|
Soricidus posted:i wish stallman was trapped in the 70s, but unfortunately he has escaped and tenderly embraces us so what was this emacs deal that happened
|
#
?
Feb 3, 2015 21:32
|
|
|
|
| # ? Jun 11, 2024 19:25 |
|
|
Deacon of Delicious posted:so what was this emacs deal that happened People asked for GCC to provide access to the AST for syntax some code reformatting plugin or something, RMS says he'd only accept specific hooks for that use-case. People say that just means there'd be a lot more churn whenever a new feature is needed, so just give us the goddamn AST. RMS says he'll think about it, because if you give the AST away then you open the door for proprietary backends to GCC, using GPL code for the frontend. People point out they'll just use Clang/LLVM these days.
|
|
#
?
Feb 3, 2015 21:43
|
|
only
|
Edison was a dick posted:RMS says he'll think about it, because if you give the AST away then you open the door for proprietary backends to GCC, using GPL code for the frontend. People point out they'll just use Clang/LLVM these days. rms himself has made this argument about other things. if you read his essay about the LGPL, he argues the LGPL is valuable when licensing a free software alternative to something that has many proprietary alternatives. if you are licensing something that has no alternatives, by all means, use the full gpl. this is why readline is GPLed -- there are no good alternatives, so it's an inducement to publish your product as fsf-free software gcc's frontend used to be a readline-like unique advantage. now it's just one among many. rms was once flexible enough to consider the competition. today he has chosen not to be. he's loving over emacs and gcc at the same time by holding fast on one thing that is no longer relevant.
|
|
#
?
Feb 3, 2015 22:06
|
|
|
prefect posted:i worked at a company where they were making java desktop apps that could also load as applets in the '90s, and it scarred me for life   worked on an all-Java browser. that poo poo will never fully come off
|
|
#
?
Feb 3, 2015 22:09
|
|
|
Subjunctive posted:
|
|
#
?
Feb 3, 2015 22:19
|
|
|
kalstrams posted:not really what u mean, but https://vivaldi.com/#Features/4 that doesn't load the right thing on my phone, but if you mean that it's built with React and such, I lived that adventure with Firefox. I'll still take JS&C++ over Java for building a browser any day. (but I'm not building a 5th browser, for love or money)
|
|
#
?
Feb 3, 2015 22:35
|
|
|
KARMA! posted:serialize is better because it supports object instances A. so does var_export() B. serialize() is full of security holes C. but most importantly serialize() is a much less human-readable format. like, it requires you to specify the length of your strings for some dumb reason. PAN is easy-peasy to write by hand, which is its #1 design goal. for example, here's some PAN i just wrote: [butt_id => 5, size => 'huge',] look at that crystal clear syntax. look at how you don't need to quote the key names. look at how you can use a comma after the last item without making it barf. now look at what happens if you serialize that same array: a:2:{s:7:"butt_id";i:5;s:4:"size";s:4:"huge";} gross!
|
|
#
?
Feb 3, 2015 22:51
|
|
|
Tiny Bug Child posted:B. serialize() is full of security holes isn't that a feature in PHP
|
|
#
?
Feb 3, 2015 23:03
|
|
|
Subjunctive posted:that doesn't load the right thing on my phone, but if you mean that it's built with React and such, I lived that adventure with Firefox. I'll still take JS&C++ over Java for building a browser any day. im not even sure if i should try it at all
|
|
#
?
Feb 3, 2015 23:06
|
|
|
b0lt posted:isn't that a feature in PHP just means that the language doesn't get in your way whining about security!
|
|
#
?
Feb 3, 2015 23:12
|
|
|
Is it still cool to complain about Java's weird OOP fanaticism? Because that's what strikes me whenever I look at Java code.
|
|
#
?
Feb 3, 2015 23:44
|
|
|
MALE SHOEGAZE posted:You can't build libraries with beautiful DSLs without evaling if only you got beautiful stack traces too it really isn't worth obfuscating the code to the compiler or the maintainer
|
|
#
?
Feb 4, 2015 00:50
|
|
|
tef posted:if only you got beautiful stack traces too sourcemaps
|
|
#
?
Feb 4, 2015 01:39
|
|
|
Ruby is starting to grow on me what have I become
|
|
#
?
Feb 4, 2015 02:27
|
|
|
ahmeni posted:Ruby is starting to grow on me have you coded in it with other people?
|
|
#
?
Feb 4, 2015 02:39
|
|
|
Data:: Dumper is a good enough serialization format any day 
|
|
#
?
Feb 4, 2015 03:18
|
|
|
just send x86 machine code over the wire and execute to unpack. why are we loving around?
|
|
#
?
Feb 4, 2015 03:39
|
|
|
Subjunctive posted:just send x86 machine code over the wire and execute to unpack. why are we loving around? That's a patented AOL technique.
|
|
#
?
