Slashdot Mirror
New Search Engine Cuil Takes Aim At Google
theodp writes "CNET reports that Cuil (pronounced 'Cool'), a startup founded by the husband-and-wife team of Xift creator Tom Costello and former Google search architect Anna Patterson, is launching a new search engine today that claims to index three times as many Web pages as Google." Running a few searches left me underwhelmed with the content of the results (hitting the next-page button on a search with a listed 62,200,000 results — for "seattle" — got me the unexpected error message "We didn't find any results for 'seattle.'"), but pleased with the actual layout of the results when it worked, so I hope the kinks are worked out. Update 7/28 18:30 GMT by SM: corrected Tom Costello's accreditation, he wasn't a professor at Stanford as the linked story suggests, just did some research there as a grad student. Thanks to the Stanford CS department for pointing this out.
Well it sure looks nice, puting pictures along with the results is a cool touch. It's a pity that the usefulness of the returned links is not on par with google.
sheep.horse - does not contain information on sheep or horses.
what I miss most is any sort of 'advanced' search, like the restriction on TLDs etc.
Idha khatabahum lijahiluna qalu salaman
seems to be fixed.
I also tried Tiananmen and was returned a blank face (I'm in China). This is many Chinese people's first benchmark at a new search engine. For me, the result is expected, since the Great Firewall is a hybrid of generic and Google/Yahoo/etc-targeted implementations.
Colorless green Cthulhu waits dreaming furiously.
The layout is pretty, the related results is nice, but the main function... the results... they suck. I was searching for an uncommon sailboat and there were 0 pages returned. Google returns results for the same query... On other searches, the domain name no longer resolved, there were 404s, I got a page that was last visited in 1997, just junk. The results summary needs to be cleaned up also, lots of funky symbols in the results are just noise.
This is the same reason I stopped using Altavista way back when. I don't buy this 120 billion pages thing. You know you can get every article on slashdot on games.slashdot.org, tech.slashdot.org, politics.slashdot.org, etc... I bet they include all those, and every other site that allows you to view message threads 8 different ways. But no results for my first query!
It could be a while before someone is going to beat google at searching. I really do like the alternative approach to displaying search results, so I will at least keep my eye on it.
There's a lot of talk about how Google is in decline, and I won't comment on that, because every company has its tipping point. But for them to have been a invaluable (and in many cases incomparable) tool in my life for the best part of a decade and to have remained almost invisible as an agent in that process takes some doing.
In fact, the most insidious thing about Google may well be that any new attempts at reorganizing the layout of a traditional search engine, such as cuil is now attempting, seem like deliberate contrivances. And probably are.
I have another observation - there doesn't seem to be any localisation to it. That may well come with time (it was a while before Google had it), but being able to go to google.co.uk and specify that it should search for pages from the UK can be extremely helpful, for example when searching for details of public holidays or shops or similar region-specific things.
And yeah, Cuil is a dumb name.
It's official. Most of you are morons.
I just "cuiled" the phrase "problems with linq to sql".. and it suggested there were no pages at all. Google however knows there are TERRIBLE problems with LINQ to SQL and served me 3,180,000 results. To say there are no problems with linq to sql is not very cuil at all.
Why does Cuil claim to have "2,784 results" to my search yet display only one? Does it mean we have to divide the impressive 121,617,892,992 claimed index web pages by 2,784 to obtain the astoundingly round number of 43,684,588? What are the odds that the result of this division would be an integer number?
1 out of 2,784.
You just got troll'd!
There are definitely a lot of bugs. For example, the first page of a search for "Cambridge University" fails to link to the University of Cambridge's official site, but "University of Cambridge" has it as the first hit.
I can see why that's happening, but that would catch a lot of people out.
Yes, the name is really bad: "Cul" is the french for "ass". And the logo even highlights the three letters c, u and l. :-O
Privacy Policy:
Way to go!
I think I would have done a closed, invite-only beta before going live with this thing... two of the four searches I just tried came back with nothing, and almost all the images that came back with the articles were not relevant.
What could be cool is the automatic synopsis they have going on... if they can make that work a little better, it could be a good place to go for some quick information on a topic.
