IRC Logs for #crux-devel Thursday, 2015-01-01

*** mavrick61 has quit IRC03:48
*** mavrick61 has joined #crux-devel03:50
diverseteK_: so why does Java need tomcat or maven?05:59
diversecan't people just install jar files themselves?06:00
*** Romster has joined #crux-devel06:53
diverseteK_: I just found out that it seems that most (modern?) programmers nowdays want a package manager for their language and not want to think about installing it on their system directly, so they rather have the language package manager do all the dependency thinking for them even on a separate project basis.07:42
diverseRomster: this is another example of how programmers are not always power users, they can be pretty ignorant users of operating systems too.07:44
frinnstprogrammers should not be trusted with anything07:46
frinnst:)07:46
diversethat is true in many ways07:48
diversefor example I don't trust myself when systems programming, because (as I said before) programmer intuition is not always correct07:48
diverseI value code-correctness from security and program-functionality point of view07:50
diversebut I guess that's just one way out of many07:50
diverseit's so complicated07:51
diverseI'll brb07:51
diverseback08:28
Romsteryeah like pip gems08:53
Romsterlike why package python stuff when you can pip install it. but then if i have a Pkgfile that needs a python dependency i have to package it.08:55
diverseyeah it makes system tracking hard09:05
diversebut as a I said, programmers != users09:06
diverseas in not all the time09:07
Romsterand generally most users wouldn't care as long as it worked.09:21
diverseyep09:21
Romsterdrives me crazy when i have untracked files.09:21
diverseI can't wait for someone to just ddos the server to prove a point some day09:21
Romsterwhich server?09:22
diversecrates.io09:22
Romsteryou do realise you can have round robin dns to many different servers.09:23
Romstereven freenode stuck in localhost 127.0.0.1 and 127.0.0.2 to help stop the ddos09:24
diverseright...09:24
Romsterbut then my client got stuck on 127.0.0.2 and wouldn't cycle to the next server.09:24
Romstereven more wtf if you got a irc server on that same port on your own machine.09:24
*** leo-unglaub has joined #crux-devel15:07
*** Feksclaus has joined #crux-devel19:09
teK_diverse: I never cared about maven oder tomcat. I hope to keep it this way19:45
diverseah, sorry, I thought I'd ask if you understood why this type of weird programming trend going on, as in for multiple languages, java being an example, in more depth19:51
diversebut I found out for myself by talking to other programmer who likes that19:53
diverseit's quite a sad site19:53
diverse*sight19:54
*** leo-unglaub has quit IRC21:02
*** leo-unglaub has joined #crux-devel21:03
leo-unglaubjust a short notice, i am going to dublicate a lot of XFCE4 ports into my own ports repo because they need a lot of work and the maintainer is not going to do it21:10
leo-unglaubhe is also not responding to my emails, so i am going to put them into my repository21:10
leo-unglaubso dont worry about upcomming dublicates, thats on purpose21:11
diverseleo-unglaub: I think Romster is attempting to fork the xfce4 ports for that reason21:12
leo-unglaubdiverse: i already talked to some people to take over the xfce ports myself21:12
leo-unglaubbut i did want to get the OK from the current maintainer first so there is no bad blood21:13
leo-unglaubhowever, he is not responding to my emails21:13
leo-unglaubbut i would love to become the official maintainer of it21:13
diversesepen is pretty tied up I suppose21:13
leo-unglaubthat would save a lot of work21:13
diversesounds fine to me21:13
jaegersome weird comments on that nvidia and e1000e thread on the ML, heh21:14
jaegerlike that "Many network drivers require being compiled as a module and not built directly into the kernel." one, hehe21:14
diverseoh?21:14
