Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!sun-barr!sun!pepper!cmcmanis
From: cmcmanis%pepper@Sun.COM (Chuck McManis)
Newsgroups: comp.sys.amiga
Subject: Re: Card Cages and extra serial port support for A500
Message-ID: <107765@sun.Eng.Sun.COM>
Date: 2 Jun 89 03:54:31 GMT
References: <2433@van-bc.UUCP> <664@madnix.UUCP>
Sender: news@sun.Eng.Sun.COM
Reply-To: cmcmanis@sun.UUCP (Chuck McManis)
Organization: Sun Microsystems, Mountain View
Lines: 55

(Wow a posting from Perry I completely agree with! :-))

In article <664@madnix.UUCP> perry@madnix.UUCP (Perry Kivolowitz) writes:
>On Christmas day I sat down to write our serial driver. On New Years
>day I had a driver which would mostly run most terminal programs. Mostly.
>Enough, in fact, that I could have easily have fooled people at trade 
>shows into believing we had a perfectly working and complete system.

This is typically the "hack" stage, a lot new developers confuse this
with the "product" stage and start advertising ...

>Then the hard work started. Hundreds of special cases. Idiosyncracies.
>Strange behaviours of the Amiga serial device which needed to be pinned
>down so they could be emulated. Undocumented behaviours. Obscure
>commands. Interactions. Stress cases, pathological cases. Writing
>programs intended to show each and every one of these cases one at 
>a time and prove that our driver sucessfully navigated them.
>
>After five months of intensive work, our driver was finished.

And this is what separates the developers from the hackers. And it
burns out a lot of people along the way. Perry was lucky that this
stage only lasted about 5months. For a really complicated system such
as a Desktop Publisher or a Music sequencer, this period can last 2 
years or more. And for those who started advertising the bills begin
to accumulate, and so does the hate mail from people who ordered the
product or called for more info. And more importantly so does the pressure
to ship that original hack even though you "know" it isn't ready yet.

>Add to this three months of intensive work going on in parallel to
>produce an effective multiple serial port solution for DOS and CLI
>users.

And that is the final "cap" to the solution, ie making it work with
the rest of the system. Complete this successfully and you actually
have a real "product" to release.

>You want my software for free? Right. You're typical of what's wrong
>with the Amiga market. This ain't no damn 64. Grow up.

Right, and you have to deal with a bunch of weenies who might even get
to the "hack" stage complaining that it isn't PD. The fact being that
if you can't do it yourself you shouldn't complain when someone decides
to do it and charge you for it. Unfortunately they are a fact of life,
more so now that we have the A500 with us, we just have to ignore them.

>The person who said ``he with the best software will win'' is
>right. We aim to win.

Way to go, show no mercy.

--Chuck McManis
uucp: {anywhere}!sun!cmcmanis   BIX: cmcmanis  ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.
"A most excellent barbarian ... Genghis Kahn!"