| 9:01 am on Jul 29, 2008 (gmt 0)|
|and QUICKLY learn from the data they're collecting |
Except, they do not collect the data. As per their privacy statement.
Hopefully they are reading the feedback here and elsewhere. That's some valuable data as well. ;-)
| 9:11 am on Jul 29, 2008 (gmt 0)|
|a sneaky HCI test. I made a rubbish system on purpose to test user behaviour and reactions once. |
Yep. But their name will be associated with "failure" and "missed opportunity" for the forseeable future. Which is not good when you want to build a business and attract advertisers or position yourself for acquisition.
I agree. IF the main rationale is not to build a business, but to do research, or to prevent future search startup launches that use a PR blitz, THEN cuil is indeed a cuil tool.
P.S.: The REAL question is - WHY DID THEY LAUNCH SUCH A CRAPPY PRODUCT?
<OT>Which brings up this memorable quote from my favorite Bond movie, Tomorrow Never Dies. Elliot Carver: "When I was sixteen, I went to work for a newspaper in Hong Kong. It was a rag, but the editor taught me one important lesson. The key to a great story is not who, or what, or when, but why."</OT>
[edited by: zett at 9:12 am (utc) on July 29, 2008]
| 9:11 am on Jul 29, 2008 (gmt 0)|
I think we would all love a serious competitor to Google but this launch really seems to have gone off at half cock.
Tried a search on the name of a London based classical musician. On G's UK searches one of my photos of her comes up on page 1 and the rest of page 1 is her own site and a number of reviews. My page (with the wrong picture!) comes up when I search on the event where I took the photo.
So a search on Jane Doe came up with three pages of various male Does in the USA at which point I gave up. I could understand an American Jane Doe pushing the results down but not the others.
| 9:34 am on Jul 29, 2008 (gmt 0)|
I'm getting a few visits from Cuil in my logs with a high pageview... more than I actually get from MSN/Live in three years. There's got to be some value in that.
| 10:03 am on Jul 29, 2008 (gmt 0)|
|But their name will be associated with "failure" and "missed opportunity" for the foreseeable future. |
If a week is a long time in politics, ten minutes is the foreseeable future on the Internet.
If they start producing consistently good results, their launch will soon be forgotten.
Heck, if Ask started producing good results, years of jokedom would be forgotten overnight.
| 10:15 am on Jul 29, 2008 (gmt 0)|
In all fairness, and in a right spirit of generosity and genuine concern, I think maybe we should look at this initial launch as nothing more than an alpha release (which technically, is what it really is), and see what we can collectively do to give positive, constructive input toward development of what could potentially be a viable contender in the search space.
| 10:33 am on Jul 29, 2008 (gmt 0)|
Agreed Marcia, but usually it's in the interst of the company to say it's an alpha or beta.
| 11:05 am on Jul 29, 2008 (gmt 0)|
>I can only imagine the long list of cuil jokes
> being passed around the water cooler at the plex.
You dig around here from the early days of Google? There were alot of jokes we passed around about them too. I bet the guys at Altavista were laughing as well. Mr Louis himself tells a story about Sergey and Larry pitching the whacked out idea of 'page rank' to him and he asked them to leave his office because it would never work.
Anyone remember Sergey's pet project, "Search Scout"?
Give the folks at Cuil some time. They have the money, the staff, and the tech to get this right.
| 11:27 am on Jul 29, 2008 (gmt 0)|
|They have the money, the staff, and the tech to get this right. |
Yeah, but given this near-perfect setup and PR it's even more surprising to see them "launch" with those very obvious and very basic problems.
From all the comments in this thread it seems that:
- the team did not do a usability test
- the team did not do a basic functionality test
- the team did not do an extended functionality test
- the team did not do a limited alpha/beta test
- the team did not plan for the launch day
They released a promising product and got their PR blitz. And they failed to fulfil their promises.
How comes, that money, staff, and tech did not prevent that release?
| 11:30 am on Jul 29, 2008 (gmt 0)|
Disastrous brand name. No-one knows how to pronounce it, no-one knows how to spell it. I thought it was pronounced "quill", but however I pronounced it to a friend, I would then have to spell it out (and, presumably, some will mis-remember the spelling).
Why pull up the curtain at this premature stage without at least attempting to manage people's expectations of how amateurish the results currently come across? (Both in terms of on-screen appearance and in terms of relevancy).
What in heaven's name is going on with the mixing and matching of competitor images and snippets? How is it even possible to get that wrong?
If they do nothing else they should change the name to something that is easy to know how to pronounce (without being instructed) and easy to remember how to spell. Not smart.
| 11:49 am on Jul 29, 2008 (gmt 0)|
You are right Brett, but things are very different then 1999.
I would love nothing more then a serious competitor to Google considering myself and everyone that depends on me lives and dies by them right now.
