[Info-vax] Open Source on OpenVMS Conference Call - 08:00 EDT 16 Sep 2021

[Phillip Helbig undress to reply]

In article <shf7r3$14pf$1 at gioia.aioe.org>, Joukj
<joukj at hrem.nano.tudelft.nl> writes: 

> > Two approaches are worth considering.  One would be getting the TeXLive 
> > distribution to build out of the box on VMS.  The other would be to have 
> > a minimal distribution which would automatically fetch anything found to 
> > be missing from a CTAN mirror somewhere.
> > 
> Good idea Phillip. Texlive on OpenVMS would be best, since than it will 
> be "compatible" with linux systems 

I looked at this a while back and seemed to remember that various GNU
utilities were used for packaging and so on.  Probably most of those
(and other non-GNU third-party stuff such as (UN)ZIP) run on VMS, so I
don't think that it is a big problem.  Disk are cheap these days, so it 
shouldn't be a problem to have the whole distribution locally.  It is 
pruned down to fit on a DVD, though; there is more at CTAN (mostly 
documentation and fonts which most users don't need).  Apparently there 
are some low-bandwidth users who prefer the DVD rather than downloading 
from CTAN.

> and one will be able to exchange all 
> documents.Writing documents is always collaborating with others 
> (probably not on OpenVMS)

Once LaTeX is up and running on VMS, compatibility is not a problem.  
I've been using LaTeX on VMS for almost 30 years and it is no problem to 
process a document from elsewhere or have someone process my documents.
I convert to PostScript then to PDF and both of those are portable 
formats.  So both input and output formats are portable.  The problem is 
an inital install on VMS, and some sort of automatic update.  The 
freeware CD with [TEXMF...] put together by Ralf Gärtner might be a good 
starting point.

One caveat is that TeX has now forked and there are various versions 
(though not allowed to be called TeX).  One has better justification.  
(Knuth's---good---idea was that a given input document should always 
produce identical output with TeX, which is why some variations can't be 
called TeX).  One allows UTF-8 input characters.  That might be a 
problem on VMS.  DEC MCS is pretty close, but probably not close enough.
But the conventional TeX with only 7-bit-ASCII input is still supported.