Forum Moderators: Robert Charlton & goodroi
According to some [slashdot.org] sites [searchbistro.com],
It's one of the best kept secrets of Google. It's a mystery on Webmasterworld. Also in Europe (France) they don't know what to expect from that odd URL [eval.google.com....] Click it and you get ...nothing. The site reveals itself only if you have the proper login and if you use a network known by Google. Residues of Eval.google are found on the web, but the full content of the mystery site has never been published before. Here it is: the real story about Eval.Google. They use... humans!
The site claims it is some kind of the secret google evaluation lab!
Google does different things at certain points in the update cycle.
EFV's website got split and it takes time for that mess to get cleaned up.
Ask g1smd about split sites etc, I think that he probably had his ticker skip a few beats when he saw all those supplemental entries enter the system again.
I know at least two others that are kind of spaced out about that exact situation right now.
Now, I've had enough for one day and am invoking rule 4 early.
Google takes a copy of every site without permission. Are you for or against that copyright infringement?
Google will argue it's a cache and not a copy, I'm not 100% for it but there are ways to prevent it. I don't currently feel the need to do that (yet).
I was actually much less impressed by the archive.org copies, so I asked them to delete it.
Whistle blower in evil company reveals secret evil plans that are copyrighted. For or against?
In general: against it. In specific cases where the evil is truly of general interest and bad enough the evil of publishing might be less than the evil being reported on. Up to the publisher to make the decision, and for him/her to live with the consequences ...
Since bloggers are their own publishers ...
The copyright question is more complicated than your response makes out.
Copyright is simple IMHO: international law, widely accepted except for some rogue countries.
Copyright was also a side issue here, if the reporter rewrote it so that the meaning was the same but the words different it would not be a copyright infringement because it would not be a copy. But Google guy would still be unhappy.
I'm neither a google lover nor hater. They provide me with good stuff for now both as a surfer and as a publisher.
Their happiness leaves me largely unaffected.
There is far worse evil in the world than Google. But it seems popular to do some Google bashing nowadays.
The copyright issue is what makes it stand out as being unprofessional from the reporter (copyright being simple to circumvent as you pointed out).
If you don't like google, feel free to tell us why. Make sure it doesn't become too insulting (slander), fine by me. But almost in a panic trying to grasp everything and doing more evil than the original is doing is pointless to start with. Moreover it is damaging to the community.
Revealing secrets is probably the thing that hurt Google (or any other company) most. If there only were world shocking facts, fine. But they aren't there: Google is just doing what is expected of anybody doing what they do.
Secret from us, perhaps: I would hate it that scrapers, spammers and other scum knew exactly how the Google Algo works. Let's keep them guessing.
Secret from their competitors: Sure, they want to be the best search engine. Don't we all want to be the best?
On the other hand, if they are actually able to influence things without oversight, then there is a real problem with the model
Understanding that the above quote was not assuming that the college students are actually influencing things, I find it surprising that more members aren't commenting on this, or discussing just what the actual connection is between the activity of these college students and the live SERP's.
Human Involvement?
Anyone who's been paying attention and reading the papers knows that human involvement has been a part of the mix when it comes to what gets displayed in the SERP's. No news there.
What is news, or perhaps just somewhat interesting, is that G has been suggesting for some time now that they are moving *away* from human intervention and towards purely algorythmic solutions. On the face of it, the use of (and apparent satisfacton with) the eval solution seems to counter those assertions.
Now, the two notions of a college student eval center and 100% algo-based SERP's can co-exist, if the human actions are purely theoretical and somehow interpreted into SERP sets via the engineers.
But I'm skeptical.
At the very least, the QC people and/or the enginners will need to test the algo tweaks against the SERP sets to determine if the Offensive pages are being purged by the tweaks. If not, there's no point to the process. So it follows that those sites/pages labeled as Offensive will in fact be nixed, and when that happens it does amount to direct effect on sites/pages, symantics aside.
Also, if as GG says, the SERP's can't be tweaked at reasonable expense in order to eliminate the Offensive pages, do the Offensive pages get to remain in the SERP's? I doubt it, but who knows.
College Students?
Sorry if I ruffle some feathers with this, but ... college students? Heck, I was a college student once. Even graduated with honors. But as much as I thought I knew a lot of stuff, what I later learned that I lacked was experience, the grand teacher.
Cavedad used to say that the best business people were 50% experience, 30% common sense and 20% guts.
These college students, as bright as I'm sure some are, have no experience, and little motivation to get things right ($10-$20/hr).
I understand the general notion that there are checks and balances here, but those of us who've been around the block know that the *content* being generated by these college students is potentially subpar, potentially dangerous, and subject not only to being overseen, but overlooked by busy editors, etc. As pointed out somewhere in this or another thread, ODP has gazillions of editors and they can't keep up.
I can't get the vision out of my head of some student in a dorm, doing who knows what while also looking at a computer screen (they multitask you know, and love to have fun!).
I guess the good news is that at least they've got that spamguide pinned to the corkboard above the computer. ;-)
To paraphrase, if these college students are influencing the SERP's in any sort of direct, or directly indirect way, then yes, there's a problem with the model. And yes, that would be news.
