Re: [Vo]:LENR-CANR.org downloads increased by ~14,000

2023-08-02 Thread Jed Rothwell
Robin  wrote:


> >I exclude robot readers after identifying them by various methods.
>
> Why would you exclude them? Surely allowing access would ensure that
> people doing searches would be more likely to find
> the site?
>

Perhaps I should make it clear I am not actually excluding anyone. That is,
blocking anyone. I just mean I am not counting their downloads in the
totals I show at LENR-CANR.org.

When you see a website or some PR person describing how many "visits" or
"downloads" there are to a website, you should bear in mind those numbers
are an approximation. The number of visits and "unique visitors" in
particular varies a great deal depending on what you define as a visit. Log
file records are poorly defined and the format is obsolete, so it is
difficult to sort out what is actually happening at a website.

Utility programs such as Weblog Expert and Awstats estimate vastly
different numbers of "visitors." I think I have seen them vary by a factor
of 4.


Re: [Vo]:LENR-CANR.org downloads increased by ~14,000

2023-08-02 Thread Jed Rothwell
Okay, I found 4,697 records associated with an AI project. I filtered those
out, bringing the July total down to 23,151. It is still substantially more
than the previous month.

https://lenr-canr.org/wordpress/?page_id=1213


Re: [Vo]:LENR-CANR.org downloads increased by ~14,000

2023-08-01 Thread Jed Rothwell
Robin  wrote:


> >I exclude robot readers after identifying them by various methods.
>
> Why would you exclude them? Surely allowing access would ensure that
> people doing searches would be more likely to find
> the site?
>

I do not want to include them because that would exaggerate the number of
readers. The Google-bot and others come by often, and they read the same
files again and again. Programs such as Weblog Expert and Awstats also try
to remove them. Awstats groups them under "Robots/Spiders visitors."

I wrote my own log analysis programs. I use these other programs from time
to time, to compare the results.


Re: [Vo]:LENR-CANR.org downloads increased by ~14,000

2023-08-01 Thread Robin
In reply to  Jed Rothwell's message of Tue, 1 Aug 2023 17:34:19 -0400:
Hi Jed,
[snip]
>I exclude robot readers after identifying them by various methods. 

Why would you exclude them? Surely allowing access would ensure that people 
doing searches would be more likely to find
the site?
Buy electric cars and recharge them from solar panels on your roof.



[Vo]:LENR-CANR.org downloads increased by ~14,000

2023-08-01 Thread Jed Rothwell
Since Jan. 2021, the average number of downloads at LENR-CANR.org per month
has been 9,085. It has been trending up recently. In July 2023 it suddenly
increased to 27,848, a level it has not reached since Jan. 2017. See:

https://lenr-canr.org/wordpress/?page_id=1213

https://docs.google.com/spreadsheets/d/1CwBlnYM7IFRjYCQcJ0OGjt-vRbbzqvm6T2Uk4YYsii0/edit#gid=0

I do not know what is causing this. I can sometimes identify the source of
an increase, for example, in a mass media article. But not this time. There
has also been an increase in the number of people viewing the YouTube video.

Most of the increase comes from readers who download all of the papers.
They are from many parts of the world. I checked carefully, and I confirmed
these are not robot readers from Google or some other web indexing robot.
They are humans from various ISPs. I checked many of the ISPs. They include
things like GOOGL-2, which is a cloud server for people and corporations,
definitely not for a Google robot. Google publishes a list of their robot
ISPs.

Some of the human readers probably used programs such as Web Reaper, which
downloads files en masse. You can tell they are not robot readers.

I exclude robot readers after identifying them by various methods. The
easiest method is to look at the records. They often include text
identifying the reader as a robot, with names such as:

www.google.com/bot.html
help.yahoo.com/help/us/ysearch/slurp
search.msn.com/msnbot.htm
www.alltheweb.com/help/webmaster/crawler

I have identified 131 of these.