#jogamp @ irc.freenode.net - 20150505 05:05:35 (UTC)
20150505 05:05:35 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20150504050535.html
20150505 05:05:35 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20150505050535.html
20150505 06:40:57 * elect (~elect@anon) has joined #jogamp
20150505 06:46:40 <elect> morning
20150505 06:58:31 * elect (~elect@anon) Quit (Quit: Leaving)
20150505 07:26:20 * eclesia (~husky@anon) has joined #jogamp
20150505 07:26:26 <eclesia> good morning
20150505 07:28:33 * monsieur_max (~maxime@anon) has joined #jogamp
20150505 08:12:33 <zubzub> if you build jogl, are the maven artifacts automatically installed in your locla maven repo?
20150505 08:13:50 <eclesia> you can remove the folder in your local .m2, and run a build in offline mode to be sure.
20150505 08:19:29 * jvanek (jvanek@anon) has joined #jogamp
20150505 08:25:10 <rmk0> zubzub: nope
20150505 08:29:41 <zubzub> can I install them easily?
20150505 08:29:55 <zubzub> or do I have to reference them manually from my ide to have a fast turnaround
20150505 08:35:23 <rmk0> well the issue is that the maven jars are produced from the jogamp-all-platforms.7z
20150505 08:35:36 <rmk0> so you'd... have to produce one of those first (and i actually don't know how that's done)
20150505 08:36:05 <rmk0> https://github.com/io7m/jogamp-scripting/tree/master/maven
20150505 08:42:11 <rmk0> think the 7z is produced by jenkins, because it has to produce a binary for each platform and then collect them into an archive
20150505 10:39:27 <zubzub> hmm
20150505 10:39:37 <zubzub> I guess I'll just reference the produced jars from my IDE
20150505 10:39:47 <zubzub> should give me the fastest turnaround
20150505 11:30:19 <eclesia> Question on directional light shadow mapping : I have the camera frustrum corners in world space
20150505 11:30:40 <eclesia> now I'm supposed to transform them in light space. but it's a directional light
20150505 11:30:57 <eclesia> there is only a direction
20150505 11:31:49 * eclesia help is welcome
20150505 11:34:19 <eclesia> or should I place the directional light in the scene like any other light and use this light space ?
20150505 11:43:08 <rmk0> eclesia: that's why i only used projective lights... directional lights don't have an origin
20150505 11:43:24 <rmk0> rather than try to deal with the ambiguity
20150505 11:48:55 <eclesia> rmk0: k, thanks
20150505 11:49:42 <rmk0> i don't think it's so bad for directional lights to be disallowed shadows
20150505 12:10:10 * doev (~doev@anon) has joined #jogamp
20150505 12:19:11 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150505 13:09:40 * elect (~elect@anon) has joined #jogamp
20150505 13:30:20 <rmk0> not too pleased about the opencl suggestion for java 9
20150505 13:30:39 <zubzub> I'm not too pleased about the way oracle is going with java
20150505 13:30:56 <zubzub> I feel a lack of vision and funds
20150505 13:31:03 <rmk0> http://www.infoq.com/news/2013/09/java-hsail
20150505 13:32:23 <zubzub> looks nice
20150505 13:35:35 <rmk0> i'd rather they not bother
20150505 13:35:48 <rmk0> they did start all this long before vulkan and SPIR existed
20150505 13:36:25 <rmk0> think both of those make it pointless now, and i'd rather not have oracle bundle some broken ancient version of opencl bindings with the jvm like they did with jogl last time
20150505 13:36:29 <rmk0> and then have to maintain all this crap for years
20150505 13:37:42 <zubzub> they bundled jogl with jvm?
20150505 13:37:49 <rmk0> yeah
20150505 13:37:56 <rmk0> as an "extension"
20150505 13:38:05 <zubzub> fools
20150505 13:38:14 <rmk0> gouessej had endless battles with it
20150505 13:38:25 <rmk0> they bundled it such that it'd override newer versions of jogl on the classpath
20150505 13:38:37 <rmk0> this was on OS X
20150505 13:38:53 <rmk0> but... oracle risk repeating that mistake if they don't do it carefully
20150505 13:39:01 <zubzub> oh they will repeat it
20150505 13:39:04 <zubzub> you can be sure of that
20150505 13:39:09 <zubzub> and they will screw it
20150505 13:39:12 <rmk0> it was apple's mistake first, to be clear
20150505 13:39:26 <zubzub> oracle is already busy completely screwing javafx
20150505 13:39:52 <zubzub> I can't imagine them screwing over other possible cornerstones of the java ecosystem
20150505 13:40:30 <zubzub> and then wondering why java doesn't sell anymore for applicatoin devs
20150505 13:42:22 <zubzub> it would be nice to have a sane, modular, nice, small, well though out java devepment eco system/jdk/something
20150505 13:42:56 * eclesia working on it
20150505 13:42:57 <rmk0> do you like osgi bundles??
