hundreds of extra hits

[DianeBouvier]

i usually average 50 - 60 hits a day. i have my faithful researchers who make ten or so hits a day, and some days if they all come and i get some new constantly refreshing person i go over a hundred. but this is very rare.

these are my stats for the last three days.

Wednesday 6th October 2004 252 7 4 3
Tuesday 5th October 2004 148 22 19 3
Monday 4th October 2004 54 13 11 2

i would expect 48 hits on tuesday and 52 maybe total on wednesday, but at this time on any day i usually have oh maybe 20.

my website is on the genealogy of the greek gods. including Cheiron, that most rightous of centaurs.

at first i thought i was getting lots of hits for the virus manufacturer Chiron but none of the remaining stats show this. i went to google and entered "chiron" and did not come up on the first couple of pages.

i spell it the greek way rather than the roman way but have the roman spelling as an alternate in his notes, so chiron would come up in google.

"chiron genealogy" does not put me in the first five pages of hits but "cheiron genealogy" puts me on the top of the second page.

so i am probably not getting all these extra hits from the chiron flu shot problem.

but 148 hits one day and then 252 hits the next? instead of my normal 48 or 52? looks like a hundred extra hits are being added every day.

we will see if i have more than 300 tomorrow.

[mhamel]

Consider looking at the "Visitor Paths" stats for your site.

It shows you the URL that someone used to find your site, which would be useful to show phrases typed into a search engine.

It also tracks where people went within your site. If there are lots of instances of just one or two hits, it's likely that people found your site by mistake and left without sticking around.

If the majority show a path of several pages, and a visit length of a few minutes, you might be able to assume that they are "real" users of your site.

Maybe someone else's web page is now featuring a link to your site, and is driving traffic there. The "Came From" statistics page is a good place to look for clues.

[DianeBouvier]

now i seem to have done something strange using the reply-with-quote, sorry, just read my mind which are the quotes.

Consider looking at the "Visitor Paths" stats for your site.

yes, this is one of my favorite parts of statcounter. not as favorite as the country/ISP, nothing will replace my affection for the ISP stats.



It shows you the URL that someone used to find your site, which would be useful to show phrases typed into a search engine.[quote]

"came from" is much more useful for me because many of my visitors come thru the Ancient World Web, which is not a search engine. people know that they can find me there, so i get many repeat visitors thru that site. i also have a family genealogy page with a link to my greek gods genealogy site, and get hits thru that site also, including many who are repeat visitors.

[quote="mhamel"]

It also tracks where people went within your site. If there are lots of instances of just one or two hits, it's likely that people found your site by mistake and left without sticking around.



yes, most of my hits go to my main page, which is designed to get rid of less serious students.

also, for example, i get a great many hits for "dystonia" which is not only a greek mythological character but also a neurological condition. the neurological condition people are not interested in the genealogy of dystonia. it is also a butterfly (just about all greek mythological characters are butterflys) and those people don't stick around much either.



If the majority show a path of several pages, and a visit length of a few minutes, you might be able to assume that they are "real" users of your site.



actually not. because my site has more than six thousand (6,000) pages and i am using the free statcounter, i only have the counter on my first and second level pages. each letter of the alphabet has a page of listing-of-names of the characters of greek mythology (second level). each character has a genealogy page (third level) and a citations page linked from the genealogy page (fourth level),

google and yahoo generally only index my first and second level pages, tho some of my third level pages have been showing up in searches lately at google.

new users tend to use my on-site search page (FreeFind) but experienced users have figured out that the letter pages are easier. transliteration of greek to english is tricky.

i tend to have visitors who look at my main index page for four seconds and go away and then my faithful guys who are around for an hour or more. the visitor length does not show this well because several of my faithfuls are on isps that don't show repeat visits. but i know my faithful visitors.

my site is a resource for researchers, and i have several who come in, check something, go away, come back in 15 minutes or a couple of hours, come back again. the visitor path stats do show this.

but since i don't have code on the individual's pages, the only way i know where someone has been is when they go back to a letter page and the path shows where they came from to go to that letter page.



Maybe someone else's web page is now featuring a link to your site, and is driving traffic there. The "Came From" statistics page is a good place to look for clues.



yes, three sites are sending me visitors. the Ancient World Web, my own Bouvier Family Genealogy site, and the mother of all genealogy link sites, Cyndi's List. but these are going to be people who know what my site is about before they get here.

i had analyzed all those ideas before i posted.

i think that the cause for these inflated statistics has three possiblities, or a combination of three.
1) i have a frequent visitor from france who is a refresher.
2) one of the main characters in greek mythology is Cheiron or Chiron, that most rightous of Centaurs, and the current problem with the flu virus is caused by a problem the company named Chiron is having. i have not had any such searches and a search for Chiron/Cheiron on google puts me out of the first five pages.
3) the statcounter counter did not reset its number for two days.

since this is the first time that i have had these unexplained numbers and the problem has gone away today, i truly do suspect that the counter got stuck.

[graham35]

since this is the first time that i have had these unexplained numbers and the problem has gone away today, i truly do suspect that the counter got stuck.

I told all my friends - not to visit your site today, that fixed for you :lol:

[DianeBouvier]

glad you are looking out for me.

don't get too many visitors from spain tho i do have one who comes in about once a month and stays an hour or so.

if that is a friend of yours, tell him to hold off until the chiron problem calms down.

[DianeBouvier]

still averaging 60 to 70 page loads a day. bunch of new visitors who don't stay a minute and my faithful visitors who stay an hour or two. i do not have one picture.

this morning when i checked about ten am my time (los angeles time, it is right now twelve midnight) i had 145 page loads in the summary.

i did have a refresher and a hundred of those showed in the path and in the page load, just like they ought to do.

tonight when i checked i had 236 page loads today. but there are only about 30 new page loads in the page load section.

so i really do think my counter gets stuck and keeps registering hits that are not happening.

i copy off the page load data. i am absolutely sure that there were not a hundred page loads between the time i copied them this morning and the ones i copied off right now.