They still have a long way to go, though.
Don't blame Google for using job market and IT industry supply and demand forces to fill whatever positions they need.
Google found a niche, exploited it for all what is worth, and are so efficient at it that they can allow themselves to get the best talent money can buy.
Please grow up, that is how a job market is supposed to work. If the rest of the IT industry can't come with innovative ideas good enough to attract new investment and bright people, it is hardly Google's fault.
When Google comes crashing (yeah, we know, all companies do, thanks for the insight genius) it will be for more important reasons than treating well, even lavishly, their employees.
At the moment it seems to be working, so I really don't see why they should change. I am not saying that all companies should provide whatever Google provides, but I am sure that morale in many companies would be increased immeasurably if they put a few pool tables around and some comfy sofas were to nap or to have a chat.
Most companies forget their employees are human and that it is important to give a degree of human empathy to your employees.
When the bad times come all those extras can be taken away, but to do so at a time when business is brisk is nonsense. It just shows why they are billionaires and some around here are unsatisfied middle managers or angry technicians.
IANAL but write like a drunk one.
The images put next to descriptions are, let's say, a little odd.
For example, I searched for 'titanite' (a titanium silicate mineral),
http://www.cuil.com/search?q=Titanite&sl=long
Not sure if it'll be fixed by the time you read this, but it had some nice My Little Pony type things next to the link to the Wikipedia article.
And, more seriously, I don't think the quality of search results on a few random tests I tried were anywhere near Google in terms of quality.
Jolyon
Please read my Canon EOS tech blog at http://www.everyothershot.com
Cuil has only just opened. Already, it is pretty decent.
I disagree. Basic one word searches return no results. I have to change my oil in my car and wanted to search out the best brand to use. Typing in "oil", let alone my original search term, yields nothing.
While I agree it would be great to have a nice alternative to Google, Cuil had better step it up a notch...
On cuili we get:
Google gives much more relevant hits
...built on a business model that is not sustainable (in Google's case, online ads, which I honestly believe are going to tank hugely when advertisers finally realize the true value of online advertisements such as those sold by Google - and it ain't alot in my opinion)
Regarding the true value of online adversing... In my case, I can say that the advertising I put onto Google is worth it. I've just enabled the online store at http://www.lillifoot.co.uk/ and started advertising on Google. It's very easy to track the metrics of how much I spend versus how much income it brings in. If the advertising wasn't covering costs, I would be looking elsewhere to spend advertising money.
My business: Farstrider Studios.
Cuil has only just opened. Already, it is pretty decent.
I disagree. Basic one word searches return no results. I have to change my oil in my car and wanted to search out the best brand to use. Typing in "oil", let alone my original search term, yields nothing.
It's just after 7AM CST, I searched for "oil" and I got 175,600,000 results for oil. Not only that, they have this "Explore By Category" menu with the first category being "motor oil" with a list of different types of oils "synthetic", "fuel", etc.
Then across the top, as a nice touch I think, they have a horizontal menu-like bar that must list the most popular oil keywords like "Olive oil", "Oil Paintings", and "Fish Oil".
Definitely interesting. I won't complain about problems with their server for the time being. I do see that clicking on "motor oil" in their "Explore by Category" sections changes the search query to "oil motor oil" which returns "no results found". There is still a lot of "learning" their algorithm needs to do. If you remove one of the "oil" from the query, it returns plenty of results.
I'll see how their doing in a week or two, once they've gotten the holes plugged out of the new system that's been in the water for a while. Looks very promising though. The search industry really does need something new like this to refine usability. Push Google back into innovating with search along with their many other side projects.
"The past was erased, the erasure was forgotten, the lie became truth." ~1984 George Orwell
Check out the yacy search engine. It's an interesting approach (based on p2p technology) and - theoretically - can't be censored.
-Dennis
Sigs suck!
Except the Cuil founders used to work at Google, and Google is constrained in what they can do because they have a big user base.
Cuil could have innovated greatly and pushed search to the next level. Cuil looks pretty incremental to me, and a bit unpredictable at that.
The fact that some common phrases result in no hits also suggests to me that they optimized at an unacceptable expense in search quality. In the end, users don't care how few machines they are using if they simply don't deliver the results.
