|Microsoft brains go head-to-head with Google|
Microsoft has hired top scientists in a quest for search algorithms that will allow it to compete directly with Google
I have said it before they need to get them to work on their own site first. Once they work out how to run an effective search there they may have a chance. :)
"It could be 50 years from now before some of this work is used."
It looks like the contest Google vs Microsoft isn't for tomorrow! :)
Even if MS develops the best search algo evr known, it won't matter because they will screw it up by displaying ads for the first 20 SERPS.
Who uses MSN anymore? Especially after they shafted us with the looksmart PPC. They will have to blow their own trumpet about their new technology if they want to succeed, and it must work. Mind you Bill Gates has got the cash
Lots and lots of people. MSN can give pretty good traffic, and it converts.
I would suggest reading that article however and doing some research. This has not much to do with Google in any case, and it's not so much interesting concerning today.
What those people are researching could very well have major influence on websearch in the future.
What Bill wants, Bill gets. If he wants the most popular search engine he WILL get it.
I'm not an active MSN follower and the Msnbot news has been around for a while, but this is the first official mentioning of replacing Inktomi by a Microsoft person? (the other pieces were more ambiguous about the future with Ink).
vitaplease, you're right. That's the most significant part of the story.
I read this article yesterday and thought about posting it, but decided not to- Because we all "kinda know" ;) already that ink is on the way out at MSN.
Still it's noteworthy to hear it from MSN. I suspect that it was a slip.
|Even if MS develops the best search algo evr known, it won't matter because they will screw it up by displaying ads for the first 20 SERPS. |
All for their best selling applications. You'll need to set the preferences to show 300 results per page.
Actually, I'd like to see MSN drop Inktomi and LS and do their own (free) results. I just think they should stop trying to compete and concentrate on making good products. They'd do much better as a company if they just played nice.
i betcha Bill Gates uses Google to surf da web :-)
That quote - from a Microsoft spokesperson no less! - is indeed huge news. Ya, we all "kind of new it" as soon as INK was acquired by Yahoo they would be out the window.
What I'm curious about is given MSN's variety of sources: Ink, LS, and Over, what will they do once they get their new engine rolling (if we guess it's about 2 years away from coming online)?
Seems to me the only logical move would be to buy Overture, with the 2nd largest advertiser base, in an effort to start taxing Yahoo as well (though Yahoo would then switch advert providers to Google) it would also give MSN a much bigger international reach. (for advertisers).
From the sounds of their team, they have some incredible brain power in the group - the Field's Medal, am I the only guy that immediately thought of "Good Will Hunting" when that was mentioned? :)
Funny, the News.com headline for this same story is "Microsoft brains take on Google brawn." I don't know. I think they have it backwards.
1.) Google's brains figured it out a few years ago. 2.) Microsoft's hired guns (brawn) still have to fight the fact they're working for ... Microsoft. 3.) Can't you count Microsoft's original ideas on one hand? Where's the brains in that?
Generally speaking, I'd take Google's brains any day over Microsoft's brawn.
I've got documentation from a book that Microsoft was one of the first USA based companies to adopt fuzzy logic in their programming, application design, etc. The quote was from 1992, I'm not sure what year they actually started using that type of mathematics.
Given that Google's founders said, in their famous, "anatomy of a large scale hypertextual search enigne" that they use a probabilistic method of indexing web pages, it seems to me that they are using a subset of the better mathematics for partial or ambigious data sets.
And then we have some research that Microsoft has done, in the search area, that is in conjunction with Chinese researchers at a university in Beijing. Since China has the world's largest number of fuzzy scientists (fuzzicists? lol) that would make Microsoft the one using the 'novel & innovative' approach, and Google the one using the old school methodology that was proved inferior about thirty years ago (with respect to probability for partial data).
Ha! Nobody, I repeat, nobody can displace Netscape --they own the browser market.
|am I the only guy that immediately thought of "Good Will Hunting" when that was mentioned? :) |
jeremy, my thought was somewhat different: it went something like "Bad Bill Hunting":)
Still, I like the news, they have enough cash to do some SERIOUS search research at M$.
You missed my point, Jeremy. If MS is so innovative, why are they just now shelling out wads of cash to hire programming geniuses so they can focus on search? How long have they had the piece of crap that is search.msn.com?
