the other end of the timescale

[Rob Lopresti <rob.lopresti@wwu.edu>]

We have read a lot of concern about getting information out 
quickly.  I worry a lot about the other end of the timescale.

There are many sites where data is simply removed when it
gets to be X years/months/days old.  If the material was
never sent to libraries (as is more and more the case) then
it can be gone forever.

There is also the fact that an agency can decide to eliminate
a publication because it is embarassing, doesn't reflect a
new policy, etc.

My (possibly incorrect) understanding is that at least 
two government agencies (Agriculture and Dept of State) have 
made arrangements with universities to make complete copies 
of those agency's websites, so that material can't be lost.

Is EPA doing anything of the kind?  Or are we relying on
EPA to decide for how long data will be available?

To avoid the building of straw men here, I will cheerfully
grant any agency the right to remove material they discover
to be factually incorrect, libelous, or dangerous.  But I
don't think they should have the right to make other categories
of material disappear from the record forever.

By the way, I am enjoying this discussion tremendously and
getting a lot out of it.