Lout (Was: PostScript?)

["Valeriy E. Ushakov" <uwe@xxxxxxxxxxx>]

On Tue, Sep 01, 1998 at 03:03:32PM +0100, Sebastian Rahtz wrote:

> Wouldnt lout be a better source of material, by the way?

I have spent some time experimenting (an year ago) and I can see two
technical problems with targeting Lout:

    . implementing inherited characteristics
    . spacing model

ICs, as I understand it, is a form of dynamic scoping within otherwise
lexically scoped Scheme.  Note that most of ICs capture "style"
information.  Similarly, Lout is lexically scoped and it also has,
though a very limited, set of dynamically scoped "style" information
which is overall very similar to the concept of IC.  Unfortunately,
since Lout's set of dynamically scoped style parameters is very
limited (font, language, paragraph style and few others), it's
impossible to map DSSSL ICs into this set.  It's theoretically
possible to implement ICs via cross-references, but this is infeasible
in practice because of limitations of current Lout implementation.

Jeff Kingston, author of Lout, wrote me that he agreed with the
concept of inherited characteristics, and thought that Lout would be a
better language if it had fully fledged ICs.  So this limitation might
well be overcome in some future version of Lout.


As for the spacing model, I don't think it's possible to implement
display-space and inline-space (12.5.4) directly in Lout, especially
display-space priorities.  This problem seems harder to resolve.


PS: Lout home page is http://www.ptc.spbu.ru/~uwe/lout/lout.html
    If you have any questions regarding Lout, don't hesitate to
    contact me.

SY, Uwe
-- 
uwe@xxxxxxxxxxx                         |       Zu Grunde kommen
http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist