[
Lists Home |
Date Index |
Thread Index
]
Michael Kay <michael.h.kay@ntlworld.com> writes:
> > >
> > > OK, so how would you archive the results of a national
> > > census? That's about as structured as you get. I would do
> it in XML.
> >
> > There are really three issues implied in this question: archive and
> > public and private access.
>
> You might read three questions but I only wrote one. I asked
> about archive.
Ok. Archive for who?
> > If one has to mix the concerns
> > then XML is
> > as good as anything. Personally, I'd rather separate them,
> generating
> > XML for public access as needed.
>
> You don't have to mix the concerns, and I didn't... I was
> asking about archive.
Archive isn't an isolated capability. Archive implies capability for
usage. Usage implies a mix of concerns. Does placing a inscribed stone
tablet in a air tight iron box meet the requirements of archive if no
one knows how to interprett the inscriptions on the stone tablet at any
point in time?
> > Personally, I'd want to know more about any application
> requirements
> > before I'd say that XML alone would be sufficient.
>
> The requirement for archiving a national census is to
> maximize the probability of being able to extract the
> information, without loss, and to minimize the cost of doing
> so, in say 100 years' time.
Given this requirement I'd tend to stay away from any computer based
media or formats XML or otherwise...
|
