Print

Print


I've started a BNF description and when I get about halfway done I'll post it. I'd say I've done about a third of it at this point.   --Ray

-----Original Message-----
From: Discussion of the Developing Date/Time Standards [mailto:[log in to unmask]] On Behalf Of Bruce D'Arcus
Sent: Saturday, March 26, 2011 9:36 AM
To: [log in to unmask]
Subject: Re: [DATETIME] Looking for a formal specification and implementation

Ray said (very recently) he's working on it.

On Fri, Mar 25, 2011 at 8:23 PM, Jakob Voss <[log in to unmask]> wrote:
> Hi,
>
> I must admit that I have not followed the discussion the last month, so
> I may have missed something. The current draft at
> http://www.loc.gov/standards/datetime/spec.html looks better than a few
> month ago, thanks to all who contributed in improving the planned
> standard.
>
> However, in the current form the draft is just a list of examples with
> some description. Some of us already asked for a formal specification in
> Backus-Naur-Form or about an implementation, but it looks like the
> current standardization process is going on detached from technical
> feasability (?). I fear that without a formal specification and some
> implementation, the format will end up as one-way data hole or as
> incomplete boilerplate of faulty applications.
>
> I started to write a parser, based on a Parsing Expression Grammar (PEG)
> - the most difficult part seem to be the gaps and inconsistencies in the
> current draft. Maybe this project is of use for anyone interested to dig
> deeper: https://github.com/nichtich/lua-edtf I stopped the work after a
> got the basic framework because my main interest is not EDTF. If you
> don't like Lua or PEG, you can also use another formalism, but EDTF
> needs a formal definition or you cannot call it a "standard" or
> "specification" at all.
>
>
>
> Jakob Voss
>
>
> --
> Verbundzentrale des GBV (VZG)
> Digitale Bibliothek - Jakob Voß
> Platz der Goettinger Sieben 1
> 37073 Goettingen - Germany
> +49 (0)551 39-10242
> http://www.gbv.de
> [log in to unmask]
>