IRC log of #maemo-ssu for Tuesday, 2012-08-28

*** nox- has joined #maemo-ssu00:01
*** toxaris has quit IRC00:03
*** javispedro has joined #maemo-ssu00:24
javispedrofreemangordon: wtf are you doing00:25
javispedrofreemangordon: http://gitorious.org/community-ssu/hildon-im-vkbrenderer3/blobs/master/src/hildon-im-vkbrenderer.c#line185400:25
javispedrofreemangordon: you cannot just paste a IDA decompiler output and relicense it under GPL300:25
javispedros/GPL3/GPL/00:26
infobotjavispedro meant: freemangordon: you cannot just paste a IDA decompiler output and relicense it under GPL00:26
*** javispedro has quit IRC00:39
DocScrutinizer06hmmmm, sounds like a valid concern01:04
*** mase_76 has joined #maemo-ssu01:09
freemangordonhmm, IDA has licensed variable names atarting with v? I was not aware of that, but if that is the case i'll rename them01:09
freemangordons/atarting/starting/01:10
infobotfreemangordon meant: hmm, IDA has licensed variable names starting with v? I was not aware of that, but if that is the case i'll rename them01:10
freemangordonwhat about unk1-unkN? is that ok?01:10
freemangordonjavispedro, DocScrutinizer06 ^^^ ?01:11
*** mase76 has quit IRC01:12
DocScrutinizer06hahaha01:17
freemangordonDocScrutinizer06: unfortunately it is not funny :(01:19
DocScrutinizer06i ndeed01:19
freemangordonBTW I actually made a mistake, but it is here http://gitorious.org/community-ssu/hildon-im-vkbrenderer3/blobs/master/src/hildon-im-vkbrenderer.c#line173101:19
DocScrutinizer06you actually think I'll waste my time on continuing with that conversation?01:20
freemangordoni'll remove that ASAP, but please have in mid, I was dooing that in late night, I am human too01:20
freemangordonDocScrutinizer06: no01:20
freemangordoncode deffinitely needs lots of cleanup01:20
freemangordonactually I think you might be of great help here :)01:21
freemangordonas I am not much in licensing shit01:22
freemangordonDocScrutinizer06: ^^^01:22
freemangordonit is just that I don't like statements like "WTF are you doing"01:23
freemangordonafter all the reason for publishing that is that someone else could take a look and contribute, help or whatever. Without pointing me "you stupid, what have you done"01:26
* DocScrutinizer06 suggests song=fopen("/allmyMP3/rolling stones greatest hits.mp3"); write(song, 0x00 0x3f ..... {~60MB bytes as hex})01:27
DocScrutinizer06and prepend that with GPL disclaimer01:27
freemangordonDocScrutinizer06: you may want to check the firmware sources in kernel ;)01:27
freemangordonsome of them of course01:27
DocScrutinizer06why? because they did it better than my syntactically incorrect and incomplete 3-liner c proggie above?01:28
freemangordonbut that is not the point, I really won't argue as I actually agree01:28
freemangordon(believe it or not :P )01:29
freemangordonthe point is that I need help, I don't know if you check it, but it is something about 150k of source code, which I know needs cleanup01:30
DocScrutinizer06first of all clean up the GPL!01:30
DocScrutinizer06it's the worst offense in all that stuff/action01:31
freemangordonDocScrutinizer06: would you rephrase? to remove the header or what?01:31
DocScrutinizer06replace by proper disclaimer like "for educational purposes only! not for copying!"01:31
freemangordonaah, ok :)01:31
freemangordonthen what?01:32
freemangordonchange the variable names?01:32
* freemangordon is not kidding01:32
DocScrutinizer06thewn nfc what to do with it. RE is gray zone anyway01:33
freemangordonbtw we have several RE things in CSSU, cbs widget being a part of them. all is LGPL licensed.01:33
DocScrutinizer06but *proper* RE wouldn't include any disaaembling (also no kidding, though I know *nobody* could do it like that)01:33
freemangordonBTW the license is not GPL but LGPL if that matters01:34
DocScrutinizer06nope01:34
freemangordonDocScrutinizer06: there is no dissasembling01:34
