* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
dreamrealtoday
nevetToday is Thursday, February 12, 2026
dreamrealtoday hebrew
nevetToday is 25 Shevat, 5786 (Hebrew)
dreamrealcalc 43*pi
nevetThe result of 43*pi is: 135.0884841043611
cess11supposedly it isn't full featured in the regex part, but i couldn't care less about it since my brain gets eczema from regex
dreamrealand so it lives again. It's not muted for #java ... yet. When javabot's back, I'll mute it.
cess11i like that it's fast compared to standard grep, quicker cadence when i'm hunting stuff in code or data files is nice
ParaHmph!
dreamrealPara: it has at last count ONE and exactly ONE factoid. :D And its syntax is not the same as javabot's.
dreamrealcess11: I'm not arguing that MY favored tool is gooder than your'n.
dreamrealjsr 52.info
nevetThe factoid for jsr 52 has the following attributes: text and urls, and was last modified at 2026-02-12T14:06:36.594004Z by unknown
bombhelp
cess11dreamreal: i didn't take it way, i just want to know if i'm missing out on something
ParaSeems the bot is helpless.
dreamrealcess11: fair. I'm used to ack and it's also fast as hell, so I use it. :D
dreamrealyeah, I don't have a help operation yet, and it's modular anyway so the help would be maddening to write AND parse as a reader. The web interface is also not stood up yet.
* B_fd joined #java
BombeHuh, what does it say about me that I was not aware that grep needed a replacement?
dreamrealIt doesn't need a replacement.
dreamrealSo: nothing.
dreamrealWhich classpath indicator is 55, btw?
ParaIt's Java 11 iirc
Parayeh, class file version 55 is Java 11
dreamreal55
dreamrealhmm
dreamreal55
Parastring minimum length check somewhere?
dreamrealno, context
lornikcess11: personally, i find it fast enough , at least sufficient for my use case. But of course it always depends on what you're trying to build
lornikSo, for the ResilientRecursiveDirectoryScanner, I've decided to use Files.walkFileTree. here , I Can apparently handle file changes well, unlike with Files.walk..
dreamreal69
nevet69 is the Java classfile format number for Java 25. Nice.
dreamrealI forgot: the factoid set process isn't "foo is bar", it's "foo=bar"
ChronosBeing the product of two primes is kind of fun too :)
dreamreal69=<reply>69 is the Java classfile format number for Java 25. Nice! ... and being the product of two prines is neat, too.
nevetok, dreamreal: updated 69.
dreamreal69=<reply>69 is the Java classfile format number for Java 25. Nice! ... and being the product of two primes is neat, too.
nevetok, dreamreal: updated 69.
* battlepope joined #java
Chronosdreamreal++
nevetdreamreal now has karma of 1.
dreamrealdamn it
Chronoshahahaha :)
Chronosdreamreal: You don't want any karma?
* nevet joined #java
* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
dreamrealIt is SPAM, but I'm testing stuff out, and, well...
dreamrealbattlepope: damn it, top and bottom karma is COMPLEX
dreamrealbut I'll have it soon :D
dreamreal(it's complex because karma isn't calculated in nevet like it is in javabot: javabot just keeps track of a simple score. nevet keeps track of the *operations* which lose value over time: a single "--" from four months ago isn't as weighty as a single "--" today.)
dreamrealChronos: ++
nevetChronos now has karma of 2.
dreamrealnoice!
Chronoskarma teckla
nevetteckla has no karma data.
dreamrealChronos: -- look, you ain't all that, dude
nevetChronos now has karma of 1.
* Chronos chuckles.
dreamrealtesting the autocomplete compatibility with the comment field
dreamrealthe comments are discarded but hey
* cheeser joined #java
dreamrealcheeser: javabot's MIA again; I have nevet running and unmuted but I'll mute it as soon as javabot's back
* javabot joined #java
dreamreal55
nevet55 is the Java classfile format number for Java 11.
cheeser javabot, of course. i don't know anything about nevet.
dreamrealjust clarifying, I wanted to make sure nevet wasn't creating an issue with javabot
* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
* nevet joined #java
* rvalue joined #java
Para~nevet
javabotPara, what does that even *mean*?
dreamrealnevet's muted here now. We have a bot; nevet is not a replacement for it.
dreamrealnevet's trigger character is ! in any event, to prevent collision with javabot should they both be unmuted.
