Welcome to WebmasterWorld Guest from 18.104.22.168
Forum Moderators: bakedjake
SAN FRANCISCO — In her two years at Google, Anna Patterson helped design and build some of the pillars of the company’s search engine, including its large index of Web pages and some of the formulas it uses for ranking search results.
Skip to next paragraph
The makers of the Cuil search engine say it should provide better results and show them in a more attractive manner.
Now, along with her husband, Tom Costello, and a few other Google alumni, she is trying to upstage her former employer.
On Monday, their company, Cuil, is unveiling a search engine that they promise will be more comprehensive than Google’s and that they hope will give its users more relevant results.
It looks as if Cuil has chosen about x images for a particular topic and is using those x images to display next to sites.
Exactly. They apparently determine the topic of a query, pull matching images for that query from their database, and splash these image files randomly across the result page next to whatever result there may be (and yes, it could also be your own site).
I found that the location of the images is random, and that they typically use the same set of images for a given query. Sometimes a new image or two comes up on page 2, but beyond page 3 there are rarely new images. They just keep displaying the same images.
It's interesting - you can find your stuff easily once you found one photo. You suddenly get an idea which parts of your sites have been scraped, and which keywords they use. A good starting point is to use the first three words on the title of a page.
To me all of this is a clear violation of copyright. I am currently running an inquiry with an IP lawyer specialized on photography. You see, Cuil's pockets still carry some VC money. Instead of spending it for muffins, they might as well hand over some of that cash to us?!
When images were appearing some of my pages had their own images (sometimes ads from the pages) while others had images from elsewhere. The external images seemed generally relevant(ish) but I was a bit annoyed to have an image of a book that I don't sell showing up.
My logo has an alt tag of "Mysite - Widget Foods"
If I search for Widget Foods any site listed which mentions the text widget foods could show my logo next to it.
I counted 8 on the first 10 pages, and the logo is even displayed all the way down to the last page on 23.
About a year or two ago my site got hammered by cuil. Reading a few posts here it was suggested that the site had a bad bot. I banned that bot - blocked the IP ranges.
I remember trying to ask the site admins (cuil) to stop indexing but the reply was curt. I seem to remember that their site splash screen at the time showed a crusty old farmer ploughing the furrows - like he was saying "Get orrff moi land"
One other point: Is it Cuil or Cuill?
I always thought Cuill but the site redirects to Cuil. If you look at the wayback machine Anna, Russell and Tom describe it as Cuill.
If they can't make up their minds what the site is called how the heck can they decide what to show for images?
[edited by: Frank_Rizzo at 4:27 pm (utc) on July 31, 2008]
If Image does not come from Site A, do not display it next to Site A listing.
The fact that they didn't figure that out LONG before their launch date is, to say the least, worrisome, and does not bode well for any kind of long term presence in the highly competitive world of search engine development...
I "could" also prevent them (the things I cannot change) from happening
I have had Twiceler blocked on all my sites since the day it first turned up, ignored robots.txt and blundered it's aggressive way into a spider trap - I don't care that my sites are not in their index.
So presumably Cuil(l) will not be using MY images to promote other people's websites.
As I understand it they host the images they use - or stole - on their own servers, so if you block their bot now it wil be too late. It stinks, but you can't put the genie back in the bottle.
I suspect that the other funny smell is $33 million going up in smoke.
If Image does not come from Site A, do not display it next to Site A listing.
That's the problem though, the images are not actually from the sites. Cuil ripped those images, placed them on their own server, and are using some kind of categorization to bring them up. That's why you see a lot of the same image over and over again for the wrong sites.
I'm sure they did all this just to save bandwidth, page load times, ineffective images due to hotlink protection, etc. But the resulting mess is not only bad because they aren't showing YOUR pictures next to another site, they are actually showing pictures they STOLE from your site on sites that come up for the search terms.
That seems to me to be even bigger of a crime. At least Google Images references the original URL in the thumbnail (yes I know along with their own).
This is the first viable and brightest search engine launch since Teoma (6+years ago). Damn - it is nice eh? I forgot there could actually be some competition in this space.
Brett? Since there was no ;) or :) after that, we'll assume that you find the results satisfactory. They must have cleaned up the area you were searching in. I've seen that happening over the past few days, lots of stuff seems to be moving about, particularly images.
I do know that a couple of days ago, you could search for specific brands and find their listing with a competitors product image totally not related to the search query.
I am willing to forget all the other bugs in the process with the exception of the images malfunction. That is one of those things that makes you overlook all the other cuil stuff that "may" be going on.
I'm still not backing off. Not until I see that "criminal mugshot" go away that has been eerily moving about the results for a particular name search. I may find myself thinking that someone is playing games with your results based on what I'm seeing.
