Too terse and too alien?
reinier at zwitserloot.com
Tue Jun 1 18:21:49 PDT 2010
Paul, where did I claim your opinions are somehow less valid?
You're flipping the argument around. I observed that your opinions are not
MORE valid than those of anyone else. That was the entire point of my
missive. If you took away from it that your opinion is LESS valid than I
must have missed the mark entirely.
Of course the closure proposal "feels complicated". It feels complicated to
you, Paul, and it feels complicated to the unnammed masses that Ruben is
referring to, because closures, the concept, is inherently complicated. Just
like generics, particularly co/contravariance, is just inherently
complicated. Trying to magic this complication out of the syntax is a fool's
errand. We could for example leave generics out entirely, but
co/contravariance doesn't go away. It just becomes invisible again, like it
used to be pre java 1.5 syntax.
The 'masses', so to speak, don't understand any of this, so if they suggest
a closure proposal is too complicated, I don't think that's meaningful.
Unless, of course, they've got some arguments suggesting alternatives which
aren't (as) complicated, making sure to check that they aren't just
shuffling the complication around. I've read countless rainbows-and-ponies
arguments where people "just want simple closures, like other languages
have!", with absolutely no idea on how to go about obtaining such a unicorn.
I don't think this means lambda-dev should continue searching until the end
of times for such a legendary closure proposal, nor should it mean that
closures should just be dropped (which would undoubtedly cause even more
whinging from the masses as implementing something that tends to be
considered as 'too complicated').
The rest of this thread is doing a great job at proving that there really is
no such thing as an almost universal sense of "java feel", as everyone is as
usual trotting out their personal syntax preferences and suggesting their
personal preference is clearly the only way that "has the java feel".
On Tue, Jun 1, 2010 at 7:48 PM, Paul Benedict <pbenedict at apache.org> wrote:
> Right on, Jesse.
> I will not be writing a dissertation to prove why something "feels"
> like Java or not. I get closures are new and so new conventions must
> exist, but I don't think my opinion is any less valid. There are
> reasons hackles are being raised over the Internet over the proposal
> -- the proposal easily violates Java's simplicity philosophy, in my
> opinion, of course. There's no "proof" for that other than one's
> internal perspective.
> On Tue, Jun 1, 2010 at 12:36 PM, Jesse Kuhnert <jkuhnert at gmail.com> wrote:
> > Though I know I'm now expressing a different opinion than I was
> > originally (thanks to Alex for pointing it out), a lot of this is
> > indeed artistic / creative and not strictly measurable in the
> > traditional sense. If we did things that way we'd just slap some blue
> > colors on it like google and call it a day right? (not trying to pick
> > on google too much, but well ...)
> > Much like traditional art, not all opinions are created equal. Not
> > saying that anyone is right / wrong, just that I'd tend to trust the
> > most knowledgeable to brainstorm and find something that feels the
> > most natural above anything else. (not that alternative syntax options
> > probably aren't welcome as long as they fit all the same functional
> > goals)
> > On Tue, Jun 1, 2010 at 11:08 AM, Reinier Zwitserloot
> > <reinier at zwitserloot.com> wrote:
> > <snipped/>
More information about the lambda-dev