ParaI was kinda hoping the reply would
Paragah
Parawould've been "Doesn't look like anything to me." or somesuch.
ChronosI feel like / is the best trigger character, since it's the hardest to typo in most IRC clients
dreamreal~nevet is <reply>nevet is like a flowing stream, a beautiful stream that flows and stuff like that.
javabotOK, dreamreal.
* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
* nevet joined #java
dreamrealokay, top karma and bottom karma operations implemented. "!top 5 karma" will give you the top five, !top 2 karma will give you the top two, etc.
dreamrealvery expensive operation if you have a large community with a lot of operations, but eh, threads am I right
dreamrealChronos: and you're right, / would be a good trigger EXCEPT that it's hard to type. So is ! compared to ~, but being actively difficult to invoke is an antipattern
ChronosIf $WORK is any indication, "just throw more threads at it" is the solution to *everything*
Chronosdreamreal: Is it hard to type on your client? It's easy on mine.
dreamrealWell, in this case, it's a deliberate choice: there are ways to allay it (calculate the karma at midnight for everyone, apply deltas to that to get the current karma, etc) but eh
dreamrealit's not hard to type but it just feels unnatural to hit // to start a line
dreamreal/test
Chronos/test
Chronos/test
ChronosOh, I had no idea you could do //. I've been doing / /.
ChronosTIL. Thanks, dreamreal.
dreamrealOf course!
dreamrealIt makes my day complete to feel like I helped someone learn something effectively useless for their irc client.
Chronoshahahaha
DoofusCanadensishooray!
* ptomli joined #java
* Nemu64 joined #java
* ChaiTRex joined #java
* kcomhnall joined #java
* stfstfm joined #java
ptomliI have decided I can have a celebratory glass of old grape juice 🎉 I have managed to coax that rando serial port into submission and am speaking to the inverter at the other end of the line ⚡
* Joel joined #java
dreamrealptomli: woot! embedded with java is fun, eh
dreamrealWhat created the java constraint?
* ErikT joined #java
ptomlinot sure, it was in place when I arrived. this is just me playing at home, but it's $work adjacent
* PyR3X joined #java
* wodencafe joined #java
* michele joined #java
* ferdna joined #java
* Casino joined #java
* ForeverDreaming joined #java
* sponkz joined #java
* ForeverDreaming joined #java
* nevet joined #java
* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
* stfstfm joined #java
p14nd4does anyone know where the behavior for allowed values of jakarta.persistence.schema-generation.database.action is specified (e.g., 'create', 'drop-and-create')? i can't find it in the JPA 3.2 spec: https://jakarta.ee/specifications/persistence/3.2/jakarta-persistence-spec-3.2 , only the names of the values
p14nd4hibernate has a little commentary on this [ https://docs.hibernate.org/orm/7.2/javadocs/org/hibernate/tool/schema/Action.html ]: "There is an ambiguity surrounding the value "create" here. The old-school Hibernate configuration interprets this as the action CREATE, which drops the schema before recreating it. The JPA standard interprets it to mean the action CREATE_ONLY which does not first drop the schema."
p14nd4...which suggests to me that there _is_ additional definition of this behavior somewhere in the JPA spec, i just can't find it
ptomliI think it's there, just in the whitespace between the words
ptomli"If a DDL operation fails, the behavior is undefined. A provider may throw an exception, or it may ignore the problem and continue." fits the Hibernate language
ptomlimore specifically, at the same method "attempt to create the schema, otherwise, assume the schemas already exist"
* rvalue joined #java
dreamrealp14nd4: I'm not even sure the spec *does* say what the options are for a given provider. Hold on.
ptomliI think, in my experience, that sometimes the specs are written in such a way that they make sense once you know them already. Typical documentation style, not really useful to learn the thing, but great to reinforce your understanding once you're on the other side
dreamrealA given provider can add more to these, as hibernate does.
