| 11:09 am on Dec 15, 2003 (gmt 0)|
Read the whole statement, not just one phrase.
Don't take out of context a perfectly clear statement. She said webmasters need to be careful who they link to and who links to you. Complete sentence/thought. It's a clear statement about linking to and from.
Some folks seem to want to make the most improbable and farfetched interpretation of a portion of a straightforward couple of sentences.
Be careful about linking to people who spam. That's the deal. In particular be careful about linking to people who spam who link to you.
Notice this word... "and"
Notice she did not say... "or"
[edited by: steveb at 11:13 am (utc) on Dec. 15, 2003]
| 11:12 am on Dec 15, 2003 (gmt 0)|
|go back and look at who you linked to and who’s linking to you. |
Looking at this logically both of these could have a negative effect now compared with pre Florida like this.
Who you link to guilt by association.
Who links to you. If you benefitted from PR by virtue of links from sites that are now filtered then you will not bennefit from it now.
In effect if you were previously benefiting from some ellement of false PR given to you bay sites that have now been filtered for whatever reason then you will have lost that ellement.
However I still think that this is a smoke screen because the over riding problem is what Google has aimed at American English commercial terms, either deliberately or as an unexpected result of algo/filter changes. This Google spokesperson is trying to shift the blame to spam and ourselves, for being so stupid, when in fact it is what they have changed that has caused the disaster.
| 11:15 am on Dec 15, 2003 (gmt 0)|
It's a dramatic improvement across *most* of the Internet, not a disaster.
| 11:34 am on Dec 15, 2003 (gmt 0)|
|It's a dramatic improvement across *most* of the Internet, not a disaster |
This implies that the results pre-florida were poor. If they were not, a dramatic improvement would be impossible.
Whilst I have seen several areas discussed (real estate, chip-no searches) where results have worsened I have not seen anyone saying that in their field the results are much improved.
If the improvement is dramatic, give us some idea of what you're searching for, what the results were like before and what they are like now. Others have backed their assertions, perhaps you should do the same.
| 11:39 am on Dec 15, 2003 (gmt 0)|
If being linked to by a "spam" site is a problem could someone define for me what a "spam" site is - I would love to see an example
| 12:09 pm on Dec 15, 2003 (gmt 0)|
the statement is absolutely clear - there's nothing ambiguous about it. Use of the word 'or' might have actually excluded the possibility of a disbenefit arising from who links to you. But the logical connective 'and' actually indicates that both are potentially harmful.
Anyway, I doubt its accuracy - and I've already posted as such (I asked for the full text / link to be stickied to me yesterday, and haven't received anything. Draw your own conclusions.)
| 12:19 pm on Dec 15, 2003 (gmt 0)|
|It's a dramatic improvement across *most* of the Internet, not a disaster. |
I wonder if you would be thinking "It's a dramatic improvement" if you had not been unaffected by what has gone on.
For those of us that have been affected it is a small personal disaster that leaves us a tad bewildered. Generalised opinions about intangible improvements will not really help until someone finds a way for sites that shouldn't have been dropped in the first place to get back.
In your position I'd probably sit on the high moral ground looking down at the innocent victims as they writhe in agony.
A bit florid but you know what I'm saying.
In the long term I'm sure that the cream will once again rise to the top, at the moment all I can see is turds floating on the surface that just won't flush away.
| 12:21 pm on Dec 15, 2003 (gmt 0)|
You certainly have a way with words Sid ;)
Sorry mods: mon chat finis
And no - my cat is fine :)
[edited by: superscript at 12:22 pm (utc) on Dec. 15, 2003]
| 12:21 pm on Dec 15, 2003 (gmt 0)|
|If being linked to by a "spam" site is a problem could someone define for me what a "spam" site is - I would love to see an example |
Google sets the rules, changes them and then possibly breaks them.
I would try and ask Google and if you get a sensible reply. Share it with others here.
| 12:26 pm on Dec 15, 2003 (gmt 0)|
I was going to sticky you a blatant example: index / doorway page with a black background with 500+ search terms in black text hidden on it, and a dozen links to his own sites, also in black and hidden.
But when I checked - he's cleaned it up, and the background is now white, with just a few search terms on it and the links for all to see.
It's probably not the reason he got bombed by Florida though, but I guess he's feeling worried and trying everything he can.
His problem was that he put all his efforts into spamming rather than content. My index was bombed too, but I still get thousands of hits (actually they have doubled overall) on my inner pages because they contain a lot of info.