What mechanism will bring about this Google crash? Unlike the famous companies in the .com bubble, Google is actually making money. And lots of it. More than a billion dollars a quarter, to go along with their $12 billion in cash and zero debt.
Google won't crash like the .com hilarities of 2000, but to consider it invincible is a mistake too.
While they expect to drag in $4.8bn this year, they are valued at $152bn. This is equivalent to a gross return on investment of 3.1% - which is not, in my opinion, an encouraging yield. Their advertising revenues (in my opinion) are unlikely to grow significantly - and (as far as I can tell) they've scant other revenue stream developed. In an economic downturn, of course, I'd also expect to see advertising spend slashed... putting even further pressure on share price.
As I see it, the only justification for today's share price is an expectation of spectacular capital growth. I think that game is over. If google returned 10% yield, I think that would be credible... that would decimate their share price - and, in so doing, would foster a complete lack of confidence among investors who would see their speculative gains wiped out. Even $12bn in cash won't keep them up forever - especially considering the size of their wage bill if they are going to stop their to talent jumping ship.
Don't get me wrong, I think Google are awesome, but I do think they are over-valued.
It seems it can't understand characters other than Latin. It gives me no results even for the simplest words written using, in my case, Greek fonts. I remember some years back Google had the same limitation.
"Sum Ergo Cogito"
There have always been few companies where the best and brightest wanted to work and could do research: Bell Labs, AT&T, MITRE, SRI, IBM, Xerox, Microsoft, Google, and a few others.
It's good that Google exists, because without them, the only corporate research lab worth talking about would be Microsoft, and that is a truly depressing state of affairs.
true, however my colleague and I both noticed 1 thing immediately about this search engine:
no f*cking links to shopping comparison sites.
Just because of that, I may come back to it in a month or so, and if it continues to filter those b*st*rds, then I may well just keep using it. That in itself makes it a Google killer IMHO, some searches in Google return a dozen pages of the same kelkoo, pricerunner and dooyoo links.
Secondly, I wonder how long it is until we see some sweet Cuil hacks? Third, the name is escaping me right now but there were two professors at Rutgers University who were working on a search engine for Ask.com a few years ago. Once that "development" engine was successfully deployed it was quickly integrated into Ask.com. I'm wondering if this might eventually wind up being a similar situation.
Cuil has only just opened. Already, it is pretty decent.
I don't know, the algorithm seems to be rather bad. They surely have indexed a decent amount of pages, but extracting relevant information seems to be done rather badly. E.g.: if you search for a language, whose name is also a place name or something, and put the name - say "Rwanda" a Bantu language - in the search field and then add "language", because you are interested in the language and not the country. Google ranks the results with a closer syntactic (or semantic?) connection between "Rwanda" and "language" higher, which gives you information about the language Rwanda or at least languages in Rwanda, while the results from cuil are not decent, to say the least. The first page has no result about the language Rwanda, only rather random pages somewhat connected to the country Rwanda. So it seems that cuil does a rather bad job in retrieving the relevant information from an indexed page.
I for one would love to have options to Google.
Yes, agreed, very much.
"Hannibal's plans never work right. They just work." Amy/A-Team
Do they really think "Cuil that for me" and "Did you cuil Jon Katz" lately is going to take off? Right off the bat I would have gone with Plex. Hell, "Plex: it's what's next", rolls off the tongue. Same for "After Google there's Plex". Anyway if they really want to take over Google they have to consider how to enter the modern lexicon and Cuil just isn't going to cut it. Today John McCain announced he was learning how to use the Cuil. nope.
Hmmm. I was just showing Cuil to someone and it returned 67,000 results for the "EngenderHealth" search. Maybe it is just database overload.
$nice = $webHosting + $domainNames + $sslCerts
My last name on Google produces over ten thousand results (prolific family, I guess), some of which are not so obscure (eg, my brother runs a major web site, my research has appeared in The Economist, on the AP wire, and other international news outlets, our mother has published a large handful of scientific papers, our father designed one of the now-standard architectures for multipliers that appears in numerous textbooks, and we have half a dozen or so patents between us). On Cuil, there are no results. None. Zero.