leo-unglaubdiverse: i think teK_ was fine with it as well21:14
leo-unglaubso i would love to do it21:15
diverseleo-unglaub: I wasn't giving an approval, I was just saying that would be great.21:16
leo-unglaubdiverse: i know21:16
leo-unglaubxfce is so awesome, but the ports don't reflect that21:16
leo-unglaubbecause so much things are broken, outdated, ...21:16
teK_leo-unglaub: did you send the mail to the ML, I must have missed it (you saidyou did, too)21:20
leo-unglaubteK_: i did21:20
teK_oops ;)21:20
leo-unglaubbut actually i did not see it on the archive21:20
teK_oh21:21
teK_which list?21:21
leo-unglaubbut then i forgot about it ... christmas and so on *g*21:21
teK_how fucking hard can correct quoting be21:22
teK_mailman just reject another post because someone wrote like "ok thanks, I'll do X" quoting several complete messages under this sentence21:22
leo-unglaubteK_: crux-devel21:22
teK_I think it is member only21:23
teK_let mecheck21:23
leo-unglaubi subscribed, thought that would be enought21:23
teK_should be, yes21:23
teK_you are and not under moderation21:24
diversejaeger: I remember a discussion you had with someone, trying to help with getting their wifi to work and it required them to compile the driver as a module instead of builtin?21:24
leo-unglaubdiverse: that was me :)21:24
teK_leo-unglaub: please resend. There are no pending messages for this ML21:24
diverseoh right21:24
leo-unglaubthe problem was that the kernel tryed to load the firmware before mounting the drives21:24
leo-unglaubcompiling it as a module helped21:25
leo-unglaubor if you want it build in you have to embedd the firmware file to21:25
leo-unglaubthere is a kernel option for that21:25
jaegermodule is nearly always the right option21:26
jaeger99.99% of the time21:26
teK_jaeger/frinnst/jue: I just added myself to: {flyspray,portdb-admin,contrib-admin}@ because actually sepen is not lifting a single finger for these things.. Ihope this will be temporary (me doing this, I have little hope sepen will return)21:26
jaegerteK_: ok. Maybe we can get him to stop by at some point to talk about whether or not he'll continue working on his stuff21:27
teK_maybe21:27
jaegerI had real trouble following that ML thread because of the formatting and HTML21:31
jaegerI *think* he fixed his xorg/nvidia problem with a reinstall and now just has the NIC issues but I'm not 100% sure21:32
teK_I didnt follow, I'm just fed up with braind-dead quoting21:32
teK_and I am not even talking about HTML *shiver*21:32
jaegerI normally don't mind HTML email, I just ignore it.... but the combination of HTML email and quoting getting all shit up makes baby jesus cry21:33
jaegerhopefully it didn't block my mail as well, I trimmed it quite down :)21:34
diversefor the wireless nic modules, why does it require having the drives mounted before loading the firmware?21:34
jaegerIf it can't see the filesystem where the firmware lives, how would it load the firmware?21:35
teK_there was only one post pending / now rejected :P21:35
jaegerheh, ok21:36
diverseI would have thought the firmware could be loaded while the kernel booted up21:37
jaegerhow would you expect it to find or access said firmware?21:37
diversemy thought would have been that it's part of the kernel runtime during boot up, but it seems to me that firmware exists as an external source to begin with.21:39
jaegerIt depends on how you build your kernel. You CAN include firmware in the kernel image if you do so explicitly21:40
jaegerI see no benefit to doing that, myself21:40
diversetoo much work?21:40
jaegerIt's more work than the default method, certainly. For no gain21:40
diversegotcha21:40
jaegerThe default method simply consists of putting firmware into /lib/firmware and loading the driver module21:41
jaegerpretty damn simple21:41
diverseyep21:41
diversethanks for filling in my forgotten blanks21:41
jaegerwelcome21:42