Sadly in todays online environment with markets entrenched as they are and Googles deep pockets mixed with a reasonable amount of savvy it will take something professional and groundbreaking to take away any market share from them and I have seen neither thus far.
| 11:58 am on Jul 29, 2008 (gmt 0)|
And then, everybody and his dog are labeling web products that are still in development as "beta" (even if the products are quite mature, really). Yet I do not find this label anwhere on cuil. How uncuil.
Some speculate that the team had spent too much time on preparations (research?) and that the VCs actually demanded the launch. They wanted to see action for their money. That is one of the very few plausible explanations for the cuil disaster. (But this does not explain why, for example, the "about" page was not present.)
| 12:16 pm on Jul 29, 2008 (gmt 0)|
What the launch of Cuil does show, however, is how quickly people would be willing to abandon Google if a better search engine came along.
As much as Google is laughing it up right now at the early failures of Cuil, it probably woke them up some and they'll need to keep putting out a superior product to keep their non-loyal userbase.
| 12:22 pm on Jul 29, 2008 (gmt 0)|
Google's success didn't happen overnight, in a few weeks, months, or more... it took years and years for Google to become a byword for searching for something.
Now all of those who thought someone else could do it in a day raise your hands and get laughed at... because from the way things look to me, Cuil could really be a serious contender.
(cool.com registrant sits back and waits for the approach, while his bank manager gives him a shoulder rub)
| 12:23 pm on Jul 29, 2008 (gmt 0)|
Keep in mind that the webmaster community isn't representative of the general public.
| 12:26 pm on Jul 29, 2008 (gmt 0)|
I am not impressed. I think it just another fauiled attempt to take over G. If anything, I think they may be onto something with the results layout, but I think that even needs to be redesigned. In any case G is most likely looking on and saying "well there is something new we can try next month!" Thanks Ciul...Cuul...Cuiool?...CRUD!
| 12:28 pm on Jul 29, 2008 (gmt 0)|
a : That name is a pathetic choice
b : design, concept and search results can not impress at this stage
Good luck guys! All the best wishes are with anyone breaking the Google search monopoly.
| 12:31 pm on Jul 29, 2008 (gmt 0)|
|Google's success didn't happen overnight, in a few weeks, months, or more... it took years and years for Google |
Yes, that's true.
But new services do not have years to develop. Today, a service that launches, be it a new search engine or anything else that competes with an existing service, has to be at least "good" right off the start. You can not approach the market in a 1990s fashion, saying "well, let's see what happens". Doesn't work.
And it does even work less when you consider the bold statements ("we're the biggest", "we're a Google killer") as well as the backgrounds ("we're ex-Googlers"). That sheds some light on Google as well (unless these guys were fired for incompentence).
Sure, they had their 24 hours of fame. Now they can go back to the lab and do usability testing, functionality testing, analyze the left-over traffic (once the topic cuils off) and try to fix their product. Will they get a second PR blitz? No. (Imagine: "Cuil now launches, for real, says :'We're the biggest!'" - Nah.)
| 12:45 pm on Jul 29, 2008 (gmt 0)|
How about opening a separate thread that focuses on technical-only observations?
Yeah, well, basically the alpha/beta testing feedback.
I'm still not sure if they weren't aiming for this right from the start, but...
none the less it seems to me there are a lot of people here who wouldn't mind if they were taken advantage of ( free testing, consultation )
...if that meant an extra percent in the chances for cuil.
| 12:57 pm on Jul 29, 2008 (gmt 0)|
You heard of testing to destruction?
It just happened.
I'll bet one dollar that they relaunch in a few months time with a new name, and better functionality - they could probably buy "mahalo.com" for $16.00
And I'll bet a second dollar that they own the right domain name next time around.
(all proceeds to charity. total payout = $2.00)
[edited by: Quadrille at 1:00 pm (utc) on July 29, 2008]
| 1:07 pm on Jul 29, 2008 (gmt 0)|
|I'll bet one dollar that they relaunch in a few months time with a new name, and better functionality. |
I'll bet a second dollar that they own the right domain name next time around.
One problem, though: they (probably) won't be able to create another PR blitz. Thus they have to rely mostly on organic visitors (plus a bit of marketing). Given the current shape of their search results, I do not see how they will be able to keep many people. So they start off almost at zero, and this will be a tough ride.
Imagine how cuil it would have been IF their PR stunt had actually worked: the world hears about Cuil, tries the product, finds it OK, and switches from Google to Cuil. But the "finds it OK" part was a prerequisite for this. (Right now webmasters find the product uncuil, and thus I see no chance for organic recommendations.)
But yep, it's possible that we're all wrong and they have a come-back after many moons.
| 1:13 pm on Jul 29, 2008 (gmt 0)|
Its comically bad, site with relevant content don't show up at all in the results, but link sites do.
