I found some interesting things:
- This new version seems to improve Wikipedia results
- Some .gov websites got lower rankings for core keywords
- This new algo is giving more weight to domains with no dashes
- I got lower rankings for some of my subdomains
Anyone noticed same as me?
|also note that its using # in the URL #q=this+that, instead of ?q=this+that |
Is everyone else seeing that? Because I'm not! Also results dont look particularly different to the live version for me, so I'm wondering if I'm even seeing the dev version.
@Seb7 - I've seen that already too. I guess they're being a bit more strict on automated requests (maybe they're anticipating lots of automated ranking comparisons?)
Personally I'm with @mack - I'm not sure the results are better or worse, just a little different.
My main site has ranked number one for several years for the top search phrase for the topic. With this new version, I went through the first 5 pages of results and didn't find my site.
I guess my new motto is - "Go Bing Go"
Think how many ebooks will be written and sold, workshops will be taught, blog posts will be posted, etc. to inform people how to manipulate the new Google search as that information is learned.
A Google provided economy stimulater?
Do they have a Canadian URL for the BETA?
And does anyone have any idea on a live ETA?
[edited by: marketingmagic at 1:49 pm (utc) on Aug. 11, 2009]
When Google starts sending me a check for doing their work by "giving feedback", I'll then start helping them out.
They don't care about me and me spending my time for free helping them won't get me a cup of coffee.
As far as the results on the supposed "new search engine" go, I don't see much of a difference.
Back to Binging things.
According to an article at PC World, Google employees have given this new engine the codename "Caffeine".
|I found some interesting things: |
- This new algo is giving more weight to domains with no dashes
I am not seeing this at all. If the sites with hyphens are well optimized, they seem to be ranking just fine.
On a number of searches, I saw hyphenate-keyword-domains doing exceptionally well- much better on a side-by-side comparison than on mainstream SERPs
I would agree with that Shaddows. I have a good number of hypenated domains and they aren't affected at all. A few of them even rank better, so I don't think that hyphens are devalued at all. Subdomains on the other hand do seem to be, based on what I am seeing. Can anyone else confirm this?
Not to derail the thread, but does anyone think this may be a publicity stunt? It's the perfect reaction to a major announcement by a competitor.
I thought this as well - bing was making a lot of "WOW" in the press, so why not fuddle something together and present it?
Got a lot of a "me-too-taste" on it either way.
I would totally think it is. Makes complete sense. The "caffine jolt" is a little bit better, based on my tests. It's like the SERP's 2 months ago, before Google's updates, + a jolt of relevancy, and speed. I am very pleased with it if they roll it out.
I suspect they're applying data they've been collecting for some time and implementing better processing ideas.
One observation - the new algo seems to like good site structure and/or fast loading for top results.
I wonder if some members aren't obsessing too much about the new engine's SERPs (which aren't much different from the current production results for the keyphrases that I watch) when the real news--according to Google--is the completely new infrastructure.
It seems to me that this announcement is less about the finished product than about the production machinery.
I wouldn't go so far as to say hyphenated domains "rank better". Their potency seems completely confined to their keyword. I didn't see a single hyphenated-keyword-domain rank for anything other than the keyword in their domain. I wasn't especially looking, though.
I'm not sure how much of a stunt it is- unless they want to see how much "viral power" webmasters have in spreading the word. I mean, as far as the rest of the world is concerened, this has been very low key.
Added- Signor-John, I thought the current chaos in the main SERPs was infrastructure related. But how do you talk about the infrastructure changes in what is essentially a black box- all you see are results
[edited by: Shaddows at 2:39 pm (utc) on Aug. 11, 2009]
In Germany this is right now going through every big online-news sites we've got.
Should we be gushing that it's fast?
There's likely not much load on it.
|There's likely not much load on it. |
*LOL* Thats right. I couldn't even see a change in speed - sometimes the new is faster, sometimes the old google. And between 0.08 and 0,12 seconds there is really no difference.
How should I imaging a change in infrastructure? I mean they have about 2 trillions of computers and now they bought another 2 trillions to store all the data? In another thread we spoke of google not having enought capazity - so did they just cut some capazity off the normale search-engine?
[edited by: cangoou at 2:48 pm (utc) on Aug. 11, 2009]
|Should we be gushing that it's fast? |
The lack of ads probably plays a factor in load/render time as well.
Ha, apparently I type too fast:
... but your query looks similar to automated requests from a computer virus or spyware application."
Presumably it actually has to do with the fact that I searched for a string of closely related keywords all in a row, but how else are we supposed to kick the tires?
In the handful of searches it let me do, I didn't see a lot of major differences -- in general the same sites seem to be dominating my niche, and some notable spammy sites have not been removed.
Some minor changes may have been, as others have noted, similar to some of the results I was seeing briefly a few weeks ago. Hard to tell though with a small sample.
I see it still has that annoying auto complete thing going on, so that was off straight away. One of the least useful things about ajax. Certainly faster though.
Ok. It's faster. (No load, no ads, etc)
Who cares? Has anyone in the past 4 years switched SEs cause it took more than .5 seconds to load the page?
Let's get into some deeper algo analysis, shall we?
- I would like to know if any of you -950 sites are seeing your sites on this SE?
- I see certain OOP filters severly lifted on this SE.
- After a cursory glance, (haven't compared this to my databases yet) I see most of the SERPS with this SE being almost identical to ALLINANCHOR results.
|I would like to know if any of you -950 sites are seeing your sites on this SE? |
No (checked this first ;-))
|I see certain OOP filters severly lifted on this SE. |
What is OOP? I only know object-oriented programming ;-)
|identical to ALLINANCHOR results |
I don't have any historic -950s, but as I say, one page that dropped for its keyword at the beginning of the current update is in a #2.
What I categorically DO NOT SEE is frequent inclusion of sites that cannot normally be found for any given query.
So either there is NOT a mass removal of -950, or the reports of -950 are grossly exaggerated (or disproportionately represented) on these boards.
OOP = Over Optimisation Penalty, often (but not always) equated with -950
|What is OOP? I only know object-oriented programming ;-) |
I think you're messing with me, but just in case. :)
OOP = Over Optimization Penalty/Filter
Ah, thanks. Sorry, hadn't got this one *blush*
I think I have some OOPs, but they are not liftet for me on the new G.
|- I would like to know if any of you -950 sites are seeing your sites on this SE? |
Does it really matter, in the long run or in the larger scheme of things? If Google is getting ready to introduce a completely new infrastructure, it's unlikely that fine-tuning filters, penalties, etc. is the #1 priority. What you see today on the "sandbox" version of Google isn't likely to be what you'll see next month or the month after.
is it called "caffeine" because they believe lots of webmaster will need lots of it when trying to make up for ranking losses after introduction of this version?
| This 201 message thread spans 7 pages: < < 201 ( 1  3 4 5 6 7 ) > > |