Right, Rich, please allow me to reply to the points you mentioned; I
declined from doing so last night as I sensed some unintentionally
irritated feelings, which I hope have eased a bit by now. I believe
all my posts in this discussion are purely technical concerns and I
believe them to be valid. I am most definitely not a troll as some
have suggested; I would've had to do a ridiculous amount of homework
over a long, long period of time and been a psychic to predict this
event (I've only found out clojurescript in the past couple of days),
and I do not believe in any way I'm making an attempt at humor in the
technical arguments I'm making.
On Jul 24, 10:28 pm, Rich Hickey <richhic...@gmail.com> wrote:
> On Jul 24, 11:19 am, James Keats <james.w.ke...@gmail.com> wrote:
>
> > Alright, to be honest, I'm disappointed.
>
> I'll make sure you get a refund then.
>
> Seriously, this is like being disappointed an action movie was an
> action movie instead of a comedy. Your expectations are a complete
> mismatch for the intentions of ClojureScript.
"clojure's rocks... javascript reaches"
>
> > First of all, congrats and good job to all involved in putting it out.
> > On the plus side, it's a good way to use the Google Closure javascript
> > platform.
>
> > On the minus, imho, that's what's wrong with it.
>
> > Google Closure is too Java.
>
> Actually, it's too JavaScript. Some JS proponents want to disavow its
> pseudo class model, but it certainly is part of the design of
> JavaScript. And it has some particular advantages over the other
> strategies, as outlined here:
>
> http://bolinfest.com/javascript/inheritance.php
Rich, the "pseudo class model" with the new keyword is a syntactic
obfuscation, semantically javascript is prototypical inheritance. It's
class free. In addition to the pseudo class inheritance advocated by
google closure and the prototypical inherent in javascript, others
like Doug Crockford advocated functional inheritance.
Now I have watched and read enough of your output and for example
Stuart Holloways talk about protocols to know that you've railed in
your adovacy of clojure against classes and inheritance, and find it
ironic that now you posit a link by an advocate of it citing it as
advantageous. In any case, as I've mentioned, I have been aware of
this article for nearly a year now, it failed to convince me back then
and it still does; most of the arguments in it concern the closure
compiler, an obeisance to which by the regular developer who doesn't
have the needs and resources of google, I feel in this day and age of
ample memory and bandwidth and fast javascript engines, is premature
optimization gone berserk (seriously, folks, people are streaming HD
video, 1.5 gbps fiber optic broadband is being rolled out in London
and soon other cities worldwide and 4G mobiles are upon and we're
fretting over mere tens of KB that gets cached after first time and
basing our development around minimizing it?!), and the remainder of
the arguments are in support of classes and inheritance.
>
> > It's not idiomatic JavaScript.
>
> There's no such thing as idiomatic JavaScript. There are a lot of
> different conventions used by different libraries.
>
The Javascript community - the vast majority of which - after a decade
and a half now of experience with the language has come to regard some
aspect of it as good and others as problematic; things like functional
programming and object literals (akin to clojure's maps/structs/
records) vs classical inheritance, which are positions you yourself
have taken and advocated.
> > I find it
> > disappointing that rather than porting from a functional language like
> > Clojure straight to another functional language like Javascript, the
> > google closure with its ugly Java-isms is right there obnoxiously in
> > the middle.
>
> In the middle of what? I look at ClojureScript code and it looks like
> Clojure to me. Google Closure is under, and it is no more annoying
> there than Java is under Clojure - an implementation detail, and a
> rich source of production-quality code.
I respectfully dispute that; for what they both do - dom, css, ajax,
events, cookies, ui, effects, animations etc - jquery does it far
better and is much more pleasant an api. What jquery itself doesn't do
the huge ecosphere of libs around it do, for example:
http://metajack.im/2009/03/13/jquery-and-strophe-made-for-each-other/
http://strophe.im/
>
> > Then, there's the elephant in the room, and that elephant is Jquery. I
> > believe any targetting-javascript tool that misses out on jquery-first-
> > and-foremost is missing out on the realities of javascript in 2011.
>
> Should it be the purpose of a new language like ClojureScript to
> orient itself around the realities of currently popular JavaScript
> libraries? I think not.
> If you want jQuery as the center of your
> universe, JavasScript is your language - good luck with it. I see
> jQuery as a tool to be leveraged when appropriate (i.e. rarely in
> large programs), not an architectural centerpiece.
As opposed to orienting itself around the realities of a currently
unpopular one? This is exactly what clojurescript does! it orients
itself around the realities of google closure and its compiler. It's
not only currently unpopular, it's never been popular, and in all
likelihood never will be.
http://www.google.com/trends?q=%22google+closure%22 (that lone
vertical line you see was its announcement, let's not do a comparative
google trends for "google closure, jquery", or even google's gwt)
>
> > Jquery is huge in its community and plugins, and it has tons of books
> > and tutorials.
>
> So what? Those people are satisfied by, and not leaving, JavaScript,
> and I'm fine with that.
>
The point is "reach", Rich; things like this:
http://jquerymobile.com/
> > Then, the Google Closure compiler is a moot point.
>
> If you seriously cannot see the benefits of Google's compiler then you
> are not the target audience for ClojureScript. In any case, for those
> interested there is an argument for Google's approach in the
> rationale, as well as this page on the wiki:
>
> https://github.com/clojure/clojurescript/wiki/Google-Closure
>
> > I'm tempted to "fork" clojurescript and redo it in javascript perhaps
> > so that seamless interop with jquery would be the main priority.
>
> Is that a threat, or a promise? I suggest you start by writing up a
> rationale like this one:
>
> https://github.com/clojure/clojurescript/wiki/Rationale
>
> making your intentions and the superiority of your approach clear.
> Then prepare yourself for messages from people who don't bother to
> read or understand it.
I did read it though. In any case, the suggested "fork" was for lack
of a better term, as github's "fork me" slogan goes; I won't now so as
not to seem like I'm here to steal your thunder. Clojure on Java, from
what I've read and watched of yours, was not your first attempt at a
lisp targeting java. It took a few attempts to get it right. We're all
human, you're not an emacs whiz and nor am i, though I do absolutely
believe you're a man of brilliance and I fully admire you. I'm merely
voicing my humble and sincere opinion based purely on technical points
that this first attempt at targeting javascript needs reconsidering.
Best regards; love you, man, and sorry again for any misunderstanding
or unintended miscommunication.
>
> Messages like yours make creating things and releasing them for free a
> really exhausting endeavor.
>
> Good luck with your fork - please start a separate mailing list for
> discussions about it.
>
> Rich
>
> p.s. note to others - if you have read the docs and have honest
> questions about the approach, I and others would be happy to explain.
> But we could do without messages about disappointment, threats of
> forks etc. ClojureScript is an action movie, and we're interested in
> helping people kick butt.Describe your new note here.
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