This is not to say I am not hacked off with the treatment of my index page though - which was totally legit.
| 12:31 pm on Dec 15, 2003 (gmt 0)|
Sorry superscript, your reading comprehension is leaves a bit to be desired.
Separating two parts from a whole makes no sense at all, especially since the obvious reading makes perfect sense and is what we have known a long time (though the degree may have changed) while the bizarre reading leads to a preposterous conclusion which has been contradicted by Google (and Google Guy too) many times. It's too silly to even be wasting discussion on, if only because it is so easy to prove. Make some crap hidden text duplicate junk and link to your competitors. The proof of the preposterousness of the bizarre interpretation is even in that no one is running out doing just that.
Occam's razor would certainly be invented after reading some of these threads if it didn't already exist.
kaled, dramatic doesn't presume poor previously. That is just another extremist, black and white, go overboard type thinking too many people have gotten into. But prior results were not nearly as good. They were based almost entirely on volume of anchor text which has almost no value at all in terms of judging quality. And, it is simply the easiest thing possible to manipulate.... Just make more pages.
At its root even, the current algo is an attempt at being an algorithm. The anchor text algo was a symbol of failure by Google, even if the results often were just fine. They were going to hell in a handbasket though in competitive areas. Webmasters in less competitive areas, who don't see thirty duplicate domains in a row in the serps, simply don't have an appreciation of the problem.
And no I'm not going to violate this board's charter just because others seem to think it is okay. Lots of people have posted about how the results have improved, but they get drowned out in the volume of new members posting the sky is falling.
| 12:35 pm on Dec 15, 2003 (gmt 0)|
|At its root even, the current algo is an attempt at being an algorithm |
Now my reading comprehension is indeed failing me.
| 12:41 pm on Dec 15, 2003 (gmt 0)|
"I wonder if you would be thinking "It's a dramatic improvement" if you had not been unaffected by what has gone on."
The problem continues to be people who were adversely affected fixating on their own sites. Get some perspective. I've had plenty of ups and down with the new algo (my number one keyword page is out to lunch right now waiting for the next big shakeup to move up to where it belongs), but it is plainly clear that the results are better, and the algorithm is no longer childishly easy to "crack".
Certain types of crap are gone. Some crap, particularly redirects, remains. Trivial content sites that ranked because of their anchor text have been blown out of the water. More authoritative sites, sometimes off-topic, have taken their place. Valuable information is far easier to find today than previously. Some sites that shouldn't have been lost have been, but that should get better with the next influx of new data which should come any hour now.
To repeat though, the main thing is that Google no longer has that algorithm a child could exploit. What they have certainly isn't perfect, but it also certainly is not instantly "crackable". That is a tremendously positive step in itself.
| 12:42 pm on Dec 15, 2003 (gmt 0)|
"Now my reading comprehension is indeed failing me."
The old algorithm was mostly just anchor text. If nothing else, this one involves a TON of factors.
| 12:51 pm on Dec 15, 2003 (gmt 0)|
In defence of the 'whingers' - if you had read my posts you would have seen that my sales are fine, hits have actually doubled over all, and things aren't looking too bad my end. I am also one of the 'new members' you refer to, apparently disparagingly, at the end of a recent post.
The 'sky isn't falling in' for me, and don't mistake a new member on WW for a new member to the Internet community, or someone new to SEO or e-commerce. I've been doing this for many years.
In short, this is no place for falling out. But I'm doing just fine Jack - and I'm a new member - but I still think the commercial SERPs are dreadful. I'm looking for solutions; not disparaging remarks from my 'seniors' regarding my reading comprehension etc.
Now let's draw a line under it and look for some answers to the strange commercial SERPs. I personally think it is an 'intelligent' filter that has gone awry. Any comments? (of the non-personal kind!)
| 1:21 pm on Dec 15, 2003 (gmt 0)|
"I still think the commercial SERPs are dreadful."
I've made a jillion posts this month, and I don't think I've made one on "commercial" serps, so you can leave me out of that.
If 5% is worse and 95% is better, then that is a very good thing. If 60% is better and 40% is worse, then the degree of the good versus the bad would make a difference.
Some areas may in fact be worse, even a lot worse. Focusing on that is a serious error though. On the whole, across the Internet, sites that an objective person would have good reason to trust are ranking well. (This does not address sites completely "lost", that is a different phenomenon.) That is a very good thing. Higher niche relevance to those sites would be an improvement, and I expect that will be a step we'll see fairly soon... but not immediately as the "lost" phenomenon seems to be taking up Google's time right now.