Lots of duplicated content, results, and out of date results. The images are mostly randomly chosen from god knows where. Search results seem skewed towards scraper sites, link sites and in one case an email list archive.
Do a search for something like widget tutorials, it doesn't even make sure sites listed on the front page contain both words !
In some cases the results were completely devoid of either word.
Anyone else notice the odd result with "& & & & & & & & & &" as the description ?
An alternative to Google and Yahoo is sorely needed, and with a junk/spam filter applied and some bug fixes this could be it.
[edited by: fabricator at 1:18 pm (utc) on July 29, 2008]
| 1:17 pm on Jul 29, 2008 (gmt 0)|
|Keep in mind that the webmaster community isn't representative of the general public. |
Yeah, we are a little harsh :)
I talked to a couple of friends about cuil last night and they said the following things: (they are non-webmasters but saw the site yesterday)
1. "How do you say it?"
2. "Why didn't they just get cool.com? Is that a site already?"
3. "I don't like the black start page."
4. "I don't know which site to click on. A list is one thing, but they [the sites] are all jumbled together and I don't know which one will be the best [result]"
| 1:27 pm on Jul 29, 2008 (gmt 0)|
Well for my own keywords rather than referencing my own site it is showing my site within a link farm set of pages - cant say I am inpressed.
And how are they going to match googles infrastructure - surely the barrier to entry has now been rainsed pretty high in terms of start up gear.
| 1:29 pm on Jul 29, 2008 (gmt 0)|
|When you search with Cuil, we do not collect any personally identifiable information, period. We have no idea who sends queries: not by name, not by IP address, and not by cookies (more on this later). Your search history is your business, not ours." |
However I believe it won't take long until they get trouble about displaying images from third party websites besides search results.
[edited by: jecasc at 1:36 pm (utc) on July 29, 2008]
| 1:29 pm on Jul 29, 2008 (gmt 0)|
|I talked to a couple of friends about cuil last night |
Four excellent points that a college-level meta-SE-clone designer would have spotted.
[edited by: Quadrille at 1:30 pm (utc) on July 29, 2008]
| 1:33 pm on Jul 29, 2008 (gmt 0)|
The quantity of bugs is unacceptable even by today's standards for Alpha and Beta releases. If part of your interface is the association of an image with the result, then it should be right more than 50% of the time!
Great to have competition, but sad to see software about to be released which is so fundamentally flawed.
| 1:34 pm on Jul 29, 2008 (gmt 0)|
|Google's success didn't happen overnight |
A valid point generally, but it happened overnight for me and everyone I know.
When Google launched it was completely different to all the incumbents and obviously superior in many ways, so much so that after using it once people tended to stay with it - it was fast, efficient, usable and pretty accurate in 1998, and none of those things have changed.
Cuil(l) has had a disastrous launch, and you only need to look at the current story in The Register to see how much worse their handling of associated images is than anything mentioned here so far. They had the benefit of a publicity coup, but it is only the fact that they had a few ex-Googlers on board that made it newsworthy, and yesterday may well be the busiest day they ever have.
Competition for Google would undoubtedly be a good thing, but right now Cuil(l) is not it.
But then neither are Yahoo, MSN, Ask or anyone else.
|the webmaster community isn't representative of the general public |
True - we have actually heard of Cuil(l) and bothered to try it.
|Give the folks at Cuil some time |
They can have all the time they want.
The day I hear of the general public using it I will take it seriously.
| 1:34 pm on Jul 29, 2008 (gmt 0)|
"When you search with Cuil, we do not collect any
personally identifiable information, period."
In other words,
"When you search with Cuil, we do collect
non-personally identifiable information, so there."
Which suggests that
(a) their search may improve as they learn and
(b) they know the power of weasel words.
| 1:36 pm on Jul 29, 2008 (gmt 0)|
Some guys working in contextual search (academics) volunteered this example:
"The thing that differentiates contextual search from free-text search is that contextual search should be able to make use of known relationships between contexts. For example, a user searching for components with both “Automotive” and “North America” contexts should receive results that include business components tagged with “Automotive Engines” and “Detroit” (but which do not have “Automotive” nor “North America” as direct contexts). Equally, if the contexts “United States of America”, “USA”, and “America” all refer to the same country, then a search for any one of those contexts should return results from all three equivalent contexts."
I don't see this in Cuil. Any thoughts?
(contextual search gets used for email a lot)
| 1:37 pm on Jul 29, 2008 (gmt 0)|
|I don't know which site to click on. A list is one thing, but they are all jumbled together and I don't know which one will be the best |
This is the main problem I had with it, I don't believe search engine results should be displayed like a magazine or classifieds newspaper section. It should be easily determined which sites are the most relevant.
Changing an algorithm or technique is okay, but reinventing the wheel (especially one that has been tested to death on user behavior) did not need to happen, IMHO.