ptomliyeah, but as he says, that's not exactly clear about what create means. the appendix has a little more info, but not too much to be honest
p14nd4yeah, i saw those allowed values -- was looking to have them defined. ptomli's link is close -- presumably create, drop, and validate are supposed to map to those interface methods
nevetyeah, i saw those allowed values -- was looking to have them defined. ptomli's link is close now has karma of -1.
dreamrealI don't think the spec can even really say "this is what SHALL happen" -- to many cooks in that particular kitchen.
nevetI don't think the spec can even really say "this is what SHALL happen" now has karma of -1.
dreamrealp14nd4: which provider are you using?
p14nd4but it doesn't explicitly say that's the case, and doesn't ever seem to actually define 'drop-and-create', since that doesn't have a corresponding SchemaManager method
ptomlithe details are usually in the appendix, marketing don't go reading that stuff too much
* ForeverDreaming joined #java
ptomlidrop-and-create isn't in the spec
ptomlidoh.. sorry, it is
* dreamreal points to the spec he quoted
* cheeser joined #java
ptomliyeah yeah, sorry, i point to my consumption of old grape juice
dreamrealof course, that's 2.2, I think
p14nd4dreamreal: i'm using hibernate in my reference implementation, but theoretically customers could use others
* javabot joined #java
ptomliso, if there's a specific "drop-and-create" then it's fair to assume that Hibernate thinking "create" means "drop-and-create" is... unexpected
* LtHummus joined #java
ptomliperhaps the very existence of the "drop-and-create" option is the uncertainty that Hibernate docs are hinting at
dreamrealyeah, hibernate treats ITS update and create as separate types. I think it'd comply with the persistence.xml definitions, though.
dreamrealhbm2ddl.auto != the schema generation property
* sa02irc joined #java
p14nd4without looking at the code, it seems to me that hibernate interprets values of jakarta.persistence.schema-generation.database.action and hibernate.hbm2ddl.auto the same, just giving precedence to the former, in case both are present
ptomlibut, if you're worried about customers using a provider which has particular effects regarding this, I would expect that (1) they're aware of the details, (2) pointing out the possible confusion should be sufficient, (3) trying hard to not delete customer data is a good step
dreamrealp14nd4: I would VERIFY that, if I were you
dreamrealI don't have any jpa persistence units lurking about or I'd try it
ptomliwe use eclipselink and I'm certain I wouldn't want "create" or anything like that anywhere near production 😆
p14nd4ptomli: agreed
dreamrealptomli: hahaha, yes
p14nd4this is mostly a curiosity as i was reading
dreamrealI use validate all the time but not create or update any more
jbosmansnothing wrong with update, as long as it's used by people that 1) understand what it does 2) understand what it doesn't do 3) make sure the whole of it keeps working
jbosmansin other words if all db changes are thoroughly tested before release, who cares who they're executed (and/or which pre/post actions are needed)
jbosmansmy 2ç :')
dreamrealI care. I want management and rollback capability, and migrations aren't done solely by altering a schema.
dreamrealschema is composed of more than table structure, after all.
jbosmansi like to be able to generate db schema's from code (when that's an option)
p14nd4also, 'update' isn't a JPA-standard schema-generation action, if you care about such a thing
dreamrealSure, nothing prevents you from doing that
jbosmansi have tools to compare schema's etc
dreamrealhell, that's what I do for initial schema generation: build a table model in entities, tell hibernate to generate it to a script, use that as an initial migration
jbosmansexactly
ptomlicode and data are too tightly coupled, farting around with one means farting around with another. it's unreasonable to think that "the tooling" will manage the migration, or the rollback when it goes nipples north
dreamreal... then turn it from create to validate :D
* fstd joined #java
jbosmanstoo much work given i know i don't need it, tradeoffs
jbosmansrigorously tested btw before release ^
dreamrealThat's fine if you know you don't need it
cheeseryou plebs with your schemas. adorable.
jbosmansall we need is TEXT
jbosmansand jackson
dreamrealcheeser: ++
nevetcheeser now has karma of 1.
ptomlioh wait, a nosql has entered the chat
jbosmans:P
dreamrealptomli: totally fair, though. SQL is what you want when you're okay with taking forever to access your data. :D
* CodeGeek!~codegeek@about/java/CodeGeek changed the topic to: Welcome! || Read Channel Rules at https://javachannel.org/ before participating. || Paste limit is two lines; ~pastebin lists options. || No applets, please. || Minecraft, Android, and Javascript all have their own channel. || You are being logged.
