This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2020-02-19
Channels
- # announcements (10)
- # aws (3)
- # aws-lambda (1)
- # babashka (24)
- # beginners (57)
- # boot (5)
- # calva (20)
- # chlorine-clover (3)
- # cider (14)
- # clj-kondo (37)
- # clojars (17)
- # clojure (200)
- # clojure-dev (40)
- # clojure-europe (9)
- # clojure-france (7)
- # clojure-gamedev (5)
- # clojure-hungary (4)
- # clojure-italy (8)
- # clojure-losangeles (2)
- # clojure-nl (9)
- # clojure-uk (97)
- # clojurebridge (1)
- # clojured (3)
- # clojuredesign-podcast (23)
- # clojurescript (13)
- # code-reviews (2)
- # component (22)
- # core-typed (7)
- # cursive (64)
- # datascript (12)
- # datomic (60)
- # emacs (6)
- # fulcro (54)
- # graalvm (11)
- # graphql (3)
- # hoplon (25)
- # jobs (1)
- # joker (85)
- # juxt (5)
- # kaocha (10)
- # klipse (8)
- # malli (2)
- # off-topic (36)
- # parinfer (1)
- # pathom (1)
- # re-frame (9)
- # reagent (4)
- # reitit (1)
- # remote-jobs (1)
- # shadow-cljs (24)
- # spacemacs (1)
- # sql (39)
- # tools-deps (10)
- # tree-sitter (18)
- # xtdb (18)
Morgen!
I think it's great that libraries are being encouraged to have their version bumped up to a 1
if appropriate
I do actually meet resistance about some libraries that seem to be 0.0.1 alpha etc... even although they stable and have been for years
It's a bit strange for me... years ago I asked about moving java.jdbc
to 1.0.0 and the core team pushed back with all sorts of things they wanted fixed first.
And now here we are moving everything to 1.0.x "as-is".
I just pushed java.data
to 1.0.64 and I've decided to move next.jdbc
to that as well (1.0.375 will likely be the next release).
I think this is where idealism meets reality. When working in a company that does software, often questions are raised about libraries used etc... and if they are all alpha, or 0.0.1 etc., the business does want to know why such a version...
so at least having a 1.0.0 version, which software people tend to think as being the first stable release (yes, I know semver etc.., can/is controversal) it does help in adoption
I've never worked anywhere that cares about that -- or about putting alpha/beta libs in production -- but I know I'm in a weird minority.
SemVer is a pile of junk.
I worked for a few financial institutions in my time that did tend to have many eyes on software and what is used (justifiably so, this is people's money!) and even although nothing formal is defined, there was some talk about which libraries to choose.
oooh, sorry, tiny chat today! the kid - he's alive!!! time to fetch, water and feed him!
Hahaha... and I'm off to bed!
I've used a very simple SemVer-compatible scheme for a while. 0.x.0 or 1.x.0 Leading 0 means unstable/alpha, middle number is release count. That's it.
I think a lot would have been solved by just being upfront about the version policy tbh
I must say I’m not entirely wild about the pointless version change across so many libraries. It’s not a big deal or anything, but it feels like arbitrary churn to me.
> I must say I’m not entirely wild about the pointless version change across so many libraries. It’s not a big deal or anything, but it feels like arbitrary churn to me. It looks like a formality.. People that give more importance to the stable nature of the library won’t mind the version change, and people that give semver more importance will feel more comfortable using a 1.x.x lib… so it may seem good for the community in general.. 🙂
It is, but I review a lot of clojure projects — so when dependencies change I have to check what actually occurred in the library, to see if there are any other consequences of the change. As not all libraries are purely accretive. For example I experienced some breakages recently upgrading some ring dependencies that were supposed to be purely accretive; but subtly weren’t. Overall I agree with the policy and direction, it just creates busy work downstream.
I think it's great - but as pointed out, it is optional to upgrade - if you're happy to keep with the version you have, no problemo! 🙂
see #announcements
It’s interesting that spec.alpha is still on 0.2.176
and hasn’t been upgraded to 1.0.
Obviously I understand why in that the design isn’t quite right; but spec.alpha is stable and shouldn’t see breaking changes it’s just not compatible with spec2
It’s certainly special 🙂 It’s a dependency of clojure itself which isn’t alpha.
and this is the technology that powers self-driving cars?
the big question @conor.p.farrell is which of fusion power and self-driving cars will we get first
I thought that's what having kids was for? For driving me around when I'm old(er) and (more) decrepit?
but anyhoo, will be moot point - by 2030 all forms of motorised vehicular transport will be banned and we'll be using teleporters - just like in star trek 🙂
I'm a big fan of anything that is enabling greater self-sufficiency for young/old/people with temporary or permanent disabilities. I just wish it wasn't done in the spirit of startups, move fast and break things.
create-a-copy then destroy-the-original : what could possibly go wrong ?
or, alternatively, tear spacetime a new one and send yourself through : what could possibly go wrong ?
Neither @mccraigmccraig, I'm afraid
I read once that in order to create a teleporter, it would require more energy output than the entire universe has produced todate.
Well of course. You don't think all those Looney Tunes animators were lying to us when they had characters shunted around as a gag, did you? 😉
it'd be just like taking the tube, but at least you wouldn't have to avoid eye contact with other passengers
guys, we can now print animations. that's it, that's the end of internet. https://twitter.com/ThamKhaiMeng/status/1229634431377342464
I wrote a short story once where we had ubiqitous teleporters to the extent that most buildings didn’t have lifts and it comes out that 1 in like, 10 million uses the person doesn’t walk out the other side & also it turns out the energy req’d is waves hand partly due to destroying the thing entering it, & this causes a global political crisis cos people are so reliant on the tech
“what, you mean I kill myself rather than use the stairs??” was I believe a line from it, it was a black comedy and I was young
> 1 in like, 10 million uses the person doesn’t walk out the other side tracks with car journeys
(...actually I may have misread the data, but I'm now curious to see what the rate is irl 😄 )
RCFOTD:
clojure.core/lazy-cat
([& colls])
Expands to code which yields a lazy sequence of the concatenation
of the supplied colls. Each coll expr is not evaluated until it is
needed.
(lazy-cat xs ys zs) === (concat (lazy-seq xs) (lazy-seq ys) (lazy-seq zs))
favourite function name evarrrr!!!!
It’s a good one…
Though I’m also partial to future-cancelled?
for the existential angst it invokes. 😀
looks good! However iirc clojure peeps don't really like other people using their clojure symbol (if its the same one as on the website)
and the source code you can chuckle at here: https://git.sr.ht/~dharrigan/rcfotd
I've always found that when people would ask me questions abut my code is when I learn the most.. With that in mind (no other intention than the academic one) and with all my respect..
.. I am asking if you are using (def system nil)
for something specific.. for my first thought was that there is declare
that would save you the nil
binding, but then I saw it doesn't seem system
is used anywhere else from (-main)
..
Anyway, just saying.. you can tell me to pipe down and I'd be happy (is what people end up doing 😁 )
No problem. A hang-over from other code. It's used at line 40 where it's stored in the namespace for ease of use later (although I don't use it later, so it could be removed :-))
Mad points for using http://sr.ht @dharrigan 🙂