[DianeBouvier]

today i had 400 pageloads. 400. four hundred.

this is a poky little academic page. i have some faithful visitors who stick around for long periods of time (hours) but most come, see there are no pictures, and leave within five seconds. i only have counters on about 30 of 3,000 pages.

the last hundred pageloads came in 42 seconds. 42 seconds.

this guy is making a mockery of statcounter. i am getting mad.

this is the drill down for this visitor


VISITOR ANALYSIS Referring Link ~http://www.csulb.edu/~dbouvier/Entities/indexS.html <http://www.csulb.edu/dbouvier/Entities/indexS.html>
Host Name mail.windjammer.com
IP Address 208.255.212.6
Country United States Region -
City -
ISP Uunet Technologies Inc
Returning Visits 0
Visit Length 42 seconds

VISITOR SYSTEM SPECS
Browser MSIE 6.0
Operating System Windows jammer
Resolution 1024x768
Javascript Enabled

[DianeBouvier]

If you are getting hundreds of unexpected page loads, couldn't it be that some search engine or marketing program is spidering through all your pages and somehow looping? I don't know the technical solution, but 'robots=no' or some such html meta tag (I'm really just guessing) to keep programs from searching your site and skewing your counts? If there is a common use of one of your God's names that could be attracting casual surfers, it could be attracting a spidering program too.

Sandy
http://www.exoticars-usa.com[/url]

now that is an interesting idea. i do get a lot of hits from people searching for dynomia (lawlessness) son of eris (strife), which is apparently a medical condition and a butterfly also. tho most of my gods are also butterflys.

but i thought that spidering programs didn't show up. i am on the first page for most searches in both google and yahoo, so they must be spidering me. also dogpile, msn, several other search engines. also i use an onsite search from FreeFind which spiders me every week, down three or four levels, and they have never shown up.

off topic, sorry, putting my genealogy hat on. do you do antique cars? i am looking for information on the lozier produced in plattsburg new york in 1910.

[DianeBouvier]

whatever it is it seems to have gone away. yesterday i had 12 hits and today about noon i have 20, just about normal for a sunday. for some reason i get very few page views on the weekends.

this seems to be the counter equivalent to spam. i mean why would anyone pay for a counter if all it showed was a thousand of these things?

i do get a lot of hits that look like echos, where the exact page is reloaded from the same previous page five or six times. someone said that that is refresing but i wonder.

about the lozier, my grandfather was the bandmaster for the company band around 1910. i have a photograph. also several family members worked for them. they moved from plattsburgh to detroit after 1910 and my uncle went with them. then they stopped making cars completely. i asked at the peterson auto museum in los angeles and their expert on antique cars had never heard of it, so give your hubby high marks.

will look for the book and get back to you. thanks very much.

PS the los angeles public library has three copies, just put it on hold. have been working on this problem for three years! thanks again.

[webado]

Diane, spiders and bots are visitors and are logged. Usually they don't have javascript enabled. Some hybrid creatures may have it too, despite what Marjolein says, since I've seen "infestations" at times of comprehensive rapid sequential hits with javascript enabled.

Bots and spiders have schedules by which they return to re-spider web sites. I have seen one coming daily at 10pm and 4am, give or take a few minutes. If it makes it happy to re-spider something that hasn't changed in a long time, oh, well.

You can control to some extent bots by placing a robots.txt in the root of the web site with instructions of which bots are allowed and which aren't Many do not bother to read this, only the big ones like Googlebot do.

You can re-invite bots to visit daily or weekly or monthly or whatever, by using a meta tag in the head section of your web pages.

[DianeBouvier]

christina, i am very confused by the whole bot issue.

i am indexed weekly by my on site search engine, FreeFind, and they never show up in my stats. i am indexed by google, yahoo, msn, and a bunch of others. they never show up in my stats.

my site changes daily, sometimes twice a day. i have six thousand separate pages, only about 30 of which have counters on them. i usually work on ten or so pages a day and ftp once or twice a day. the basic organization does not change, but the details definitely do.

the whole problem seems to have gone away again.

one thing i am wondering about is if this is a hacker either trying to hack into the school computer thru my site or trying to hack into something like the pentagon thru the school server. is there any way to know this?

.

[webado]

christina, i am very confused by the whole bot issue.

i am indexed weekly by my on site search engine, FreeFind, and they never show up in my stats. i am indexed by google, yahoo, msn, and a bunch of others. they never show up in my stats.

my site changes daily, sometimes twice a day. i have six thousand separate pages, only about 30 of which have counters on them. i usually work on ten or so pages a day and ftp once or twice a day. the basic organization does not change, but the details definitely do.

the whole problem seems to have gone away again.

one thing i am wondering about is if this is a hacker either trying to hack into the school computer thru my site or trying to hack into something like the pentagon thru the school server. is there any way to know this?

.
Well, Diane, I assume and hope that you have a security tech team at the school, whose job is to oversee suspicious activity and secure the server from any possible attacks. At least in an ideal world.

I think I'm remembering vaguely seeing that Statcounter is ignoring bots and spiders it can identify. Those would be the official ones that everybody knows about and who identify themselves as such. For instance Googlebot answers to that name when a reverse lookup is performed on the IP addresses it uses (lately it's been swarming the web with multiple bots in an effort to reindex the entire web I read somewhere). Besides the big, known bots and spiders, who obey directives in robots.txt and in met tags, there are many others out there. They don't identify themselves in any particular way, there may or may not be a host name associated with the IP address. Some of them are ISP's own bots, building quick indexes of all the web sites to be used by their own search engines. Many are also those cursed spambots who relentlessly spider all sites they can lay their hands on looking for email addresses to add to their lists. Also other critters looking for unsecured email scripts they can exploit to send spam out faking its origin to make it look like it originates on the web site's server. Those are to be kept in check. But you'll find more about them from server logs, because their traces are to be seen there more than on your own web pages. If you find an unusual number of hits the start at your home page and only go to your contact page, then that's either a spambot looking for the email address or one of those looking for typical email scripts (like the cgi email script). On my server logs I see lots of the latter ones, because they trigger error 404 - we don't have any of those unsecured scripts any more exactly because they are unsecured.

If you feel uncomfortable, talk to the network security staff and show them your logs. I doubt the Pentagon's security rests on this, but you never know. The school's server ought to be protected at a level above where you web site is stored.

[DianeBouvier]

i was not making up this idea of someone hacking into the pentagon thru a school server. this happened about five years ago thru the junior college in los angeles where i learned HTML. quite exciting to have the fbi swarming the unix lab. with guns drawn. they shut us down for the rest of the semester. we were running a 386 as our server.

i have my school e mail on every single one of my 6,000 individual pages on the site. i do not get any spam on this e mail. the school server does very little to prevent spam and has no way to report it.

windjammer is back today. today i have 315 hits.

the last 60 or so are my regulars, the austria guy, the french guy, the redding virginia guy. many of that 60 are newcomers who click on a few pages and go away. my visitors tend to stay less than five seconds or more than an hour.

but on my last page is windjammer, up until about eight thirty this morning. so i would assume that about 250 of that 315 hits were windjammer.

i have been indexed in google and yahoo for some time. this never happened before.

surprised that it is not happening to lots of other people. from what you are saying i begin to suspect that this is another bot that statcounter isn't not counting yet, if that made any sense. but why they are boting at me of all sites is a mystery. i mean who cares about the genealogy of aeolus but me?

