> Javilk wrote:
> > * Search Engines -- Selecting your keyword neighborhood
> > and analyzing your log files for optimum placement.
> > http://www.mall-net.com/se_report/
> > (Under Construction!!! Better if viewed Thursday...)
> Well, I took a peek early...very interesting. A mirror of the interests of
> the general population of the planet Earth! I am surprised that beanie
> babies beat spanking, though:
> # 15 0.0011 beanie babies
> # 16 0.0011 personal
> # 17 0.0010 panties
> # 18 0.0010 spanking
> What search engines are you querying in order to compile your stats? I
> remember you saying something about this project when we were talking about
> Magellan's Search Ticker: http://voyeur.mckinley.com/cgi-bin/voyeur.cgi . Is
> that one of them?
Several. That is one of them. I am looking for more to poll, and
offer a free subscription to anyone turning in one that I have not already
found. Send the suggestions to me directly.
> Also, are the figures for "Matching Web Pages" parsed out of the "hits
> found" results from your various sources? If so, how do you calculate the
> total you show from results like:
> keyword "barney"
> Yahoo: 50000 pages
> Hotbot: 15234 pages
> Lycos: 75009 pages
>
> ? Since each engine has a different catalog, I'm not sure how you come up
> with your figure. ?
I had my systems spend nearly a week polling some of the search
engines. Alta-Vista came out ahead on most of them, so I used the largest
number. In some of the custom derivative reports, I will probably
indicate the differences. An article by IBM suggested that using a
package of five search engines, and running down all the URL's, yielded
substantially more. Unlike IBM, however, I do not have the bandwidth or
compute time to verify each and every page... so I go with the largest
number. We know those numbers are obsolete by the time the engine
reveals them.
> Interesting and useful stats you're providing. Are you going to offer any
> "one-off" deals for the $95 "Weekly list of the Top 500 co-keywords and
> queries using one or more of five words of your choice". I can see having a
> need for an occasional report.
The 95 is a suggested retail price suggested by several people. Some
suggest lower fees and monthly reports. I plan to offer the ability to
send custom reports to your clients under your own label at a nominal
rate, with your ability to add/change some text blocks as a newsletter.
Anyone interested?
I have not really looked at individual one-of-a-kind reports much
yet. I did set up a CGI based stats engine as a trial; but some of the
queries took one to 45 minutes to run... (My SGI Indy workstation took 45
minutes vs three hours for an older Pentium. After adding two higher
speed Pentiums with more RAM, it it is usually under three minutes per
query now. Not fast enough for the web.) The trade off is disk space vs
compute power, and I have not really balanced that yet. I am in the
process of adding another Linux machine to my network so I can have some
computer time to myself...
My thought is to have a query engine send most reports via e-mail, at
least till I can afford to add enough machines or balance in enough disk
space to do it real-time.
What do you think the service would be worth? How vast a database
would you want to consult, a week's worth? A month's? What other
services are you interested in?
I can see quite a few derivative products once this gets going,
assuming I get enough subscribers... But right now, I am just sending out
the weekly reports. (And quaking in fear that the peep holes will be shut
down. Last night's network slowdown really woke me up! Two of the larger
engines seemed to fail to respond sporadically, making me think this whole
effort was doomed! But they are responding ok this afternoon. I should
probably keep some response time statistics, too...)
One rather interesting thing that I have turned up, is that some pages
have rather long canned queries that are run over and over often enough to
show up on the stats.
A problem I have with that site, (or possibly with my browser,)
is that the nested tables don't seem to line up despite TD VALIGN=TOP
tags. I'll probably be uploading another set of format changes late
tonight / early Thursday AM. Do any of you see the tables radically
misaligned, or is it just me?
Suggestions, ideas, etc. are most welcome! On-list or off-list. The
best ideas will likely get free subscriptions to the services they
suggest.
[EMAIL PROTECTED] ---------- [EMAIL PROTECTED]
------------- Every mouse click, a Vote ---------------
------ Do they vote For, or Against your pages? -------
-------------------------------------------------------
Hit and Link Analysis: http://www.mall-net.com/webcons/
-------------------------------------------------------
Web Imagineering -- Analysis, Architecture, Automation,
Advanced CGI-BIN Programming, and Content Development.
<A HREF="http://www.mall-net.com"> www.Mall-Net.com</A>
* Anti-Spam FAQ: http://www.mall-net.com/spamfaq.html
____________________________________________________________________
--------------------------------------------------------------------
Join The Web Consultants Association : Register on our web site Now
Web Consultants Web Site : http://just4u.com/webconsultants
If you lose the instructions All subscription/unsubscribing can be done
directly from our website for all our lists.
---------------------------------------------------------------------
