Live data from Hacker News

On Leaky Abstractions and Objective-J

cappuccino.org

21–30 of 50 posts

Re: On Leaky Abstractions and Objective-J

#21

Earlier quoted context omitted.

"1. goofy/eyes-bleed syntax" This is almost always an indication that a programming language critic has nothing really thoughtful to say. Same with Lisp and S-expressions. If I may paraphrase: "It's slightly different than what I'm used to, therefor I despise it."

I agree, the middle part is weak, but the bookends are compelling arguments.

You mean:

"impossible to find it being put to use outside of places apple forces it."

That's a huge caveat. iPhone development is extremely popular. Maybe the point is that people just put up with Objective C in order to do iPhone/Mac development. But I don't think that is true. NextStep had a small but rabidly devoted following, and part of that was because they really liked Objective C. Apple tried Java bindings to Cocoa at one point, but found everyone used Objective C anyways.

In general, I think it's true that most developers don't know much about ObjC until they want to develop for iPhone/Mac. But once they learn it, they tend to like it, from what I can tell.

Re: On Leaky Abstractions and Objective-J

#22
I don't think languages have to change every few years, in fact, I prefer they wouldn't in many cases. I also think that the JS libraries kind of define language extensions, but it also feels better to me to stick with the JS syntax than to redefine the syntax.

Even with LISP, I am not sure if I am so happy about the new trend to add syntactic sugar. Having just brackets is pure, in a way, throwing in square brackets makes me feel a bit uneasy (not that I am much of a LISP specialist, so maybe my judgment is completely off base).

Re: On Leaky Abstractions and Objective-J

#23
post #9

Earlier quoted context omitted.

And again, since Objective-J is a proper superset of JavaScript really? what can it do that js can't do?

Read his first sentence again: "There is also the issue that implementing things like method_missing and import are prohibitively difficult without new syntax."

"difficult" has nothing to do with the power of languages

TeX macros are fundamentally as expressive as java.

Re: On Leaky Abstractions and Objective-J

#24
post #7

Earlier quoted context omitted.

Separately, Objective-C is a great language, and plenty of people love it so much so that its almost impossible to find it being put to use outside of places apple forces it. and before you say "gnustep"...no one uses that Can you give me three reasons why you don't like it? 1. goofy/eyes-bleed syntax 2. i don't care about OO 3. i'll think of something later Finally, claiming OO programming is dead is nonsensical. Ja…

"1. goofy/eyes-bleed syntax" This is almost always an indication that a programming language critic has nothing really thoughtful to say. Same with Lisp and S-expressions. If I may paraphrase: "It's slightly different than what I'm used to, therefor I despise it."

This is almost always an indication that a programming language critic has nothing really thoughtful to say. Same with Lisp and S-expressions.

i didn't mention lisp or s-expressions. of course taste is subjective, but the "objective c is fugly" meme has lots of adherents

Re: On Leaky Abstractions and Objective-J

#25

Earlier quoted context omitted.

I agree, the middle part is weak, but the bookends are compelling arguments.

You mean: "impossible to find it being put to use outside of places apple forces it." That's a huge caveat. iPhone development is extremely popular. Maybe the point is that people just put up with Objective C in order to do iPhone/Mac development. But I don't think that is true. NextStep had a small but rabidly devoted following, and part of that was because they really liked Objective C. Apple tried Java bindings to…

Apple tried Java bindings to Cocoa at one point

oh come on, it is well known that apple intentionally dragged their feet on their java support for years

probably because they realized that as gross as java was, very few people would bother with objective c if they could get first-class support for java on cocoa

Re: On Leaky Abstractions and Objective-J

#26
post #8

Earlier quoted context omitted.

to me jquery feels less heavy-handed about the abstraction. every abstraction has a price, i tend to avoid them unless they have a demonstrated value. to me, OO has no demonstrated value to be honest, i see the basis of objective-j born from some notion that apple does things right so naturally we all want to do things they way they do...but objective c sucks (having coded in it), and i have no idea why people would…

"to be honest, i see the basis of objective-j born from some notion that apple does things right" The word "apple" DOES NOT APPEAR IN THE ARTICLE. (I just checked.) Please, read the article and reply to the very thoughtful points made there. Ascribing some vague, unsubstantiated motives to people who have carefully laid out their entire reasoning behind their decision does not reflect well on your ability to engage i…

wow i seem to have got your irish up. typical fanboy response. this site is full of moonies

Re: On Leaky Abstractions and Objective-J

#27
post #25

Earlier quoted context omitted.