20150505 13:43:10 <rmk0> ... didn't mean to have two '??' there
20150505 13:43:14 * rmk0 mashes keys
20150505 13:44:27 <zubzub> do program OSGi as my daytime job
20150505 13:44:46 <rmk0> urhur... is why i asked
20150505 13:44:59 <zubzub> no I don't like OSGi
20150505 13:45:01 <zubzub> I hate it
20150505 13:45:03 <rmk0> hehe
20150505 13:45:12 <zubzub> it's nice and all
20150505 13:45:13 <zubzub> on paper
20150505 13:45:21 <zubzub> in practice it's quite the nightmare
20150505 13:45:27 <rmk0> what happens?
20150505 13:45:30 <rmk0> i've not used it
20150505 13:45:39 <zubzub> 90% of java software out there doesn't support it
20150505 13:45:47 <rmk0> oh, right
20150505 13:45:52 <zubzub> so you're constantly busy making it work on OSGi
20150505 13:45:52 <rmk0> the peer to peer classloading?
20150505 13:45:59 <zubzub> not just that
20150505 13:46:17 <zubzub> resource URL representation is also different in OSGi
20150505 13:46:30 <zubzub> and there's the classloading nightmare
20150505 13:46:48 <zubzub> that's just one aspect
20150505 13:47:03 <zubzub> the other one is the fact that has dynamic runtime dependency injection
20150505 13:47:08 <zubzub> which is great
20150505 13:47:11 <zubzub> if you need it
20150505 13:47:17 <zubzub> and you have a couple of very skilled programmers
20150505 13:47:34 <zubzub> 90% of java devs are dumb monkeys unfortunately
20150505 13:47:45 <zubzub> ok
20150505 13:47:47 <zubzub> 80%
20150505 13:47:48 <zubzub> still
20150505 13:48:05 <zubzub> so that creates lots and lots of really bad code
20150505 13:48:15 <zubzub> code that would've been bad without OSGi
20150505 13:48:31 <zubzub> but the dynamic aspect has now made it 10 times worse
20150505 13:48:35 <rmk0> /o\
20150505 13:48:48 <zubzub> race conditions everywhere
20150505 13:48:51 <zubzub> deadlocks
20150505 13:48:54 <rmk0> ugh
20150505 13:48:58 <zubzub> stray services
20150505 13:49:09 <zubzub> all that crap
20150505 13:49:30 <zubzub> if you start programming OSGi, you need a very skilled team to make it work nicely
20150505 13:49:42 <zubzub> in which case, it *is* really nice
20150505 13:49:43 <rmk0> i'm unlikely to
20150505 13:50:14 <zubzub> but in the real world your team is 80% clueless button mashers, and 20% skilled people
20150505 13:50:20 <zubzub> that's just how it is
20150505 13:50:28 <rmk0> if i was looking to develop a jvm language, i'd likely copy the peer to peer classloading
20150505 13:50:32 <zubzub> and OSGi just introduces more complexity
20150505 13:50:36 <eclesia> that much skilled peoples ?
20150505 13:50:45 <zubzub> 2 out of 10
20150505 13:50:45 <zubzub> yeah
20150505 13:50:48 <zubzub> sounds about right
20150505 13:51:00 <zubzub> programmers
20150505 13:51:11 <zubzub> don't do it rmk0
20150505 13:51:16 <zubzub> it's nice idea
20150505 13:51:28 <zubzub> but a lot of libraries assume everybody uses the same classloader
20150505 13:51:36 <zubzub> and fail if that's not the case
20150505 13:51:44 <rmk0> bear in mind that i'm not particularly bothered about any existing code
20150505 13:51:46 <zubzub> you can argue that it's the library's fault
20150505 13:52:00 <zubzub> but I'm not going to fix every library that sins
20150505 13:52:09 <zubzub> yeah in that case, go ahead :p
20150505 14:20:15 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 14:46:36 <elect> so
20150505 14:47:05 <elect> shall I be worried as a (nub) java dev or not
20150505 14:47:10 <elect> for the future I mean
20150505 15:02:09 <zubzub> not for the next few years
20150505 15:02:17 <zubzub> java is big
20150505 15:02:25 <zubzub> it will take a while before it's dead
20150505 15:02:56 <zubzub> also, android is not managed by oracle
20150505 15:02:59 <eclesia> I think monsanto will kill you before java ends
20150505 15:03:23 <zubzub> but java will fail to live up to it's full potential
20150505 15:03:34 <zubzub> thanks to the incompetence of oracle
20150505 15:06:10 * elect reads about Monsanto
20150505 15:06:11 <elect> Monsanto is a sustainable agriculture company. We deliver agricultural products that support farmers all around the world.
