Dr. Jiang Li joined Microsoft Research, Asia as Researcher in January 1999. His research interests include video compression, image processing, video broadcast and communication, peer-to-peer networking, realistic image synthesis and image based rendering. Read the interview at ActiveWin.
Interview with Jiang Li, Project Leader at MS Research
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I was sure they were always buying company full of ready to commercialize product only lacking the last marketing push by buying company in cashflow problem that already did restructuration so that they don’t have to look bad by doing them themself.
…hummm perhaps the research they talk is reading lot of venture capital prospectus to seek eventual prey for the hunt (insert deep troat predator noise here).
Still
i enjoy interview like that.
…in PHP form (took some ‘R’ of myself to discover this).
M$ R&D = $bought_out_company_technlogy + $drm
Actually, Microsoft have a lot of R&D, but we don’t really see much of it because Microsoft generally don’t release something because it is cool and new and different, but rather something that would make money. A stark contrast to the likes of Apple, Be, NeXT, Sun, etc.
Can Microsoft release the coolest most newest idea around? Sure they can, look at their R&D budget. But would they? Doubt it.
You’ve pointed out why Microsoft isn’t innovative.
How about them do something radical. Microsoft could easily go out, do something completely radical and make a name for themselves and contribute something to the over all “technology marketplace”. Take a look at Apple with Rendevoz (easy networking), and has opensourced it. Why hasn’t Microsoft done something like that
As for the .NET, you want to wow me, how about Microsoft develop a 100% compatible version for FreeBSD and MacOS X instead of the crap, aka, Rotor, they have spat out to the gullable masses.
If Apple to the Microsoft approach, they would never have bought NeXT, they would have kept slashing off the business bit by bit in the hope to “cut costs”.
Instead, they took a gamble, and the gamble paid off.
“Take a look at Apple with Rendevoz (easy networking), and has opensourced it.”
This is not only Apple’s doing. This is the work of the zeroconf workgroup which Apple is a member of. Apple can take no more credit for making it opensource than any other partaker in that workgroup.
MS have had Universal PnP (also the result of a workgroup effort http://www.upnp.org/ ), that does just about the same thing as Rendevous, since 1999.
“instead of the crap, aka, Rotor”
Rotor is certainly not crap but it is not a full .NET implementation which is limitation that I’d love to see rectified.
“If Apple to the Microsoft approach, they would never have bought NeXT, they would have kept slashing off the business bit by bit in the hope to “cut costs”.”
“Instead, they took a gamble, and the gamble paid off.”
Buying NeXT IS a MS approach. “If you aint got the tech, buy it!” That’s waht Apple did and that’s it.
Sofar I have not seen the gamble pay off. Infact, I think Apple’s marketshare is still declining. And the users are certainly not embracing MacOS X. MacOS X have potential but pieces are still missing from the puzzle.
Sofar I have not seen the gamble pay off. Infact, I think Apple’s marketshare is still declining. And the users are certainly not embracing MacOS X. MacOS X have potential but pieces are still missing from the puzzle.
—
Personally, the first version of their MacOS X, aka, MacOS X Server 1.0 was excellent. It was fast and responsive. Unfortunately they had to stuff it up by adding unneeded crap to the desktop.
“Microsoft have a lot of R&D, but we don’t really see much of it because Microsoft generally don’t release something because it is cool and new and different, but rather something that would make money”
Maybe there is an other reason too … how many people here read research papers? Yeah … I guessed so. But if one look, one can find a lot of papers from MSR scientists.
Cheers,
A.
1) Actually, Rendervous isn’t a new radical idea lik you suggest. Instead, it is a commercialized idea. At least MIT have been going on and on with a idea like this for at least a decade, so it couldn’t be completely radical.
2) As for .NET, Rotor only reimplements parts under the EMCA. Personally, I don’t see any business reason for Microsoft to make a completely compatible version of .NET fo FreeBSD and Mac OS X when neither platform would use it extensively.
3) Actually, if Apple is like Microsoft, they would have bought NeXT, which they did. But they won’t jump straight into OS X, instead trasitioning to it ala Windows XP (how long did Microsoft have Windows NT?) (Besides, since when
4) Microsoft takes gambles. Like XBox, it is completely different from other consoles because it uses PC hardware (Celeron, GeForce, suprised?). Or Pocket PC, whom the concept is very different from Palm’s or Symbian’s solutions. However, they don’t make huge gambles.
5) Taking huge risks in creating something radical is not the best thing to do. How many companies actually start up with a brand new radical idea and survive for at least a decade. Little. Microsoft is just doing what’s best for their investors, taking currently successful ideas and adding to them.
However, whatever the reason why current OS X don’t really get the amount of users it wanted to get, I guess it is still not yet paid off. But yeah, I agree with you. If Apple just resort in a little eye candy and didn’t go overboard with Aqua, OS X would be far more responsive. But the main reason why people don’t really want to move from OS 9 to OS X is that all their investments are in OS 9. They can move from OS 8 to OS 9 fine, but for OS 9 to OS X, that is way harder.
They have to buy new native apps for OS X, and for Photoshop (and perhaps other apps), they have to port their filters to OS X. Either that, or use Classic, which is dog-slow. That is the main reason.
Explain why there are JVM’s for:
OpenVMS
IRIX
Linux
MacOS
HP-UX
AIX
OpenServer
UNIXWare
Tru64
DigitalUNIX
Windows 95/98/ME/NT/2000/XP/.NET
You comments regarding .NET hold no water. Either make it cross platform, or stop lying to the public about .NET capabilities.
You can side step the issue and bitch and whinge, but the fact remains Microsoft has pushed .NET as the full .NET runtime environment that is available for download off their website (20MB) and claimed that it is cross platform.
Microsoft also completely neglect to tell publicly and openly for all and sundry to hear, that only a small, microscopic piece of .NET is actually an openstandard and the pathetic excuse for a implementation, Rotor performs WORSE that the Windows counterpart because they were too bloody lazy to make a native port abd instead they used Mainsoft, which is a UNIX implementation of core win32 features, and as a result, provide pathetic performance.
That is reality. Not the “fictional world” according to Bill Gates and Balmer.
Rajan R, instead of being a cheerleader, how about learning a bit of schepticism or hasn’t the weight of the world crushed your spirit yet? The fact remains that until .NET does everything Java can, you might as well locate in the bin along with the pet rock, mood rings and the lotto ticket you buy each week.
Matthew, stop bashing me when you miss my point almost completely. I’M NOT A SUPPORTER OF .NET! Rather, I’m a supporter of Java for enterprise usage (J2EE), while completely don’t support using either Java nor .NET for desktops and embedded stuff (like handhelds).
So let me say it one more time, I’M NOT A SUPPORTER OF .NET! If I was given a project that requires either .NET or J2EE, I would use the latter. Only because it is far more mature and has far more tools for it.
Besides, Corel, a company not all that famous for good software, made Rotor, not Microsoft. Microsoft hired Corel. So don’t blame me what Corel’s decissions. Besides, Microsoft hired Corel to implement the EMCA standard. Never said to make it fast, never said to make it 100% compatible. Like every other smart business, they do what is nessecary, and didn’t go the extra mile. Why? Corel makes no money from Rotor other than from Microsoft.
I’m not a cheerleader. I’m a pragmatic person, I don’t just bash a company mainly because they told half truths or didn’t tell the whole truth. Otherwise, bashing Microsoft only would be wrong for me, I would spend hours here bashing every single company that don’t tell the whole truth to their customers.
So, I’m very sorry I neglected to go off-topic and promote Java and diss .NET like you.