It's true that my main area is hypercompetitive with enough spam to feed Uganda, and the drastic improvement is plain to see. It makes sense though that areas that weren't hypercompetitive, but were *semi*-competitive might suffer quality-wise by these changes, as some babies are thrown out with the bathwater. Also it makes sense that non-spammy areas would improve slightly or not change much.
Probably shouldn't make broad statements before going to bed, but maybe it is that Google was addressing broad matching, major term, super competitve spam, and the more middling competitive stuff like "location widget" got overwhelmed.
| 1:25 pm on Dec 15, 2003 (gmt 0)|
|I still think the commercial SERPs are dreadful. |
Depends on the search term. I just searched on the name and model number of a popular digital camera, and the first result was the manufacturer's page for that camera. The other results on the first page were from review sites. I suppose a merchant or affiliate site that was selling the camera might not be happy with the search results, but I thought they were pretty good. (BTW, when I searched on the same term with the word "buy" appended to it, about half the results were results from comparison-shopping services, one was from Amazon, and the others were reviews.)
Results were less good when I searched on the name of several famous European luxury hotels. The top results in many cases were boilerplate affiliate pages, but there's nothing new about that. If anything, the results were better this time than the last time I checked, because at least I could find the hotels' own sites in most cases (and in a few cases, there were at or near the top of the SERPs).
|Now let's draw a line under it and look for some answers to the strange commercial SERPs. I personally think it is an 'intelligent' filter that has gone awry. Any comments? (of the non-personal kind!) |
Not from me. There's been plenty of speculation on this forum about "what went wrong." Some things obviously did go wrong (the disappearance of index pages but not inside pages even on non-SEO'd information sites), but every time someone comes up with a hypothesis, somebody else comes up with an exception. There's really no way of knowing--and if there were, the cycle of "Here's what I think happened..." would just have to repeat itself a few weeks or months from now with Google's next major update.
| 1:30 pm on Dec 15, 2003 (gmt 0)|
|Some areas may in fact be worse, even a lot worse. Focusing on that is a serious error though. |
The biggest assumption made by the "complainers" is that if some areas are worse, the whole google's SERPs have suddenly became garbage.
there will never be a perfect algo..you can aim for it, though.
| 2:15 pm on Dec 15, 2003 (gmt 0)|
|I just searched on the name and model number of a popular digital camera... and the first result was the manufacturer's page for that camera |
You provide both the name and model number - and you get a relevant result. What is this example designed to illustrate? That if you know exactly what you are searching for, you can still find it on Google?
It's tempting to be sarcastic about this, but I won't, it is probably best to allow this particular search engine revelation to speak for itself.
| 3:07 pm on Dec 15, 2003 (gmt 0)|
|You provide both the name and model number - and you get a relevant result. What is this example designed to illustrate? That if you know exactly what you are searching for, you can still find it on Google? |
No, it's intended to show that all top 10 search results were completely relevant information results rather than the previous mixture of information and affiliate or merchant pages. That may be bad news for sellers, but it's good news for people who are seeking information.
As for the name and model number, they weren't anything exotic like catalog numbers: Think "Canon Coolpix 9600" (to use a made-up phrase that illustrates my point).
| 3:18 pm on Dec 15, 2003 (gmt 0)|
|At its root even, the current algo is an attempt at being an algorithm. The anchor text algo was a symbol of failure by Google, even if the results often were just fine. They were going to hell in a handbasket though in competitive areas. Webmasters in less competitive areas, who don't see thirty duplicate domains in a row in the serps, simply don't have an appreciation of the problem. |
I think that there is a lot of truth in what you say here and I guess that if I'm painfully honest much of the crap that has dropped from my target searches term SERPs does fit the description that you give here. I got used to being able to compete with those guys because as you say even a baby could work out how to do it. Now I have a big problem.
Whatever I was doing in the past that pleased Google also pleased Alltheweb, Inktomi and to a slightly lesser extent Teoma as well as all of the other secondary search engines. The net effect of what Google has done in my case is to consign one of the market leaders in the niche to oblivion on Google only.
The confusing thing for many of us who are still searching for answers is how do we change what we have been doing in order to get back to where we believe we should be. If I could find a definative answer to why my sites are doing as well as ever for every term that I track except one (the most important one) and why one site appears to be filtered entirely for one term but not at all for another closely associated one. I would go away and fix them.
Answer me these two questions please Steve. Why is the allinanchor algo still working for me with the search term specific widget insurance but not at all with widget insurance.