teK_jaeger: yes, IF the kernel wants that. I remind you of my  screwed wlan firmware (intel)21:42
teK_it will load after excatly 60s of uptime21:42
diverseI remember there was a reason, but I had to has ask why again21:42
teK_but no moment earlier21:42
jaegerteK_: that's definitely not a normal case, though21:42
jaegerthere's something wrong with that one21:42
teK_for shizzle21:43
jaegerMy 7260, for example, has no such issue, and it loads both wifi and BT firmwares21:43
jaegeruses iwlwifi21:43
teK_they surfaced 4-5 major releases ago21:44
teK_before these, itworked just fine21:44
jaeger:(21:45
jaegerconsidered replacing it with another chip? if possible21:45
teK_I could use a USB dongle21:46
teK_but I'm too tightfisted and lazy21:47
diverseteK_: by lazy you mean efficient?21:47
jaegerhttp://www.amazon.com/gp/product/B00E7QGHE6/ref=oh_aui_detailpage_o01_s01?ie=UTF8&psc=1 <-- this is what I bought... not too expensive21:47
teK_diverse: hm depends on the POV ;)21:47
jaegerefficiency would improve the situation :)21:47
jaegeror should, at least21:48
teK_I'd have to check if  that fitted into my netbook21:48
teK_I could post my thingy to the lkml, too21:49
teK_still: laaaaazy21:49
jaegerhehe21:49
teK_also: wicd sucks too, as I have to manually connect  the WLAN. On the other hand I have to do that once per  boot, so what..21:49
diversejaeger: I decided I'm going to get the same board again, which I can replace my asus21:49
diverses/which/when/21:49
diverseI'm going to take another nap, ttyl everyone21:56
teK_take care21:56
diversethanks :)21:56
jaegerlater21:57
*** jue_ has joined #crux-devel21:57
teK_did you know that arch linux holds several trademarks?23:15
teK_did you also know that one of them is:23:15
teK_"A simple, lightweight Linux distribution"23:15
teK_Per (?) coined: CRUX is a lightweight Linux distribution23:16
frinnstcool :)23:27
frinnstgood luck enforcing that. well the foundation might try to go to the hosting provider23:27
frinnstbtw jaeger, nice reply on the mailinglist. that thread was *painful* to read :)23:28
jaegerthanks. :) It was23:28
teK_pricks :>23:28
jaegerteK_: heh23:28
teK_also.. happy new year folks :-)23:31
jaegeryou too :)23:31
teK_thanks23:32
teK_I just realised that we are setting the *local* repo dir in each .httpup file23:32
teK_like: ROOT_DIR=/usr/ports/mechaniputer23:32
teK_srsly wtf? :)23:32
teK_I never liked httpup :p23:33
frinnstROOT_DIR=/lib23:34
frinnst:)23:34
frinnst"check out my awesome port ld-2.20.so! its just a file though"23:35
teK_${PORTBASE:-/usr/ports}/{REPONAME} in /etc/ports/drivers/* ?23:35
teK_omg23:35
teK_PORTS_DIR="/usr/ports"23:35
teK_from /usr/bin/ports23:36
teK_duh :)23:36
frinnstyeah something like that would be more sane probably.. with the correct path :)23:36
frinnstbut it might be painful to implement?23:36
frinnstor not since we could ignore ROOT_DIR in the .httpup file23:37
teK_the only issue I see is that ports(1) needs some configuration file so users could set PORTS_DIR23:37
teK_we could stop exporting ROOT_DIR or comment it out23:37
frinnstwanna open a bug for it?23:37
teK_sure23:37
frinnst\o/23:37
teK_jaeger and jue should tune in for some opinions  :)23:38
frinnst*back to binge-watching the wire*23:38
teK_haha :D23:38
frinnstim on s04 :)23:38
teK_I've been asking myself what frinnst was doing (recovering from new year's eve^Walcohol?)23:38
teK_when did you start23:39
teK_I heard rumors that we will be getting HTTPS support for crux.nu quite soonish23:39
teK__btw_23:39
jaegerI just started season 2 of the wire23:40
jaegerregarding the ROOT_DIR thing, I honestly don't care as long as it works. :) If it's cleaner without, I won't object23:41
teK_echo "$COMMAND: only root can update ports"23:44
teK_^_°23:45
teK_Lennart acting on mailing lists: http://img.pr0gramm.com/2015/01/02/ec17d2c8df42cc2e.webm23:48
teK_(sorry :p)23:48

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!