> } Yes, dump and all other utilities that are commonly used to read and
> } write tapes should handle end-of-medium intelligently, but of course we
> } must not assume that all tape hardware does the same, so there still
> } must be support for media specification flags.

> Also, dump has traditionally written an end of volume record (a special
> record of type TS_END) and an EOF marker at the end of the tape.  This
> is tough to do if you it's already banged into the hardware EOM.  I
> presume this is to tell restore to ask for the next tape volume and that
> that dump just didn't stop writing.

It's been a quite a while since Stu Grossman gave me his lecture about
how tapes, tape drives, and tape drivers are supposed to work and how
TOPS-20 had everything right... :-), but shouldn't tape drives signal
logical end-of-tape before physical end-of-tape?  The theory being
that programs can write any end of medium record that might be
necessary on the last N% of tape reserved for that purpose?

	--jtc