freemangordonjust a couple of variables named v1-vN01:34
DocScrutinizer06you transformed (C) materiel to another representation, that doesn't relieve the (C) from it01:34
freemangordonreally a couple01:35
freemangordonwhich one is (C)01:35
freemangordon?01:35
DocScrutinizer06it looks like a disassembler listing that gives us bit identical binary when assembled01:35
freemangordonhave you ever seen hexrays output?01:36
DocScrutinizer06I've seen disassembler output01:36
freemangordonand it is not bit identical01:36
DocScrutinizer06the point is, it's not your work, it's derived work from a binary01:36
DocScrutinizer06that is (C) aiui01:36
freemangordona lot of the drivers in kernel are the same01:37
freemangordonand there is no problem kernel to be GPL licensed01:37
freemangordonso WTF is the problem with FKB?!?01:37
DocScrutinizer06well, really I know where it's from, I know there's probably no other way, and I think you just shouldn't put a GPL header in there. No idea what else could be done with reasonable effort to form a proper genuine intelectual property work from that01:38
freemangordonwell, maybe I should wait javispedro to appear again and give some more info besides "WTF"01:40
DocScrutinizer06to have your own licence header (whatever flavour) *you* have to write the code01:40
freemangordonbecause if it is not LGPL licensed, I don't see a way it to be distributed01:40
freemangordonDocScrutinizer06: yep, (L)GPL :D01:40
freemangordonaah, sorry01:41
freemangordonmisread what you have wrote01:41
freemangordon*written01:41
DocScrutinizer06that's the problem with (C)Nokia blobs: we MUSTN't distribute them, no matter if in original binary form or as disassembler listing01:41
freemangordonDocScrutinizer06: hmm, incorrect, we have the "qgil's exception" for community driven projects01:42
*** arcean has quit IRC01:42
DocScrutinizer06but as mentioned before, I think Nokia won't care as long as you don't claim it's "yours"01:42
freemangordonwell, I can't just (C) Nokia :(01:42
DocScrutinizer06yep, that's why I think you might get away with it, if and only if published without GPL header01:42
DocScrutinizer06well, why not? :-)01:43
*** M4rtinK has quit IRC01:43
freemangordonhmm, I am ok with that actually, I don;t give a shit whether my name is in the header or not :)01:44
DocScrutinizer06you could write header "(C) Nokia. This is derived work to allow owners of the binary to user their right to modify it"01:44
freemangordonDocScrutinizer06: if you gimme the exact license, i'll replace it tomorrow01:44
DocScrutinizer06"you mustn't use or distribute that code unless you own a N900"01:44
freemangordonbut i think javispedro was talking about IDA, not about Nokia01:44
DocScrutinizer06nah01:45
DocScrutinizer06He meant exactly this01:45
freemangordonwell, I'll ask him tomorrow (if his majesty is int the mood to gimme some meaningful answer)01:45
DocScrutinizer06:-)01:45
freemangordonI told you, it is not funny :(01:46
freemangordon:P01:46
DocScrutinizer06he probably is when you don't come like you did above01:46
DocScrutinizer06I actually thought it was a lame attempt to be funny01:46
DocScrutinizer06or sarcastic01:46
freemangordonlame?01:46
freemangordonit was funny :D01:46
freemangordonwell, half sarcastic, half serious01:47
freemangordonActually I didn't get what the problem is, now it is clear01:47
DocScrutinizer06try to leave out the sarcastic part, will not help to solve the issue ;-D01:47
DocScrutinizer06o/01:47
DocScrutinizer06afk01:47
freemangordonI will, I was a bit angry on someone (even he is javispedro) coming and pointing at me "you stupid",without much info. Having in mind I spent the last month on that01:49
DocScrutinizer06awesome work btw :-)01:50
freemangordonDocScrutinizer06: I am going AFK too, will appreciate if you prepare that license for me, I am ok with whatever license you find appropriate01:51
