RE: CVS corrupts binary files ...

[Paul Sander]

>--- Forwarded mail from address@hidden

>[ On Thursday, June 17, 2004 at 13:09:09 (-0700), Paul Sander wrote: ]
>> Subject: RE: CVS corrupts binary files ...
>>
>> I've done revision control by backup, and I've done revision control
>> by naming convention.  Both have proven to be disasters.

>Obviously you tried to use these tehniques in isolation.

>> Introducing uncontrolled sources into your process is not the answer.

>Obviously you missed (or conveniently ignored) the step where the build
>system's configuration source file(s) are checked into the CVS repository
>and thus becomes a very much _controlled_ part of the SCM process.

Nope, I got it.  The thing is, you can control pointers (e.g. makefiles
containing references to files stored in a library somewhere) all you
want, but that buys you nothing unless the targets of the pointers are
also tightly controlled.

Just grabbing stuff and throwing it in a directory, maybe giving it a
unique name, simply isn't enough.  You still need to prevent uncontrolled
changes from being introduced, reproducing it if something fails, repeating
the configuration when upgrading, and all the other stuff that a full CM
system does.  That means the stuff that you can't create from other sources
must be kept under some form of source control, and you need a build
procedure to make it useful, among other things.

>--- End of forwarded message from address@hidden