* nevet joined #java
ptomliI think, though I might be wrong, that most of the "you don't need relations" stuff we see these days, is entirely because we have more compute than we know what to do with
jbosmansdreamreal, i violated your rule in a way that avoids me to create 10ish tables to fetch data for a few very related use cases that always require all data
* stfstfm joined #java
jbosmans"if one part of the data changes, all related data changes"
jbosmansptomli, i use relations very strictly 99+% of the time
dreamrealptomli: I don't think so
jbosmansi prefer error to faulty data
dreamrealnosql can have relational models
ptomliit all comes down to one group of people thinking "you don't know how you will need to query this" and the other group saying "I know how I will query this". one group is old, the other isn't
dreamrealwhy are you limiting yourself to "nosql = a hash table of json"?
dreamrealRedis is NOT THE ONLY MODEL OUT THERE
jbosmanspick & choose ftw
dreamrealand it's not even a very good example of that model
ParaDocument DBs have all the indices, relations, transactions and all that jazz as well.
dreamrealIMDGs too
jbosmansgraph dbs same
jbosmansimho least PITA = RDBM
jbosmansexcept when not
dreamrealand an IMDG will take a document db to the woodshed, beat it raw, slather it with butter, and draw rude things on its forehead before a document DB even realizes an IMDG has entered the room. And by the time a document db scrapes the rudeness of its forehead, an RDMS is going "hey, look, something happened"
jbosmansin which case good to have "escape hatches" (to put it into react terminology?)
ptomliI think people who *choose* a graph DB know enough to understand the tradeoffs, those that drift towards the document db are those that just don't want to deal with the underlying limits. In general. Vague handwaving. Don't moan at me
ParaMy favorite thing about postgres is that you can do relational data perfectly with it and then all the other types as well and when you finally hit a wall, you can migrate that part to something more specific.
jbosmansPara, well put
* dreamreal shrugs in IMDG. Except sssssslllloooooowwwwwwwwwllllyyyyy so you guys can notice it.
dreamrealpara isn't wrong, though
dreamreal~para++
nevet~para now has karma of 1.
javabotpara has a karma level of 24, dreamreal
* Square2 joined #java
dreamrealpostgres is a good baseline. Still slow, but quite reliable and quite capable.
dreamrealaight, I'm off.
ptomlicapability > speed
ptomlican do, eventually is more important than can't do
NeXeNthe speed at which it does what it does is more important than the speed at which something else doesn't
jbosmansagree ptomli
jbosmanswell, depending on what matters for the business
ParaI haven't hit a performance wall with postgres in...hm. Almost a decade? Although that speaks more about the systems I usually do than the db itself.
NeXeNthere's performance limitations to any thing designed
ptomliwoooo, i'm not pasying postgres is slow, i'm saying havingn a tool that does the job is nore important than ... well, anything
ptomlipasying... damn
NeXeNbut the DB ultimately will be the first bottleneck
jbosmansptomli, generally speaking, which DBs would you prefer?
jbosmans(purely out of interest)
ptomlii'm quite partial to sqlite, because it working in my realm
ParaI once lived a life where ingress traffic link was the bottleneck. We saturated parts of a specific data center.
jbosmansptomli, your realm being?
dreamrealPara: have you ever gotten it to respond in < 2ms?
jbosmansPara, wow :)
ptomlii work in energy management, so inverters, batteries, local management. it's not about distributed data, it's about local data
Parajbosmans: In fairness, this was 14 years ago. It was "easier" in a sense back then :) Now everyone has 25 gig links at minimum.