freemangordonDocScrutinizer06: thanks :)01:51
DocScrutinizer06I don't think I'm the right one to find the proper "licence" for that01:51
freemangordonI really don;t give a shit if my name is there or not01:51
DocScrutinizer06first best approach: no licence at all01:51
freemangordonas long as it works and can be distributed to n900 users01:52
DocScrutinizer06it for sure can be distributed, the question is how01:52
freemangordonDocScrutinizer06: well, hildon-input-method is GPL01:52
freemangordonand it uses those libs ;)01:52
DocScrutinizer06hmmm01:53
DocScrutinizer06might be a vector to tackle that problem01:53
DocScrutinizer06but I haven't got the head for that, right now01:53
DocScrutinizer06need a beer and then sleep01:53
freemangordonCopyright (c) 2005-2007 Nokia Corporation.01:53
freemangordonhildon-input-method is licensed under the terms of the LGPL version 2.1.01:53
freemangordonok, bb, i have to go too01:53
freemangordonnight01:53
DocScrutinizer06n801:54
*** peetah_ has quit IRC01:59
*** mase_76 has quit IRC02:02
*** povbot has joined #maemo-ssu02:27
*** ChanServ sets mode: +v povbot02:27
*** DMBoyCloud has joined #maemo-ssu03:08
*** gri has quit IRC03:12
*** DMBoyCloud has quit IRC03:14
*** taziff1 has quit IRC03:14
*** taziff has joined #maemo-ssu03:15
*** gri has joined #maemo-ssu03:23
*** taziff1 has joined #maemo-ssu03:52
*** taziff has quit IRC03:52
*** nox- has quit IRC05:30
*** taziff1 has quit IRC05:34
*** taziff has joined #maemo-ssu05:34
*** amiconn_ has joined #maemo-ssu05:57
*** amiconn has quit IRC05:57
*** amiconn_ is now known as amiconn05:57
*** peetah has quit IRC06:48
*** LaoLang_cool has joined #maemo-ssu07:47
*** LaoLang_cool has quit IRC08:00
*** taziff has quit IRC08:18
*** taziff has joined #maemo-ssu08:18
freemangordonDocScrutinizer06: A work created by observing how the original functions, and then09:26
freemangordonindependently writing code that performs the same function, is not a09:26
freemangordon"derivative work" in the sense defined in copyright law09:26
freemangordonaccording to internet :)09:26
*** peetah has joined #maemo-ssu09:50
DocScrutinizer06exactly09:54
freemangordonwell, I don't get the problem then, really09:55
DocScrutinizer06that's a proper reverse engineering, from observing how stuff *works* and creating your own code to do the same09:55
freemangordonyep09:55
freemangordonand there is no definition/restriction afaik on what tools one should use to do the observations09:56
DocScrutinizer06while decompiling is not looking how stuff *works*, but how it's *built*09:56
freemangordonyeah, right, don't tell me gdb is the only way09:56
DocScrutinizer06particularly if you're using parts of the decompiler output and copy it to your own source - but you already said you didn't do that09:57
freemangordonDocScrutinizer06: I used some parts, javispedro pointed to one of them09:58
freemangordonthus my question about "unkN"09:58
DocScrutinizer06hmm, then that's his concerns09:58
freemangordonand that is why I asked about variable names starting with v, no matter that you took it offensive09:58
freemangordonit was not meant to be09:59
DocScrutinizer06even partial copy of any representation of original code is an act of (C)-violation and counterfeit code09:59
freemangordonwell, could be, but I still cannot grok what needs to be done. I mean - is variables name change enough?10:01
freemangordonhave in mind those names are not part of the binary10:01
freemangordonthere is no debug info or whatever10:02
DocScrutinizer06sure, the problem is that it suggests that you copied the code as seen by disassembler, rather than writing your own code10:04
freemangordonDocScrutinizer06: http://pastebin.com/NSgzvu6v10:04
freemangordonhere is how hexrays outpu looks like10:04
freemangordoncompare that to what I published10:05
