Subject: Re: XSL Trans From: Paul Prescod <papresco@xxxxxxxxxxxxxxxx> Date: Mon, 28 Sep 1998 07:53:42 -0500 |
Daniel Glazman wrote: > > Paul Prescod wrote: > > > An extreme example is a stylesheet that > > removes information. How can you edit documents that use this stylesheet > > in a "WYSIWYG" view? Clearly we are going to need different stylesheets > > (and perhaps a different stylesheet *language*) for word processors. > > [ A stylesheet never removes information. A stylesheet may define a null ] > [ rendering for a given substructure. It makes a slight but important ] > [ difference. ] The transformation removed information. It may not have deleted information in the original tree, but it removed information during the transformation. (as a conversion from GIF to JPEG removes information) > 2 stylesheets ? Why not ? And so what ? Grif for instance already did that > ten years ago ! CSS linking mechanism into HTML also offers this possibility. The problem is not that it is hard to support two stylesheets. The problem is that it is no longer "WYSIWYG." People will have to give up expectations of WYSIWYG. The best they can hope for is "conveniently rendered." I'm curious: how powerful are the transformations that Grif allows? I haven't yet seen a truly transformation based word processor. > To answer your point : if stylesheet or transformation rules make unavailable for > editing purpose a subtree, this subtree must not be edited in a wysiwyg > environment. If author wants to provide a way to edit this subtree, *he* has to > adapt transformations/styles according to this need. Providing two different > stylesheets is a common answer to the problem. Forcing word processor users to tweak transformation rules does not seem like a good way to make XSL popular. Let's just say that if I was a word processor vendor, I would be thinking hard about whether XSL is really worth the effort. A non-transformational style language is probably more appropriate for that environment and easier to implement, to boot. Perhaps the other style language could be called "monastic XSL." In monastic XSL, nothing would ever be removed or duplicated in the tree. Node insertions would be allowed, but the inserted nodes would be non-editable. CSS or FOSI would be appropriate for this purpose, but monastic XSL might be more popular if it was a proper subset of XSL. If I was going to provide full XSL in a word processor, I might do it in a separate "rendered view" window. > All MsWord users already do that : if you insert columns in a text, you have to > switch to paginated mode to see them in wysiwyg. In page layout view, Word provides a completely editable, completely wysiwyg view of columns. Paul Prescod - http://itrc.uwaterloo.ca/~papresco How many of the Congresspeople who voted for the CDA do you suppose also voted to release the report that reads like a borderline por- nographic dime-store romance written by a Texas preacher's son? - Keith Dawson, TBTF http://www.tbtf.com/archive/09-14-98.html http://www.tbtf.com/resource/hypocrites.html XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: XSL Trans, Daniel Glazman | Thread | Re: XSL Trans, Dave Peterson |
Re: XSL Trans, Daniel Glazman | Date | Re: XSL:USE ?, James Clark |
Month |