[webado]

Maybe they are building an encyclopedic dictionary and your site has tons of words - and the perfect examples to be used in their definitions - only half kidding actually :lol:

So you didn't get spammed.... lucky you.

[DianeBouvier]

no spam on my school e mail, but i get enough on my hotmail and my yahoo to make up for it!

i suppose someone could be downloading each of my 6,000 pages one by one. sometimes i think someone is doing that, because they are there all the time and then when i do my monthly new upload they realize that everything has changed and give up.

but i don't even have any words or definitions. just names, transliterated from ancient greek, and the names of their parents, siblings, and children.

is there anyway that i can thru statcounter block seeing the visits of windjammer?

[webado]

If this visitor has the same IP address all the time, then you can block the IP address from being counted, from you project setting.

Or you can even block this IP address from accessing your site by placing (or modifying an existing) .htaccess file in the root of the site:


order allow,deny
allow from all

deny from nnn.nnn.nnn.nnn

where nnn.nnn.nnn.nnn is the IP address to be banned.

[DianeBouvier]

and more aha.

today windjammer again filled up my log. according to the drilldown he spend more than 7 hours on my site. lets see 6000 seconds are about an hour and a half?

he is hitting my top tier pages sometimes six in one second.

something is wrong here. this is not a good thing.

clearly this .htaccess file is the way to go if i have permission to do it. i don't have unix permission to write, only thru an ftp client. will look around in wsftp and see what i am allowed to do. only have read permission for my own subdirectory tho i can see the names of the others.

does the IP address have to be the numerical one?

there is no .htaccess in my root, only .cshrc .login .logout .profile and they are the generic files (all dated 1993)

otherwise i suppose it is time to go to the administrators. have been avoiding this because i am over my quota and they are carefully not noticing it. but if this may comprimise the server guess i better be a good citizen.

[DianeBouvier]

and here comes another one.

my stats are totally consumed by this thing. i have decided its name is godzilla.

the last several days i have had about three hundred visits a day from godzilla.

this is not fun. i enjoy reading my stats very much, and now godzilla has taken all the fun out of it.

tho without statcounter i never would have known about it, so this really is a big service statcounter is providing and not mentioning.

how to get rid of godzilla is the problem.


christina, i had a brainstorm when i woke up this morning that i can wsftp files into my root and i just tried it and it let me ftp a little txt file. so i can probably ftp a .htaccess file also.

questions:

can i block several isps? what happens if godzilla keeps changing isps?

i need to look this up, haven't done anything with UNIX for several years because the FBI closed down our UNIX server. at csulb they are real picky about who gets access.

so since i just noticed that in the drilldown statcounter is giving me the numerical isp address i will start making a list.

would appreciate very detailed instructions on how to do this.

[webado]

You just create a test file called .htaccess (NB: it starts with the dot).

Using Notepad, put in these lines:

order allow,deny
allow from all

deny from nnn.nnn.nnn.nnn


Instead of nnn.nnn.nnn.nnn use an actual ip address. (NB: every nnn is in the range 000 to 255).

If you want to ban a range of addresses you can leave out the last octet (after the last period). I don't know of any way to ban a range based on one of the other octets.

You can repeat this line as many times as you want to, for all the ip addresses you want to ban.


Save this file and upload it to your root folder.

There's a pretty good article on .htaccess at http://www.outfront.net/tutorials_02/adv_tech/htaccess_magic2.htm .

[vanfruniken]

I too see multiple counts of the same page with no referrer, sometimes up to 3 times per second, 4, 6, 8 or more occurrences in a row, all within one minute
It seems that MSIE 6 is always involved.
I use no frames
Other people seem to experience similar problems
It also seems to happen to my home page most if not always (and to another page that also has redirects to it)
Many times this is an exit sequence

http://users.pandora.be/_nl/home.html. /_nl/index.html redirects and /index.html redirect to it.

http://users.pandora.be/_nl/Birdsexing/geslachtsbepaling-vogels.html. _nl/Birdsexing/index.html redirects to it.

Is there anything I can do to prevent this behavior, which overestimates the actual page counts.
Or is there anything StatCounter could do to ignore these reloads, because I believe they are caused by MSIE 6.0

Thanks

Van Fruniken

[webado]

I too see multiple counts of the same page with no referrer, sometimes up to 3 times per second, 4, 6, 8 or more occurrences in a row, all within one minute
It seems that MSIE 6 is always involved.
I use no frames
Other people seem to experience similar problems
It also seems to happen to my home page most if not always (and to another page that also has redirects to it)
Many times this is an exit sequence

http://users.pandora.be/_nl/home.html. /_nl/index.html redirects and /index.html redirect to it.

http://users.pandora.be/_nl/Birdsexing/geslachtsbepaling-vogels.html. _nl/Birdsexing/index.html redirects to it.

Is there anything I can do to prevent this behavior, which overestimates the actual page counts.
Or is there anything StatCounter could do to ignore these reloads, because I believe they are caused by MSIE 6.0

Thanks

Van Fruniken
Sorry, I can't figure out your links at all. Even the domain alone doesn't work.

Redirects will always generate more traffic. Try to streamline your site.

[DianeBouvier]

thanks christina, this is a big help.

my test txt file that i ftped last night is still there, so i guess it is going to let me do this. sometimes at this school you ftp stuff and it looks like it works and then after 24 hours it disappears.

today so far things have been quiet, only 29 hits so far, which is about normal.

thanks for that link. i did a search last night for tutorials and found a couple but this is much better. do wish it was more handicapped friendly. i would complain to its main site about how text based browser unfriendly it is but i guess something dedicated to FrontPage is not going to be interested in the W3 conventions for the disabled.

i may need the redirect part if the school says i have too much up and i have to move it to a professor's web site. we were trying to avoid doing that until we are finished, which will be several years, because giving me permissions on his space will be tricky.

this tutorial also includes an example of how to prevent people from linking to your graphics. does this prevent right clicking on an image? this makes everything very difficult for those of us who surf with graphics turned off and only open an image when navigation is impossible without it. W3 conventions are very much against both situations.

off the soapbox. i should be grateful for the information about .htaccess

vanfruniken, i found earlier that i was getting what i call "echo" hits when i had a link to the current page on the current page. for example, i have pages for each letter of the roman alphabet with lists of names. i had the entire alphabet clickable on each page. so when the A page loaded, there was a link to the A page on the A page. this caused some odd statistics. after i removed the link that was looping back on each page this problem went away.


this problem i am having now is different. i was wondering if anyone else was having it, because if it is hackers or e mail spammers using my server to bounce things off of it may be a problem for all of us.

i think statcounter should count this as a feature. maybe call it Security Alert?

