(Another topic I can't miss:)

That's I've tried to explain during ESUG 2011: classic + mockist TDD (which is very similar if not identical to BDD) provides a lot of benefits including solid process, flexible design��� and free documentation of course. "Seamless" TDD is really cheap and very efficient in Smalltalk (compared to other development environments, like Java/C# at least), but it still has zero adoption in Smalltalk community, unfortunately.

--

Best regards,


Dennis Schetinin



2014-06-21 2:09 GMT+04:00 Sean P. DeNigris <sean@clipperadams.com>:
kilon alios wrote
> Pharo has zero official reference documentation

Well Pharo By Example and Deep into Pharo are close. IMHO, we've chosen to
pour as many of our limited resources as possible into shrinking and
redesigning the system into something one person can master, rather than
thoroughly document things that may soon change drastically or disappear.
Real BDD (Behavior Driven Design i.e. tests from the perspective of the user
of the library) would help here as the tests would provide documentation for
free. But BDD doesn't seem to have caught on in the Smalltalk world, and I
find many (most?) tests to be useless from a "how does this library work"
perspective. Also, when the system stabilizes and we start documentation
"for real", the fun part will be active essays; traditional paper-like
documentation is the kind of thing Smalltalk was invented to replace.



-----
Cheers,
Sean
--
View this message in context: http://forum.world.st/Where-is-the-postcard-with-syntax-tp4763281p4764026.html
Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.