alternatives to the concept of a page, Gutenberg press vs LCD screen

[Mike Marchywka]

_____
>From: Doug McKenna <doug at mathemaesthetics.com>
>Sent: Thursday, August 29, 2019 1:32 PM
>To: Mike Marchywka
>Cc: texhax
>Subject: Re: alternatives to the concept of a page, Gutenberg press vs LCD screen
>
>Mike Marchywka asked:
>
>>| Has anyone had a reason to typeset a scroll or
>>| other unnatural page formats?
>
>Two thoughts:
>
>1. My recently released eBook/app, "Hilbert Curves" for iOS, which I demoed at the recent TUG conference a few weeks ago, self-typesets each time the app launches, using its own re-implementation of a TeX language interpreter.  The typsetting relies on plain.tex and opmac.tex, not LaTeX, for the markup on a TeX source file for the book's text and illustration placement.  Because I wanted most illustrations to be as large as possible on any page, I purposefully set the logical page size to a longer-than-usual amount compared to that of physical paper, so as to help prevent the otherwise somewhat chaotic placement of large illustration boxes (a well-known problem with TeX's float algorithms).  The user can scroll vertically because pages that fit horizontally don't fit vertically (and the app has to work on a variety of devices all with different screen sizes and orientations).  In some respects, I would have liked to get rid of pages altogether, in favor of one long scroll.  But this would have interfered with the need to keep the eBook/app's memory requirements under control, since nearly all scrollable views have a backing store to them these days (at least on Apple devices), especially views that require deeply zoomable, resolution-independent graphics.
>

Do you have a paper and link that describes this? I think when you mentioned the figures before they were regenerated
or something directly from the equations you wanted to illustrate- is that true or how close could you get to that?
Image generation while not strictly a part of tex does seem to be a big deal and SVG makes adaptation easier but if there
is some predictable and tractable way to regenerate a class of illustrations that may be nice to have. Can Tex integrate "R" code lol?


>2. In the legal profession, there is a constant annoyance of having to cite to a legal case using page numbers.  The problem is that legal cases get published in different volumes with different pagination.  Many forward thinking courts are doing away with page numbers in favor of numbered paragraphs.  Because paragraph numbers are embedded in the text, rather than imposed externally by a changing page or font size, numbered paragraphs are just a better and higher-resolution way of citing to somewhere in a published legal decision (or any other kind of writing, I suppose).  Eventually, pagination might become irrelevant in legal writing, which means using a single-scrolling interface for one super-long "page" would work well (assuming memory requirements can be handled).

I have mentioned this in my other posts, with my play code users define blocks that have serial numbers appended and the
viewer can pick a group of blocks to view, cite, or share etc. I guess at issue is how different this is from WYSIWYG Tex.
Presiumably users don't want to edit Tex but the view/navigate code is sort of giving them limited options of this type.
This is also nice as it pretty much means people need to distribute Tex source and if properly written it is human readable
in "vi" ( without the evils of proportional space fonts and if you want the equations there is emacs and auctex lol). 

I like the dvi as it is easy to modify but curious what is already supported in pdf viewers.



>
>Doug McKenna
>Mathemaesthetics, Inc.