dreamrealI mean, dow jones had < 0.2ms granularity for distributed transactions
jbosmansPara, i year you :-D
Square2What is the minimal boilerplate / manual charade coding approach to exposing spring server methods to a react frontend. Sure it's REST, but how to avoid the tedium of replicating server model object and hand coding invocation logic?
jbosmans<2ms is always attainable, given proper time to do all the work
dreamrealand 2ms is still pretty damn slow
ParaThat would be an interesting test to figure out.
jbosmansmoney performance quality, choose two (iirc)
NeXeNSquare2: sounds like a good project. probably hundreds of docker containers out there, even in the hardened line, that probably could get you from 0 to 60 in a few seconds
dreamrealI've never seen MY servers go under 3ms for roundtrip from postgres; I don't think I tune it especially well, though. But I haven't seen anyone else get a lot better.
Square2NeXeN, how is docker images helping here?
DoofusCanadensis0.2ms is fiiiiine
jbosmansdreamreal, i'd say, how many tables with how many changes over how many years ?
NeXeNSquare2: to eliminate the tedium
dreamrealjbosmans: in an IMDG? :D
NeXeNto get the boilerplate out the way. there's already initalizr which is on fleek
dreamreal"How many wings does your roof have? How many blades of grass does your bottle of paint contain?"
jbosmansi've made a fool of myself? :)
dreamrealyou're asking the wrong questions, in other words
jbosmansyeah i was thinking of an RDBMS
Square2NeXeN, I mean that if I have a set of server side services, how do I get them available in React with least amount of effort?
NeXeNtRPC usually
ptomlidepends on definition of effort, though "consultants" is a good start
jbosmansptomli, same for me regarding local data, but evolving requirement-wise and quantity-wise
NeXeNnext you can use react server components
ParaGave it a thought, I think the fastest DB to respond I've ever worked with was Riak once one of my colleagues fixed its abysmal Java driver.
jbosmansNeXeN, never
jbosmans:P
NeXeNs/next/nextjs/
ParaI don't remember the exact numbers but the fix was in order of hundreds/thousands times faster.
ParaOf course Riak really is just a distributed hashmap after all the fancy vector clock stuff soooo
NeXeNPara: well lmdb is fast and acid
ptomlijbosmans: yeah, my line of work is odd, but not interesting. I'd like to skip the weird mysql db we have to data and move to quest, but we have installed base, and that means effort.
Square2NeXeN, as I see it, the information is there to autogenerate all of this.
ptomliChanging stuff after the fact is just a pain
jbosmansptomli, it always is
jbosmansmaking those changes might make life better tho, for many involved stakeholders?
jbosmansin my experience, the pain increases over time
jbosmansusually it ends up at "someone needs to make a business case"
jbosmansthis will cost you $$$$$ unless adressed NOW
ptomli20k installed devices, a few GW power generation between them, not hard to think about, but mess that up, things burn
NeXeNSquare2: i'm just fooling with you :^)
jbosmansso that's important, which is good/great
ptomliyes, we look to fix things, but you have to pick your battles
jbosmansptomli, yeah totally, not easy with 20k "clients"
jbosmans"not easy" being the understatement :)
ptomlimeh, clients is a bad word for it
jbosmansconsumers?
jbosmansusers?
jbosmansit all depends on context/language/point of view
NeXeNSquare2: vaadin hilla might interest you for least amount of effort
jbosmans"from a certain point of view" as they said in some star wards
jbosmans*wars
ptomliwe sell batteries, and 40ft containers with batteries and inverters. You can fit a lot of MWh into a 40ft container, and a couple of MW inverters on the side of them makes you think twice about "just send it"
jbosmansptomli, wow yeah, "consequences"
ptomliwell, i was trying to figure out an energy meter today, because a hospital kept tripping over onto grid power. it's technically not fancy, but the outcomes are important
jbosmanslike hardware integration?
jbosmansas i understand it, it is technically fancy ^
ParaTransformer maintenance videos on Youtube have that special wtf aura. People wearing weird rubber suits, backup person with a long pole and a hook ready to pull the other guy away...
ptomliyeah, modbus, canbus, serial232 when needed, whatever the site has/needs to make it happen. we have plumbing code to handle water works at damns, it's all just numbers really
ParaElectricity in general is wild, from the very physical and hands-on to...well, just managing it, software etc.
