Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 beta 3/9/83; site cca.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!harpo!decvax!cca!cfh
From: cfh@cca.UUCP (Christopher Herot)
Newsgroups: net.micro.pc
Subject: development environments
Message-ID: <348@cca.UUCP>
Date: Thu, 26-Apr-84 22:29:20 EST
Article-I.D.: cca.348
Posted: Thu Apr 26 22:29:20 1984
Date-Received: Sat, 28-Apr-84 09:26:20 EST
Organization: Computer Corp America, Cambridge
Lines: 36



We are in the process of starting several projects for the IBM-PC,
after having lived in the relatively comfortable environment of VAX's
running UNIX.  It appears that there are at least five different ways
on could go about writing software for a project of significant size:

1.  Have each programmer use a PC with the 2 floppies.  This approach
will usually result in almost anyone giving up in frustration.

2.  Have each programmer use PC-XT's with their (slow) hard disk.  More
like the timesharing environment, but hard to share files.

3.  Buy PC's and outfit with a faster (e.g. Tallgrass) hard disk.
Faster compiles but still no info sharing.

4.  Link together the PC's with a disk server such as 3Com's.

5.  Use the VAX for compiling and linking, loading the object code
(perhaps through LAN) into the PC for execution.

Does anyone out there have any interesting observations, especially
on approaches 4 and 5?  Are there cross-compilers which are as good
at optimization as some of the newer PC native mode compilers?

As usual, I will summarize responses for the net.


-- 

 Christopher F. Herot
 Computer Corporation of America

 (decvax,linus)!cca!cfh
 cfh@CCA-UNIX