[webado]

Diane, I didn't read that whole thing on .htaccess, but preventing right-clicking on an image wouldn't be handled that way. I put out something recently on how I'm doing this myself (and yes, it's a bit of a headache to to set it up and remember to use it all the time).

.htaccess is used to disallow hot-linking to the image file from another site than the one that has it (or from another directory even). It can also disallow using the url of the image file to access directly through the browser.

Thsi would not prebent somebody who's already viewing the image afetr having reached it "legally", from right-clicking on the image and saving it, or from using the little toolbar that is available in browsers and opos up when you hiove over the image.

To disable that toolbar, your .... tag has to include galleryimg="no". This will not disable it if the image is opened in its own browser window addressed by its url however.

For that I sometimes replace a hyperlink to an image file with an onClick even handler where I open a pop-up window with no browser toolbars, no status bar, nothing and just an html page containing the <img...> tag for that image plus the no right clicking script from below.

To disable right-clicking I use a script that I place in the <head> section of any web page where I need it:

[code]
<script language="JavaScript1.1" type="text/javascript">
// distributed by http://www.hypergurl.com
<!-- var debug = true;
function right(e)
{
if (navigator.appName == 'Netscape' && (e.which == 3 || e.which == 2))
{
// this does not work simply as an alert in Netscape: alert('This Page is fully protected in Netscape!');
document.location= "noclick.html";
return false;
}
else if (navigator.appName == 'Microsoft Internet Explorer' && (event.button == 2 || event.button == 3))
{
alert('This page is fully protected!');
return false;
}

return true;
}
document.onmousedown=right;
if (document.layers) window.captureEvents(Event.MOUSEDOWN);
window.onmousedown=right;
//-->
</script>


This works with a page that I called noclick.html which is where I want to send users of Netscape or Firefox who right-click. My page simply says: "No right clicking!" . In IE it only displays an alert box.

None of this does anything to prevent people from using the browser menu and select View Source eventually, or take snapshots of the screen. But it makes it harder for the majority of web surfers :lol:

Incidentally, since you are an adept of W3C, it's possible some or all of this may not agree with those rules.

[DianeBouvier]

i am definitely not an adept of anything except maybe putting my foot in my mouth! i am however handicapped. i have seizures which are sometimes triggered by flicker, flash, blink. i always surf with graphics turned off. also i have blind friends who surf using either a screen reader or a screen brailler, and they tell some stories. so i do suppose you could describe me as an adApt.

navigating any web site that is not designed according to the W3 standards for access for the handicapped can be very dangerous for me, and very frustrating for the blind. we have to right click on the image placeholders (or guess where the image is on the page if there are no placeholders) and show picture to figure out where to click to go where we want.

look at that front page website with graphics turned off in your browser and you will see what i mean.

one of the things i love about statcounter is how disabled-friendly it is. i never need to turn on images to use the site. and now that he has sized the buttons the place is perfect. was just messing around with the Surprised emoticon. i can right click, and then "show picture" but i can't open the link. so obviously it is possible to allow this, altho it may just be that only people as bright and helpful as the statcounter guys know how and are willing to do the extra work to make it happen.

off my soapbox.

in the twelve hours since i last looked i have had about 150 new visits. most of them i think are legit.

but something is definitely wrong here.

[vanfruniken]

Dear chrisooc,
Thanks for your speedy response

Sorry, my mistake, I meant

http://users.pandora.be/Life-ID/_nl/home.html
... /_nl/index.html and .../index.html redirect to it.

http://users.pandora.be/Life-ID/_nl/Birdsexing/geslachtsbepaling-vogels.html
.../_nl/Birdsexing/index.html redirects to it.

Van Fruniken[/code]

[webado]

Ok, I'm not sure if this would help, but if you use this in the <head> segment of the index.html pages and instead of the javascript that redirects, it might.


<META HTTP-EQUIV=Refresh CONTENT="0; URL=home.html">



Is there any particular reason why your main page is called home.html instead of index.html? If you change it, of course you'll have to change all references to it on your other pages, but it would be so much cleaner.

The same principle applies to the page /Birdsexing/geslachtsbepaling-vogels.html - it really should be simply renamed /Birdsexing/index.html instead of redirecting to it. And if you do redirect, maybe the meta-refresh may work better than a javascript location.

[DianeBouvier]

have any of you seen the new google scholar? i wonder if that is who has been all over my website. this would make sense. and christina did say she thought it was a bot indexing me. if it was this new bot statcounter might not have gotten it excluded yet.

i probably have more than ten thousand citations from Homer and that much or more from apollodorus. just what a scholar google would index.

very quiet today, only 29 hits in 20 hours. about normal for a friday.

[webado]

have any of you seen the new google scholar? i wonder if that is who has been all over my website. this would make sense. and christina did say she thought it was a bot indexing me. if it was this new bot statcounter might not have gotten it excluded yet.

i probably have more than ten thousand citations from Homer and that much or more from apollodorus. just what a scholar google would index.

very quiet today, only 29 hits in 20 hours. about normal for a friday.
Diane, do you have the IP address of this mysterious visitor?

[DianeBouvier]

VISITOR ANALYSIS
Referring Link ~http://www.csulb.edu/~dbouvier/Entities/indexA.html <http://www.csulb.edu/dbouvier/Entities/indexA.html> Host Name hide104.suffolk.edu
IP Address 192.138.214.104
Country United States
Region Massachusetts
City Somerville
ISP Suffolk University
Returning Visits 1
Visit Length 3 hours 18 mins 56 secs

VISITOR SYSTEM SPECS
Browser MSIE 6.0
Operating System Windows XP
Resolution 1280x1024
Javascript Enabled



VISITOR ANALYSIS
Referring Link No referring link
Host Name mail.windjammer.com
IP Address 208.255.212.6
Country United States
Region - City -
ISP Uunet Technologies Inc
Returning Visits 2
Visit Length 7 hours 15 mins 8 secs

VISITOR
SYSTEM SPECS Browser MSIE 6.0
Operating System Windows jammer
Resolution 1024x768
Javascript Enabled

[webado]

Are these 2 mysterious visitors visiting just one or 2 pages or are they going through many of them in fairly rapid sequence? This would indicate a systematic crawl (or maybe copying) of your site.

That first visitor appears to be from a university, so it kind of makes sense. Perhaps another scholar who visited one page, left the computer on and came back after a few hours to do some more browsing and looked at another page.

