RFC on routines and mode Pos

[Nelson H. F. Beebe]

Chris Hermansen is working on implementing the Algol 68 transput model
and asks today:

>> Would anyone prefer to mostly or entirely chuck the whole (p,l,c) ....

The idea of fixed-length lines, page markers, file content page sizes,
and preallocated fixed-length files are traits of (mostly) long-gone
filesystems.

I personally don't see much need to support the (p,l,c) concept in van
Vliet's implementation, or the Algol 68 Revised Report.  Filesystems
even in those days were highly operating system dependent, and the
earlier Algol 60 left I/O entirely unspecified, and up to compiler
implementors (which killed code portability).

The Unix and C model of files as streams of 0 or more characters seems
far more sensible today.  Any further structuring into "lines" or
"pages" or "blocks" is an issue for a higher level.  The key
primitives are C's fgetc() and fputc().  Everything above that can
build on them.  Yes, C also has [f]read() and [f]write() for handling
multiple bytes at a time, but that is really only an efficiency issue
(albeit an important one) to reduce system call overhead.

-------------------------------------------------------------------------------
- Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
- University of Utah                                                          -
- Department of Mathematics, 110 LCB    Internet e-mail: beebe@math.utah.edu  -
- 155 S 1400 E RM 233                       beebe@acm.org  beebe@computer.org -
- Salt Lake City, UT 84112-0090, USA    URL: https://www.math.utah.edu/~beebe -
-------------------------------------------------------------------------------