Cuil: Is this the Beta?

[Gangster Octopus]

OK, so I went and tried the hyped cuil search engine at http://cuil.com (pronounced "cool") and typed in "Straight dope message board" and it got me nothing olike this place. But it did say there were 1,060 matches, but only showed me two and I had no way to see the others.

This is supposed to challenge Google?

[Diogenes the Cynic]

I agree. Either I'm not doing it right or it completely sucks ass.

[bannerrefugee]

I played with cuil this morning. It was disappointing. Maybe with time it will improve. I liked the way it displayed search results though.

I have a hard time imagining any search engine that can really rival Google.

[Sunspace]

Odd. It's got the pictures of two of the Straight Dope books, but the text is random spam text in one entry and something unrelated in the other.

[Quartz]

Well, it does show that this site may be ripping off the SDMB.

I like the way results are shown better than Google.

But it's slow, which is a death-knell for a search engine.

I'm wondering if they're simply hoping to be bought out by someone like Microsoft or Yahoo.

[Diogenes the Cynic]

How are you supposed to access the hits other than what's displayed?

[Diogenes the Cynic]

The info link gets an error message. What kind of an amateurish operation is this? This is supposed to be their big launch day? This thing is ASS.

[Diogenes the Cynic]

A search for "Straight Dope" gets zero results. I'm going to assume that they're still working out some bugs, but as of now, I think Google is laughing its ass off.

[Duckster]

I entered "Straight dope message board" (with the quotes) and received 4,610 results. Using the two-column format, the real SDMB shows on page two, column one, second link. The Perfect Master would not be amused.

Cuil was exceedingly slow, and highly inaccurate. I tried several other search phrases and found web content attributed to everyone but the actual owner. I'll give them the benefit of the doubt their search engine needs tweaking, and the end of the week. If they haven't got it working by then, their investors can kiss their $33 million goodbye, not to mention their customers.

[Musicat]

Could be it's heuristic, and needs time to get refined.

[Mr. Excellent]

My first search for "ACLU" got zero results. The second time I tried it, it actually showed some reasonable hits - the ACLU national page, the texas page, and so on. But honestly, now - how many people are going to try their search twice? Most would probably just assume the page stinks.

[Musicat]

Originally Posted by Gangster Octopus

This is supposed to challenge Google?

Not yet, anyway. A random check of some of my web sites shows that it compiles a search results page by combining pix from different sites than the text next to it, and in most of the cases I saw, the pix have absolutely nothing to do with the text except if you think like a stoopid computer.

Example: one site (link below) is named doorbell.net. The name comes from the County (Door), and has absolutely nothing to do with wooden doors or door chimes. But Cuil thinks that it does because of the similarity of words.

And some of the links tied to pix have no relation at all, that is, the pic doesn't even come from the page it's linked to, but some unknown site.

[The Hamster King]

Wow, epic FAIL.

I google'd myself. The first page contained link to different professional organizations I belong to, some online interviews I've done, my blog from a couple of years ago, and my NY Times wedding announcement from 1996. All in all a pretty good snapshot of my online existence.

I cuil'd myself. The only thing that turned up was my name in the guest list of a random science fiction con from ten years ago. Utterly worthless.

Slow and inaccurate. Just what I want in a search engine!

[Whack-a-Mole]

I took a quick peek at Cuil and was not overly impressed.

Some of it may have to do with simply being used to Google. While Googles layout is uninteresting it is functional. Cuil’s results layout I find harder to navigate as simply.

Additionally Cuil does not hand you news results as part of your search as Google does. I find this feature of Google of great use as often I am looking for a media report on a subject and not the webpage of the subject itself.

Finally, Cuil's results seem off a bit. I Googled a website I am involved with and while Cuil returned links it did not return the basic homepage link at all on the first page (returned articles more deeply embedded in the site). I used the name of the website as the search so this was odd. Additionally, I searched on the term “cuil” on Cuil’s site and it returned nothing about them at all opting to show me links to Ireland and windmills and restaurants.