Feb 4, 2015 03:55
|
|
|
Suspicious Dish posted:That's a patented AOL technique. I had it happen more often with Adobe stuff, I think.
|
|
#
?
Feb 4, 2015 03:57
|
|
|
Subjunctive posted:just send x86 machine code over the wire and execute to unpack. why are we loving around? we can call it NaCl Object Notation
|
|
#
?
Feb 4, 2015 03:59
|
|
|
pseudorandom name posted:we can call it NaCl Object Notation except someone might use this
|
|
#
?
Feb 4, 2015 04:01
|
|
|
Subjunctive posted:except someone might use this hopefully on iOS
|
|
#
?
Feb 4, 2015 04:13
|
|
|
lol I can make x86 secure, just control branch targets to be %32
|
|
#
?
Feb 4, 2015 04:59
|
|
|
memory is an array and bigints are arrays which means memory is a bigint so just bitshift memory to move eip around.
|
|
#
?
Feb 4, 2015 05:25
|
|
|
wish someone would move ur posts around move them right out of this forum
|
|
#
?
Feb 4, 2015 05:27
|
|
|
MALE SHOEGAZE posted:have you coded in it with other people? yes and it turns out that some people just write ruby like an rear end in a top hat but some people write it nicely
|
|
#
?
Feb 4, 2015 05:41
|
|
|
Deacon of Delicious posted:wish someone would move ur posts around General Bit Shift is thataway >>
|
|
#
?
Feb 4, 2015 05:41
|
|
|
Shaggar posted:irrational java hatred esp wrt memory usage is another great way to detect people who have no idea what they're talking about and/or are still stuck in the 70s. not surprisingly most of these people use a Linux and plangs. shaggar was right
|
|
#
?
Feb 4, 2015 06:01
|
|
|
ahmeni posted:General Bit Shift is thataway >>
|
|
#
?
Feb 4, 2015 06:18
|
|
|
irrational java hatred comes from starting jira or stash or any atlassian product and waiting ~5 minutes for it to be ready while it pegs all the server's cores and then it's slow as gently caress to use
|
|
#
?
Feb 4, 2015 06:26
|
|
|
the people who think java are slow on the desktop have probably used it on the desktop the people who think java is slow on the server have probably never used it on the server
|
|
#
?
Feb 4, 2015 13:39
|
|
|
tef posted:the people who think java are slow on the desktop have probably used it on the desktop it's this thanks swing thwing
|
|
#
?
Feb 4, 2015 14:05
|
|
|
I had a professor in college who insisted that we create every frame and dialog our program was ever going to display ahead of time. Like, when the program started. So, instead of: *click a button* *slight pause while dialog is built*, it became: *run the program* *long-rear end pause while everything is built*. I get that, sometimes, JIT is Just Too Late, but man, that was some obnoxious code style. Meanwhile, the other members of my group made their dialogs appear when the program started, so like twenty windows would immediately pop up.
|
|
#
?
Feb 4, 2015 16:19
|
|
|
CPColin posted:I had a professor in college who insisted that we create every frame and dialog our program was ever going to display ahead of time. Like, when the program started. So, instead of: *click a button* *slight pause while dialog is built*, it became: *run the program* *long-rear end pause while everything is built*. i am pretty sure this is why eclipse has a splash screen
|
|
#
?
Feb 4, 2015 16:44
|
|
|
CPColin posted:I had a professor in college who insisted that we create every frame and dialog our program was ever going to display ahead of time. Like, when the program started. So, instead of: *click a button* *slight pause while dialog is built*, it became: *run the program* *long-rear end pause while everything is built*. probably because a) students created them on demand and before in a mix and ended up thrashing memory, or having lots of duplicates and never getting the right events handled b) it's easier to have one init method that creates them all and assigns them to singletons
|
|
#
?
Feb 4, 2015 17:28
|
|
|
Notorious b.s.d. posted:i am pretty sure this is why eclipse has a splash screen and is then slow as gently caress because
|
|
#
?
Feb 4, 2015 19:48
|
|
|
i made the gui for a reasonably large java program once and it was relatively responsive and not terrible (although unpleasant to make)
|
|
#
?
Feb 4, 2015 19:49
|
|
|
now i just use wpf and xaml, god's own ui thing
|
|
#
?
Feb 4, 2015 19:49
|
|
|
swing is ok, swing performance is fine as long as it's done properly with worker threads and stuff. slower than native guis, faster than 100% of web apps i've ever had the misfortune of using. the swing api is pretty lovely though, i'm not gonna defend that.
|
|
#
?
Feb 4, 2015 20:05
|
|
|
|
| # ? Jun 11, 2024 19:25 |
|
|
http://cr.openjdk.java.net/~jrose/pres/201502-JVMChallenges.pdf this is a cool slide deck about what they want to do with the jvm now
|
|
#
?
Feb 4, 2015 22:17
|
|