W3C home > Mailing lists > Public > public-rif-wg@w3.org > November 2006

Re: proposed: use abstract syntax notation (asn06)

From: Sandro Hawke <sandro@w3.org>
Date: Mon, 13 Nov 2006 22:49:00 -0500
To: Hassan At-Kaci <hak@ilog.com>
Cc: Dave Reynolds <der@hplb.hpl.hp.com>, Chris Welty <cawelty@frontiernet.net>, public-rif-wg@w3.org
Message-Id: <20061114034906.10C834F2D1@homer.w3.org>

> Good point. Seriously - why reinvent a YAAST (Yet Another Abstract
> Syntax Notation)?
> > ASN.1 is pretty complex but I guess it might be possible to make asn06
> > strict subset of ASN.1 then one could use the various ASN.1 tools for
> > generating concrete serializations using the standardized or custom
> > encoding rules.
> I think this is not as far-fetched as it sounds - with the caveat that
> ASN.1 lacks detailed descriptors for URI's. Still, we could express
> quite a chunk of what we need for RIF using a very small subset of ASN.1
> (http://www.itu.int/ITU-T/studygroups/com17/languages/X.680-0207.pdf).

Interesting.   <ponder, ponder>   asn06 seems simpler to me, and I like
the way it feels kind of like an ontology language, but ASN.1 looks more
applicable than I remembered it.   I don't think I'd object if everyone
else wants to use ASN.1.

> > Though I detect a lack of fondness in people's voices when they talk
> > about ASN.1 ...
> Yes - but is this any worse than when we hear people speak of XML Schemas
> or any of the W3C Language Mill's output? ;-) This being said, Dan Conolly's
> point about having to learn, and get familiar with, yet another notation is
> a valid one. But do we have any choice? One way or another we will have to
> bite the bullet and write down a sufficiently complete specification for
> rule interoperability - are we not? Any formal notation that is sufficient
> "up-to-isomorphism" will do. So why ahould be care so much more about HOW
> we describe - (E)BNF, AST, ASN.1, ... or whatever - than about WHAT we
> need to describe?
> Thus, I share Michael Kifer's frustration with this WG's relentless lack
> of focus on essential issues due it being constantly side-tracked on form
> rather than contents. Indeed, like the [in]famous mythical "Buridan's Ass"
> who died of hunger and thirst between a water bucket and a heap of hay just
> because he could not make up his mind on whether to eat of drink first, so
> are we talking past each other and running in circles ... :-(
> http://www.mon-expression.info/index.php/etre-comme-l-ane-de-buridan.
> PS/ Yes - this is a rethorical point - shall we move on? :-)

I also find the progress of the Working Group frustratingly slow, but I
feel like we've turned a corner recently.

I also don't have any desire to be having this e-mail discussion, but I
(obviously) think it's important to use something more like asn06 than
BNF, so I'm trying to make this as easy for everyone as I can, in the
hope that we'll have a quick and easy consensus path forward.  I am
willing to support asn06 and try refining it if people find
shortcomings.  So far, people have just been asking for clarification, I
think, and expressing broad frustration.

    -- Sandro
Received on Tuesday, 14 November 2006 03:49:34 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:47:41 UTC