IRC channel logs

<nalaginrut>mark_weaver: I think TCO in ES6 spec, so it's the standard of ecmascript now. Actually it's added few years ago, but recently ES6 has been released finally, http://www.reddit.com/r/programming/comments/167js5/javascript_es6_has_proper_tail_calls/

***michel_mno_afk is now known as michel_mno

<zacts>huh

<zacts> http://www.gnu.org/software/coreutils/manual/html_node/su-invocation.html#index-fascism-2365

<zacts>^ gnu su seems to not support a wheel group for philosophical reasons

<zacts>I wonder if these reasons still apply today?

<zacts>anyway, I meant to post this on #guix

*nalaginrut think wheel is bad idea

***michel_mno is now known as michel_mno_afk

***michel_mno_afk is now known as michel_mno

<lloda>it's great how in the Guile REPL C-P, C-R, etc. work by sexpr and not by line. But if I go in history before the current session, I do get lines instead of sexprs. Is there a way to fix this?

<ijp>hmm, I don't see that

<lloda>is it always by sexpr in your repl?

<ijp>it is

<lloda>hm. where should I look for settings? I have this in .guile:

<lloda>(use-modules (ice-9 readline))

<lloda>(activate-readline)

<lloda>(readline-set! history-length 10000)

<lloda>

<ArneBab_>lloda: could you check if this persists in stable-2.0

<ArneBab_>(what will become 2.0.12)

<ijp>it's possible that readline and geiser interfere, but I couldn't say how

<ArneBab_>lloda: there were adjustments to the newline handling

<ijp>ArneBab_: it can't be just recent. my history goes back a good way

<ArneBab_>ijp: do you use stable-2.0?

<ArneBab_>the fix was a few months back

<ArneBab_>s/fix/change/

<ijp>generally I do, but I'm on master at the moment (which doesn't work with geiser, but anyway...)

<ArneBab_>then you have to go back many months to find the old behavior :)

<ijp>and I have many months of history

<ijp>find me a log of when someone was asking about a styled-metronome-markup

<ijp> https://lists.gnu.org/archive/html/bug-guile/2014-06/msg00030.html means my log goes back a year

<ArneBab_>wow…

<ijp>that's nothing, bkuhn has over a decades worth of bash history

<ArneBab_>I should have that, too

<ArneBab_>I often hit the limits of my emacs command history…

<ijp> https://lists.gnu.org/archive/html/bug-bash/2011-06/msg00040.html

<ijp>"I have all my bash history going back to 2003-10-15 accessible to me."

<lloda>ArneBab_: I'm on 2.1.0, I cannot build stable atm, but I'd be satisfied if it's fixed there

<ArneBab_>lloda: on 2.1.0 it should be per expression I think (as on 2.0.12)

<ArneBab_>except if it wasn’t merged…

<lloda>yeah I think that might be the case

<dsmith-work>Morning Greetings, Guilers

<mark_weaver>sneek: later tell nalaginrut: again, TCO is not useful for us. if it's merely an optimization, we cannot rely on it, so it's useless for us. However, it appears that ES6 actually has "proper tail calls", which sounds like what we need. See http://web.archive.org/web/20131018025351/http://bbenvie.com/articles/2013-01-06/JavaScript-ES6-Has-Tail-Call-Optimization

<sneek>Got it.

<mark_weaver>sneek: later tell nalaginrut: and notice that the author of that article is also confused about the crucial difference between TCO and proper tail calls, but he did at least put an "Update" at the beginning of the article, where he conveys the difference.

<sneek>Will do.

<mark_weaver>sneek: botsnack

<sneek>:)

<dsmith-work>Bah. Corporate overlords prevent access to web.archive.org

***michel_mno is now known as michel_mno_afk

***karswell` is now known as karswell

***dje is now known as xdje

<paron_remote>hey civodul, if I try submitting the https stuff to guile

<paron_remote>which branch should that go to?

<paron_remote>2.2?

<civodul>good question!

<civodul>maybe 2.0 is acceptable

<civodul>though of course we'll have to be careful not to break things incompatibly

***karswell` is now known as karswell

<paron_remote>civodul: okay cool thanks

<paron_remote>time to read docs on the module system :)

<daviid>hello guilers

<daviid>sneek: seen wingo

<sneek>wingo was here Jun 24 at 08:05 pm UTC, saying: and how long prerelease will take, no idear.

<daviid>sneek: later tell wingo I'd like to release guile-gnome 2.16.3, but it would make user's life [and mine too] much easier if we also release g-wrap, which I can do, and guile-cairo: could you release guile-cairo 1.10.x?

<sneek>Okay.

<mark_weaver>paron_remote: my preference would be to add it to 2.0.x, but as civodul said it has to be done in a compatible way.

<mark_weaver>and we cannot add hard requirements that weren't there before.

<mark_weaver>hi wingo!

<mark_weaver>wingo: is it true that ES6 has guaranteed proper tail calls? how widely implemented is that among browsers today?

<paron_remote>mark_weaver: makes sense

<ArneBab_>Dedicated to Jürgen Hesser.

<ArneBab_>I once tried to convince you to teach Python instead of Scheme. You were right in the end. It seems those 30 years of experience amount for something after all.

<ArneBab_> https://bitbucket.org/ArneBab/py2guile/downloads/py2guile.pdf

<ArneBab_>^ contains all the corrections I found

<ArneBab_>(though I’ll still need to add final polish to the pdf)

IRC channel logs

2015-06-30.log