You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Maj Justin Seiferth <se...@www.disa.mil> on 1997/09/26 03:45:28 UTC

[Fwd: [Fwd: Browser capabilities?]]

Do we need to get Netscape's ear?  Hear it is- it someone will forward
the suggestion to me I'll work to get it into the feature list.

justin

>   Presume you're aware of this issue if not let me know.  Hate to see
> > MSIE getting three cheers for anything.
>
> It sounds as if you're asking if Communicator could send out one user
> agent when doing normal web surfing, and one when doing off-line
> browsing (i.e., what Netcaster does).  It that what the issue is?
>
> If so, I'd be more than glad to send a note to the Netcaster product
> manager and submit this as a request for enhancement to the next
> Netcaster release.  However I'm interested: did the discussion clarify
> exactly why this was a desired capability?  (I can guess, but I thought
> you'd be able to give me more info.)
>


Re: [Fwd: [Fwd: Browser capabilities?]]

Posted by Rob Hartill <ro...@imdb.com>.
On Tue, 30 Sep 1997, Dean Gaudet wrote:

> The desire is to be able to distinguish between clients which are being
> controlled by a human at that moment, and clients which are uncontrolled. 

[snip]
 
> But mostly it's 'cause of ad impressions (at least for the folks I deal
> with).  Rob probably wants it mostly for robot control reasons. 

I do, and I'm not alone in wanting it for that reason. Uncontrolled
crawling can do enormous damage if it doesn't comply with the
voluntary robot exclusion protocol and the guidelines for responsible
crawling.

Ok guys, what else is on your Netscape wish-list that we can cheer
MS for already doing ?  :-)

--
Rob Hartill                              Internet Movie Database (Ltd)
http://www.moviedatabase.com/   .. a site for sore eyes.

 


Re: [Fwd: [Fwd: Browser capabilities?]]

Posted by Dean Gaudet <dg...@arctic.org>.
The desire is to be able to distinguish between clients which are being
controlled by a human at that moment, and clients which are uncontrolled. 
The former are considered for "ad impressions", the latter are considered
for robot control rules (such as "hey you crawler, I told you not to go
there and now you're recursing infinitely! stop!").  Yup the latter can
also be done somewhat well enough with robots.txt ... but if the
user-agent distinguishes between the two modes then a bug in a distributed
browser implementation does not become a problem.

But mostly it's 'cause of ad impressions (at least for the folks I deal
with).  Rob probably wants it mostly for robot control reasons. 

If you can forward that on to netscape's techies that's great.  I know
folks at one of the content partners that have been trying to get their
rep (who I'm guessing is not tech) to understand the issue. 

Dean

On Thu, 25 Sep 1997, Maj Justin Seiferth wrote:

> Do we need to get Netscape's ear?  Hear it is- it someone will forward
> the suggestion to me I'll work to get it into the feature list.
> 
> justin
> 
> >   Presume you're aware of this issue if not let me know.  Hate to see
> > > MSIE getting three cheers for anything.
> >
> > It sounds as if you're asking if Communicator could send out one user
> > agent when doing normal web surfing, and one when doing off-line
> > browsing (i.e., what Netcaster does).  It that what the issue is?
> >
> > If so, I'd be more than glad to send a note to the Netcaster product
> > manager and submit this as a request for enhancement to the next
> > Netcaster release.  However I'm interested: did the discussion clarify
> > exactly why this was a desired capability?  (I can guess, but I thought
> > you'd be able to give me more info.)
> >
> 
>