| 4:32 am on Mar 11, 2005 (gmt 0)|
I got this using WebCEO and Googlebar. It is likely that it was WebCEO that triggered it. I sure did not have any trojans on my machine.
| 6:01 am on Mar 11, 2005 (gmt 0)|
The single greatest way to drive users away is to annoy them with this new trigger.
Happened to me several times today, I wear a glove at all times, there are no trojans, viruses, bho's or any other unwanted software on my pc.
Thankfully, firefox also has Y in their search bar and with 1 click I was getting the searches I wanted.
So drive the power users away, let us find and become comfortable with something else and we will start reccomending it.
| 7:27 am on Mar 11, 2005 (gmt 0)|
>He means that there is alot of information in your server logs that can help you improve your site. You can see where your referrals are coming from, you can see what people are using, what paths through the site they are taking, etc.
>Page rank info is not available in the logs but you can see what keywords people are using that find you in given search engines.
>What communitynews said. I'd avoid getting obsessed about backlinks/PR and concentrate on useful content/services you can add to your site that will attract links naturally. Server logs give you a great picture of what words/phrases people are using to find your site. Usually looking through those logs will give you ideas about other pages would be good to add to your site, for example.
Excellent, thanks communitynews and GoogleGuy! Is there any thing else besides the above that one can mine from the server logs?
| 5:16 pm on Mar 11, 2005 (gmt 0)|
Has anyone seen any response when using Ranking Manager?
| 4:42 pm on Mar 12, 2005 (gmt 0)|
Thanks communitynews! You've been wonderful.
| 5:06 pm on Mar 12, 2005 (gmt 0)|
I typically check my ranks by crawling the scrapers. ;)
assuming no robots.txt block of course
For broad terms, it is interesting to check the start= parameter in the referer. You can track a lot of upcoming and under-utilized kwds that way. If you are getting a lot of searches with start=30 or more, you may want to focus some content on it.
| 10:08 pm on Mar 12, 2005 (gmt 0)|
I think you've added a lot to a rather simple statement, communitynews, but feel free to run with it. Using the start= parameter to check ranks is one that has been posted here before, although I'm not sure where or when it was. I'm pretty sure there is at least one stats package out there that uses it already. I'm not sure, but I think GoogleGuy may also have a claim on it.
| 11:53 pm on Mar 12, 2005 (gmt 0)|
For me, just about any search for terms containing .php calls up this virus report screen.
I'm on a Mac, it's clean and I can do searches for just about anyghing else.
So this is clearly a residue from the worm that was used for finding and infecting phpBB forums.
A crude, however probably effective method for finding a specific kind of queary. But hey. For some searches one go is enough to generate the screen.
| 5:16 am on Mar 13, 2005 (gmt 0)|
I for one have learned a lot from this start= thing. After looking at it page one is start=08 page two is start=108. The only way you can get start=08 is if you go to another page ln the same search like page 2 or 3 and go back to page one. If somebody searches and clicks on your site before going to another page you get htt*://www.google.com/search?hl=en&lr=&q=my+search. Google changes the order of the h1= and the q= depending on what page you go to. Hitting a number and next does not seem to matter. This is all very interesting.
| 8:56 pm on Mar 14, 2005 (gmt 0)|
Are you sure you don't mean start=0& and not start=08? All the start variable is, is a pagination variable that tells where the results will start at on that page.
|After looking at it page one is start=08 page two is start=108. The only way you can get start=08 is if you go to another page ln the same search like page 2 or 3 and go back to page one. If somebody searches and clicks on your site before going to another page you get htt*://www.google.com/search?hl=en&lr=&q=my+search. Google changes the order of the h1= and the q= depending on what page you go to. Hitting a number and next does not seem to matter. This is all very interesting. |
| 3:51 am on Mar 15, 2005 (gmt 0)|
Oops your right. It does not change the meaning of what I said. So instead of start=08 it is start=0 then start=10 and start=20.
| 8:26 am on Mar 23, 2005 (gmt 0)|
The start idea could also be combined with the Google API. I have been using it to check postion but it is very limiting to only be able to see 10,000 results. You could take the start postion from the logs and use it to guide your postion checking script as to where to start.
One little caveat you should also check the num= parameter as start=0 could mean the result was anywhere in the first 100 results if the searchers preferences are set to show 100 results/page.
| 6:25 am on Mar 24, 2005 (gmt 0)|
Where do you see num=?
| 6:30 am on Mar 24, 2005 (gmt 0)|
num= can be set to change the number of results per page.
I'm not sure all the cases when it is set, and it doesn't have to be set to show alternative number of results per page - it definitely dirties the data slightly, but not much. (looks like num is set about 2% of the time)
One case it is set is when a user with an alternate number of results set goes to the second page of the results. That is, if the user has the number of results set to 100, you will see num=100 variable when they are coming from the 2nd or 3rd page or when they click 'previous' from page 2.
| 7:01 am on Mar 24, 2005 (gmt 0)|
I see it even on the first page when I have my preferences set to 100 resultes/page (my normal setting). When I change to 10 results/page I don't see it any more. Also, if I do a search from the google toolbar (Firefox) I don't see the num even though it still shows me 100 results; however, it does have sourceid=mozclient in the referrer field of the log entry. It doesn't put that in when I do a search using the box on the google home page. It looks like a little bit of log analysis would show which entires in the log file are safe for finding the SERP position.
| 11:26 am on Mar 28, 2005 (gmt 0)|
I developed a web based statistics application some time back which utilised the start and num values in the query string of search engine referrals.
There are two things that this can be useful for:
1) A statistics package can utilise these variables to work out what search engine result page number a visitor found your site on. Which combined with the keyword they typed in can be very useful. This can never be 100% accurate due to some instances when these values are not set but in 90% of cases/users it is accurate.
2) You can manually modify these values to increase the number of search results you see. Which I am sure is what many rank checkers use and set the count to 100 instead of requesting 10 pages of 10 results and if they don't then that's what they should be doing. If I remember rightly some search engines let you increase the value to 300 results per page! (maybe msn cant remember though)
| 3:53 pm on Apr 20, 2005 (gmt 0)|
According to backroom reports optilink is the latest to get the Captcha challenge.
| This 77 message thread spans 3 pages: < < 77 ( 1 2  ) |