IRC channel logs

2016-06-15.log

back to list of logs

<floor13>should I be using ggspec for my tests or srfi 64?
<floor13> http://srfi.schemers.org/srfi-64/srfi-64.html
<galex-713>What is average and medial time for bug report handling in guile?
<dsmith>galex-713, Depends. For master, usually spurts. Like they might queue up for months and then most all get fixed in a few weeks. I think whatever is currently stable is more even.
<dsmith>galex-713, Just my (possibly incorrect) observations.
<dsmith>galex-713, More people work on stable bugs I think.
<dsmith>But since master is usually really in flux, it's hard to fix things.
<dsmith>galex-713, But it all really depends on who has time to work on whatever.
<galex-713>what do you mean by in flux?
<dsmith>changing rapidly
<galex-713>ah ok
<galex-713>then things are fixed in releases you mean?
<dsmith>Mostly. Yeah.
<galex-713>Like devs wait bug to accumulate for months then fix everything at the release time?
<galex-713>Okk
<dsmith>Hmm. Not "waiting" so much as too busy adding to features to spend time on bugs.
<dsmith>Again. Just the way it seems from my perspective.
<galex-713>yeah, not waiting, that’s an image, I suppose if there would not so much features to add they would work on bug but that doesn’t happen because each time a dev has free time she has ideas to implement
<galex-713>dsmith: yeah but your persective is certainly better than mine since I just suscribed to bug-guile and had no idea of this :)
<dsmith>Heh
<dsmith>galex-713, Also, because the codebase can change so radically in master, the bugs might not even be there anymore...
<galex-713>oh I see
<galex-713>dsmith: are there guile people in rnrs/srfi comitees?
<galex-713>or is guile “not stable enough” to try export stuff for the outside world?
<galex-713>*not stable enough yet
<dsmith>galex-713, I don't know. Don't think so. I think mark_weaver was active there at one time.
<galex-713>dsmith: if not so, any idea of why?
<dsmith>Time and exaustion more than anything I would guess. Standards work it tedious.
<galex-713>dsmith: btw I was asking myself, is there a way to estimate schemes communities comparatively? (like size of guile, racket, etc. communities)
<galex-713>dsmith: why is it tedious?
<galex-713>(say it me if I ask too much question/at a too big rate)
<dsmith>Because people are idiots. And its hard to get them to see reason. ;^)
<dsmith>And you have to fight tooth and nail to get something reasonable.
<galex-713>Wouldn’t sociology of programming and reasonning upon memeticly evolutionistly human hardwired irrationality be a better explanation that “people are idiots”? :p
<dsmith>galex-713, Yeah. Idiots is too strong. But you get the idea.
<galex-713>dsmith: also I heard there’s r7rs and r7rs-large + srfis, if some scheme people disagree on certain things, wouldn’t be efficient to make larger substandards with other implementation we could partially agree with?
<dsmith>I couldn't say
<galex-713>ok
<dsmith>galex-713, I have been involved with a few standards bodies (LXI, IVI) and it can be very tedious dealing with conflicts. People can be very opinionated.
<galex-713>even without technical arguments/with counter-arguments?
<galex-713>dsmith: btw, do standard making work by consensus or by most-people-are-ok-then-let’s-do-it?
<dsmith>The groups I;ve been involved in have been mostly by consensus.
<galex-713>dsmith: then what if someone come and disagree with everybody?
<dsmith>galex-713, Sorry, got to go the store...
<galex-713>see you later then?
<dsmith>galex-713, Well, most people want the best. Some people are just, well... The word used was "obstreperous".
<dsmith>galex-713, It's not *that* big of a deal. People will disagree about things. Working though those takes time and energy. Especially if you *care*.
<galex-713>dsmith: but, do that mean languages comitees are sybil-attack fault?
<dsmith>galex-713, Dunno. I've never been on a language committee
<galex-713>open consortiums at least
<amz3>Héllo!
<galex-713>Hèllo :p
<amz3>what's up galex-713 ?
<galex-713>Well, I’m re-getting interested into gnunet&cie again x)
<amz3>wah!
