W3C home > Mailing lists > Public > www-tag@w3.org > September 2006

RE: Proposed versioning diagram change for Language Descriptions

From: David Orchard <dorchard@bea.com>
Date: Tue, 12 Sep 2006 20:01:16 -0700
Message-ID: <E16EB59B8AEDF445B644617E3C1B3C9C023FD261@repbex01.amer.bea.com>
To: <noah_mendelsohn@us.ibm.com>, <www-tag@w3.org>

I don't see a lot of difference, and in fact I think it's almost going
back towards diagrams I had 2 years ago.  Which btw is what this whole
latest round of discussion is painfully feeling like.  

A couple of years ago, I had a diagram that had languages have
constraints, the constraints can be syntactic or semantic.  Seems an
awful lot like a "description" as you are using it.

Also, you removed the relationship from language to syntax, where syntax
constrains the text set.  Why did you not remove the relationship from
language to semantics?  You've kept languages relating to both semantics
and information set, which seems duplicate to me.  It seems to me like
you can have either: text set + information set, OR those 2 plus syntax
and semantics.


> -----Original Message-----
> From: www-tag-request@w3.org [mailto:www-tag-request@w3.org] 
> On Behalf Of noah_mendelsohn@us.ibm.com
> Sent: Tuesday, September 12, 2006 3:29 PM
> To: www-tag@w3.org
> Subject: Proposed versioning diagram change for Language Descriptions
> On the call last week we had some discussion of how language 
> descriptions would best fit into the UML diagram in the 
> versioning finding.  I've hacked up an edit that's pretty 
> close to showing what I've been thinking. 
> Basically, it tries to show that there can be multiple, 
> alternate (though hopefully consistent) descriptions of the 
> same language, and that none of those descriptions are 
> inherently "part of" the language in the way that the text 
> sets are.  Each language description is intended to include a 
> syntax (I.e. a specification of the set of legal texts) and a 
> semantics (which, pending agreement on the right way to say 
> this, I'll loosely describe as determining the information 
> that can be gleaned from each legal text).
> A few caveats:  (1) I'm not a UML expert so there may be some 
> misuse of UML conventions and (2) I don't have the diagram 
> editing tool, so I had to do some really ugly stuff to import 
> a bitmap image of the diagram into a bitmap editor, and hack 
> it up from there.  So, while it is being saved as a .png, in 
> fact it's probably just pixels and won't scale.  If it looks 
> weird, make sure your viewing tool is zoomed to 100%.  If 
> this or something similar meets with approval, someone with 
> the tool should redraw it properly (or point me to the tool 
> and I'll do it, if the learning curve isn't too bad.) 
> Thanks!
> Noah
> --------------------------------------
> Noah Mendelsohn 
> IBM Corporation
> One Rogers Street
> Cambridge, MA 02142
> 1-617-693-4036
> --------------------------------------
Received on Wednesday, 13 September 2006 03:02:02 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:50 UTC