>And if you wan't to suggest anything to the FAST team,
>or you'd like to report spam, then you can do that at ...
Yep, and it(they) work(s) like a charm! Quick response from the tech team and quick removal if the spam is really obvious and disturbing the results!
BTW: you're right - the results look really good, much better than before. Even my page climbed from #90 to #37 after 1 year :) Fast also returns latest news for the search on top of the results ... is this new, also?
>>... is this new, also?
Nope have been there for a long time, they had it before Google had it too.
I also think there are still nonsense and fishy results but 40% garbage compared to 70% is a big improvement ... could be called "better than before". ;)
And the improvement is obvious ... for the searches i did.
I'm not sure for what terms you are looking for, but when it comes to very competitive search terms (top 1000), i see a clear improvement.
(- all adult/casino business, don't watch them)
Ok, just checked a few more searches and it seems that the index didn't get improved but the algo. For one 2 word phrase they list at #1, #2 and #3 pages from a formerly well known german experts portal that is dead since ages (last year). So, i guess, they tweaked their algo but didn't recrawl the index since a long time. That's why i love google! :)
had a good look at the SERPs, do agree with you LZ, the results, are alot better than i have seen before.
Sites that i didn't even know existed are showing up as well, before G i hasten to add.
I will probably get shot down in flames for this, but i think they are even better than G's, as the first thing that becomes immediatly poingant is the a near opposite weighting on linking, per se, ATW is letting the site do the talking rather than its inbound links, which i personally believe is where its main comp, on the SE front comes completely unstuck.
|Also seems like the focus on the domain have gone down which is good in most cases. |
... now that, I don't see. I agree that results have been getting steadily better for a while, but still far too many (peeve mode) hyphen-junk domains in there, as well as assorted totally off-beam results. If I want France then I'll search for France, but I searhced on Italy.
Better, but I don't think G need to be looking over their shoulder just yet.
I'm liking what I see. The NFP site I service is coming up very well under a variety of word combinations ... still some work to do in this area though. Some combinations I think could be used simply aren't functioning as well as I hoped, but maybe that's a lack of visits since last updates.
But most importantly, there is not a trail sign one of the not-so-nice person(s) out of a far southestern state (think butterfly ballots if you're real curious) who/which had glommed up information from my NFP's site and shackled it to a couple of sites that had nothing whatsoever to do with the mission and goals of either the local council or the national organization. It's NICE to not see junk sites with a description that matched our description at the 99.99999-100% level.
>I will probably get shot down in flames for this, but i think they are even better than G's, as the first thing that becomes immediatly poingant is the a near opposite weighting on linking, per se, ATW is letting the site do the talking rather than its inbound links, which i personally believe is where its main comp, on the SE front comes completely unstuck.
Yep. ATW definitely is running the algo heavily based using on page text. ATW is definitely giving better results for certain kinds of searches.
steveb, when it comes to an algo ATW's tends to do better when dealing with non-commercial searches. Remember, the vast majority of searches are non-commercial. I almost never do commercial searches. Non-commercial sites don't tend to try and play SEO games much. ATW seems to be sacrificing commercial SERPs in favor of non-commercial ones. Which for me is a good thing.
Looks good to me as well and something is in the works at FAST for sure.
Brett just reported back [webmasterworld.com] from SES and revealed a few nuggets from FAST's Tim Mayer. Could be some of that we're already seeing?
"ATW's tends to do better when dealing with non-commercial searches"
Then I'm glad I never search for commercial stuff there because their non-commercial results are pitiful for subject areas I am intimiately familiar with. Their results are much worse than AV or Ink, and mentioning them in the same breath with Google is like mentioning a Porsche with a Model T.
FAST seems to have no way to understand the value of words. Many of the sites I see get excellent search positions are just laundry lists of words. Commercial sites that sell games, for example, might have a long list of game names (same with bookstores). These sites manage to get high ranking for those game terms, even though the content for each specific game is miniscule.
FAST is not valuing content. It is valuing words, and it appears to have no way to interpret those words, neither by linking or anything else. I can understand that they are trying to craft results that benefit commercial sites, but doing so is pure anti-user.
If people who type in widgets get results that favor a generic store that sells widgets as one of a hundred things instead of a 1000 page site dedicated to the history, explaination and understanding of widgets of every kind, that is an engine that won't get many users and deservedly so. FAST touts the pages in its index. Until it figures out what to do with them and present even Ink or AV quality results, they will continue to be a footnote... which is too bad for webmasters since we should all be wanting more than one search engine that delivers relevancy.
May have a lot to do with the specific search. Note lazerzubb started this thread with "and i was must say i was very positive about the results that were presented to me, it's certainly the best i've seen, since yes i can't remember." What I see on the searches I do with ATW look quite good to me. May have to do with the sorts of topics we search on.
"May have to do with the sorts of topics we search on."
Sure, but look at the other thread running now about FAST. A site gets a #1 ranking without *any* evidence of quality content -- no human directories, no "votes" via linking, and no hint at all of "quality votes" from links from other solid content sites on the same topic (aka, themes).
It's true that all these things can be manipulated, but just ignoring them and handing out a rank based only on words without content, that is just really bad.
Random results will lead to some searches being fine and some hideous and most somewhere in between. FAST does best, great even, for complicated multi-word searches and obscure searches because it has a large index. It does very poorly for one-word searches or common phrase searches because it has no way to sort the results content-relevantly.