20150505 15:07:46 <elect> anyway
20150505 15:08:08 <elect> which language will take place of java if it is gonna die
20150505 15:08:42 <elect> Why Does Everyone Hate Monsanto? http://modernfarmer.com/2014/03/monsantos-good-bad-pr-problem/
20150505 15:13:49 <zubzub> elect: lots of possibilities
20150505 15:14:18 <elect> such as?
20150505 15:14:21 <zubzub> go, c#, any of the scripting languages (python, ruby, perl, ...), javascript, ...
20150505 15:14:50 <zubzub> or maybe a new kid on the block
20150505 15:15:03 <zubzub> lots of things can change in the next 10 years
20150505 15:15:58 <zubzub> maybe oracle will sell java as non profitable, google will buy it and make it one of it's most profitable components :p
20150505 15:16:24 <zubzub> somehow microsoft manages to make money out of C#
20150505 15:33:34 * doev (~doev@anon) Quit (Ping timeout: 245 seconds)
20150505 15:46:02 <rmk0> cold sweat at the possibility of a future with python, ruby, perl, go, javascript
20150505 15:46:20 <rmk0> those weren't even the best 1996 had to offer
20150505 15:58:38 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150505 16:00:50 <eclesia> rmk0: have faith, imagine an hydrib between C++ and javascript
20150505 16:01:21 <elect> when you say python, ruby, perl, javascript I always think about scripting
20150505 16:01:30 <eclesia> templated, memory addressing, untyped code, awesome
20150505 16:01:31 <elect> I admit I dont know them
20150505 16:01:45 * eclesia (~husky@anon) Quit (Quit: Leaving.)
20150505 16:02:18 * elect made eclesia quitting
20150505 16:02:40 <rmk0> hehe
20150505 16:03:00 <rmk0> i think he was trying to give me a heart attack
20150505 16:03:06 <elect> :D
20150505 16:03:07 <rmk0> fire it off and ran!
20150505 16:03:09 <rmk0> *fired
20150505 16:04:58 <elect> im also out
20150505 16:05:02 <elect> ciao belli
20150505 16:05:05 <rmk0> byee
20150505 16:09:13 * elect (~elect@anon) Quit (Ping timeout: 250 seconds)
20150505 16:25:56 * jvanek (jvanek@anon) Quit (Quit: Leaving)
20150505 16:39:16 * monsieur_max (~maxime@anon) has joined #jogamp
20150505 16:39:26 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150505 17:03:13 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 17:54:46 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150505 17:56:21 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 17:57:35 * badshah400 (~badshah40@anon) Quit (Client Quit)
20150505 17:58:07 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 17:58:40 * badshah400 (~badshah40@anon) Quit (Remote host closed the connection)
20150505 17:58:57 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 18:54:33 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150505 19:30:17 * Eclesia (~eclesia@anon) has joined #jogamp
20150505 19:33:59 * Eclesia is back
20150505 19:34:39 <Eclesia> rmk0: you survived the assault ^^
20150505 19:35:14 * rmk0 gleams
20150505 20:10:39 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 20:31:11 <zubzub> its not about which language 'is the best'
20150505 20:31:25 <zubzub> it's about which language can give you the most productive program when used by total idiots
20150505 20:34:35 <zubzub> money and idiots go in, program comes out
20150505 20:34:57 <zubzub> the language that gives the fastest result while still running somewhat ok is the language that wins
20150505 20:50:44 <Eclesia> even then, you can make a difference between one shot program, and long shot libraries
20150505 21:31:46 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150505 21:44:02 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150505 21:49:28 * badshah400 (~badshah40@anon) has joined #jogamp
20150505 22:04:38 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20150505 23:08:01 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150506 02:35:34 * badshah400 (~badshah40@anon) has joined #jogamp
20150506 03:31:43 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150506 03:31:57 * badshah400 (~badshah40@anon) has joined #jogamp
20150506 03:36:17 * badshah400 (~badshah40@anon) Quit (Ping timeout: 246 seconds)
20150506 05:05:35 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20150506050535.html