[rfc-i] Comment on Holistic archiving (section 2.4.3 of draft-flanagan-rfc-preservation-01)

Henning G Schulzrinne hgs at cs.columbia.edu
Tue Oct 28 12:37:35 PDT 2014


Also, unlike MS Word or PDF, the structure of XML is documented, in
sufficient detail to write a parser, in hundreds of widely-published books.
(Leaving aside that modern Word files are also XML.) If all of these books
were to disappear without a physical or electronic trace, I suspect at that
point, deciphering old RFCs may not be exactly the highest priority for
humanity and we're in a "The Road" scenario.

On Tue, Oct 28, 2014 at 3:00 PM, Robert Sparks <rjsparks at nostrum.com> wrote:

> The paragraph that starts out with "Consider a future where XML has been
> obsoleted for half a century" has a rough edge I would like to try to make
> better.
> It talks about the need to keep an environment (programs and OSes) around
> to be able to use whatever bits you have stored. While that's true for the
> general case, it overstates what would be needed to be able to read an XML
> file in the future.
>
> Unlike some compressed or specialized binary format, the bare XML source
> is relatively directly accessible by humans. The environment you need to
> keep looks more like the environment you need to keep to be able to read
> the current ascii RFC files than it does one you would need to keep to
> read, say PDF,
> as long as you keep the XML files stored in a way that is amenable to
> letting people see the characters easily. That is, the target is more like
> 'cat' than it is 'adobe reader'.
>
> We're choosing the element names in the XML to be relatively self
> descriptive. In an archival-recovery situation, a person would have a
> relatively easy time determining what the bits mean, _especially_ if we
> keep the definition of the elements around in the same easy to get to
> format.
>
> Please be careful with choosing what the archived bits really need to be -
> I think we have an opportunity to avoid much of the complexity the current
> text warns about in this section.
>
> RjS
> _______________________________________________
> rfc-interest mailing list
> rfc-interest at rfc-editor.org
> https://www.rfc-editor.org/mailman/listinfo/rfc-interest
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rfc-editor.org/pipermail/rfc-interest/attachments/20141028/002530d9/attachment.html>


More information about the rfc-interest mailing list