<amz3>because of the latest news about decentralized web?
<galex-713>amz3: what decentralized web?
<galex-713>No because I re-gain interest into routing and low-level stuff on a regular basis
<amz3>galex-713: Tim Berners Lee the creator of the web is thinking about making a new web which will be more decentralized
<amz3>I mean he started working on it more loudly
<amz3>there is several recent article on hackernews about the subject
<amz3>it even made it do new york times
<galex-713>amz3: people did that before him
<galex-713>i.e. retroshare, gnunet
<amz3> http://www.nytimes.com/2016/06/08/technology/the-webs-creator-looks-to-reinvent-it.html
<amz3>galex-713: yeah I know, IPFS and other stuff he is not involved, but maybe if there is a standard something like the web but decentralized can emerge
<amz3>nobody can compare the web to things like gnunet or IPFS
<amz3>I don't say IPFS is the solution, just the name I can recall
<random-nick>didn't he just work on the WWW system?
<galex-713>omg, please say me he’s not basing his stuff on bitcoin…
<amz3>I tried one solution recently which is based on Bittorrent DHT and bitcoin for replacing DNS which works great as far as the website is not too big
<amz3>galex-713: of course, they are think about bitcoin
<amz3>ad based revenue is one of the failure of the web according to TBL
<galex-713>amz3: that’s a common thing that’s people with social capital/influence/power say they will revolution something while attempt already are coming on their way since long time ago and they don’t care and do as if it were their idea and they were so smart they had this idea alone and the first and etc. to get all credit while, at the end, not doing anything useful
<amz3>galex-713: I'm not sure who will get the credit, but I think they work together
<galex-713>amz3: bittorrent is the basical example, but suffers from a LOT of security holes
<galex-713>bitcoin cannot work and it’s broken
<amz3>galex-713: which security holes?
<galex-713>DHT poisoining, sybil attack, bad encryption, metadata tracking, etc.
<galex-713>of course I’m talking about the case you take the trackers apart the equation, because they’re even worse
<galex-713>amz3: the only way to base a name system on bitcoin is to base it on ad, as did Twister, then if he want to use bitcoin and remove ads he’s facing a wall
<galex-713>bitcoin/blockchain is all about money/transactions
<galex-713>money has to come from somewhere, people won’t mine for nothing
<galex-713>if that don’t give them power, influence, etc.
<galex-713>that’s why bitcoin is inherently broken, among other things
<amz3>money is broken
<amz3>galex-713: I don't know what they plan to do to fix those issues, I guess we will have to live with those issues as much as we live with bad web
<wingo>meep
<dsmith>wilo, pthup pthup pthup
<dsmith>Bah
<amz31>héllo :)
<wizardedit>hi there, I'm trying to compile guile-1.8.8 with clang on gentoo, which fails with ERROR: Wrong type argument in position 1: #<freed cell 0x7f69d9d8f180; GC missed a reference>, similar to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=650761 (debian ia64) http://pastebin.com/9MCL2wjK. How can I debug this?
<wizardedit>I've tried the freebsd patches mentioned at https://bugs.gentoo.org/show_bug.cgi?id=427346, but that made no difference
<dsmith>wizardedit, I wish you well. Debugging GC stuff is hard. Unfortunately, I don't thing there is much motivation for fixing things in 1.8. 2.0 has been out for years, and there are rumors that 2.2 might be released soon. "soon" as in maybe in 2016.
<daviid>wizardedit: guile-1.8 has been unmaintained for such a long time that you will hardly find any help from any one I'm afraid. I suggest you you grab guile-2.0 instead, the latest stable is 2.0.11
<wizardedit>daviid, my goal was to fix the bug in the gentoo tree, but thanks for the info. Maybe we should focus on dropping 1.8 instead
<wizardedit>was hoping there was some easy fix I was missing :/
<daviid>wizardedit: you definitively want to get rid of 1.8 in all distro
<wizardedit>daviid, okay. Thanks for clear statement :)
<dsmith>wizardedit, Yeah.
<daviid>welcome!
***heroux_ is now known as heroux