ParaI'm kinda jealous of your work :)
jbosmansdamns or dams? sorry could not resist :)
jbosmansactually i may be wrong there, english is like my third language at best
ptomliwe 3d print covers for the high voltage battery connects. my introduction to the company was writing integration to a 125kW inverter. "Just turn it on" :-)
ptomlidams :(
ParaDid you look up to see how many covers were embedded into the roof already?
NeXeNi wonder how many batteries one could put in a container and still keep cool. how many times could i charge my scooter? hrmm a few million cycles?
jbosmansno worries, it was sort of funny in context
ptomliwe have aircon units installed at the end of the containerss
jbosmansfwiw clearly important work, which is nice
ParaNeXeN: Isn't heat more of a function of (dis)charge rate? Not that I know anything, really.
ptomliit is nice work, i'm not saying it's not
NeXeNi guess, i only expect if heat got too high then it would set off a chain reaction of some sort
ptomliit happens both ways. charge/discharge both add to the energy waste, which is heat
ParaThere was just a news article the other day that in silence a French company had built about 500 MW worth of container batteries all over Finland in a year.
NeXeNmaybe one could actually use those contact cooling rails like in the milspec stuff and need no air circulation
ParaIt was kinda like...that's cool but who ordered them?
ptomlicontainers have burnt... :sad face:
NeXeNprobably electrical services, who have to balance out their production and the amount of usage, it would be highly valuable in arbitrage
NeXeNi actually want a nuclear battery. one that'll run for 11k years down to halfway
ptomlinot one of ours, but basically the same
NeXeNjust have it in my back yard.....yeah never pay for power again
ParaNeXeN: RTG, Radioisotope Thermal Generator
ptomlithat one has doors to put in/replace cells from the side, we have a central corridor. but in the end there's a limit to the Wh you can store in a 40ft
NeXeNi could also give the neighbors free power, who cares
Paraof course my energy pipedream would include one of these for every ~10k people community https://polarnightenergy.com/
NeXeNPara: interesting, but heat exchange loses a lot of power, but the arbitrage saves the costs
ptomliPara: in the EU, the users for storage are often market movers. You can download the next 24h of market prices from ENTSOE, so if you have storage (regardless of PV) you can margin the market. We have lots of customers who simply time-shift energy for what is essentially residential markets. Buy cheap, put in battery, sell high. I've even added
ptomlicode recently to manage predictions of load vs PV vs market pricing, so the "system" just figures it out for you, and those fancy buggers can just go chill on their yachts in the Med. I do question my life choices
* jamezp joined #java
Paraptomli: Yeah, the whole offsetting is really the thing that everyone seems to be focusing on now.
ParaWhich is great, as the amount of renewable energy we could harvest during summers could offset most of winter's energy needs, there's just the pesky problem of storing it somehow.
Para(plus we also have that one neighbour so all forms of added resilience is also a big plus)
ptomliI'm not sure it's a bad thing, in general, but the people doing it are making a total killing. I keep feeling like I'm just "helping them along", and I should have though of this myself years ago
ptomliCATL have sodium in production now, I suspect it'll change things in the next few years
ParaCorrosive superheated liquid :)
ParaOf course as kids we used to stare at particle accelerators for entertainment, so science is stuff and such.
ptomliI'm trying to figure out where you ares, Sweden? I think you're a bit closer to our next best friend than I am (South Africa)
ParaFinland
ptomlioo, I like
ParaYou know, the Internet, there's always a Finn somewhere.
ptomliI used to work for a Danish company, I enjoyed my time up there
* ptomli apologises to all the people wondering "what the hell has this to do with Java"
NeXeNi'm sure they have java installed on one of their machines
ptomli😆
ParaTatu Suoranta is Finnish and he created Jackson.
