--- In [EMAIL PROTECTED], "Ken Boucher" <[EMAIL PROTECTED]>
wrote:
>
> Warning: I spent a good chunk of last night sitting in a hot tub
> reading Joel on Software. This is probably a direct result.
>
> Ok. We have this GUI layer. It doesn't do anything, because all that
> stuff is down in the model. At this point, this is so essential as to
> almost be XP dogma by now. The GUI is hard to Unit Test. It's hard to
> Acceptance Test. We want it thin, so thin as to not matter.
[rest of interesting post snipped snipped]
Phlip makes an interesting distinction between
programming and authoring. I don't think he's
brought that distinction up in this thread, but
he's made it in a number of other threads, and
it makes perfect sense to me.
GUI design is different enough that it takes
a substantially different skill set. The support
tools we have for the task are primitive at
best. Most toolkits don't come with a declarative
designer that will either do a completely
hands-off job of code generation for the entire
GUI layer, or better yet generate it at run time
from the specification.
That doesn't mean that there aren't any,
but they aren't among the standard tools we
expect to be there, like the compilers and
debuggers.
I think that this explains why we have such
a hard problem with testing them. Frankly,
we not only shouldn't have to test them,
but the question of testing them shouldn't
even come up (other than in the context
of whether the code generator works, of course.)
John Roth
To Post a message, send it to: [EMAIL PROTECTED]
To Unsubscribe, send a blank message to: [EMAIL PROTECTED]
ad-free courtesy of objectmentor.com
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/extremeprogramming/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