Search has been Google's focus since inception, and it's been others' focus, as well. MS is last to the table. That's innovative? That's brains? Not from where I sit. They know they missed the boat, so they're flexing their wallet to hire people to build a new boat. That's brawn -- "we didn't figure this out before, so let's spend some money and muscle our way in now."
Yep, rc, just like they did with the browser market. :)
I hope they figure it out and build the greatest search engine yet seen on the web, but like others in this thread, I won't be holding my breath.
There was a movie in the USA called, "Good Will Hunting" in which a guy played an MIT professor who had also won the field's medal. ;) Rather good story, too - if you're a geek.
Thus, the connection between the news story about Microsoft's team (who has a member that won the Field's medal for mathematic's) to the movie.
Sorry about being off topic, just wanted to clarify - that movie was the first place I read about the Field's medal for mathematics, and it kind of stuck.
>>>Deja Vu - nobody can take Netscape
Lol, in some ways, an apt comparison. However, Google has a huge advertiser database, and distributes their ads over the net through adsense, the Google syndication, etc. Even if they start to 'slip' as a user destination, I highly doubt all the folks paying them $$$ for advertising (which is the bulk of the revenue) will jump ship unless the adverts stop yeilding a successful ROI for the marketer's.
Somehow, I do believe MS will squish Google search wise, but it will take a few years to do so -> who knows what tricks Google has up their sleeve between now & then?
Google's Achilles' heel is that it is below MS when it comes to the desktop food-chain. Same problem NS had.
Now, what would G's be if it were expressed as a percentage of MS' revenue?
|Microsoft's team (who has a member that won the Field's medal for mathematic's |
There's at least one other who won that medal, Michael H. Freedman.
I was just clicking around that page reading their bio's and stuff. These people are like the X-Men of Mathematics.
>I was just clicking around that page reading their bio's and stuff. These people are like the X-Men of Mathematics.
If these are X-Men of Mathematics, the start-up I worked for previously had Z-Men of Mathematics, including the most cited mathematician of the past decade and one of the most IEEE-awarded professors, as its investors, managers and advisors.
Problem was that was all for hype, just like I am afraid, is the case with MS. These people will do their own work, and other people will do the actual work related to search engines.
um, IITian - you are aware that Microsoft has a bigger cash reserve than most countries in the world, right?
And that Bill Gates is the richest man in the world?
Better yet, that in the past, every single company that Microsoft has gone up against has fallen over time?
They have yet to lose market share to somebody when they wanted a piece of the market.
Besides, any company that employs two or more people who've won Field's medals (there can't be that many out there) is bound to have something going for it. Perhaps you are a bit off on your recollection of their mathematical skill?
>>Google's Achilles' heel is that it is below MS when it comes to the desktop food-chain. Same problem NS had.
nail on the head right there. The more MS can slap into the OS and set it at the default the more dangerous they get. If you doubt this go and do some reading on the LongHorn system they are rolling out. The lines between searching your harddrive and searching the web are becoming increasingly blurry. That's exactly the way MS wants it. Buy your new computer - hook up the modem to it and start surfing. Your default homepage is msn.com and your default browser is IE. Most people don't have the first friggin clue how to change their homepage and for as much as we love Opera the general public (the public that fills MS coffers with $$$$) has never heard of Opera and most of them don't see the need to install Netscape.
IMHO MS doesn't need to have a better search algo than Google to win. It's not really about the algo when you're playing at that level.
I think an open source search engine is needed. Linux is competing as an operating system, and something like that could be done with search.
I didn't write about the odds of Microsoft doing well - I think it will do well in the long-term, as it has always done. I was just mentioning that the big names are used to hype up one's image and as you rightfully pointed out Gates is the richest men on the earth, he will be able to hire very brilliant people as he has done.
However, most of the brilliant people have big egos and their own pet projects and believe me or not, search engine algorithms are not that high on many of these people's objectives. I suspect they are "trophy" researchers Bill Gates has been able to buy, but the real work will be done by more mortal developers at Microsoft.
PS Regarding comparing mathematical skills, I won't know much but I recall my ex-company's backers as being profs at mainly Berkeley and Stanford and being on boards of Field's committee. ;)
MS have got 150 million users just from hotmail. How hard to you think it will be to channel them to search there as well? Not very. As Heni says - MSN traffic converts like crazy. None of us think windows is up to much, but that doesn't change the fact that most of our customers use it. This is the situation MS will be aiming for with their search services as well.