Cuil's lying about indexing more pages than Google in one way or another. Either they aren't in fact indexing that many, an outright lie, or they have indexed them, but can't do anything with that information, a lie of omission.
Cuil's image-with-each-hit idea is cute, but personally, I find it to be fluff. Give me text-only when I want information, images-only when I want images. Forcing a pairing makes it look like USAToday, rather than an information source.
Put my fist through my alarm clock with its ding-dong death inside my ear. - The Blackjacks.
The price per earnings ratio may be a little out of whack, agreed. However, you also have to look at the other factors. Having a debt-to-equity ratio of 0 is nothing to cough at.
Yes. Its was instantly rememberable. Goo-gle. Done. Hey, that's a play on googol. Awesome. Hey, their results rock. Screw yahoo. never been back
I was consistently getting "No results" for common search terms, until I turned off SafeSearch, I then got expected search results.
If you aren't blessed with particularly tender sensibilities, I'd suggest you turn off the "Safe Search" option. I entered "rhinoceros" in the search box and came back with virtually nothing. After turning off "Safe Search" and "Typing Suggestions" in Preferences, I got back about what you'd expect from a decent search engine.
There's obviously some bugs to work out, but Cuil looks like it could be a winner. I especially like the fact that they emphatically DON'T keep track of searches and personal information. That aspect of Google has always made me a bit uneasy. True story: about six years ago, my sister was having trouble with her new dog. I thought I'd find some information to help her train the beast, and typed "Dog Instruction" into Google's search box. What came back was, um, not exactly what I expected.
I can imagine Google linking that search up with my computer in some way, and stuffing it away in its little electronic filing cabinet. For that matter, I'm not pleased with ANY of my searches being tracked, and not because I spend my life searching for pornography or anything else shameful. It just isn't anybody's business.
I've calculated my velocity with such exquisite precision that I have no idea where I am.
You say Google will come crashing down, yet you also say no one will "ever" be as good in search. So I ask again, what mechanism or event do you foresee in your crystal ball to bring about such an unlikely crash?
Cuil has only just opened. Already, it is pretty decent.
No it's not. Maybe it's the load but it's totally screwed up. Results return with images that have nothing at all to do with the actual result. (Someone posted a screenshot of a result about Giant Squid, the image with it was two WWE wrestlers). Some queries come back and say no results. I personally tried anime blogger and got nothing. Google returns 3,910,000 results for the same query. Bizarrely searching for anime blog got results, however nearly every single image was wrong, most coming from sites not even remotely connected to the site of the actual result. Google was never this bad. I can't remember any search engine performing so horribly.
And I'm sure someone will say that it's just cause of the increased load and it'll be better later. There might not be a later for them. This is their public unveiling, there's even an article on CNN.com about them. And what are they doing? Totally blowing their time in the spotlight. Irrelevant results, no results, slow responses, etc. Many people won't come back. Even if they get reminded months from now they'll probably remember their bad experience and not bother. Cuil's going to have a VERY steep slope to climb to overcome this, much more than just competing against the behemoth Google. They've basically shot themselves in both feet and broken an arm.
-A INPUT -s 64.1.215.160/255.255.255.224 -j REJECT --reject-with icmp-port-unreachable
whois returns:
network:Organization;I:CUILL, INC. (259492-1)
network:IP-Network:64.1.215.160/27
Why? because despite multiple attempts to prevent the aggressive spidering of my sites by this abuser they did not stop.
Buh-bye, losers.
-- @rjamestaylor on Ello
Hmmmm....
"""
Cookies
Cookies are small files on your computer that websites create to store user preferences, such as language settings. Each time you visit a Cuil page, your computer's cookies automatically provide Cuil with your preferences. You can change or delete your cookies anytime via your Web browser options.
We do not record the information in your cookies on our servers; your browser sends your preferences to us with each search request. This way, we do not store any personal information about you on our servers.
"""
While the 'suggest' and 'safe' cookies do indeed simply store my preferences, I'm a little bit suspicious of the 'TRACKID' cookie. I don't remember indicating any preference for being associated with an ASCII nonce.
Also FatPhil on SoylentNews, id 863