Back to the channels

#java

10:55 deebo anyone know of any "official" spring (boot) discussion channels with actual devs/contributors? need to ask about some extension options or possible changes and using stack overflow as a platforms seems just plain stupid
10:55 deebo there used to be links to matrix or similar services but can't find them anymore
11:02 Para Maybe they deprecated all that.
11:02 Para Apparently https://spring.io/community points to Linkedin now :D
11:02 Para and few other such
11:11 deebo i get their reluctance to not have anything like that but damn is it annoying, no projects even have discussions open on github
11:20 dreamreal deebo: hrm
11:23 whaley vmware wants your money for such things
11:25 whaley though seriously, I have no idea but I'm willing to bet you aren't going to get direct access to actual dev without a support contract.
11:26 deebo yeah i'm just interested in the possible user extensions to spring-data-relational, or possible contributions to get some stuff we need available
11:27 dreamreal deebo: I've had to reach out to individuals to get information, which really kinda sucks
11:28 deebo yeah i guess i can check the contributors profiles if they have some way to contact them to get further info, just an annoying way to do it
11:28 deebo at least i'd hate it :)
11:29 dreamreal who knows, maybe you'll be so annoying that they'll go "dang it, slack, discord, irc are all there, maybe we should use one of them like normal people"
11:31 deebo time for an opeclaw bot that shames them into starting some sort of open channel
13:23 dmlloyd here's more drivel from me: https://github.com/smallrye/smallrye-ffm
13:23 nevet GitHub - smallrye/smallrye-ffm: A utility to simplify usage of the Java FFM API
13:23 javabot dmlloyd's title: "GitHub - smallrye/smallrye-ffm: A utility to simplify usage of the Java FFM API"
13:23 dmlloyd replaces my earlier "autolinker" project with a (hopefully) safer and more performant model
13:28 dreamreal dmlloyd: ++ holy shit
13:28 nevet dmlloyd now has karma of 1.
13:31 dmlloyd works on linux, mac, windows, aix (theoretically), maybe z/os (no tests yet) - CPUs supported are x86, x64, arm32, aarch64, ppc32/64, probably risc-v
13:31 dmlloyd mips not supported for errno handling on linux
13:33 dmlloyd (so this won't work on playstation 2, sorry)
13:38 dreamreal heh
13:39 dreamreal which arm processors, though? v7, v8, v6?
13:39 dreamreal I'm thinking of all those raspi0 machines running FFM...
13:43 Para PPC support theoretically means XBox 360 as well.
13:46 dreamreal dmlloyd: more seriously, that's awesome
13:46 dreamreal I have a project that REALLY wants FFM but is targeting an older JVM :(
13:47 dmlloyd all ARM including raspi :-)
13:47 dmlloyd well all ARM that you can get a JVM for anyway...
13:48 dmlloyd Para, yeah and playstation 3... but it's mainly because IBM big iron has PPC as one of its CPU types
13:48 dmlloyd but you know, bust out your mac g4 I guess :-)
13:55 dmlloyd I wonder if anyone still has one of those playstation 3 clusters up and running... I wonder if their GPU is any good for AI
13:56 dmlloyd AI says no :
13:56 dmlloyd :-)
13:56 deebo it's just lying so the nvidia-openai-oracle-etc circle jerk goes to the moon
14:01 dreamreal dmlloyd: the raspi's different processors support different instruction sets - you can't run the official mongo images on some of the raspis because of the instruction set even though it's "compatible"
14:02 dreamreal so it's a real observation but may not apply :D
14:02 dmlloyd it basically comes down to: can you get a (recent-ish) JVM running on it, and can you get your target library compiled for it
14:03 dreamreal yeah, I know.
14:04 Para There was that one guy a short while ago trying to make Java run on that...whatever OS it was, recruit that guy for the effort.
14:04 Para was it hurd? something like that
14:34 dreamreal dang it, jackson 3 is a pain
14:35 dreamreal I get that configuring a mapper wasn't perfect but not providing a distinct migration... or even repeating the features... feels dumb
14:55 deebo spring-boot-jackson2 for lyfe, or hopefully at least until someone else deals with it
14:55 dmlloyd was it a full rewrite or something?
14:56 dreamreal I'm getting damn close to doing that
14:56 dreamreal dmlloyd: I dunno, it's just amazing: "oh, you want no nulls and trailing commas and dates as strings like jackson 2? Sounds easy! So about that... ummm..."
14:56 deebo we use jersey so it doesn't even support jackson3 yet, and we also have some weird jackson (de)serialization cache warmup tasks, wonder if those will be compatible
14:57 dmlloyd is it stuff that was un(der)specified? or were those defined behaviors?
14:57 dreamreal I'm trying to futureproof this as much as I can, so I WANT to target jackson 3 if I can, but I dunno if I can, do I win a prize for the most uses of "if I can" in a single message, I'd like to have that prize if I can
14:58 dmlloyd bugs get you when you're young, but when you're older, it's unspecified behavior that kicks your ass
14:59 dmlloyd iffycan
14:59 dreamreal and eventually bugs get you when you're old
14:59 sbalmos munch munch munch
14:59 dreamreal it's the circle of liiiiiiiiiife
15:00 sbalmos dmlloyd: my favorite 5 words in my previous position: "For the avoidance of doubt..."
16:18 vitaliy just gotta make enough bugs when young... to build immunity
16:18 * dreamreal files that away as justification
16:19 dreamreal "I would have fixed all that stuff, maybe even tested, but... you know... immunity"
16:25 vitaliy <3
16:26 vitaliy i really admire Uncle Bob Martin, from the agile manifesto days
16:26 vitaliy he's pretty active on twitter
16:26 vitaliy and not afraid to get his hands dirty and make all sort of bugs
16:27 vitaliy hope i age professionaly as he does
16:39 dreamreal I like bob