Re: Lycos' HEAD vs. GET

Klaus Johannes Rusch (e8726057@student.tuwien.ac.at)
Sat, 23 Nov 1996 23:47:56 CET

In <9611231843.AA30830@panther.middlebury.edu>, otisg@panther.middlebury.edu (Otis Gospodnetic) writes:
> I looked at our access logs here and noticed that Lycos' Web Robot(s??) uses
> GET to request all non-image files/documents (.html. .txt, .ps....) and uses
> HEAD to request various image files (.gif, .jpeg...)
>
> what I don't understand - why doing even HEAD on the image files ? What is
> the benefit of verifying the existance of images on the Web ? I can't think
> of any .....
>
> Does anyone know the reasoning behind Lycos' HEAD requests for images ?

Not sure about Lycos, but some engines do index images also, based on links to
the images, e.g. if one of your documents contains

<a href="http://www.atmedia.net/KlausRusch/images/klaus.gif">Klaus in Finland</a>

a search for "Klaus Finland" will return the image also.
Looking through my logs I see some people access individual images but not the
referring web page so images must be indexed.

On the other hand GETting the whole image wouldn't make sense, so HEAD seems
perfectly fine.

Klaus Johannes Rusch

--
e8726057@student.tuwien.ac.at, KlausRusch@atmedia.net
http://www.atmedia.net/KlausRusch/
_________________________________________________
This messages was sent by the robots mailing list. To unsubscribe, send mail
to robots-request@webcrawler.com with the word "unsubscribe" in the body.
For more info see http://info.webcrawler.com/mak/projects/robots/robots.html