[arc-discuss] When should interfaces be private?

[James Carlson]

Philip Brown writes:
> Having the original engineer spend 5 minutes on documentation, can save some 
> later person looking at the code, 5 *days*.
> 
> That later person, may have an interest in doing something "innovative" with 
> the code. However, ... finding someone with sufficient motivation to do 
> something, even WITH documentation, is challenging.
> Finding someone with sufficient motivation to spend DAYS reverse-engineering 
> something that should have been documented in the first place... is 
> near-impossible.
> 
> The feelings that come about are; "These people couldnt spend 5 minutes to 
> write a little documentation to clarify this jumble. If they dont care 
> enough to spend 5 minutes writing a little doc, why the hell should I care 
> to spend my time on it?!"

Hence the reason for architectural and design review.  It forces
people to write these things down and expose them to the world.

That doesn't mean you can build software that _relies_ on those
details, but rather that the details are not secrets.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677