|
lamentable dustman posted:cmd.exe is really bad and power shell isn't much better
|
# ? Sep 3, 2012 14:29 |
|
|
# ? May 31, 2024 19:28 |
|
MononcQc posted:2. ORMs are okay but if you use them instead of a real model layer you're in for some trouble because that sucks. that's why core data is so ftw
|
# ? Sep 3, 2012 15:51 |
|
Shaggar posted:shells serve no function on windows aside from legacy support. lol
|
# ? Sep 3, 2012 17:05 |
|
if you're using a shell you did it wrong.
|
# ? Sep 3, 2012 17:09 |
|
Shaggar posted:if you're using a shell you did it wrong. this but if your trying to shoehorn windows into any server level equivalent role
|
# ? Sep 3, 2012 17:10 |
|
windows is great as a server
|
# ? Sep 3, 2012 17:11 |
|
Adventures in Clojure Day #420: Clojure has two completely different and largely incompatible systems for doing polymorphism, and the clojure community apparently prefers the less powerful of these. I seem to be the only person who thinks this is weird and dumb. (also windos is great as a server for me to poop on etc, etc)
|
# ? Sep 3, 2012 17:49 |
|
Police Academy III posted:Adventures in Clojure Day #420: Clojure has two completely different and largely incompatible systems for doing polymorphism, and the clojure community apparently prefers the less powerful of these. I seem to be the only person who thinks this is weird and dumb. multimethods and protocols?
|
# ? Sep 3, 2012 18:09 |
|
Police Academy III posted:Adventures in Clojure Day #420: Clojure has two completely different and largely incompatible systems for doing polymorphism, and the clojure community apparently prefers the less powerful of these. I seem to be the only person who thinks this is weird and dumb. this is why lua is ftw
|
# ? Sep 3, 2012 18:16 |
|
Police Academy III posted:Adventures in Clojure Day #420: Clojure has two completely different and largely incompatible systems for doing polymorphism, and the clojure community apparently prefers the less powerful of these. I seem to be the only person who thinks this is weird and dumb. clojure is terrible crap
|
# ? Sep 3, 2012 18:36 |
|
trex eaterofcadrs posted:multimethods and protocols? yeah. i understand a lot of the reasoning behind protocols, but there's no reason they couldn't have added those features into the multimethod system instead of making an entirely new one completely from scratch. also given that you can't specify a superclass in deftype/defrecord, you've pretty much got no way to do inheritance with protocols. (caring about programming languages itt )
|
# ? Sep 3, 2012 18:49 |
|
i'm having to do a neat embedded project using a horrible toy programming language. don't ever work on a project that requires you to use toy programming languages can't remember if i've already posted this or not but it bears repeating.
|
# ? Sep 3, 2012 19:03 |
|
Mr Dog posted:i'm having to do a neat embedded project using a horrible toy programming language. what language
|
# ? Sep 3, 2012 19:19 |
|
Police Academy III posted:yeah. i understand a lot of the reasoning behind protocols, but there's no reason they couldn't have added those features into the multimethod system instead of making an entirely new one completely from scratch. also given that you can't specify a superclass in deftype/defrecord, you've pretty much got no way to do inheritance with protocols. for better or worse ( i think better) clojure really tries to keep you from doing abstract/concrete superclass polymorphism at all costs. i actually like defprotocol/deftype myself so i guess i'm one of the dumb ones but i use a lot of spring so maybe that's why i use it so much
|
# ? Sep 3, 2012 19:28 |
|
vapid cutlery posted:what language nesC basically i'll sum it up as follows: it's like C, but you have to explicitly declare every single caller-callee relationship for no particular reason. also every time you do something that causes a callback to happen, every single thing that can receive that particular callback gets notified, so you have to make sure the callback was actually intended for you it is so god drat bad that's not even the worst thing about it, the worst thing about it is that it's loving impossible to debug because its compiler shits out this giant C file and 90% of the functions in that C file do nothing but call another function. Everything eventually ends up getting condensed into this gigantic inlined pile of sludge around each interrupt vector which is supposedly ~space efficient~ but it also makes setting breakpoints virtually impossible and control flow utterly incomprehensible due to the compiler going on a CSE/DCE rampage and line number information not meaning anything in particular remember this thing is intended for bare-metal embedded projects Sapozhnik fucked around with this message at 19:35 on Sep 3, 2012 |
# ? Sep 3, 2012 19:29 |
|
do it in java instead and dont tell them.
|
# ? Sep 3, 2012 19:47 |
|
Mr Dog posted:nesC write a compiler that does all this for you
|
# ? Sep 3, 2012 21:21 |
hey shagger, what is the deal with asp.net and asp.net mvc, i know that they're two different things but it seems like there aren't that many resources on the latter and/or if any companies actually use it
|
|
# ? Sep 3, 2012 21:30 |
|
gucci void main posted:hey shagger, what is the deal with asp.net and asp.net mvc, i know that they're two different things but it seems like there aren't that many resources on the latter and/or if any companies actually use it for asp.net mvc just read the ruby on rails documentation and then add a bunch of extra syntax
|
# ? Sep 3, 2012 21:47 |
|
asp.net is the platform that connects your code to your markup and runs it through IIS. It also kind of refers to the basic layout engine in asp.net which is very jsp/php-like. asp.net mvc is an model view controller framework built on top of asp.net. mvc3 came out like last year and is the first "good" version of asp.net mvc. it has the new razor engine. its still lovely web stuff but its atleast better than regular asp.net. I have a couple of mvc3 projects. It makes certain types of simple projects braindead easy. If you're data model is an object model then its gonna be way easy to tie it directly to the entity framework and have mvc do most of the work for you. If you have a proper data model you'll need to do some work converting the data model into an object model so you can present it to mvc. imo it sucks less than most web frameworks but it still loving sucks.
