🚀 go-pugleaf

in article <1136@atari.UUCP>, good@atari.UUCP (Roy Good) says:
> I was going to let this one go by, since a Developer Release (softloaded)
> is imminently available - details are being mailed to all registered 
                                                        ^^^^^^^^^^^^^^
	I've been a registered Atari developer since day one.  In that
time period Atari *HAS NOT* sent me a single letter or package.  Back
when I was still writting commercial software for the ST I complained
several times.  No change.  Strange way to treat a registered developer.

> developers. But then I realized this is just the sort of unthinking posting
> that starts off a rumor chain and network explosion.
                    ^^^^^
	Myron's comments on bugs are not rumors.  I've seen both of the
problems he described occur on the Beta ROMs.  (No, I don't have a set.)

> The mature way to handle this concern would have been to call/email me or
> someone at Atari, rather than posting what is clearly rumor and hearsay to
> an already sensitive group of readers. But no... it seems it is better to
> to launch another missile that can only do damage, either to reputations
> or to morale. The closing comment is entirely unnecessary, regarding "fix a
> few, create a few..". Why on earth do you think we have have an extended 
> Beta with the subsidiaries? Why do you think we are making the Developer
> Release before committing to ROM? This version fixes some bugs found in
> Beta, and also includes even more performance improvement.
> 
> I have tried to keep a conservative tone to my responses, but the lack of
> foresight and thought behind this posting got to me. That and a teething
> baby [:-)].
> 
> PLEEEEEEASE: before putting this sort of muckspreading out for public view,
> get the FACTS!!
> -----------------------------------------------------------------------------
> Roy J. Good
> Product Development, Atari Corporation
> Views expressed are my own. Atari may agree or disagree; they have the right.
> -----------------------------------------------------------------------------

	Roy, you don't seem to really understand the problem Myron was
trying to point out.  The problem isn't the two (three if you count the
free space calculation bug he mentioned) bugs that Myron mentioned.
The bugs aren't all that major, I'm sure they have very simple, quick
fixes.  The problem is with Atari's method of developing and testing
code.  These are the kinds of problems that SHOULD have been found back 
in unit test (testing by the developer).  Didn't Alan have an IBM PC with
a 3.5" drive sitting beside him when he wrote the new format code?  Or
even an ATARI PC?  I could accept a problem with the format that only 
occured with a single version of DOS or if you had a strange number
and arraingement of files on the disk.  Those are the types of bugs that
you should be finding in Beta test code.  Not fundamental bugs that 
indicate that the code was never tested at all.

	When you say something is in Beta test you are saying that it is
almost ready to ship.  The only bugs should be small and realatively
obscure (eg. program A breaks if it is run with desk accesories B, C and
D on a machine with 2.5 meg of RAM).  But the primary features of the
code (eg. formatting PCDOS compatible disks) should already work.

	Please listen to what Myron is trying to tell you.  One of his
jobs is working on development methodologies for large projects.  There
are lots of differences between developing a small piece of code, like 
the 192K ST OS ROMs, and a multi-megabyte piece of code.  But there are
also a lot of similar issues, such as how you should test the code in 
order to develop a quality product.


					Dan Moore
					AT&T Bell Labs
					Denver
					dlm@druhi.ATT.COM
					   or
					ihnp4!druhi!dlm