EFV shows the way...LOL.I wonder does the guy have a lot of isiders in Gplex or a lot of money gambling in the NYSE at GOOG.
Unfortunately, the answer is "no" on both counts. :-)
Your site is singled out in the eval docs as a "fine" site (congratulations and certainly deserved!), yet this suggests clearly that, at least in your case, the algo was performing in very unintended ways.
I think your experience supports the idea that Google does NOT use the raters actively but also that the algo has some major challenges separating the good from the bad.
EFV - do you think your G traffic problems were caused by "splitting your PR" between www and non www as suggested above? I think you posted elsewhere that you were skeptical of that explanation.
I think it could have been a contributing factor in cases where I slipped a bit in the rankings and the sites around me didn't change much. But I think changes in the algorithm played a bigger role. (For example, the top 10 places for some destination keywords were dominated by template-based pages from massive "deep but shallow" travel networks and e-commerce sites after March 23; that changed with Bourbon.)
I think your experience supports the idea that Google does NOT use the raters actively but also that the algo has some major challenges separating the good from the bad.
I'm inclined to agree. Having human evaluators play Whack-a-Mole with suspected spam would be a doomed cause from the start. It's more likely that Google's software would "learn" from human input (in this case, human spam or quality ratings) in much the same way that an e-mail spam filter tracks a user's input to learn which e-mails are likely to be legit and which are likely to be spam.
I wrote this message a while ago and at the time I didn't know eval.google.com existed. I tried to post it to this forum a few times but it never made it for some reason.
This is the original message:
For probably the past number of months I have constantly been getting emails and posts to my forum which appears to be from the same person or few persons. They ask questions that are either very basic or very specific about the industry that I'm in.
An example would be: ”I need tires what are good tires?” (too general) to ”I have X3000 Super Turbo and want to know if a new Johnson Rod PT-47 with mutate capacity would give me access to the E-56 extensions?” (very specific)
Most people (I deal with anyway) dont ask questions in this fashion. Most users also vary the sentence structure, punctuation, and have misspelled words which is normal but the emails/post Im talking about are different.
They usually come in groups of two or three at a time so its pretty easy to know when this is happening (within 2 or 3 minutes). They sign up to the forum in the same way two or three at a time and post.
The IP Addresses where they originate are usually very different but the email usernames are usually the same layout such as: john_doe_234@domain.com (comcast yahoo hotmail lycos etc...) and the language and style of writing is also the same (from the same person). It again is very easy to know when these communications are occurring.
Im now getting posts to another site I have that are more like pop quizzes than anything else. Im wondering who these people really are and why anyone would waste so much time doing this. For a long time I thought it was from the typical hackers trying to get information but now think that it is something else and I hope that I am wrong.
Has anyone else here had similar problems or noticed such a thing?
END:
I stopped replying to these messages a little bit before getting booted out on Feb 3, 2005. I felt at the time that it was some type of email scam and a complete waste of time.
Perhaps they have links to a website in their forum profile or something and they just send a bot to automatically post questions that will lead to their target site getting links.
PatrickDeese said
I would venture to guess that your forum is being subject to automated spam, and that ths has nothing to do with the Google rater hub.Perhaps they have links to a website in their forum profile or something and they just send a bot to automatically post questions that will lead to their target site getting links.
Thanks for your reply. Yes this is what I thought when the forum messages started but that doesn't explain the emails. Those who have been booted may want to check their emails and responses. I'm becoming more aware of the possibility that these two (eval.google.com and emails) things are connected.
But what do I know :)
So this thread had dropped way back - I'm sure the guys at google were glad about that - but can we start another discussion here on whether we feel, knowing what we do about eval.google.com, that Bourbon is another example of what they are using this for.
Hands up who run's a site that uses affiliate programs and got booted by Bourbon...
Hands up who run's a site that uses affiliate programs and got booted by Bourbon...
There are plenty of sites that use affiliate programs and are doing well in Bourbon.
For that matter, I can think of pure affiliate sites that are doing well in Bourbon (and which were doing well before, during, and after Florida and Allegra, too).
a site of mine came back on bourbon. Of course I have more than just a few affiliate links slapped on each page, but that's how I make my money.
Now why any poster here cares about nda, copywrite of the documents etc is beyond me, it's just more interesting information about how google is ticking today, info that google clearly didn't want to get out, I could care less if it should have gotten out or not, this is as noted hardly an algo code publication or anything, it's just some screen shots and a document that contains little new to anyone here.
Something like this should be of interest to any seo type, who cares where the information came from, gg's posting tone removed any doubt about their authenticity, that was nice of him... nice posts walkman, several others too.
"Google is not god, it is a blind-deaf person with a poor grasp of language and no awareness of culture."
great post ncgimaker, classic
Again, I'm just guessing, but I say that Google would NEVER accept anyone into the Rater program that:Has a site; Had a site; Designed a site; Hosted a site; Is/was employed by a site; Contemplated developing a site; Registered a domain; or Read Webmasterworld (yuk, yuk, yuk)
Note to self: add some photos for the students free hand! :o