| 9:53 pm on May 26, 2005 (gmt 0)|
|...penalty in google for gaining too many links too fast. |
There is definitely not a penalty in "all cases" for getting too many links too fast.
| 1:53 am on May 27, 2005 (gmt 0)|
|Isn't it more likely that Google would say "Oops--that filter didn't work the way we thought it would" and try something else? |
I don't think this will happen anytime soon. For this to happen, Google will need to get back to the drawing board and completely ditch the link evaluating/counting process. It is inaccurate and no longer valid nowadays due to the massive growth in junk.
Someone on this thread (or earlier thread) has mentioned
that maybe google should ignore links age and simply count the number of incoming links to every site regardless of their origin or age. IMO, this wouldn’t work as well. If they do, then the G serps will be directly influenced by Yahoo serps (e.g., if you enjoy top positions on yahoo then your link end up on thousands of yahoo scrapper sites which will directly affect the number of incoming links you have).
As i said earlier, to my opinion google's algo is busted due to it's reliance on determining relevancy and quality via the quality and number of incoming links. This no longer works.
What they should be doing realy is:
1. First determine a site's age, note the date a site was first discovered. This will help to determine who is copying content from who. With the original site getting more "credit points" then the sites copying it’s content (your good old scrappers).
2. Give much more credit to incoming links from quality sources like dmoz and other human review directories.
3. count the number of incoming links to a site on a page by page basis and not domain wide.
4. Google should start their very own human review process. I would have started an online webmasters recruiting site. Recruit around 4000 webmasters and let one thousand of them review 10 sites per day paying something like 50 cents per site reviewed. Rotate the group, hence, let a different group of thousand webmasters review 10 sites per day and then use a different group every week to ensure a democratic process of reviewing and ranking. With googl's almost unlimited monetary resources i don’t see what is the problem of starting this type of human review process ALLREADY, they can use it for their own ranking process as well as sell access to results to other companies.
5. Better design the scarper (keyword stuffed, no content, full of outgoing links) detection mechanism. Penalize an discount all the outgoing links detected on these pages (emphasis on pages, page by page basis).
6. Put more resources into HUMAN review
7. Put more resources into HUMAN review
8. Put more resources into HUMAN review
9. Put more resources into HUMAN review
10. Put more resources into HUMAN review
11. Put more resources into HUMAN review
And last but not least, Put more resources into HUMAN review.
Their current algo is BUSTED and doe’s not work the way intended. It is massively being influenced by Yahoo serps right now.
Smile Yahoo PHDs, smile.....you are dragging google down. How simple that was?
| 2:16 am on May 27, 2005 (gmt 0)|
Did you read the TrustRank thread a while back that talked about "seed sites"?
| 2:46 am on May 27, 2005 (gmt 0)|
|Did you read the TrustRank thread a while back that talked about "seed sites"? |
Nope, can you please post the link.
| 3:03 am on May 27, 2005 (gmt 0)|
Forgot to add another point to my post #113.
Re the human review directory:
Charge something like 10$-20$ from every site who wish to get listed in this directory (and undergo a review process).
This resource/service alone could increase google's bottom line by a few more billions each year (coming directly from review fees).
I don't think anyone will have a problem with paying such fee (per year) to have their site reviewed and properly ranked accordingly.
From google's point of view they will be charging the review fee which doe's not guarantee listing but doe's guarantee ranking according to site quality.
Hell, we are paying for domain name and hosting each year. I have absolutely no problem in paying a review fee each year as long as it is affordable and reasonable (yahoo’s $299 listing fee per year is a joke and way to expensive), I can leave (and I am sure that many others) with $10-20 per year though.
The democratic ranking process in googl’s mantra (we use each link as a vote) can no longer be applied to the web in it’s current form. It is a good idea but it passed it’s due date and is easily abused and manipulated nowadays.
| 3:39 am on May 27, 2005 (gmt 0)|
Here's the TrustRank thread that refers to seed sites:
| 7:34 am on May 27, 2005 (gmt 0)|
I remember a search engine named Alta Vista. It was a fine one and everyone wanted to be listed in it UNTIL...their algo got so messed up they could not tell a "clean" site from a porn site therefore it became a search engine that SEEMED to have only porn sites, no matter what term you searched on...hummmmmmmmm
Maybe a search engine can only get so big before the monster becomes uncontrollable and bites the hand(s) that feed it.
| 8:34 am on May 27, 2005 (gmt 0)|
I used Alta Vista last night for the first time in years (in my continuing quest to find a replacement to Google for my own personal searching) and found the results to be pretty darned good.
The Sponsored Results sometimes fill the area above the fold, but scrolling past them, I liked what I saw.
I guess the AV results are based on Yahoo results (?) but the cleaner, Google-like interface of AV feels better to me than the busier, sometimes graphically intense Yahoo.
I've been using Yahoo alot lately. The only times I still use Google is if I am searching for very fresh info or info I expect to be found very deep on a site. Google still beats the other engines in these two areas, but I have to wonder what I didn't see because it got scrubbed in some filter or other.
| 5:37 pm on May 27, 2005 (gmt 0)|
this topic should be brought back to the top.. and.. why has webmasterworld removed this from the front page teaser list?
| 10:54 pm on May 27, 2005 (gmt 0)|
this topic should be brought back to the top..
| 12:51 am on May 28, 2005 (gmt 0)|
|4. Google should start their very own human review process. I would have started an online webmasters recruiting site. Recruit around 4000 webmasters and let one thousand of them review 10 sites per day paying something like 50 cents per site reviewed. Rotate the group, hence, let a different group of thousand webmasters review 10 sites per day and then use a different group every week to ensure a democratic process of reviewing and ranking. With googl's almost unlimited monetary resources i don’t see what is the problem of starting this type of human review process ALLREADY, they can use it for their own ranking process as well as sell access to results to other companies." |
Max, this is a great idea, and I agree, human interaction is necessary on top of whatever search algo one's using. It's the DMOZ idea with micropayments. I think you should start such a company yourself.
I think some webmasters, mostly indepedents, hobbyists, bloggers and 'net junkies would go for .50 a site, if all it meant was giving a site a 10 page look, and grades for quality content, non-spamming, technical reliability and useable navigation. If sites simply use good site management and SEO techniques - which of course, your new company would promote - the whole search process doesn't have to be very complicated. And guess what? Charging $10 per site would eliminate 90% of the junk and human review would wipe out the rest.
I actually think you could pay the webmasters $1.00 per site. I'm sure nobody would complain about an extra $50 a week. If the demand was strong (and it should be), some guys could work extra and probably would want to.
Not only would this produce a quality experience, but it would put honest people to work (always a good idea).
You might even make a buck or two on the idea.
| 12:56 am on May 28, 2005 (gmt 0)|
Surprised you like his idea... Nobody seemed to like mine here:
| 1:24 am on May 28, 2005 (gmt 0)|
I think if Google were to do this they would use their own employees, keep the program a secret, keep the results confidential, and build the system into their algorithm as though it were just more automation.
That may or may not be better but as a long-time G observer I think that's what they'd do.
| 2:33 am on May 28, 2005 (gmt 0)|
You ever actually bothered looking through DMOZ I mean realy at every link in a certain category. I've been doing this just to see what other sites are like in my own area.
From just over 600 links so far 75 point to holding pages. 6 had been hacked/defaced. And a good 40% have obviously been left on the shelf.
| This 44 message thread spans 2 pages: < < 44 ( 1  ) |