Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!snorkelwacker!apple!mips!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!cica!iuvax!maytag!watstat!dmurdoch
From: dmurdoch@watstat.waterloo.edu (Duncan Murdoch)
Newsgroups: comp.lang.pascal
Subject: Re: old .TPUs under TP5.5???
Message-ID: <1591@maytag.waterloo.edu>
Date: 27 Feb 90 14:36:38 GMT
References: <19557@mephisto.UUCP> <1968@bucket.UUCP> <90055.172954GILLA@QUCDN.BITNET>
Sender: daemon@maytag.waterloo.edu
Reply-To: dmurdoch@watstat.waterloo.edu (Duncan Murdoch)
Organization: U. of Waterloo, Ontario
Lines: 30

In article <90055.172954GILLA@QUCDN.BITNET> GILLA@QUCDN.QueensU.CA (Arnold G. Gill) writes:
>
>     If this is the case, that .TPUs are *NOT* upwards/downwards compatible,
>then the user community should seriously think about bitching to Borland in
>earnest and in great numbers, or else switching to someone else's software.

I agree with this part completely, but...

>It sounds to me like an idiotic ploy to keep the user community locked into
>using Borland software and forcing them to buy new versions whenever they
>come out.  With so much public domain material around, this is really a bad
>practice.

It seems like just the opposite to me:  a strong disincentive to upgrade.  
After all, if you upgrade your compiler, you lose all your sourceless .TPUs.

Or, from an optimistic point of view, a strong encouragement to library writers
to release source code. 

(Actually, there are a couple of ways to avoid releasing
source code while allowing multi-version compatibility.  Anything written as
an external .OBJ need not be recompiled; thus there's no need to release
the low-level assembler code.  If a unit is mostly written in TP, then it
can be made unreadable by removing all the comments and renaming all the 
local variables to meaningless junk names.  I've heard that this is done
by TurboPower in their Turbo Analyst package, which uses parts of their
Turbo Professional package.  They don't want to give away those parts of
TPro, so they garble them.)

Duncan Murdoch