The windjamer one I'm more concerned about. If you notice the operating system is reported as being Windows jammer. Somehow I don't see how the word jammer got in there. It's certainly not a normal echo of a Windows OS. So something is running on that computer that changes the HTTP information that is echoed. Maybe that's common. It may be a from a program that copies an entire site or parts thereof. It depends on how many pages were viewd by this visitor in all.

The visit length doesn't actually matter. If 2 pages were viewed and that happened say 7 hours apart it may be that they are considered part of the same visit and the visit duration is computed as being that long.

[DianeBouvier]

no, it really was seven hours. about 300 hits. i do not have a counter on all 6,000 of my pages, only on the top letter pages.

each letter of the roman alphabet has a page with links of names of ancient greeks transliterated starting with that letter. each name has a separate genealogy page with links to parents, siblings, children. each name has a separate source file with a link from the name page to the source file.

only have statcounter code on the letter pages.

google and yahoo have indexed the individual name pages but not the source file pages.

so it looks to me like someone went thru all the name pages and their source pages.

he was back i believe three times, more than five hours each time, more than 200 hits on the letter pages each time.

i have faithful visitors who leave me up and check things. i also have people who try to copy off pages. since i put up a new version once a month they quit after the new version. i am familiar with both these patterns. my purpose after all is to be a reference site.

what is noticeable about windjammer is how fast he is moving thru things. five or six hits per second, and that is on the letter pages. so if he was going down two levels from the letter pages, well no one can mouse click that fast.

[webado]

Windjammer must be a bot, designed perhaps to copy entire sites.

[DianeBouvier]

i just did a search on the scholar.google for homer genealogy which if they are indexing my citations ought to turn me up and nothing did.

a search for any obscure character will put me on the top of the list in the regular google. try Teutamus.

so apparently either this was not the scholar.google bot indexing me or they searched and decided that i didn't have enough of homer up there for them to index. i only have citations on genealogy. but i have every single citation from the iliad on genealogy.

but i do think this ought to be a new feature, because it certainly does alert us to hackers.

now of course i am trying to figure out whether i should block windjammer. if it is a google bot i don't want to, if it is a hacker i certainly do.

[webado]

Diane, I really think it's a rogue bot, safe to block.

[CaribbeanChoice]

christina, i am very confused by the whole bot issue.

i am indexed weekly by my on site search engine, FreeFind, and they never show up in my stats. i am indexed by google, yahoo, msn, and a bunch of others. they never show up in my stats.
I think I'm remembering vaguely seeing that Statcounter is ignoring bots and spiders it can identify. Those would be the official ones that everybody knows about and who identify themselves as such. For instance Googlebot answers to that name when a reverse lookup is performed on the IP addresses it uses (lately it's been swarming the web with multiple bots in an effort to reindex the entire web I read somewhere). Besides the big, known bots and spiders, who obey directives in robots.txt and in met tags, there are many others out there. They don't identify themselves in any particular way, there may or may not be a host name associated with the IP address. Some of them are ISP's own bots, building quick indexes of all the web sites to be used by their own search engines. Many are also those cursed spambots who relentlessly spider all sites they can lay their hands on looking for email addresses to add to their lists. Also other critters looking for unsecured email scripts they can exploit to send spam out faking its origin to make it look like it originates on the web site's server. Those are to be kept in check. But you'll find more about them from server logs, because their traces are to be seen there more than on your own web pages. If you find an unusual number of hits the start at your home page and only go to your contact page, then that's either a spambot looking for the email address or one of those looking for typical email scripts (like the cgi email script). On my server logs I see lots of the latter ones, because they trigger error 404 - we don't have any of those unsecured scripts any more exactly because they are unsecured.
Bots that do not load images and do not load javascript will not be logged by StatCounter. This includes Google and Yahoo! search engine spiders. If the bot loads images and/or loads javascript, it will be considered a visitor, unless StatCounter is set-up to identify it as a bot.

[vanfruniken]

Hi Christina

Thanks for the suggestion to use a refresh metatag. I haven't used it because of a bad experience I had when I started learning Javascript...
Shouldn't I also provide a reasonable refresh time of, say 2 secs to allow people to back out with the back button?

I also haven't had much success with the window.replace(...) function. (Is there a chance StatCounter would be able to retrieve the referrer to the redirect page instead?)

Also, my dns provider hosts my page in a single frame window (see www.birdsexing.be, if you are interested at all).
In order for my pages to be bookmarkable, I need to do get rid of the hosting frame by scripting top.location.href="home.html".
Otherwise google listings would all be titled with the title they provide (advertising for them:).

As to the extra layer(s) of redirects (DNS home->index.html->home.html->geslachtsbepaling-vogels.html

[webado]

Hi Christina

Thanks for the suggestion to use a refresh metatag. I haven't used it because of a bad experience I had when I started learning Javascript...
Shouldn't I also provide a reasonable refresh time of, say 2 secs to allow people to back out with the back button?

I also haven't had much success with the window.replace(...) function. (Is there a chance StatCounter would be able to retrieve the referrer to the redirect page instead?)

Also, my dns provider hosts my page in a single frame window (see www.birdsexing.be, if you are interested at all).
In order for my pages to be bookmarkable, I need to do get rid of the hosting frame by scripting top.location.href="home.html".
Otherwise google listings would all be titled with the title they provide (advertising for them:).

As to the extra layer(s) of redirects (DNS home->index.html->home.html->geslachtsbepaling-vogels.html
The meta refresh will, unlike a redirect, permit using the browser arrows properly. If you provide a refresh higher than 0, then it will not be a smooth transition. Best to have it switchd as fast as possible.

I have no idea whether or not Statcounter will even consider dealing with pop-up windows. I did my reasearch on it, I tested my findings, I am satsified that that's one way, maybe the only way to approach it. The rest is up to Stactcounter. I personally don't have a pressing need for this kind of counter, but I can see the need for it.

You can get out of the frameset by <a href="...." target="_top"> I believe.

Best would be to get better hosting that does not force you to have your domain pointed to via a parked page and framest. I think it's because they don't have domain name servers.

If you are paying for hosting right now, you can find other solutions where you probably will not pay any more than what you're paying now, and which offer a full complement of webhosting features. Check out www.solidinternet.com for instance.

If it's a problem with the domain registrar who will not provide a way to specify your own domain name servers, then by all means change the registrar. Perhaps the .be tld is a problem? It's really infuriating how sometimes these country tld registrars establish a monoply and not only do they offer a poor service, but they are also very expensive. One really has to be very patriotic to go through all the hoops and expense just to acquire a domain with their own country's tld!

