[no subject]

[Rob Austein <SRA@XX.LCS.MIT.EDU>]

    Date: Wednesday, 23 September 1987  17:10-EDT
    From: "Patrick G. Sobalvarro" <PGS@AI.AI.MIT.EDU>
    To:   BUG-EMACS@AI.AI.MIT.EDU, bug-system@XX.LCS.MIT.EDU

    OZ seems to be running EMACS 165, where XX is running a version 169.
    However, the version running on XX doesn't seem to be able to read JCL; from
    the outside, it looks like it may never have gotten the changes that GZ made
    to implement JCL-reading.  When I say JCL-reading, I mean that the sucker
    should be able to see altmodes in the JCL and run it as TECO code, so that
    one can do, for example,

    emacs $m(m.mRead Mail⊗◊)⊗◊⊗◊

    which works on OZ, but blows out on XX.

    I guess it looks from here like there's been some sort of version-splitting,
    if XX has a later version number that doesn't include earlier fixes.  Can
    someone make everything all better?

The changes are to OZ's EMACS.INIT file, which is probably why nobody
ever noticed.  If anybody announced the new EMACS.INIT to BUG-EMACS, I
missed it.

The scheme seems to be that if the first character is a dollar sign,
the JCL is scanned for ^V quoted altmodes.  This seems a little risky,
given that dollar is a perfectly legal twenex filename character.  A
safer scheme would be to start JCL lines with "# " (<pound><space>) or
something else that is obviously not a filename.  Does anybody have
any idea who is using this hack?  Would anybody be horribly upset if I
rewrote it?

BTW, OZ's SITE.INIT needs to have some stuff added to it before OZ can
run the current merged (XX+OZ+ITS) version of BABYL.  See
XX:<EMACS>BABYL.OZ-INIT-KLUDGE for details.