*** jon_y has quit IRC10:05
*** jon_y has joined #maemo-ssu10:05
DocScrutinizer06honestly, I can't help here, other than tellung you I thought you copied code from doisassembler when I looked at that URL javispedro posted here10:05
DocScrutinizer06and I got no time to discuss this now, and also I'm no expert10:06
DocScrutinizer06please use wikipedia and/or other publicly available source of info about (C) and RE and GPL implications10:07
*** taziff has quit IRC12:00
*** taziff has joined #maemo-ssu12:04
*** taziff1 has joined #maemo-ssu12:07
*** amiconn_ has joined #maemo-ssu12:08
*** amiconn has quit IRC12:08
*** amiconn_ is now known as amiconn12:08
*** taziff has quit IRC12:16
*** DocScrutinizer51 has quit IRC12:16
*** DocScrutinizer51 has joined #maemo-ssu12:28
*** arcean has joined #maemo-ssu13:38
*** ivgalvez has joined #maemo-ssu13:56
ivgalvezfreemangordon ping13:56
ivgalvezYou can use any binary blob copyrighted by Nokia in Open Source projects for Nokia devices, as long as you include the proper copyright notice.The same applies for derivative work such as RE. Do not relicense it as (L)GPL but put a notice saying saomething like "This is a derivative work from copyrighted material by Nokia. It's only intended for educational purposes on Free Open Source14:00
ivgalvezprojects and can only be used in Nokia devices"14:00
ivgalvezQgil was clear at that respect, there is no problem in using Nokia stuff (but not from 3rd parties) with that limitations14:01
*** lizardo has joined #maemo-ssu14:23
freemangordonivgalvez: thanks, could you point me to the place I can get the exact license?15:11
*** arcean has quit IRC15:39
ivgalvezfreemangordon: what do you mean by exact licence?16:00
*** arcean has joined #maemo-ssu16:00
*** arcean_ has joined #maemo-ssu16:07
*** krayon has quit IRC16:11
*** arcean has quit IRC16:11
*** NIN101 has joined #maemo-ssu16:31
freemangordonivgalvez: sorry, I meant the exact wording. Is there a place I can get an example from?16:32
ivgalveznot really16:40
ivgalvezbut you could use the paragraph I wrote16:41
ivgalvezthe question is that you cannot change original Nokia license or copyright notice16:41
ivgalvezwhatever they were16:41
freemangordonok16:41
ivgalvezbut you can add a simple fair use notice with that sentence16:41
arcean_merlin1991: https://gitorious.org/community-ssu/hildon-desktop/merge_requests/2516:47
keriohm, what's that supposed to do?16:53
keriomake the thumbnail gray before making it disappear?16:53
arcean_kerio: yes, something like that :)16:53
kerioneat16:54
arcean_also blurless desaturation option produces sharper textures16:54
kerioblurless desaturation is the shit, i agree16:58
*** ivgalvez has quit IRC17:00
*** jon_y has quit IRC17:10
*** jon_y has joined #maemo-ssu17:12
freemangordonarcean_: you're my hero, at last we have that :D17:20
*** krayon has joined #maemo-ssu17:21
*** krayon is now known as Guest1068017:21
*** Krayon has joined #maemo-ssu17:22
*** DocScrutinizer05 has joined #maemo-ssu17:50
*** DocScrutinizer06 has quit IRC17:53
*** toxaris has joined #maemo-ssu19:03
*** MrPingu has joined #maemo-ssu20:17
freemangordonmerlin1991: ping20:37
*** BCMM has joined #maemo-ssu20:48
merlin1991freemangordon: pong20:49
freemangordonmerlin1991: I know you are on holiday and with bad internet, but is there any chance to create a repo for glibc on gitorious?20:49
merlin1991why do we need one?20:50
merlin1991afaik we don't change stuff on glibc, or is this thumb related?20:50
freemangordonhmm, ppoll/pselect?20:50
freemangordonwell, the package is named 2.5.1-1eglibc27+0m520:51
merlin1991it's a recompile of old sources20:51
merlin1991no change whatsoever20:51
freemangordonmerlin, patches for pselect (and co) should be applied to both kernel and libc20:52
freemangordonwe have the kernel in CSSU, but not libc20:52
