> >
> > o To assist in identifying the profile of information flow
> > requirements from HEIs to the Hero portal (i.e. are most
Universities
> > like Cambridge?)
>
> The problem is... what does 'like Cambridge' mean? The survey, as
> presented, doesn't tell us whether Cambridge has one physical server
with
> 318 DNS aliases, one physical server hosting 318 virtual servers, 318
> different physical servers or (more likely) something in between.
>
> Andy
Is it possible for a robot to differentiate? Presumably an ethical
robot which wishes to spread its load across physical servers would have
to know this? Or, if, as I suspect, the robot exclusion policy and
guidelines for ethical robots was developed before virtual web server
standards were widely used, perhaps ethical robots will share the load
across different domain names?
If robots can't differentiate presumably this will mean that robots will
duplicate the indexing of site which have aliases for the same file
store (i.e. www.oxford.ac.uk and www.ox.ac.uk)? Is this a problem? I
have a feeling that it may be, as the use of automated agents, B2B
applications, etc. grows.
If nothing can be done from a remote user agent, do we need
organisations to produce collections / site maps of their servers?
Brian
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|