You mean: "impossible to find it being put to use outside of places apple forces it." That's a huge caveat. iPhone development is extremely popular. Maybe the point is that people just put up with Objective C in order to do iPhone/Mac development. But I don't think that is true. NextStep had a small but rabidly devoted following, and part of that was because they really liked Objective C. Apple tried Java bindings to…

Apple tried Java bindings to Cocoa at one point oh come on, it is well known that apple intentionally dragged their feet on their java support for years probably because they realized that as gross as java was, very few people would bother with objective c if they could get first-class support for java on cocoa

No, if they "dragged their feet" it was simply because Java's static OO-ness doesn't map well to Objective-C's fully dynamic nature. Objective-C is basically just Smalltalk semantics (minus control structures) bolted on to C.

It works surprisingly well, but you can't and won't believe it until you try it yourself.

Re: On Leaky Abstractions and Objective-J

#28
post #22

I don't think languages have to change every few years, in fact, I prefer they wouldn't in many cases. I also think that the JS libraries kind of define language extensions, but it also feels better to me to stick with the JS syntax than to redefine the syntax. Even with LISP, I am not sure if I am so happy about the new trend to add syntactic sugar. Having just brackets is pure, in a way, throwing in square brackets…

Why would you NOT want to redefine syntax if that new definition could make you dramatically more productive and or your code more readable? Sure if you are working on a large project for a large company (or even a large open source project) of course you want to make sure your code can be understood/maintained by johnny random but when you are talking about your journeyman type project - why would you compromise - and better yet why is it you aren't tackling problems which make you "up your game" in terms of abstractions and tools available to you?

Mechanics have wrenches with the edges ground off or thinned down, even heated and bent to fit special applications. So why shouldn't I have my own syntax which sits on top of a readily available platform?

Re: On Leaky Abstractions and Objective-J

#29
post #27
post #25

Earlier quoted context omitted.

Apple tried Java bindings to Cocoa at one point oh come on, it is well known that apple intentionally dragged their feet on their java support for years probably because they realized that as gross as java was, very few people would bother with objective c if they could get first-class support for java on cocoa

No, if they "dragged their feet" it was simply because Java's static OO-ness doesn't map well to Objective-C's fully dynamic nature. Objective-C is basically just Smalltalk semantics (minus control structures) bolted on to C. It works surprisingly well, but you can't and won't believe it until you try it yourself.

That's inconsistent with how quickly Apple had Carbon ready using C. Just a matter of effort.

Re: On Leaky Abstractions and Objective-J

#30
post #25

Earlier quoted context omitted.

You mean: "impossible to find it being put to use outside of places apple forces it." That's a huge caveat. iPhone development is extremely popular. Maybe the point is that people just put up with Objective C in order to do iPhone/Mac development. But I don't think that is true. NextStep had a small but rabidly devoted following, and part of that was because they really liked Objective C. Apple tried Java bindings to…

Apple tried Java bindings to Cocoa at one point oh come on, it is well known that apple intentionally dragged their feet on their java support for years probably because they realized that as gross as java was, very few people would bother with objective c if they could get first-class support for java on cocoa

> oh come on, it is well known that apple intentionally dragged their feet on their java support for years

Apple (or more, exactly, the part from NeXT) did initially bet a lot on Java adoption. They ported the entire WebObjects stack from ObjC to Java, only to watch the technology get abandoned by those in the banking and eCommerce communities. The conversion was so total that when the Cocoa-To-Java bridge in OS X was deprecated, it was no longer possible to continue using the original WO tool chain. (What WO development outside of Apple exists made a new chain based on Eclipse, from what I read.)

Post reply on HN