Anyway, I'm sure you'll find a good way out of this maze soon.

[vanfruniken]

Remainder of message

Question: if I would revert to a site hosted entirely in my DNS provider's single frame, would StatCounter still work?

BTW in those dummy pages I have been putting lots of keywords (both in the metatag and in text made invisible by a style) to get a high Google ranking, but I am not so sure how effective this is.

I do believe, that something in MSIE 6.0 (a bug, or some GUI quirk that causes people to hit the Back button more often, or some user preference that MSIE6.0 allows people to specify) is to blame for the repeated reloads that I have been experiencing. I only happens occasionally, though. Most often 6-10 reloads of the home page with no referrer, but sometimes 16 and more (number seems to be even), most often approx 1/sec

Thank you so much for looking into this.

Van Fruniken

[DianeBouvier]

i don't have any images whatsoever (yes, 6,000 pages without one image) and i do not have any javascript.

if it was good enough for homer it is good enough for me. one reason i am so big on being accessable to the blind is because homer was blind.

so if this bot is looking for images it must be very disappointed.

sounds like i better block it.

tho everything has calmed down the last couple of days, about 25 hits both days. typical for the weekend. most of my visitors come from schools.

[webado]

Remainder of message

Question: if I would revert to a site hosted entirely in my DNS provider's single frame, would StatCounter still work?

BTW in those dummy pages I have been putting lots of keywords (both in the metatag and in text made invisible by a style) to get a high Google ranking, but I am not so sure how effective this is.

I do believe, that something in MSIE 6.0 (a bug, or some GUI quirk that causes people to hit the Back button more often, or some user preference that MSIE6.0 allows people to specify) is to blame for the repeated reloads that I have been experiencing. I only happens occasionally, though. Most often 6-10 reloads of the home page with no referrer, but sometimes 16 and more (number seems to be even), most often approx 1/sec

Thank you so much for looking into this.

Van Fruniken
With all your redirects I'm not surprised really. Some indexing spiders may be getting confused also while trying to follow links. Rememebr, some bots and spiders may well be logged. Also if you are using things like FrontPage and publush directly to the web and also test, this may also register, and quite possibly with no referrer.

Not sure how good are those keywords and site description in the frameset page, but it's all you can do now anway, unless you change registrars and/or hosting.

ALl I can say is stay away as much as possible from frame hosted domains. If you do use them use the frames versions of Statcounter. The pop-ups will continue to not have a referrer until and unless Statcounetr changes the counter to track them.

Let me say here I've long been in favor of having only one version of the counter, combining the regular, the frames and the xhtml and now also the pop-up windows versions into one script. It is not difficult and it would make maintenance easier as a bonus.

[vanfruniken]

As a test. I removed one level and put in a refresh metatag elsewhere. I'll see how this will affect things.

As to our domain(s), they only cost 9.68 euros a year (2easy.be). It is a lure into full service for $49/yr. .be domains (w. more features) typically cost 30EUR/yr, which seems on a par with other countries. (My regular ISP does charge 15EUR to get OUT of the agreement:)

Thanks again for your thoughts. I will keep them in mind -- in particular when business will pick up, we will surely go to a full-fledged domain.

Van Fruniken
www.life-id.be
www.birdsexing.be

[DianeBouvier]

this is getting worse and worse. now i think it is not a bug but that this is a feature, alerting me that something is very odd here. perhaps this should be included as a feature.

for the last week i have several hundred visitors a day who don't seem to exist. cannot figure out if they are not showing in my log or if they are just getting overwritten quickly. if they are visiting during the middle of my night i suppose they can be overwritten soon but it just looks like too strange a pattern.

273 hits, 271 unique visitors? almost every day i have something like this in my summery, usually a difference of 2 between the visitors and the hits.

in the other stats people show up normally. i have a few faithful visitors, some with more than a hundred visits, who visit for an hour or more, i have a bunch of people who come and look at the index page for less than a second and leave, and then there are the middle people who stick around for about 20 minutes.

i used to recognize most of my people.

but this last week i don't even want to look at my stats because they can't be real. 350 hits, 348 unique visitors. yet when i look at other places i will have an old timer with 25 hits on one visit. today i had the same difference of two but one guy who is always around for an hour or so with a bunch of hits.

something is definitely wrong here.

[seiko]

Hi to All

I know this is an old tread but I've just had the same thing happen to me over the last two days.

I've gone from a very modest 5 to 6 visitors a day to 1790 visitors a day, this has now blocked my log so I cannot get much detail about the visitors.

The only info I can gather is they are running a Mozila 4.04 Browser the IP address range from 82.110.109.208 to 82.110.109.215, which is an Easy.net ISP.

There is no referring address or leaving address plus no hint of where in the UK
The visitors came from.


The strange thing about this is I also have Real Tracker installed on the same pages, which doesn’t pick up this extra traffic, Has any one got any other Ideas on this subject since it was last discussed?

Regards Seiko

Edit:

The only info thats given about the visitor.

VISITOR ANALYSIS
Referring Link No referring link
Host Name
IP Address 82.110.109.212
Country United Kingdom
Region -
City -
ISP Easynet
Returning Visits 0
Visit Length 3 mins 3 secs
VISITOR SYSTEM SPECS
Browser Mozilla 4.04
Operating System Windows NT
Resolution Unknown
Javascript Disabled

[DianeBouvier]

the professor i do this site for just switched from AOL dial-up to DSL. i have been logging his visits on purpose just to compare what i know he is doing to what the stats show. up to now this has not produced much of interest.

except now he has DSL and many amazing things seem to be occurring. the other night i put up a new, expanded version of the site, as i do about once a month. he went to check it, looked at the main index page, clicked on the A index page, then went off and read e mail. this is what statcounter showed for that visit:

Date Time WebPage
7th April 2005 01:06:51 AM www.csulb.edu/~dbouvier/Entities/
No referring link
7th April 2005 01:06:51 AM www.csulb.edu/~dbouvier/Entities/
No referring link
7th April 2005 01:06:51 AM www.csulb.edu/~dbouvier/Entities/
No referring link
7th April 2005 01:06:55 AM www.csulb.edu/~dbouvier/Entities/
No referring link
7th April 2005 01:07:01 AM www.csulb.edu/~dbouvier/Entities/indexA.html
www.csulb.edu/~dbouvier/Entities/


notice three mouse clicks within one second.

this is the kind of pattern i am getting from other visitors, sometimes i have ten mouse clicks within one second.

but mostly when i have a couple of hundred extra visitors they cannot be found in any of the statistics. just in the summary.

and no matter how many visitors i have in a day, there are always two less returning visitors than unique visitors.