I really, really like that Cuil will not track users searches as Google does. Just on principle I very much dislike Google for doing this but so far I am not seeing Cuil as a Google killer.

Here’s hoping though that Google finally gets some decent competition!

ETA: Tried doing a search (several times) 30 seconds ago on Cuil and this is the response I got (that is something Google never does and does not engender faith in Cuil):

“We’ll be back soon…

Due to overwhelming interest, our Cuil servers are running a bit hot right now. The search engine is momentarily unavailable as we add more capacity.

Thanks for your patience.”

[Duckster]

It's DOA now:

We’ll be back soon...

Due to overwhelming interest, our Cuil servers are running a bit hot right now. The search engine is momentarily unavailable as we add more capacity.

Thanks for your patience.

[Quartz]

I tried a vanity search and it was dreadful.

[Kid_A]

My dad likes it...it won't make it past next Tuesday.

[An Gadaí]

I did a vanity search and once I used the ""s it was sorta useful, otherwise scheisse.

[mnemosyne]

I tried searching for the company I work for... Google gets it as the "I'm feeling lucky" link. Cuil gets it on the third page with a useless summary, and I had to check the web address to see if it was actually correct.

Also, I keep reading it as Cul.... ass, in French. Sounds about right.

[Quartz]

And, according to The Register www. culi. com is an Italian porn site. Worse, according to this - http: //www. theregister. co. uk /2008/07/29/ cuil_launch/ - NSFW article, they paired up an academic first with a soldier and secondly with gay porn.

[Green Cymbeline]

Originally Posted by Pochacco

Wow, epic FAIL.

I google'd myself. The first page contained link to different professional organizations I belong to, some online interviews I've done, my blog from a couple of years ago, and my NY Times wedding announcement from 1996. All in all a pretty good snapshot of my online existence.

I cuil'd myself. The only thing that turned up was my name in the guest list of a random science fiction con from ten years ago. Utterly worthless.

Slow and inaccurate. Just what I want in a search engine!

Yep, I did the same thing with the same results. When I google my name, the first hit is my web site. I got nothing relevant with cuil. When I looked up another page of mine, it had an accompanying thumbnail photo of some strange woman who I had never seen in my life. WTF?

[control-z]

I don't understand the images they display with the search results. Sometimes they're relevant, sometimes they're not. Sometimes they're banner ads from the site, not very useful. On our site, somehow they extracted the text from the top of our home page and made an image out of it. I don't get how or why they did that...

[Giles]

I searched for "cuil" on Cuil -- and got something about Cuil Mhuine, Ireland. When you search for "google" on Google, the first hit you get is what you expect. Oh dear -- when a search engine can't even find its own home page, you have to wonder

[jk1245]

Originally Posted by Giles

I searched for "cuil" on Cuil -- and got something about Cuil Mhuine, Ireland. When you search for "google" on Google, the first hit you get is what you expect. Oh dear -- when a search engine can't even find its own home page, you have to wonder

Even better, type "cuil" into Google and click "I feel lucky".

As I said in another thread, Cuil has about 12 more hours to get this fixed, or forevermore "Cuil" will be shorthand for "epic failure"

[Duck Duck Goose]

Epic fail. Type in "bezoar" and you get a weird mishmash of websites selling "bezoaar" talismans, and offhand mentions of the word "bezoar" in other documents. Put "bezoar" into Google, and up comes Wiki, Medline, and a host of other--informative--hits.

[Whack-a-Mole]

Part of me wants to give them a bit of a break as a startup is bound to have some issues but this thing is so screwy one has to wonder how they ever missed it and thought opening for business at this point was a good idea.

[Derleth]

Cuil is a crappy search engine because it's 2008 and we know how search engines are supposed to work. The world has been using search engines for over a decade now (yes, the Internet has been mainstream for roughly a decade) and the interface portion of a search engine is a solved problem: Alta Vista, Yahoo!, and Google all did and do the same thing, to varying degrees of usefulness. Cuil is simply a really crappy version of the technology going head-to-head against an entrenched version that's a whole lot better. There really is no competition and six months from now Cuil will have been forgotten.

