Re: Proposed new requirement: Ability to produce textual output
"Dmitry" <[email protected]> Wed, 2 Jul 2003 00:18:55 -0400
| Newsgroups | gmane.text.xml.xtm.tmql |
|---|---|
| Message-ID | <00cd01c34051$0d5ea160$7d01a8c0@dbn> |
From: "Lars Marius Garshol"To: <[email protected]> Sent: Tuesday, July 01, 2003 11:20 AM Subject: Re: [tmql-wg] Proposed new requirement: Ability to produce textual output > Here's an attempt to summarize the thread: > > - There seems to be agreement that it would be good for TMQL to be > able to produce output. Unless anybody screams I'll put this in as > a SHALL requirement (other alternatives would be SHOULD or MAY, but > there seems no reason to choose those if we all agree). It depends on what you mean by "output". I personally in favour of "basic" TMQL which transforms/filters TM into TM based on SAM. As soon as we have filtered TM we can apply different plug-ins to get real "output". Some TM manipulation languages can have these features bundled together. In other cases we can rely on regular languages "glue". I also see huge value in developing something like TMTL presented in http://www.ontopia.net/topicmaps/materials/tolog-for-tmql.pdf . I just think that well defined TM->TMQL->TM transformation is much more important than TM->TMQL->XML, TM->TMQL->HTML or TM->TMQL->Text. Dmitry