- From: Florent Georges <fgeorges@fgeorges.org>
- Date: Thu, 14 Mar 2013 17:23:46 +0100
- To: Jirka Kosek <jirka@kosek.cz>
- Cc: Christian Grün <christian.gruen@gmail.com>, EXPath <public-expath@w3.org>, Michael Kay <mike@saxonica.com>, XPath <w3c-xsl-query@w3.org>, XSL Working Group <w3c-xsl-wg@w3.org>
On 14 March 2013 11:05, Jirka Kosek wrote: > On 13.3.2013 15:18, Christian Grün wrote: >> - Most EXPath specs and vendor-specific modules avoid >> redundancies in the prefixes and the function names. This is >> why I would suggest to remove the "binary-" string in some of >> the function names and instead rename the prefix to "binary". >> The resulting names could be called "binary:length", >> "binary:or", etc. > This looks tempting. However my only concern is that if this > binary module is adopted as built in for some future version of > XPath/XQuery/XSLT then it would be more natural to allow to use > functions without namespace prefixes. By removing binary- > prefix we will get naming clashes with several existing XPath > functions. My bet is that if the module is adopted in F&O 3.1 or in XSLT 3.0, they will remain in their own namespace, like math:* and map:* functions. Even if it is not the case, as soon as the namespace changes, the WG can change the local name as well, the effect is the same: the name is not the same anymore. So I feel like the possibility of being adopted by F&O and/or XSLT should not affect the naming conventions for the EXPath- version of the module. I personally like bin:length, bin:or, bin:and, etc. Regards, -- Florent Georges http://fgeorges.org/ http://h2oconsulting.be/
Received on Thursday, 14 March 2013 16:24:42 UTC