[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
RE: Transclusion Issues
- To: xanadu@xxxxxxxxxxxxx
- Subject: RE: Transclusion Issues
- From: Ted Nelson <ted@xxxxxxxxxxxxxxxxx>
- Date: Fri, 22 Aug 1997 00:18:12 -0700
- In-reply-to: <01BCAE17.BF20A600@xxxxxxxxxxxxxxxxxxx>
- Reply-to: xanadu@xxxxxxxxxxxxxxxxx
- Sender: xanni@xxxxxxxxxxxxxxxxx
I expect to present views on all these matters
in my forthcoming article, "Embedded Markup Considered Harmful."
You should get it within a month.
[address at bottom]
>I have been looking at some of the issues involved surrounding
>transclusion lately and I thought I would bounce some of the issues off of
>the list to see what you other Xanies think.
>If text is transcluded from another point, there seems to me to be a
>number of problems surrounding the markup from the original text.
>1) Should the markup from the original source be maintained and exibit its
> attributes in the text that transcludes it? For example, if the text
> is in normal face font and the text transcludes another text that
> happens to be bold, should the bold be maintained?
>2) Should the author which creates a text which transcludes another text
> be capable of modifying the attributes of the transcluded text? (I.e.,
> should the author of the primary text be able to add bold, italic, etc.
> to the transcluded text?) One argument would be that if a author can
> add attributes to the transcluded text, the author is simply modifying
> its format to fit the current textual presentation. OTOH, this could
> intefere with the "art" of writing the transcluded text. (Some people
> might get a bit sensitive of people bolding, italicizing, changing
> the fonts, whatever, of the transcluded text.
>[Jay Osako] This is a difficult one in regards to HTML. If I'm correct,
>in Xanadu the issue is one of front ends - since markup is handled
>as links, rather than inline, it the decision of the documenter (or at
>least the programmer writing the viewer) whether to keep the font and
>sizing properties, by keeping or eliminating the links. With in-band
>markup, its harder to allow that flexibility; you'd have to either set a
>fixed policy, leavingone side or another dissatisfied, or else have a
>complex system of explicit enable/disable markups. Moreover (apropos
>your question below) if your including the markup, you have to process
>the whole document at least once to get the correct markup (the server
>will probably cache both the transclusion and the markup for it, but
>that just move the issue to a different level and to some extent defeats
>the purpose). It probably would be best to have no carryover as a default,
>as in Xanadu; the content, after all, should be the message, MacLuhan
>notwithstanding. What's *really* complex is the handling things like
>Java applets and other forms of active pages; I can't really think of
>a suitable solution in the general case for HTML.
>3) What granularity should the transcluded text maintain? What I mean by
>this is that there is a strong movement towards marking up the structure
>of a document and then assigning the fonts and other textual attributes
>based on the structure based on this -- SGML like (For example, all
>headers are Palentino, 25 pt, Bold.) Should the transcluded text have to
>maintain an entire document structure? For example, if you want to
>included part of a header, should you be able to included only part of it
>(say, 3-4 words) or should you have to include all of it since you are
>including the "header" and to prevent the meaning from being manipulated
>in the presentation? On one side, you could argue that by forcing a author
>to include an entire document structure (header, paragraph, footnote,
>etc.) it avoids words being taken out of context and if someone includes
>something they are including a "whole" something. On the other side, it
>could become quite combersome to have to do so.
>[Jay Osako] My view is that, if possible, only the transcluded material
>should be picked up, but that it should have a backlink to the source.
>Most browsers would probably pre-cache the source (esp if it had to
>process it for markup anyway), and the link should spawn a seperate
>browser window to allow comparisons, but those are front-end issues
>and different browsers will probably handle them differently.
>There are a number of other issues along these lines.
>I would be interested to hear if anyone else has thought about these
>issues. (And what they thought. :)
In USA until 10 September 1997:
Ted Nelson, Project Xanadu, 3020 Bridgeway #295, Sausalito CA
Tel. 415/ 331-4422, fax 415/ 332-0136
In Japan after 10 September 1997:
Theodor Holm Nelson, Visiting Professor of Environmental Information
Keio University, Shonan Fujisawa Campus, Fujisawa, Japan
http://www.sfc.keio.ac.jp/~ted/ PERMANENT E-MAIL: ted@xxxxxxxxxx
Home Fax: 0466-46-7368 From USA: 011-81-466-46-7368