>FAST is not valuing content. It is valuing words, and it appears to have no way to interpret those words, neither by linking or anything else.
Steve, of course onpage factors are, just like at Google, an important part of the equation.
Assuming Fast does not factor in links, does not factor in overall topics anyhow is way off. There's a lot more than just onpage factors.
From my end I can say I find ATW and Google usually on par with results.
I use both for daily searches, both informational and commercial.
In fact there are travel related terms where Fast gives me better results on top.
The only thing where I see Fast lagging behind is updates. I really think a more frequent overall update cycle would put them in a head to head race with Google.
>>In fact there are travel related terms where Fast gives me better results on top.
That's one of the topics i also checked.
Much Better :)
All mights sites that were dropped to about page 50 for no reason are back up to #1
Smiling all the way to bank :) :) :).
What clever thing can I tell my clients I've down ;)
Just tell them to look at Fast, and don't say a word.
What is mystical in their minds is money for you.
Looks a little bit more like my Google listings now.
I guess the way of checking these updated differences is comparing Lycos with Alltheweb.
One thing that occurs to me is that Fast seems to put more emphasis on proximity in title and less emphasis on anchortext, compared to Google.
(just got a heavy crawl yesterday as well)
i do believe FAST is *very* close to Google at this point, a matter of 3 to 6 months.
my opinion is FAST is about close to where Google was july this year in algorithm terms, when Google gave (in my opinion) too much weight to anchor text and domain keyword stuffing. the first often led to googlebombing and with cheap domains the second led to spammy results all over.
then google starting giving more value to singleworded www host/domains to fight subdomain spamming, this around august/september 2002.
IMHO FAST needs to fight the subdomain spam, award less value to internal linking and keep using link popularity as a strong measure of a page's relevance, which is obviously what caused their huge improvement this year.
>From my end I can say I find ATW and Google usually on par with results.
I'm actually seeing some better SERPs on FAST for non-commercial SERPs than Google. By no means is FAST cleaning Google's clock or anything. However, FAST definitely is in the major leagues when it comes to search engines.
A site that had slipped out altogether is now back in, and doing much stronger than it was before. Amazing how it's doing for the secondary keyword phrase, it was buried for it before (the homepage is optimized for two phrases) - with very little text on the page, it's very graphics intensive. It's doing several notches above Google for the secondary phrase - can't quite figure out why. Identical position for the main one.
Others are about the same, not unlike the Google rankings at all. All in all this update did beautifully for me.
Comparing Fast and Google is like comparing apples and oranges since they run different algorithms. They're both good IMO, and I use them both.
I also want to call attention of all webmasters who have had a site permanently trashed by a search engine to take a look at Fast's spam policy:
Of particular interest is that they aren't threatening to remove an entire site for a spam violation. It appears they say they just remove or downgrade the offending page's ranking. Anyone who has had two years worth of work trashed by an inconsiderate search engine will appreciate if Fast is spending their energies developing on-page spam filters that work rather than callously blacklisting the sites that may not have realized they were breaking the rules.
It probably takes more effort in the short run to develop effective on-page filters than to dump entire sites, and spam may find it's way in in the meantime, but in the long run they should have a better search engine. Rather than scare site owners of condemned sites into playing by the rules, the act of getting a whole site permanently banned probably steers more than one site owner toward more advanced and search engine specific spamming and sends them underground with their advanced techniques. All one has to do is to look at Google's high public profile in acting out against spammers, and coming down on them heavy handed, to realize they are being hurt by the advanced optimizers they have unwittingly cultivated by trashing the ranks of legitimate web sites.
Maybe I'm wrong in interpreting how Fast deals with spam, but I haven't heard people reporting that their whole site has been trashed by Fast.
Maybe it's fitting to note that several hundred years ago the Vikings left Scandanavia and Norway to conquer much of the civilized world. And conquer they did by being adaptable. Then they spread their genes all over the world by assimilating into the society of the peoples they conquered, rather than trying to dominate them. Maybe the builders of Fast know something the Wild West gunslingers of the US engines haven't yet figured out.
Certainly an improvement in a dozen searches I just tried. One of my old standby tests is to put in all the other search engines names and see what comes up in the results.
Press release [fastsearch.com]
I thought the word.doc feature was always included?
It's all academic. They provide little traffic of any value.
nell, I will have to agree. After seeing this thread, I looked at fast SERPs. For a two word competitive phrase with 6 million+ results, our site is ranked on page1 .. Yet hardly any traffic ...
I dunno, they are not on par with Google IMHO for a few select searches I generally use to test out an engine. (Just stuff I'm familiar with).
It's full of doorways, keyword-keyword-keyord domains and titles and that seem to be driving the rankings. It could just be appearance though since Googkle clips relevant text from the copy of the page, but I'd put em' on par with Ink.
Not bad, but not in any kinda shape to give Google a run for the money. I dn't expect to find forms in the top 30 results, it should be content.
Fast still has many problems. Here is an example: search for "some keywords" . After the first page of results all subsequent serps ( for as deep as I cared to go) list one site which appears to be a forum of some sort.
[edited by: heini at 3:29 pm (utc) on Dec. 19, 2002]
[edit reason] no specific search terms , see TOS please ¦ thanks! [/edit]
| This 48 message thread spans 2 pages: 48 (  2 ) > > |