[webado]

You have

Return to top of page (#top of page)


So anybody clicking that too would refresh the page and score another hit.

All the bots that visit your pages follow that link as well, until they get tired of it. Some bots will get logged, as not all are either javascript disabled or image disabled.

If your log size is 100, then you won't see details of more than the 100 last pageloads, of any nature.

[DianeBouvier]

well, this could certainly account for some of it. and there certainly is an easy way to find out.

some time ago i discovered that having a link on a page to itself sometimes but not always caused a sort of echo effect. for example i have 20 some pages for separate letters of the alphabet, with a clickable alphabet on them. when i removed the S link from the S page that cut down on the echo but did not eliminate it.

my record for someone going from page A to page B repeatedly in the same session has been 26 times within 3 minutes. my record number of clicks within one second has been 5.

i will go take the top of page links off and load them. if that solves the problem i am going to go eat a lot of chocolate.

[webado]

Stock up on some really GOOD dark, bittersweet REAL chocolate and have some for me too :lol:

[seiko]

What I cannot work out, as I have two trackers on each page, StatCounter picks up this extra 1790 visits a day but Real Tracker Doesn't ? :?

[webado]

Spend some time evalauting what each picks up. What do those extra hits have in comon? javascript disabled browsers perhaps?

Does real Tracker only log new visitors and ignores returning ones? Is javascript an issue?

[DianeBouvier]

we haven't got our chocolate yet.

since i took the top of page links (both parts, by the way) off of all the pages that had them and ftped them up 24 hours ago my repeated stuff has slowed down a little. today i have someone going from one page to another ten times every ten seconds instead of three times in a second. but he is still going from page H to page G ten times within one minute. and never from page G to page H.

got a reasonable number of total hits for a sunday morning but again in the summary one less unique visitor. this cannot possibly always be true. there seems to be some sort of ratio here, that is set so that the ratio of unique visitors to first time visitors is always exactly the same?

i am now indexed in Clusty and getting hits from it. all three thousand people clusterd, by the way, quite well. so they must have indexed me last week sometime. but this did not show up in my statistics. i mean if someone looked at all six thousand pages (each guy has an individual page + citation page) and the links between the three thousand internally linked pages (anyone want to figure out how many links it would take for all three thousand to be linked to each other three ways? no, parents, siblings, spouse, parents for spouse, children so for three children and three siblings oh my. and some have a hundred children.

well this would have shown up noticeably. and it didn't.

so i do not think that any of this stuff is from search engines.

[DianeBouvier]

still no chocolate

today i had the same thing, about a hundred hits overwritten if they actually did happen and i don't believe they did.

summary completely out of whack with any other statistic.

visit length completely out of whack with path.

people going from page A to page B ten times in one minute but never going from B to A. and there is no logical reason for anyone ever to go from A to B more than once. in their lives.

so whatever it is, it was not the return to the top of page links.

[DianeBouvier]

it has been a bad week for phantom hits. couple of days with 200 and one day with more than 400 hits that i don't believe are really there.

lots and lots and lots of the visitor going from page A to page B ten or fifteen times in ten or fifteen seconds. sometimes five times within the same second.

then today i have one visit counted as three. this is a web tv isp which seems to be changing during the visit. what the visitor is doing makes perfect sense, there are no hits repeated over and over. but there are three overlapping visits.

the person comes in at 2:10 on one number and stays until 2:17. this visit has a gap between 2:10 and 2:17:38.

on slightly different number the visitor appears at 2:14:14 and stays until 2:17:26. on a third number the visitor appears at 2:13:26 and stays until 2:14:41. so all three visits overlap.

my search page does open in a second window, but this has never shown up before as a separate visit.

will post the actual paths.

pool-141-149-116-51.pghk.east.verizon.net (Verizon Internet Services)

New York, Millbrook, United States, 0 returning visits

Date Time WebPage
30th April 2005 11:07:33 AM www.csulb.edu/~dbouvier/Entities/
www.google.com/search?hl=en&lr=&q=greek mythology genealogy
30th April 2005 11:07:45 AM www.csulb.edu/~dbouvier/Entities/indexA.html
www.csulb.edu/~dbouvier/Entities/
30th April 2005 11:08:29 AM www.csulb.edu/~dbouvier/Entities/indexB.html
www.csulb.edu/~dbouvier/Entities/indexA.html
30th April 2005 11:08:34 AM www.csulb.edu/~dbouvier/Entities/indexS.html
www.csulb.edu/~dbouvier/Entities/indexB.html
30th April 2005 11:08:44 AM www.csulb.edu/~dbouvier/Entities/indexB.html
www.csulb.edu/~dbouvier/Entities/indexA.html
30th April 2005 11:08:44 AM www.csulb.edu/~dbouvier/Entities/indexA.html
www.csulb.edu/~dbouvier/Entities/
30th April 2005 11:08:45 AM www.csulb.edu/~dbouvier/Entities/
www.google.com/search?hl=en&lr=&q=greek mythology genealogy

bay-6me-tv-1a-natpool-nc3.bay.webtv.net (Webtv Networks Inc)

California, Mountain View, United States, 0 returning visits

Date Time WebPage
30th April 2005 02:10:50 AM www.csulb.edu/~dbouvier/Entities/
No referring link
30th April 2005 02:17:38 AM www.csulb.edu/~dbouvier/Entities/f.htm
www.csulb.edu/~dbouvier/Entities/i338.htm
30th April 2005 02:17:51 AM www.csulb.edu/~dbouvier/Entities/indexOrphic.html
www.csulb.edu/~dbouvier/Entities/
30th April 2005 02:18:03 AM www.csulb.edu/~dbouvier/Entities/
No referring link

bay-6me-tv-1a-natpool-nc4.bay.webtv.net (Webtv Networks Inc)

California, Mountain View, United States, 0 returning visits

Date Time WebPage
30th April 2005 02:14:14 AM www.csulb.edu/~dbouvier/Entities/f.htm
www.csulb.edu/~dbouvier/Entities/i338.htm
30th April 2005 02:15:30 AM www.csulb.edu/~dbouvier/Entities/g.htm
www.csulb.edu/~dbouvier/Entities/i665.htm
30th April 2005 02:17:26 AM www.csulb.edu/~dbouvier/Entities/indexJ.html
search.freefind.com/find.html?id=82040005&pid=r&mode=ALL&n=0&query=&sitemap=Site Map

bay-6me-tv-1a-natpool-nc2.bay.webtv.net (Webtv Networks Inc)

California, Mountain View, United States, 0 returning visits

Date Time WebPage
30th April 2005 02:13:26 AM www.csulb.edu/~dbouvier/Entities/
No referring link
30th April 2005 02:13:38 AM www.csulb.edu/~dbouvier/Entities/indexOrphic.html
www.csulb.edu/~dbouvier/Entities/
30th April 2005 02:14:41 AM www.csulb.edu/~dbouvier/Entities/indexJ.html
search.freefind.com/find.html?id=82040005&pid=r&mode=ALL&n=0&query=&sitemap=Site Map

[DianeBouvier]

today more than half of my hits show as duplicates, triplicates, what is the word for eight?

unless 15 people have all joined together and conspired to spend time on my website going from one particular page to another particular page eight times in 30 seconds there is something wrong here with the program.

