On Tue, 20 Jun 1995 20:12:44 GMT Michael I Yawn said:
>Jeff Kell ([log in to unmask]) wrote:
>: I have created a new list, [log in to unmask] (to subscribe, send a
>: message to [log in to unmask] with text 'subscribe hp-posix Your Name'.
>
>Will postings to the hp-posix list be echoed to comp.sys.hp.mpe?
>(The answer I'm looking for here is "yes")
 
But the answer here is "no".  Based on public postings to the list, it seems
that the majority wishes the discussion to be carried out here on hp3000-l (or
comp.sys.hp.mpe).  But based on private replies I have received, almost all of
the respondents are grateful (in the words of one, "It's about time").  So now
I'm not sure which avenue to take; I don't want to annoy any hp3000-l readers
who have no Posix detailed interest, nor do I want to alienate any potential
Posix contributors by requiring them to subscribe to another list.
 
Perhaps I was not specific enough... HP-Posix was intended for people that
meet or exceed the following criteria:
 
* You have MPE/iX 5.0 at a minimum,
* You have the Posix Developer's Kit,
* You have obtained (or plan to) software from jazz
 
This is not "cut-and-dry" Posix, but rather, umm, Posix++ to coin a term.  The
basic Posix functionality is standard HP-3000 and supported by HP; most if not
all of the jazz ports are "officially unsupported" by HP, and these "gifts"
are courtesy of the Cupertino lab engineers' spare time.  Consequently, some
of us have obtained the software, tweaked it a bit for our purposes, but now
there are growing numbers of "mutated" releases floating about.  Consider the
httpd server... Stan Sieler tweaked it with new features, Eric Schubert added
an AFS authentication exit, and even the original source is not incorporated
into the "authoritative" sources of the code.  Same with Lynx -- David Greer
has done many modifications to make it well-behaved on the 3000, but we still
don't have this code readily available (even via jazz, let alone "official"
sources).
 
I'm suggesting very detailed, technical discussion along these lines.  Despite
the public postings suggesting we keep the discussion on the list, I still
find it hard to believe that everyone here (or even a majority) fits the
criteria (MPE/iX 5.0, Posix Developer's Kit, jazz experiments).  I am not in
any way suggesting that the Posix list be totally "isolated", but rather that
I feel that the majority of the readership is only interested in the end
result -- executable code -- and not the development details.  Again, if I'm
wrong, I'll trash hp-posix and ramble on with excessive verbosity to this
list (I'm intimately involved with httpd, and have current runtime problems)
but I don't want to annoy the "majority" of the list.
 
Perhaps I should post the initial "charter" of hp-posix to both lists, and
you can decide from there; but I still have the feeling that detailed Posix
discussion is much like Powerhouse, Manman, or other areas which HAVE their
own lists.  If you are a heavy Posix user (as most of the public replies
have been) then I understand you wanting to keep the discussion on hp3000-l.
But for the other (majority) of users, I don't want to bore them with details
of the Posix development environment (and I'm just as guilty if not more so
than the next person posting long diatribes on Posix issues).
 
If you don't want to hear this drivel, please post your opinion as a public
post.  I have made note of the private replies, but the "opposition" seems to
be a silent majority.  Again, any object-code-ready code that we come up with
will be posted and described in hp3000-l; I'm only trying to isolate the
detailed issues.  As one reply stated, "As I float alone in the sea of Posix
portability, it's good to know that others are sinking beside me".  Well,
that is a bit negative, but it also illustrates my primary purpose -- to
make the porting issues global and to foster communication between the various
developers and standardize our porting efforts.
 
Bottom line -- sorry for the hp-posix diversion, but consider the other
hp3000-l subscribers as well.  I didn't want to split the list, but I was
beginning to feel guilty posting my diatribes on httpd here as well.
The general readership (hp3000-l) will not suffer from lack of information,
simply spared the boring details of the port.
 
[\] Jeff Kell, [log in to unmask]
>-----------------------------------------------------------------