Para(I think he lives abroad though)
ptomliSo I can blame him for the Jackson 3 changes?
ptomliI don't think there's more than 1 or 2 java devs that can do anything than thank him for his input
* Square2 joined #java
ptomli'd not seen that before, thanks
ptomliI'd not seen that before, thanks
ptomliSo, things like "Automatic subtype detection", sounds lovely. But I need to see how that works in practice, because I don't want it randomly trying to class-scan on a Allwinner H616 reading from an CD card. 🤷
ChronosPara: My favorite OS is FinnOS!
Chronos(I think most people call it Linux)
jbosmansleenooks
ParaYeah well.
ParaIt's kinda funny how the whole software chain for being able to chat in real time originates from Finland. Linux, SSH, IRC, irssi.
ChronosThis will age me, but I was chatting online in 1979, which predates all those technologies by a little bit.
ChronosChatting online real time.
jbosmanscrazy
jbosmansdarpa?
ParaOf course we've done more modern stuff. NVIDIA's core research which led to tensor cores was also done by a few Finns.
Chronosjbosmans: Not DARPA, but other online entities, like MECC and bulletin board systems.
jbosmansChronos, what kind of protocols?
ParaChronos: So you're one of the increasingly few people who know why lowercase usernames implied terminal features and also why email has always and always will be a pita to route.
Para+been
jbosmansunless MECC was/is a protocol
jbosmansi'm guessing not BBS etc?
Chronosjbosmans: When I started? It was old school analog modem stuff -- ASCII in and ASCII out.
nevetjbosmans: When I started? It was old school analog modem stuff now has karma of -1.
NeXeNSquare2: jhipster is also worth mentioning, i googled it earlier and thought it was neat
ChronosEven more basic than Telnet.
* x1bncwn joined #java
ChronosPara: haha yep :)
jbosmansChronos, lucky you :)
ChronosBeing old both sucks *and* has its advantages!
jbosmans"everyone wants to get old, no-one wants to be old"
Chronosjbosmans: Oh, MECC was the Minnesota Educational Computing Consortium (later Corporation). They had CDC (Control Data Corporation) Cybers with banks of modems you could dial into.
jbosmansi refuse to be sorry i'm not dead already :P
ChronosAnd people had written chat programs so you could dial in and talk in real time.
jbosmansi love that
jbosmansway ahead of my time
jbosmansi just captured network traffic to fetch credentials to get more capacity on my school's network
Square2NeXeN, Hmm.. just changed computer here. If you mentioned other libraries before jhipster can I bother you to repeat them? Fear I'm missing scrollback
jbosmans"once upon a time", approaching 25 yrs ago
* Ragnor joined #java
ChronosA lot of people don't know just how much stuff was invented way way WAY earlier than they think. In 1979 we had email and Reddit-like discussion forums real-time chat and the whole works!
NeXeNi have not tried it it just looked neat. i did some stuff with vaadin a while back but it seems to be newer versions, might be cool
jbosmansi remember my connection dropping when someone called in or out
Chronosjbosmans: haha that's awesome. I would write programs and trick people into running them. The program made it appear as if they got logged out and a new login prompt would appear. I would capture their user names and passwords and store them in files.
jbosmanshaha awesome
jbosmansi was btw quite considerate of who etc, we lived together
jbosmansit was "unused bandwith"
jbosmans*"bandwith"
jbosmansiirc ARP poisoning, way back when
ChronosOh my goodness, I thought I was in a different channel, my apologies to everyone for the off topic stuff
jbosmans&& likewise (?) :-)
jbosmansjava is 30+ now so
jbosmansit least we're not talking about: "Anthropic Raises $30 Billion at $380 Billion Valuation, Eyes IPO This Year"
jbosmansat*
ChronosHeh, indeed :)
NeXeNonly 10x plus tax?
jbosmansno worries, it'll al work out just fine
jbosmansgrmbl, *all
jbosmansfwiw i think we're still all good
jbosmansassuming we all care and invest in our craft
jbosmanswe'll see who maintains the generated codebases over time :')
jbosmansaddendum: "Anthropic claims more than 500 customers spending over $1 million a year on its workplace tools."