[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re2: Mother ship App.(flame)
- To: firstname.lastname@example.org
- Subject: Re2: Mother ship App.(flame)
- From: email@example.com
- Date: Sat, 23 Jan 1993 09:39:05 +0100
>The fact is, MCL makes B I G CLUMSY and BUGGY applications
I believe John has a point in that there are a lot of buggy MCL
applications out there, but that isn't necessarily MCL's fault.
As Howard at UK0392@applelink.apple.com points out, MCL bugs
which crash the Mac are fairly rare. That may be part of the
"problem": Bugs in MCL programs don't cost programmers enough,
so they are too easly ignored. At least in my case, I am a lot more
careful about tracking down bugs and writing correct code in
the first place when using an environment where the price of
an error is heigh. I am quite new to Lisp, but I can write 10 pages of
Lisp before I start testing - and still make the program work within
a reasonable amount of time. That is an important part of the reason
for increased programmer productivity. It is also the cause of a lot
of little bugs in little used parts of my code.
Add to this the fact that many MCL "applications" are prototypes
and probably quite few go through anywhere near the kind of testing
applied to traditional large market commercial applications.
The Lisp tradition is to make many of the "prototypes" available to
fellow programmers. I really like that tradition, but for obvious reasons
what is given away can not be supported in the same way as what people
pay for (although some people do a good job trying).
The "problem" of buggy MCL applications can be viewed from another
angle: Most of those buggy MCL applications wouldn't have been completed
at all if they had been implemented in another language, and they most
certainly would not have been less buggy at the same level of fuctionality.
Norwegian Institute of Technology