home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Nebula 2
/
Nebula Two.iso
/
SourceCode
/
MiscKit1.7.1
/
MiscKitArchive.mbox
/
mbox
/
000211_misckit-reques…aska.et.byu.edu_Thu Jun 2 15:12:06 1994.msg
< prev
next >
Wrap
Internet Message Format
|
1994-10-30
|
2KB
Return-Path: <misckit-request@alaska.et.byu.edu>
Received: from alaska.et.byu.edu by darth.byu.edu (NX5.67d/NX3.0M)
id AA02764; Thu, 2 Jun 94 15:11:34 -0600
Received: from yvax1.byu.edu by alaska.et.byu.edu; Thu, 2 Jun 1994 15:13:28 -0600
From: don@ngt.sungard.com
Received: from DIRECTORY-DAEMON by yvax.byu.edu (PMDF V4.3-8 #4169)
id <01HD2KDGP88G01B4A0@yvax.byu.edu>; 02 Jun 94 15:08:05 -0600 (MDT)
Received: from hamlet (hamlet.ngt.sungard.com)
by yvax.byu.edu (PMDF V4.3-8 #4169) id <01HD2KBSYLE801B1G3@yvax.byu.edu>; Thu,
02 Jun 1994 15:05:49 -0600 (MDT)
Received: from lear.ngt.sungard.com by hamlet (4.1/SMI-4.1) id AA18534; Thu,
2 Jun 94 17:04:50 EDT
Date: Thu, 02 Jun 1994 17:04:50 -0400 (EDT)
Subject: Re[2]: MachKit
To: ernest@pundit.cithep.caltech.edu, misckit@byu.edu
Message-Id: <9406022104.AA18534@hamlet>
Content-Transfer-Encoding: 7BIT
> [Don Radvan writes:]
> How about a wrapper around some of the Mach stuff (threads,
> message queues, kernel garbage?,...)
>> Seriously, I would like to see some classes to help support OS
features like these.
Ack! Too many Dons to keep track of! :-)
The obvious answer to this question is:
"Wait, there is a MachKit already!"
Oops! I forgot all about it (most <all?> of my thread stuff was done pre 3.0).
The better answer to this question is:
"Wait, OpenStep will change a LOT of this stuff."
Coming up with an OS-independant way of accessing threads, queues, and
the kernel is one if the big problems facing the OpenStep spec
writers.
>From what I've been able to gather, they are going to throw out most
of
MachKit, and subsume that functionality into something more general.
OSKit ???
It could very well be that NeXT's new "Foundation Classes" will cover
that, but it could be something else.
Yes, this means that NeXTSTEP 3.2 will *not* be OpenStep compliant (I
have had this confirmed). It is not clear whether NS 3.3 will or not.
Either way, I would recommend against putting any more work into
something like this until we see what OpenStep and the Foundation
Classes contain. i.e., table this for after Expo.
Okey-dokey.
-- Ernie P.
Don (not Yacktman)