|
# ? Sep 3, 2012 21:53 |
|
if i had the time i'd check out http://www.zkoss.org/ which claims to do mvvm which would make me way happier.
|
# ? Sep 3, 2012 21:54 |
|
I gave a talk this weekend called 'Programming is terrible. Lessons learned from a life wasted'. It went down far better than I expected, for a talk I cobbled together the afternoon before. Afterwards I found out people had been quoting me on twitter, and a friend described it as 'stand up comedy'. I also got to call Joel Spolsky, Jeff Atwood and Paul Graham shitlords. Feels good man. After the talk, a bunch of people came up to me and gave me a hug and told me i'd inspired them
|
# ? Sep 3, 2012 23:01 |
|
It's gonna appear on vimeo real soon now and you can all enjoy my drunken sleep deprived ranting and my funny accent.
|
# ? Sep 3, 2012 23:02 |
|
if only youd seen the light of objc
|
# ? Sep 3, 2012 23:16 |
|
i'm always down to watch anyone call atwood a shitlord
|
# ? Sep 3, 2012 23:58 |
|
but joel is a cool guy aside from his "leaky abstractions" nonsense
|
# ? Sep 4, 2012 00:02 |
|
Gazpacho posted:but joel is a cool guy he's a a decent writer. he seems to be a decent product manager. fogbugz is poo poo though, and he seems to make some really weird technical decisions and his advise typically boils down to "just hire great programmers, hth"
|
# ? Sep 4, 2012 00:20 |
|
Gazpacho posted:but joel is a cool guy aside from his "leaky abstractions" nonsense heh read peopleware from tom demarco
|
# ? Sep 4, 2012 00:49 |
|
Mr Dog posted:nesC
|
# ? Sep 4, 2012 01:01 |
|
tef posted:read peopleware from tom demarco
|
# ? Sep 4, 2012 01:08 |
|
Mr Dog posted:nesC I presume this one: http://nescc.sourceforge.net/ The rational makes sense even though it sounds like it makes complex tasks quite cumbersome.
|
# ? Sep 4, 2012 01:21 |
|
Gazpacho posted:why peopleware owns and basically can be boiled down to "nobody gives a poo poo about your methodologies and savings. If you crush the souls of your workers your business will suck", where soul crushing can be anything from bad work environment (physical or social), hiring people who solve problems (that's what programmers do) and then not letting them solve them, etc. It's a short, good book that states the obvious, but in a more credible way than most people would.
|
# ? Sep 4, 2012 01:24 |
|
Also read Characterizing people as non-linear, first-order components in software development on why people matter and nobody gives a poo poo about methodologies.
|
# ? Sep 4, 2012 01:30 |
|
trex eaterofcadrs posted:for better or worse ( i think better) clojure really tries to keep you from doing abstract/concrete superclass polymorphism at all costs. i actually like defprotocol/deftype myself so i guess i'm one of the dumb ones but i use a lot of spring so maybe that's why i use it so much man, has the entire clojure community been infected with some sort of memetic virus or something? i think you're a smart guy trex, but polymorphism without inheritance has never been a thing anywhere. the only good reasons that i've seen to use protocols are speed and java interop, neither of which explains why everyone seems to regard multimethods as semi-deprecated or why protocols have to have a completely seperate syntax. they could've just made a special multimethod constructor that used java dispatch behind the scenes, but instead they decided to build an entirely seperate and redundant polymorphism system. i really don't get it
|
# ? Sep 4, 2012 03:04 |
|
MononcQc posted:We also got the same publisher. Did you get the sw8 cover material that nostarch used for the cover of land of lisp?
|
# ? Sep 4, 2012 03:18 |
|
Otto Skorzeny posted:Did you get the sw8 cover material that nostarch used for the cover of land of lisp?
|
# ? Sep 4, 2012 03:27 |
|
Police Academy III posted:polymorphism without inheritance has never been a thing anywhere. Wha?
|
# ? Sep 4, 2012 03:41 |
|
Police Academy III posted:man, has the entire clojure community been infected with some sort of memetic virus or something? i think you're a smart guy trex, but polymorphism without inheritance has never been a thing anywhere. the only good reasons that i've seen to use protocols are speed and java interop, neither of which explains why everyone seems to regard multimethods as semi-deprecated or why protocols have to have a completely seperate syntax. they could've just made a special multimethod constructor that used java dispatch behind the scenes, but instead they decided to build an entirely seperate and redundant polymorphism system. i really don't get it u mad
|
# ? Sep 4, 2012 03:42 |
|
what the heck is duck typing???
|
# ? Sep 4, 2012 03:42 |
|
|
# ? May 31, 2024 19:28 |
|
vapid cutlery posted:what the heck is duck typing??? duck typing doesn't force you to use inheritance, but it's still an option. clojure's protocols literally provide you with no way to specify any kind of inheritance. if you want two object to share the same behaviour you literally have to copy/past the same code. e: literally
|
# ? Sep 4, 2012 05:07 |