Why is allinanchor still the dominant algo factor for widget clubs but not for widget insurance?
I can see that you can rationalise that all that has happened in those searches where things have changed dramatically that all that has happened is allinanchor has been dropped from the algo. But why has it not been dropped for closely associated searches and how does the algo know when and when not to stop using it.
Its like the dealer can decide when to deal the aces and when not to deal them ;)
| 7:12 pm on Dec 15, 2003 (gmt 0)|
is 3rd top for "televisions"
I haven't really tested many other popular terms but if "televisions" is so bad, you would expect the same on other major terms. Google's witchhunt against affiliate based sites has really hit the relevancy of commercial serps from where I am standing.
It's odd how affiliate driven sites such epinions, dealtime and kelkoo etc don't get hit? Maybe it's because of the advertsing money they pay Google. Just a guess.
I mean check out Kelkoo for example. Just about every backlink it has is from cross linking with it's German / UK sites. But because of it's advertising money it never gets hit. Google doesn't manually alter the serps, dream on.
| 8:50 pm on Dec 15, 2003 (gmt 0)|
|This site, |
[molecular.com...] is 3rd top for "televisions" completely irrelevant
What you neglected to mention is that the other top nine search results are relevant for "televisions."
Obviously, it would be nice if Google were perfect, but 90% isn't too bad a score.
| 10:07 pm on Dec 15, 2003 (gmt 0)|
>Some areas may in fact be worse, even a lot worse. Focusing on that is a serious error though.
Why? Doesnt this still point to a serious flaw?
| 10:18 pm on Dec 15, 2003 (gmt 0)|
>What you neglected to mention is that the other top nine search results are relevant for "televisions."
>Obviously, it would be nice if Google were perfect, but 90% isn't too bad a score.
Right. Best way to think of this is at what point does Joe Searcher find irrelevant results in the SERPs materially annoying? In the case of this SERP, I'd just mentally filter out this bad one, and focus on the other returned sites. This SERP is quite usable. Now if the majority of the results in a SERP were consistently irrelevant, Google would be in big trouble.
And, check out the competition. 1 or 2 bad results out of the first 10 aren't unusual. ALL SEs are imperfect.
| 10:26 pm on Dec 15, 2003 (gmt 0)|
>Now if the majority of the results in a SERP were consistently irrelevant, Google would be in big trouble.
You just summed up the "sky is falling' update threads. The definition of 'irrelevant' is what is at the core of the debate regarding the quality of serps. It is a lot like trying to describe the color blue; it comes in many shades and everyone sees it differently.
Forget quality. Too subjective. The focus of the discussion should be why are the serps different.
<last paragraph added>
| 10:37 pm on Dec 15, 2003 (gmt 0)|
>The focus of the discussion should be why are the serps different.
Good point. Almost all SERPs I have run have been as good or better than before Florida. I'll concede I have seen a handful of rotten ones. For someone where the SERPs are bad for the keywords that are important to their site, figuring out what the new algo is so they can do well the next update should be the priority.
| 11:17 pm on Dec 15, 2003 (gmt 0)|
"Doesnt this still point to a serious flaw?"
Of course not. Barry Bonds strikes out sometimes. Tiger Woods hooks one into the trees sometimes. Focusing on one specific can lead to drastically incorrect general conclusions.
| 11:20 pm on Dec 15, 2003 (gmt 0)|
Hissingsid, allinanchor hasn't been dropped for some searches. Don't look at this stuff as so black and white.
allinanchor is less important in some areas and more important in others. I don't see why that would strike anyone as strange.
| 9:09 am on Dec 16, 2003 (gmt 0)|
|Hissingsid, allinanchor hasn't been dropped for some searches. Don't look at this stuff as so black and white. |
allinanchor is less important in some areas and more important in others. I don't see why that would strike anyone as strange.
The thing is that very closely related searches are and are not effected.
widget financial (effected)
widget financials (not effected)
widget clubs (not effected)
Why not? Well because the plural of the word I'm refering to isn't used in the American dialect of English and you can't make money out of clubs (unless widget =golf). The same reason tire is affected tyre is not, etc etc etc.
Read paper here [webmasterworld.com ] referred to by Philosopher.
When they built the list they weren't even that clever Dallas is affected Huston is. I bet Cleckheaton isn't in so folks in West Yorkshire can rest easy.
| This 180 message thread spans 6 pages: < < 180 ( 1 2 3 4 5  ) |