[Moredial]

I have not read every single post in this thread but how is this for a theory.

You have mentioned that some of your keywords also refer to medical conditions and manufacturers. There is every possibility that these key words have a high price tag in the PPC world of advertising.

Net effect is to make these key words very attractive to the people who run the 'paid to read' programmes. It could well be that the pseudo directories that create the lists that are viewed and clicked by the PTR affiliates includes your url, especially if you appear in the top #20 on any search engine.

Why do they do this? The PTR affiliates earn money for every site link that they click on - these are the highly targetted visitors that some 'optimisation' programmes offer to send you. The PTR programme orgainisers set up special pseudo directories that group together links to sites bidding on high value key words and they get paid by the PPC programmes for every click recorded on their sites.

How can you find out if your site has been included in the pseudo directories and are being clicked by people who have not understood the difference between paid clicks and other listings. Well, you can't really. But you can get a feel of the number of pages that are including you by setting up your domain name as a google alert.

The other possibility is that you are being hit by one of the scripts that is used in place of real visitors to mimic real clicks from the pseudo directories.

If you think this is the cause of your problem, you can find out who is bidding on the search terms by looking at the sponsored ads in google, yahoo, overture, etc when you use only the name of one of the gods as search term.
You could try asking them to not use your spelling of the name if they use a different spelling so that you cease to be plagued by false hits. Unfortunately there is little that you can do with words that are part of everyones language.

Good luck.

[DianeBouvier]

hum. interesting ideas altho needs some fine tuning.


You have mentioned that some of your keywords also refer to medical conditions and manufacturers.


each separate individual has a separate page showing its genealogy and another separate page showing the sources for that individual. the name for the individual is a keyword on the sources page but not on the (automatically generated) genealogy page. every name is also on the pages of the individual's parents, children, and siblings, but is not a keyword on these pages. the genealogy page is indexed by search engines but not the sources page.

i currently have 3019 individuals. one of them is Cheiron, that most noble of centaurs, and another is Dysnomia, or lawlessness. i get lots of hits from people searching for the medical condition dysnomia, and during the flu shot problems got a lot of hits from people searching for the flu shot manufacturer Cheiron.

a great number of things are named after greek gods, so you are definitely on to something here.



There is every possibility that these key words have a high price tag in the PPC world of advertising.

Net effect is to make these key words very attractive to the people who run the 'paid to read' programmes. It could well be that the pseudo directories that create the lists that are viewed and clicked by the PTR affiliates includes your url, especially if you appear in the top #20 on any search engine.


most of my pages are in the top ten at both google and yahoo. especially the obscure ones are the first both places. one name i get a unexplained number of hits on is Ucalegon. this might be why.



How can you find out if your site has been included in the pseudo directories and are being clicked by people who have not understood the difference between paid clicks and other listings. Well, you can't really. But you can get a feel of the number of pages that are including you by setting up your domain name as a google alert.


hum. well clearly i need to do an analysis of which search engine my weird hits are coming from.



The other possibility is that you are being hit by one of the scripts that is used in place of real visitors to mimic real clicks from the pseudo directories.

If you think this is the cause of your problem, you can find out who is bidding on the search terms by looking at the sponsored ads in google, yahoo, overture, etc when you use only the name of one of the gods as search term.
You could try asking them to not use your spelling of the name if they use a different spelling so that you cease to be plagued by false hits. Unfortunately there is little that you can do with words that are part of everyones language.

Good luck.

this is very interesting, and i didn't know anything about it.

the problem is not so much the first hit in the series but groups of them. i will frequently have repeated hits, someone going from page A to page B twenty times in one minute, sometimes several times in one second. this cannot possibly be happening. i have regular visitors (one guy from austria is up to 127 visits as of today) and i know what a normal pattern looks like. these endless repeated hits make no sense.

but everything you are saying is something i should explore. i always feel sorry for people looking for information on the medical condition dysnomia who wind up looking at greek gods.

so i will investigate this. thank you.

[DianeBouvier]

people who run the 'paid to read' programmes.


did some searching on these tonight. pyramid schemes, amazing. alta vista did something like this years ago and i never got any money from it.


But you can get a feel of the number of pages that are including you by setting up your domain name as a google alert.


just did this. very interesting, i didn't know about this, thank you. set up an alert on my domain name, on "genealogy of greek mythology", "genealogy of greek gods" which are the two things i get the most hits on from search engines.

my most strange thing today involved someone who hit on arene eight times in one minute. arene does not have a very interesting genealogy altho she did marry one of her brothers. but this visitor did not come from anyplace, so they must have had it bookmarked.

when i do a search on one of my names + genealogy i do not get any ads.


suspect i am going to be working on the information in your message for weeks.

[Moredial]

my most strange thing today involved someone who hit on arene eight times in one minute

If you search on 'arene' and 'arena' on Google you will see that the same ppc adverts appear. With 32 miliion web pages to compete against plus music and maganize publishers being involved, who know how much is bit on the top spot.

Espotting only charge £0.14 for top spot on arena so there is not big money to be made.
Overture offer $0.05 for arene and $0.15 on arena

Neither strike me as being enough money to be of interest to the ppc fraudsters.

Until later

[DianeBouvier]

still getting all those hits that aren't there. endless lists of someone going from page A to page B, from page A to page B, from page A to page B, sometimes a hundred times in less than one minute. they are not doing that, they CAN'T do that. even if they were the fastest clickers alive, they would still have to go from page B back to page A to get there.

also i consistently have exactly two more unique visitors than first time visitors. always. exactly a difference of two. no matter what the number is, there is exactly a difference of two. can anyone calculate the probabliity of this happening every single day for a year?

the reports here are absolutely splendid. would be very nice if they were based on real numbers.

set up a google alert a month ago and it has just started sending me reports. will try to see if there is any correlation between what they see and what statcounter is reporting.

as far as i can see, my numbers are entirely mythological.