«
»

cuil.com search ranking

I was intrigued to read about a new search engine, cuil.com, which is being launched by former Google employees. So I tried it out by searching for XULRunner. The top hits were:

There were also suggested “categories”:

Mozilla

Mozilla Foundation, SpiderMonkey, MozillaZine, Mitchell Baker, SeaMonkey, Mozilla Corporation, Mozilla Public License, Bugzilla

Mozilla Developers

Asa Dotzler, Mitchell Baker, Daniel Glazman, Window Snyder, Brendan Eich, Mike Shaver, Ben Goodger, Tristan Nitot

Mozilla Extensions

ChatZilla, Adblock, Greasemonkey, CustomizeGoogle, DOM Inspector, Venkman, Flashblock, ColorZilla

Netscape

Mozilla, Netscape Navigator, Netscape Browser, The Book of Mozilla, Daniel Glazman, Ben Goodger, Netscape Communicator, Tristan Nitot

W3C Standards

MathML, Resource Description Framework, XSL Transformations, Document Object Model, Cascading Style Sheets, Scalable Vector Graphics, XHTML, … (hidden)

I’m a bit disturbed by the pattern here: a search for XULRunner turns up interesting results, but the primary page on XULRunner (http://developer.mozilla.org/en/docs/XULRunner) is not present. Suggested terms for Mozilla has interesting results, but the most important (Firefox) is not present. The suggested categories are interesting, but the sub-results aren’t relevant to the specific topic I searched for: what I really want is “Mozilla Developers who do XULRunner”.

Is cuil.com built on a strategy of “list all the search results except for the one I really wanted”?

Oh, and a search for “tamarin javascript” turns up 0 results… something must be fishy with their search index.

This entry was posted on 28 Jul, 2008 at 2:20 am and is tagged with , , . Trackback.

Atom Feed for Comments 6 Responses to “cuil.com search ranking”

  1. Sebastian Says:
    July 28th, 2008 at 3:11 am

    The index is very out-dated (it’s probably the most expensive part of a search engine), which of course makes ranking difficult. It would be interesting to see their ranking algorithms on a more recent version of their index.

  2. david harrison Says:
    July 28th, 2008 at 3:52 am

    Holy shit! That (“list all the search results except for the one I really wanted”) is _EXACTLY_ what I thought when I ran a few simple tests through it. There was a heap of variety but the “main” (or authoritative) link I was expecting to see at the top was nowhere to be seen.

  3. Aqualon Says:
    July 28th, 2008 at 6:12 am

    A search for “tamarin javascript” finds results now. I also had the problem that sometimes nothing is found and after a reload I get lots of result. Perhaps somethings broken with their load balancer.

    And yeah, “list all the search results except for the one I really wanted” sounds really like the track they are following ;)

  4. owenmpk Says:
    July 28th, 2008 at 9:44 am

    Cuil misses basic marks for me.
    Cuil is a new challenger to Google and when I read about it Monday AM I checked it out. I liked the clean first page at http://www.cuil.com but that went away with the first search results. The search results are show in 3 column format which I find very hard to read. I based on my experience with my clients and how they prefer files and folders to be shown list or detail view Windows Explorer they have missed the mark. Second is on their pages describing product they use 10 point font. What is it with web designers today do they think everyone has perfect vision? Or every viewer of the web under 30? I can work with the small font but until the search results page gives me the option to view in list format I will not be using this service.

  5. Neil Pfister Says:
    July 28th, 2008 at 10:10 am

    I could not agree more. Every search term I entered returned nothing I actually wanted. Back to Google we go.

  6. Vic Says:
    July 28th, 2008 at 2:25 pm

    I was very interested when I heard there was a new rival to Google, as I dont like Google’s quasi monopoly, but unfortunately on trying Cuil it turned out to be not so cool afer all. First is the black page – aaagh! (personal preference you may say but I suspect I am not alone) but second and more important is the speed – Cuil is sloooooooww – a coupe o fminutes to list results that Google does in seconds is not impressive. Third it’s obviously missing information somwhere – I did a search for “someone’s name” translation (I used the person’s name but prefer to not to advertise it here without her permission) – first Cuil said there were no pages whereas Google has 191 results (on the second attempt for some reason Cuil now found 2 results) – in addition Cuil put up a incorrect picture of a person with the same name that I was searching for alongside an article translated by the correct person – so if you are looking for exampe for Roger Federer Tennis – Cuil would be quite capable of putting up a picture of a random Roger Federer alongside an article about ‘the’ Roger Federer just because they have the same name – all very odd.
    So not a rival to Google at all as far as I can see – which is a pity.

Leave a Reply