[NurseCarmen]

I entered "tits" and only got a few results. Very few were what I would prefer.

Total failure.

[Diogenes the Cynic]

Originally Posted by NurseCarmen

I entered "tits" and only got a few results. Very few were what I would prefer.

Total failure.

My first search yesterday was "free porn." The results were highly unsatisfactory but I actually got more hits with safe search on than off.

[troubledwater]

The search for my website's keywords returned the homepage - and apparently every single other page on the site- of which there are hundreds, as it is a genealogy site for which I used a program to generate a separate page for every individual in my database. And of course completely random images, including one the looks like the work of Ralph Steadman, for the page about my great-great grandmother.

[Musicat]

Originally Posted by Ms. Meta

... And of course completely random images, including one the looks like the work of Ralph Steadman, for the page about my great-great grandmother.

That seems to be a common thread in many complaints, including mine. Cuil's attempt to match images with text just doesn't work much of the time, producing bizarre results. They seem to ignore the obvious associations -- a pic on the same web page as text probably has some connection, but they don't use that piece of information.

[control-z]

Originally Posted by Musicat

That seems to be a common thread in many complaints, including mine. Cuil's attempt to match images with text just doesn't work much of the time, producing bizarre results. They seem to ignore the obvious associations -- a pic on the same web page as text probably has some connection, but they don't use that piece of information.

I guess in fairness, Google News has the same sort of problem. At least Google doesn't try to associate images with the search engine results though.

[h.sapiens]

Weird. I searched for my company, and most of the results on the first page were not from my company's website, but product pages from the site of one of our distributors. The images are of our products, but they're all mixed up, and don't match the product in the link.

Then, when I clicked on the link for the second page of results, I got a page saying there were no results matching my search term.

[bordelond]

Originally Posted by Quartz

I like the way results are shown better than Google.

To me, that's by far Cuil's biggest detriment. I don'f feel I get an added benefit from seeing images, nor from previewing random content from random pages.

[Philster]

The pics next to some links sometimes have nothing to do with the link. I am sure they are connected in some roundabout way, but I can't figure it out. I CUILed my company name and the pic was of an EA Sports game...and we are in the credit business!

[msmith537]

fauil (pronounced "fail")

[Astroboy14]

I'm gonna give it a few weeks and try it again... I know nothing of how search engines work, and maybe it needs a breaking-in period or something?

But, yeah, so far FAIL.

[control-z]

Google is so good that I doubt I'll try Cuelle again unless I start hearing that it's absolutely great.

[Quartz]

Originally Posted by bordelond

To me, that's by far Cuil's biggest detriment. I don'f feel I get an added benefit from seeing images, nor from previewing random content from random pages.

You mistake me: it's the organisation of the text that I like. It's very much like red-top computer news sites like The Inquirer and The Register.

[hajario]

Back in 1993 or so I was at a small party on X-mas Eve with some friends. We had all kinds of gourmet food that we were making. One of the guests decided to try to make Oysters Rockefeller. We were ready to go when we realized that none of us knew how to open an oyster. It was X-mas Eve and there was no one to call because everywhere was closed. Luckily my hosts were on-line and I got my first taste of the power of the Internet. We searched for "how to open an oyster" on Alta Vista or something and instantly had instructions. It was an epiphany for me as I realized the way of the future.

Yesterday I put "how to open an oyster" into Google and got millions of hits. Right on the first page were a couple of YouTube vidoes as well as sites with pictures and great instructions. I put the same thing into Cuil and it didn't return a single hit. It was worse than the first search engine that I ever used nearly fifteen years ago.

Cuil=Fail

[Savannah]

Wow, it's bad. Vanity search on myself took two pages in to find 'me'. Searching for the company I work for brought up pages of crap, spam-laden links using the company name. Apparently, our female employees sell a lot of misspelled herbal Viagra.

