LivelapBot

[keyplyr]

UA: LivelapBot/0.2 (http://site.livelap.com/crawler)
Protocol: HTTP/1.1
Robots.txt Yes
Host: OVH (multiple ranges)

IMO a good traffic source

[lucy24]

IMO a good traffic source

Good enough to merit poking a hole? Detour to raw logs suggests they have no interest* in me-- but coming from a shoot-to-kill range, I wouldn't have noticed them anyway.

:: unhappily envisioning a string of environmental variables along the lines of "really keep out, I don't care who they are" vs. "ask for ID" vs. "mobile, so look the other way" vs. "I know these guys, so ignore all anomalies" vs. etcetera, with ensuing need for cookies if any rules involve acceptable referers ::


* Where "no interest" = I found one occurrence in logs. The surrounding log entries suggest that someone had just tweeted-- or, uhm, whatever'd-- a particular page.

[keyplyr]

I let them through.

[keyplyr]

more...

Yes, a social syem bot = one of hundreds getting their links from social media instead of the WWW.

IMO some of these bots are bad agents, some beneficial & some very beneficial.

But just because a site does not engage in social media activity doesn"t mean these social stem agents can't bennefit a site.

[lucy24]

just because a site does not engage in social media activity doesn't mean these social stem agents can't benefit a site

Hey, I even let the Twitterbot visit my (otherwise 100% roboted-out) test site :) If someone blundered in and was amused enough to tweet, more power to 'em.

:: irritably wondering why facebook can't just fetch an image once and be done with it, instead of having to re-download every single time, just as if they were still hotlinking -- do they think that normal websites change every five minutes? ::

[keyplyr]

I agree, FB doers do that. For me, sometimes FB will re-request all the images from over a dozen pages of mine which were posted a month ago at FB, but they send me so much traffic I pretty much allow them to be bandwidth hogs. On average they keep things withing reason.

Just a FYI - I found it necessary to filter FB by IP range. I was finding impersonators using:
facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)
or just facebookexternalhit with valid referrers from FB posts. It seems FB allows bots to crawl even though they say they don't.

However most of these posers were coming from AWS, OVH, etc... so they got caught in my other filters prior to me installing the FB IP range condition.

[blend27]

--FB IP range condition

+ Santa!

keyplyr, sorry for a personal question, but do you have a "like" buttons on the sites where you let FB thru?

[keyplyr]

@ blend27 - I tried the FB Like API but removed it when I went responsive because:

1. It loaded slow
2. There was really no point unless you displayed the count (people like to like what other people like) and on popular pages that number grew to 4 or more digits interfering with the space between tap points.
3. Didn't like how they ended up looking.

Instead I use the FB *share* code but host my own static FB icon. When tapped/clicked the respective page is shared on the user's timeline along with a custom image designed for that purpose.

Note - apologies for the typos in earlier posts. When I post from my phablet I have trouble with the WW text editor. What I see is sometimes not what ends up posted (WYSINWYG)

[lucy24]

What I see is sometimes not what ends up posted

I recently got an email containing the word "shallowness". It seemed bizarre and inexplicable in context (we were talking about cats, and when would you ever describe a specific individual cat as "shallow"?). All was made clear when I got a follow-up email with the "### autocorrect" boilerplate and the explanation that she'd meant to say "mellowness".

This is really true.

It also raises the suspicion that autocorrect has gone right off the deep end.

[keyplyr]

when would you ever describe a specific individual cat as "shallow"?

Well, as a musician...

[lucy24]

Touché. (Assuming "musician" = "jazz musician".)

[keyplyr]

Is there any other? (that's rhetorical)

[blend27]

..... I tried the FB Like API but removed it when I went responsive because:

1. It loaded slow
2. There was really no point unless you displayed the count (people like to like what other people like) and on popular pages that number grew to 4 or more digits interfering with the space between tap points.
3. Didn't like how they ended up looking.

Instead I use the FB *share* code but host my own static FB icon. When tapped/clicked the respective page is shared on the user's timeline along with a custom image designed for that purpose.

That above should go as a BIG MEMO to all Webmasters.

[keyplyr]

That above should go as a BIG MEMO to all Webmasters.

Isn't that the popular sentiment with *all* my posts?
(that's rhetorical, no need to answer)

[blend27]

We are getting there :)