Comments on: Broken Promises /code/broken-promises/ sealed abstract class drew {} Sun, 27 Mar 2016 22:51:38 +0000 hourly 1 https://wordpress.org/?v=4.4.15 By: jayprich /code/broken-promises/comment-page-1/#comment-10771 Mon, 26 Aug 2013 16:58:29 +0000 /?p=1635#comment-10771 Excel’s lazy evaluation tree is pretty efficient and parallel. The functions are supposed to be stateless without side-effects, you can break that with VBA of course.

The scattering of the execution path is a core issue for this programming style that can only be solved with better tools, e.g. ability to display the ‘trace’ for debug.

It seems your complaint is not about a “promise” per-se but that Objective-C and it’s libraries do not guarantee a functional style outside Grand Central Dispatch and Blocks (closures).
Do you think iOS 7’s multitasking will help address the issues that you say “promises” don’t?

It does seem a number of people haven’t given up trying this direction of development:
https://github.com/mogeneration/functionalkit
https://github.com/stuartervine/OCTotallyLazy
https://github.com/leuchtetgruen/functional.m
( OCaml used in an iOS app http://psellos.com/ocaml/example-app-slide24.html )

]]>
By: Joe /code/broken-promises/comment-page-1/#comment-10320 Mon, 12 Aug 2013 15:55:00 +0000 /?p=1635#comment-10320 @Tane That just isn’t true. I’ve been quite annoyed by the fact that Python doesn’t have do-while loops numerous times. As a matter of fact, I have a Visual Studio window open next to the browser and there is currently a do-while loop on screen. They have their uses.

]]>
By: Marek Sieradzki /code/broken-promises/comment-page-1/#comment-9934 Mon, 05 Aug 2013 19:52:54 +0000 /?p=1635#comment-9934 What you’re talking about sounds a lot like space leaks in Haskell caused by laziness. It seems that you’re explaining it using different words but mean exactly the same thing.

]]>
By: Lior Bar-On /code/broken-promises/comment-page-1/#comment-9141 Sun, 21 Jul 2013 07:09:27 +0000 /?p=1635#comment-9141 Thanks for a Great Article!

It does indeed seem Promises are a bit overrated, and it good you nailed down some of its downsides.

I don’t feel I’m the right person to do that, but I think some of the issues you raised (e.g. memory consumption) would also happen in naive callback-based implementation.

All in all this doesn’t change the overall conclusion, that Promises are not silver bullets.

Lior

]]>
By: Tane /code/broken-promises/comment-page-1/#comment-9075 Fri, 19 Jul 2013 05:53:19 +0000 /?p=1635#comment-9075 “The war between promises and callbacks is kind of like the war between while loops and for loops.”

I love the unspoken understanding that no-one uses do-while loops. 😛

]]>
By: Andy Matuschak /code/broken-promises/comment-page-1/#comment-8416 Wed, 10 Jul 2013 20:38:44 +0000 /?p=1635#comment-8416 This is an excellent case study for API authors, Drew; thank you. I really appreciate all the detailed explanations of the failure modes and the paths you took to get there, rather than the hand-waving hate one typically reads.

]]>
By: Jason Mulligan /code/broken-promises/comment-page-1/#comment-8048 Fri, 10 May 2013 16:20:33 +0000 /?p=1635#comment-8048 Haters gonna hate.

You clearly haven’t figured out the appropriate place to utilize the pattern, but you nailed the “I don’t like this! rawr!” part common to devs these days.

Congrats.

]]>