Not impressed. Especially since the company name came up in the list of suggestions as I was typing; you'd think the links provided would be better than dubious online drug sellers.

[Sage Rat]

I've oddly been on something like the Cuil side of things in this.

I was working for a company in Japan that wanted to do word ranking based on lexical analysis of blogs, so you could like, look up a book and it would give you a score based on what people had said about that book. So you had an instant review from normal people summarized into a single number, and then of course you could click through to see the blogs themselves.

The problem was that as a startup, while we knew what we wanted to do, we of course had little money and no real good programmers to do this with (except for myself, woo!) Plus, before they had hired us on, they'd hired a company to do the front end, webpages, which ended up being huge, half of it black boxed, and all of it buggy. And then management had decided up front of even hiring anyone that we'd have a product ready and on the web in two months. That was entirely unfeasible, but the issue was that they'd already gone out and announced the ship date and sold it to the investors and whatnot. So overall, the instant the thing worked well enough to generally display a result instead of an error message, we were told to make it public.

It sucked and was buggy and totally shouldn't have gone out the door, but I guess when you're running on borrowed money, there's a pretty strict timeline that you've got regardless of any link to practicality. You just have to hope that it doesn't ruin every prospect that you have when the first batch of viewers summarily trash the product. Ours was different enough from anything out there that we were able to carry on and get better (I think), but Cuil might be toast just based on this.

[Capt. Ridley's Shooting Party]

It seems to be improving. I did a vanity search last week, and found myself on about page three (from Google searches, I have a pretty unique name, only one guy in the world shares it). Today, I'm at the top of the first page.

The results still aren't as good as Google's, but it does seem to be getting better.

[bannerrefugee]

Originally Posted by Capt. Ridley's Shooting Party

It seems to be improving.

I re-did the searches I did earlier, and it's a major improvment.

Before it found nothing for cyclingnews, now it found it on top of the list. When I searched for TTUHSC before, it only found inappropriate sites, now the the home page was the second on the list.

[Sunspace]

It seems to be working better now. It hinted 'Straight Dope' when I had typed 'straight d', and the resulting finds were mostly actual Straight Dope pages.

I put in my own name. I'm still an evangelist in Alabama, it seems, as well as an aspiring songwriter. No hints of my own web pages in the first page.

[control-z]

Check out this article: http://www.pcpro.co.uk/blogs/2008/07...fins-and-porn/

Lunch is ordered in every single day. Huge fridges burst with snacks and drinks. Bowls of strawberries and muffins lie around the rest area. The company pays for a personal trainer and gym membership for everyone. A doctor calls round each Friday, after the weekly barbeque, to see if everyone’s in good health. Employees drift in an out at times that suit themselves,” she continued, before temporarily expressing worry about the spending.

[Tastes of Chocolate]

These folks really need to figure out how to move a sites main page to the top of the list.
I did a search on my employer.
Page 1: Link to our Hong Kong Branch.
Link to our investor relations page
Link to our China branch
Link to another random page from our site.
Wikipedia entry
Link to one of our divisions
Wikipedia entry for someone who's last name is the same as the company name
Link to an artical mentioning the company, from an industry magazine.

I went through the first 4 pages of links. There are probably 20 links to pages on the company site, but none of those are the main page.

[Gangster Octopus]

Originally Posted by control-z

Check out this article: http://www.pcpro.co.uk/blogs/2008/07...fins-and-porn/

Ridiculous.

[Huerta88]

I just checked and it still blows.

I'm bumping this as an example of why I do not trust the media. This could be a case study in how easy it is to dupe major press outlets into giving you millions and millions of dollars of free promotion regardless of how half-baked, almost sociopathically-unprepared, your alleged business is.

On the bright side, if any of you have a legit startup, by all means take this as an object lesson in how to get free front-page advertising.

[Quartz]

Originally Posted by Huerta88

I just checked and it still blows.

Heh. I just tried it too. What a disaster area!