As a commercial site using those images in a commercial product, that's just asking for a lawsuit.
How do they think they can get away with it?
Look at all the fuss Google had to go through with keywords containing trademarks, and that was just the words. Imagine if it had been images.
Fact is, they made a HUGE screw-up on their launch and it has to be down to VC and marketing department pressure.
I agree that their results are not great (but for Joe Public they are fine as far as they have commented to me).
I completely agree that the image association is the probably the greatest foul-up I've ever seen (and a greater copyright infringement than Google's cache).
I also agree that this has got to be in the "Edsel" arena of marketing screwups.
But if- and it is a BIG if - the average user got to this site - I think they would like it.
Crap results for us - are not always crap results for the surfing public.
Just my opinion (and I'm often wrong) - but then, I was a great supporter of Inktomi!
[edited by: makemetop at 7:46 pm (utc) on July 31, 2008]
This is the first viable and brightest search engine launch since Teoma (6+years ago). Damn - it is nice eh?
I probably miss the part where this launch was "bright". The name? The availability? The layout? The relevance of results? The images?
I understand that anyone (including me) was excited about the appearance of a Google competitor, but the launch was, er, far from perfect. They could have done sooo much better! Sad.
That is the next phase of the marketing campaign. Remember, this is a "negative press campaign"
Nah. Sure, it may drive traffic. But it is expensive traffic. The damages for infringing copyright protected works is up to $150,000 per infringement. I found 32 images so far (and I did not look closely), hmmmmm, that's up to $4,800,000 in damages. Now, what was that "negative press campaign" again...? ;-)
I see what everyone is seeing, but it is a baby search engine learning to crawl. Obviously, they have some work to do. Bitching and moaning about it, is not going to change that. They have little traffic at this point and we shouldn't throw out the baby with the bath water. I don't care what they are showing right now. I haven't done more than a hand full of searches and probably won't give them another look for a few months - I'm going to give them some time to work it out.
Although some of my larger sites do quite well for some terms (and I've even received a little traffic from Cuil), the vast majority of searches for more obscure terms are simply full of spam. And the result count, even for one page, is always wrong.
Even on a normal browser like Firefox wheel scrolling doesn't work if the cursor is in their fixed bars at top and bottom, making the user wonder why the wheel broke.
These interface problems though would be unimportant if the search actually worked, but the results are simply unusable for anything other than major keywords (which are maybe hand tuned). Plus, the snippets shown are months old.
Pity, a Google rival is really needed.
... it is a baby search engine learning to crawl. Obiously, they have some work to do. Bitching and moaning about it, is not going to change that.
And as regards to their layout/format -- they have every right to present that any way they want. All responses here or anywhere else -- positive or negative -- would be subjective, not objective.
But as to the complaints, much of what I'm reading are simply observations/feedback to an over-the-top PR campaign. Someone should have told them:
"Be careful what you say -- you may be held to your own pronouncements."
much of what I'm reading are simply observations/feedback to an over-the-top PR campaign
That's the part that got me. You expect some hype but the level of outright BS and unfounded arrogance was just too high.
Follow that with, possibily, the biggest collection of mistakes at launch in the history of search and it's no surprise that people don't want to cut them any slack.
Yeah we all want to see more competition but it seems unlikely that Cuil is going to fill that role any time soon.
it is a baby search engine learning to crawl.
more like a purportedly fully-fledged search engine, that hopped out of the nest, roared "Hello world, I'm huge and I'm gonna soar!", then plummetted to the ground, joining the remains of startups that never really flew.
Obviously, they have some work to do.- same could be said of all alternative search engines, indeed any company that has less than prime or even lousy product/service. Doesn't mean the work'll get done.
A bird emerges from the nest, no one expects it to swoop n soar n do barrel rolls, but if you can't flap as far as the next tree, start being able to feed yourself, you won't be around long.
Not sure if folk overly patronising re Joe Public; if so incapable of telling when searches decent, so-so or pap, I don't think google would have become so dominant. Plus, cuil not in zone of being so-so, so need to look hard and be smarty pants web expert to know things could be better; it's terrible.
They take the premise that any new search engine will eventually have to reverse engineer Google and conclude
Web spammers are looking for an audience. As the most popular search engine, Google controls what its audience sees, so the junk jockeys generate their pages in ways that game the system. Everyone else legitimately seeking those same eyeballs for their content or customers for their business want the better search rankings too, so the SEO crowd works to make legit sites dance to Google's tune. Techniques such as adding needless internal links, creating PageRank-friendly URLs and distorting normal grammar are all widely deployed with varying degrees of dastardliness.
And so it goes. Gradually the structure and content of the web becomes at one with the Google data centre. Disrupting such a tight, interconnected mutualism seems impossible for would-be "Google Killers". The best others can hope for is to imitate Google results.