, like the damascene process (copper on ICs), SOI, etc., we see it in chips pretty soon after. It was only about 10 years ago that the copper process was invented by IBM, and now every CPU has it to my knowledge, as has for quite some time. Intel invented a "strained silicon" process, and their CPUs have it now.
So where are MS Research's efforts paying off? Research isn't any good if it isn't actually applied somewhere. Basic research with no obvious course to application has its place, such as with fundamental science like quantum physics, exploration of Mars, etc., but software isn't one of them. If you can't find a place to use your findings, you've wasted your time. Back in the 60s-70s, researchers invented new programming languages and operating systems, and pretty soon industry and academia were all using C on UNIX machines. But we haven't seen anything come out of MS Research that's made a significant difference in anyone's lives.
Nobody's saying "Everything Microsoft produces is crap". (Or they shouldn't be, because it's not true).
What is true is that Microsoft do not - indeed have never - innovated. They've taken existing ideas, either bought them or copied them then marketed the hell out of the result.
Examples:
Flight Simulator - bought from SubLogic. (You said this yourself!)
FoxPro - Originally produced by Fox Software, which was bought out by Microsoft in 1992.
Outlook/Exchange - Lotus Notes was a groupware product well before then.
Access - Originally plagiarised from Borland Paradox.
Excel - Plagiarised from Lotus 1-2-3. The two were basically playing leapfrog in feature sets before 1-2-3 bit the dust.
Word - Plagiarised features from WordPerfect. Won the battle primarily by being sold to the boss rather than the secretary who was actually typing the letters.
Windows - Most graphical operating systems of the 1980's-1990's were shamelessly taking ideas from each other. The bar across the bottom of the screen, for instance, was seen in RISC OS and CDE long before Windows '95 hit the shelves.
XBox Live - the PS2 offered online play, but Sony never really exploited this. Frankly, it was a little early because it predated ubiquitous broadband.
In fact, Microsoft can't even innovate at the very simplest level.
Microsoft Paint (yes, that crappy little paint tool which has come free with Windows since the Windows 3.x days) - Take a look at this. It's PC Paintbrush for DOS - developed by a company called ZSoft.
Not only are Clippy and Bob so incredibly horrible that they will be remembered forever in the annals of stupid computing, Microsoft stole the ideas behind them from Brenda Laurel, and got them all wrong.
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
T-SQL always used to annoy me with it's fussiness about the order you specified tables when using JOIN's
I wasn't good enough to notice when I was using SQL Server 6.5, but I've never noticed such a thing in 7, 2000 or 2005.
On the one project I used MySQL for, I was relieved to discover that it finally supported subqueries, but they ended up being unusably slow because the optimizer couldn't seem to do any optimization between the inner and outer queries. I ended up using Java code for what I would've just done with a subquery in SQL Server. Of course, now I'm mainly working in Oracle, and I have an almost opposite complaint; subqueries (and frequently several of them) seem to be the only way to accomplish a lot of things that wouldn't have taken much thought in SQL Server.
T-SQL always had the edge by allowing you bypass its annoyances by using stored procedures and views but this has now changed since MySQL 5.
I've only done stored procedures in SQL Server, Oracle, and barely in Informix. Informix procedures just suck unreservedly. Oracle PL/SQL is a decent procedural language, but the interface to regular SQL can be a bit awkward, and there's entirely too much iterative code needed for my taste. T-SQL is rather limited as a procedural language, but seems to do a lot better at letting you stay within set-based logic.
What are MySQL procedure like?
I think I'd be even blunter than you. Microsoft's profits come from a small range of technologies and philosophies that are often old at deployment, often weak by design, and fixed by the 'vision' of a small number of powerful people with strong personalities but extremely limited technical competence. Its strategy is to protect those profits, by limiting the extent to which innovation reaches the marketplace. This can be accomplished by destroying competition financially, by acquiring and dismantling competition, by obtaining (through patent law and, if necessary, research) exclusive rights to the technology that would enable competition, by hiring the researchers who might otherwise help the competition, and by actually innovating only as an absolute last resort. This is because change, any change at all, in a Microsoft product, is a tacit admission that the existing version is not already the best it could be. While to normal people progress is just the nature of the technological world, to the personality types who reign at Microsoft it means that they were ignorant and they were wrong, and this is very, very hard for them to accept.