Cuil - long on marketing, short on effectiveness

'Cuil' is pronounced 'cool' or more likely 'kewl'. It's never a good sign when you have to start by telling people this, but at least it's not gone down the double 'o' route. Cuil has been around in stealth mode for a while, but it's exploded onto the general public scene in the last few days with a blaze of (unwarranted) publicity. It's making great claims, and some people are talking about it as the latest Google killer. I've played around with it for a while, so lets take a look at some of the features and functions.

It indexes 120 billion pages. I think that I'm supposed to be impressed by that, but I think we left the macho posturing of 'We index more than you' behind a long time ago. It is interesting to note that Google released their little note last week about being aware of a trillion webpages - they didn't actually say that they indexed that many, just that they knew about it. Even if it's the largest index that simply means that there's even more pages that I'm not going to look at, with the whole long tail business. Having said that, having run a few test searches against Google, Big G still comes out with more results. So, the size of the index doesn't mean a great deal to me. However one could argue that if Cuil does have a bigger index it behoves it to use that effectively. This brings me onto my second point, one of functionality.

I'm quite frankly astonished that there is no indication on the site of search functionality. As regular readers know - one of my hobby horses is help pages or cheat sheets. Google's pretty good on this, with examples, cheat sheets and so on. Cuil has *nothing*. They are expecting their users to either whack in a few words and thats it, or they're so confident of getting good results they think they don't need to provide help. However, I've got to play around and see what Boolean operators work, proximity, what file formats are searched, does it limit by country or language, can I limit by time/date - I have no idea. I'm going to have to waste MY time trying to do THEIR work for them, and quite simply it's not going to happen that quickly. So, functionality is limited to heavens knows what and there's no advanced search function. Which, if the index is as big as they say it is, is pretty poor.

The display of results is a little different, in that it's in magazine style. This isn't unique - Silobreaker for example takes that approach as well, and Exalead allows for different displays of results. Users can choose if they want 2 or 3 columns. We're expected to guess which is the relevant result - is it in the top left hand corner? The top 3 results? Who can say - it's just part of the Cuil guessing game. There are two further search/display options. The first are a series of tabs across the top of the page to focus on particular categories. Great in concept but poor in execution. A search for 'Everton' for example brings me up tabs for Everton FC and Everton Football. Quite what the difference between them is I'm not entirely sure, and I'm not going to waste time getting into a guessing game. The other option is an explore by category approach. Cuil gives us the option of exploring Premiership teams, and of the teams they list in this section 80% are not IN the Premiership. This is not right - it's not even wrong! It's hopelessly inept, and quite frankly inexcusable.  This does not give me any kind of trust in their system at all.

It gets worse. Next to a lot of the results entries are images. These apparently relate directly to the result but they don't. I ran a search for me (the way you do) and up popped a link to my Flickr collection. Sensible and good so far - except that the image shown isn't one of my images! I have no idea where this image comes from and have no way of finding out. Disgraceful. This isn't a single error - it happens time and time again - images do NOT relate to the pages that are returned. Total shambles.

Cuil is hot on the concept of privacy - to the extent that they don't keep any kind of information on who is searching for what. No problem with this, but it does mean that they've tied themselves into a system where they're not going to be able to offer personalisation, and given that this is going to be an increasingly hot topic in the future means that they're sidelining themselves from the start.

I hope that Cuil improves - the more, and better the engines, the more we'll all benefit. However, this is so ineffectual that it's not going to be competing with Google - ever. If people are looking for the elusive Google Killer they'd better get used to the idea that it's not going to be this one. If you're looking for an alternative to the major players - Exalead is still (in my opinion) your best bet.

Comments

[Scott]

I too searched for myself (as you do) and got a random photo that isn't mine next to link to my blog. Going through a few pages of results I got my blog, and my claimID page three times in 5 pages, and some my name as Ning user's friend amongst the odd finds. Interestingly, aside from a US Librarian that shares my name there is also a professional snowboarder, whihc will explain why top of the expolore by category section is Native American Writers? Er, what?

I too, really hope that they can do something with Cuil, but I think the fact they went live with it, when it is this poor, was a bad move - not helped by proclaiming "Welcome to Cuil—the world’s biggest search engine. The Internet has grown. We think it’s time search did too." With Cuil itis more groan, than grown, sadly

[Tom]

There also seems to be some doubt over the name which according to Cuil itself (but no-one else) is Irish for knowledge: http://languagelog.ldc.upenn.edu/nll/?p=420

[Sarah Washford]

I had a play with Cuil and have to agree with the negatives. I also searched for me and either came up with very old comments left on other peoples' blogs or adverts for twitter. Usually a search will at least bring back something recent or relevant in the first page or so! It also didn't seem to make any difference when I tried to create a phrase with speech marks or any of the other helpful extras I use with google, ask etc. Gave up and went back to the usual suspects pretty swiftly...

[David Larch]

I'm for the privacy. I don't need a site to remember my search preferences at the cost of a cradle to grave file for the government peruse when they are looking for the next bunch to put in the boxcars!