What Bill wants, Bill gets.
> It's not really about the algo when you're playing at that level.
With MS it's about playing "courtroom". I hope other SE's are keeping tabs on how longhorn is being put together. My guess is that unless you get some kind of API out of MS, you don't play.
Hopefully some pre-emptive strikes before they own the market, and they *do* plan on owning it.
|I recall my ex-company's backers as being profs at mainly Berkeley and Stanford and being on boards of Field's committee. |
Michael H. Freedman
"The work for which he is best known is the solution of the long-standing Poincare conjecture in four dimensions, for which he received the Fields Medal, the highest honor in mathematics. Mike has received numerous other awards and honors including Sloan and Guggenheim Fellowships, a MacArthur Fellowship and the National Medal of Science. He is an elected member of the National Academy of Sciences, the American Academy of Arts and Sciences, and the New York Academy of Sciences."
"Christian Borgs is a senior researcher at Microsoft Research since April 1997. Christian Borgs has been a long-term visitor at Princeton, Harvard, and UCLA. He has twice been a member of the Institute for Advanced Study in Princeton, first in 1994/95 and again in the winter/spring term of 1996/97."
The Startup versus Micro$oft?
I can see a startup buying brains in order to feed the venture fund beast, but MS is in a different position. If you click around the above mentioned web site you'll see that these folks are in a league of their own. But I can see where you're coming from. :)
You win. ;) Anyway, this small start-up had David Donoho of Stanford Statistics Dept.(world's most cited mathematician of the past decade) on its team, Terry Winograd - have you heard of him - he is currently an advisor to a company called Google, Robert Oliver - heard of FICO credit scores? and many others. A brief profile of one of this start-up's advisors/investors/directors follows:
Elwyn Berlekamp, Professor of mathematics at UC Berkeley
IEEE Information Theory Society
1998: Golden Jubilee Award
1993: Shannon Award
1969: Best Research Paper Award
1991: R. W. Hamming Award
1990: Koji Kobayashi Award for Computers and Communications
1984: Centennial Medal
Scientific Advisory Boards
1999-now National Academy of Sciences, Computer Science Section
1977-now National Academy of Engineering, Computer Science and EE Systems Sections
1997-1999 Institute for Defense Analysis, FOCUS Review Panel
1984-1988 Livermore-Los Alamos National Laboratories, UC President's Science and Academic Advisory
1974-1976 Army Mathematics Steering Committee
1972-1974 Institute for Defense Analysis, chairman
Non-profits Governing Boards
2000-now International Computer Science Institute, Berkeley, CA (Chairman since October 2001)
1999-2002 College Preparatory School, Oakland CA
1994-1998 Mathematical Sciences Research Institute, Chairman
1986-1992 Head Royce School (K-12), Oakland, CA
1979-1982 American Mathematical Society
1970-1973 IEEE Information Theory Society, President
Academic Review Board
1999-now Princeton Mathematics Department
1997-now MIT Mathematics Department
1980-1981 The Technion EE Department
1971-1974 MIT EECS Department
2003-now Internet Mathematics
2000-2002 Proceedings of the National Academy of Sciences
1993-2000 Theoretical Computer Science
1982-now Utilitas Mathematicae
1967-1985 Information and Control
1979-1982 Mathematical Reviews
1970-1973 IEEE Transactions on Information Theory
1968-1971 American Mathematical Monthly
In regards to the numbers MS puts out concerning both their Hotmail subscribers and MSN search users, let's keep some things here in mind that are really, REALLY skewing those numbers.
First, with hotmail, based on a cross section of all the mail I've ever received from a hotmail account, about 5 out of 500 are real legitimate users of hotmail and not automated spambots creating account after account with which the send junk mail anonymously. From the recent articles regarding the problems with hotmail spam, my experiences are all too common. 150 million users is a gross exaggeration.
Second, with the claims MSN makes regarding that they are the #1 search destination (or whatever that quote on their site use to be) and receive X amounts of visits, let us realize that with their browser dominance, 9 times out of 10 when someone mistypes a URL, MSN search gets a hit. Think about how often your fingers slip, or you forget that address, or you leave out a W in WWW (or add in extra), etc... and compare that to how many times you actually intentionally go to MSN search. The numbers they present are, again, grossly exaggerated. If all browser errors were directed to my site I'm positive I could crack the Alexa top 5 with ease.