[WEB4LIB] web cache and remote access

[Dan Lester]

Tuesday, May 30, 2000, 2:43:22 PM, you wrote:

BHD> Their argument:
BHD> They argue that 80% of web page requests are for static pages. The web
BHD> caching system is therefor supposed to increase network speed by serving up
BHD> a cached version of that page the next time it is requested, thus reducing
BHD> network traffic.

They are correct in general.  They're not correct in the case of
libraries, but you are.

BHD> It seems to me that, for most libraries, the opposite is true -  80% or more
BHD> of a libraries' Internet activity goes to online subscription databases
BHD> where users perform dynamic, unique searches every time.  Even as I type I
BHD> am getting calls from our users who are being prompted for passwords where
BHD> they should be authenticated by IP.

We faced the same problem at Boise State and they solved the problem
in a much simpler way.  They don't cache anything with a referrer of
library.boisestate.edu/*

BHD> Has anyone else encountered this problem. Should it even be a problem?

Well, should be or not, caching is the way of the world, for pretty
good reasons from the ISP's point of view.  Other people to whom it
causes problems are the websites that don't get a full count of the
number of hits on their sites.  If you're Yahoo or CNN or someone like
that and want to be in the "Top Ten", you care.  Of course all of them
face the same issue, so it may all come out even in the long run.

As far as not notifying you, I assume you aren't surprised. First,
they don't think in terms of customers, and particularly libraries.
Second, they probably don't have a clue that anyone is doing IP
authentication.  Neither of these are excuses...just descriptions.

cheers

dan

-- 
Dan Lester  dan at RiverOfData.com
3577 East Pecan, Boise, ID 83716-7115 USA
www.riverofdata.com  www.postcard.org  www.gailndan.com