[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Mon, 21 Oct 1996 10:55:58 +0000 (GMT)
I wonder if it is possible to develop a text browser that can interpret
SGML documents. My experience of SGML is that the browser fetches about
4-6 documents, and forms references among them. This creates a very
navigable system. On the left panel in many implementations I've seen, an
expandable/collapsable outline forms, which points to areas of the
document on the right.
An equivalent may be to have a web browser that takes a document and forms
two documents with it. The first would list any attribute summaries you
wanted, while the second would contain the text in full. For example, you
may want to see all <H#> tagged data, and use that as a basis for linking
to the actual text. This browser would pluck those headers, and allow you
to select any of them, taking you to the corresponding text.
Is this what is meant by CSS?
It may be that GUI browsers will leave lynx behind in substantive ways at
some point. At present I don't feel deprived that I have to go through a
rigamarole to see 'frames' or that I can't see Java or VirusX. But if GUI
browsers start helping you to organize downloaded pages automatically,
then I think the difference becomes considerable.
But the difference is unnecessary. The fact is that UNIX text-mode
programs can easily support meta-organization. It's just text after all.
In fact we ought to be able to do better than the GUI people.
; To UNSUBSCRIBE: Send a mail message to address@hidden
; with "unsubscribe lynx-dev" (without the
; quotation marks) on a line by itself.
- LYNX-DEV SGML,
Carl Reimann <=