W3C home > Mailing lists > Public > www-zig@w3.org > September 2001

RE: ZNG dicussion

From: Pieter Van Lierop <pvanlierop@geac.fr>
Date: Thu, 27 Sep 2001 13:50:42 +0200
Message-ID: <91C02F76882CD2119EAE00805F851D850137E295@paris.geac.fr>
To: "'Sebastian Hammer'" <quinn@indexdata.dk>, www-zig@w3.org
Well Yes and No.

Yes because as Sebastian says, language is not only a linguistic or a
computer problem, but also a cultural problem. One of the reasons for the
popularity of Z39.50 in Europe is certainly that Z39.50 has no language.
Apart from an additional diagnostic text here and there, the Z39.50 messages
are completely language-free. Isn't that great?

No, because "Title" is more comprehensible than 4, even for French or Dutch
people. And if you want to do something "quick & dirty" it is easier to be
able to read what is being sent.

Thinking this over -- is the mystifying complexity of Z39.50 not one of it's
attractions?

Pieter

> -----Original Message-----
> From: Sebastian Hammer [mailto:quinn@indexdata.dk]
> Sent: jeudi 27 septembre 2001 13:22
> To: www-zig@w3.org
> Subject: Re: ZNG dicussion
> 
> 
> At 11:59 27-09-2001 +0100, Mike Taylor wrote:
> 
> > > * Numeric values for attributes is ugly - there is no 
> standardized human
> > >   readable query language.
> >
> >Totally agree, that's a part of the reason that I am keen on the
> >current initiative to design a human-readable query langauge (BUT not
> >to compromise on expressiveness in doing so!)
> 
> This argument popped up when poor old GRS-1 got the axe in 
> the Bath profile 
> as well... and while I as a fluent English-speaker can 
> sympathise, I also 
> get a feeling that "human-readable" is a relative term. 
> There's a lot of 
> people (including software programmers) out there for whom 
> English is a 
> third language or less... I try to imagine if I would still 
> enjoy working 
> with Z39.50 if the Bib-1 attribute set had been cast in 
> Spanish or German 
> (both languages I have kind of a painful acquaintance with) 
> instead of 
> numbers. The jury is still out.
> 
> I think in a global world, there is a lot to be said for 
> linguistic and 
> cultural neutrality, at least when broad interoperability is the goal.
> 
> Personally, I find the readability of the query language to 
> be of little 
> consequence.. while SQL is human-readable, it's still not 
> something you 
> normally subject your users to, and since Z39.50 (any 
> generation) is not 
> about database management, we spend a lot less type pounding 
> queries into 
> low-level tools than your average RDBMS jockey. What you want 
> is a language 
> that's easy to parse and easy to render.. to me, that's a big 
> plus with a 
> "Polish" notation over something like CCL, even though prefix 
> or suffix 
> notation are definitely not end-user friendly.
> 
> --Sebastian
> 
Received on Thursday, 27 September 2001 08:00:54 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:13:27 UTC