*** M13 has joined #maemo-ssu20:53
merlin1991freemangordon: my understanding was that we only have to recompile glibc from source and it will pick up the pselect support in the kernel headers?20:54
freemangordonhmm, could be, but don't we want the history? at least changelog and such20:55
freemangordondebian/control, etc20:55
merlin1991well I'm atm fighting vs my bad internet20:56
merlin1991might get a gitorious repo in a few mins :D20:56
freemangordon:D ok20:56
merlin1991need to reload pages all the time because of timeouts :/20:56
merlin1991freemangordon: what would the repo title be?20:57
freemangordonglibc?20:57
freemangordonaccording to debian/control the source package is glibc20:58
freemangordonmerlin1991: BTW did you see the discussion on VKB license and such?20:59
merlin1991nope I didn't have the time to read the backlog20:59
freemangordonok20:59
merlin1991freemangordon:  https://gitorious.org/community-ssu/glibc20:59
freemangordonthe fuck, source package contains .svn directory21:00
merlin1991lulz21:00
merlin1991somebody didn't invoke dpkg-buildpackage properly :)21:00
merlin1991but that hints, that there should be svn source avaiable somewhere on maemo.org21:01
freemangordonyes, I am trying to find it21:01
freemangordonjuri.research.nokia.com21:02
freemangordonWT?21:02
freemangordonF21:02
freemangordonsvn+ssh://juri.research.nokia.com/maemo-sdk/svn/tools/glibc/glibc-2.5.121:03
kerioooh, is that the nokia repo with the updated flash player?21:03
merlin1991kerio: that's glibc nothing flashy whatsoever :D21:04
kerioi meant the whole repo21:04
merlin1991nah21:04
freemangordonwell, cannot find anything, lemme try to svn browse it21:04
merlin1991host not found for the dns here21:05
keriomerlin1991: and where would *that* repo be?21:05
merlin1991hidden places ;)21:05
keriohow hidden, exactly? (;21:09
keriosharing is caring!21:09
merlin1991actually I don't have the url here21:09
kerioalso, does that flash player work properly?21:10
merlin1991never tried it21:10
keriohm, would that be cssu material?21:12
*** M13 has quit IRC21:24
freemangordonmerlin1991: could you rename kernel-cssu to kernel?21:26
freemangordonjust for the sake of consistency21:26
freemangordonmerlin1991: any objections against "Backport c99/gnu89 inline support patch from glibc-2.6 to support new gcc."21:36
freemangordonwell, I will push it, it can be easily disabled through series21:36
freemangordon(if needed)21:37
keriofreemangordon: "kernel" should be the omap1 kernel with bugfixes, though21:40
freemangordonkerio: feel free to do it21:41
keriono u21:41
MrPingukerio: same idea I had about "cssu" kernel21:41
freemangordonI am going to do what was agreed with the CSSU maintainer21:41
keriosrsly though, kernel-power as "kernel" is probably not a good idea21:41
MrPinguhowever I don't mind having a kernel with more features in CSSU ;)21:42
*** EdLin has left #maemo-ssu21:42
freemangordonand who told you that KP is goingt o be "the kernel"21:42
freemangordonkerio: ^^^21:42
MrPinguSince cssu3 is kp51?21:42
keriobecause kernel-cssu is kp5121:42
kerioyep21:42
freemangordoncssu3 is not distributed with official CSSU21:42
MrPingutrue ^^21:43
freemangordon"kernel" will contain those patches from KP which are considered usefull and safe21:43
MrPinguMay I ask what features are left out in official cssue kernel if one compares to kp21:44
freemangordonMrPingu: it is still not cleaar21:44
MrPingufreemangordon: Thanks, actually everything is fine to me as I am probably going with KP...21:45
freemangordonmost probably USB hostmode will not be21:45
freemangordonbut it is still to be decided, part of my job will be to prepare a list of the patches and my opinion on whether we should include them or not.21:46
freemangordonthen the others could take that and agree/not21:46
kerioyeah, usb hostmode as pali made it requires bq24200_charger.ko which is borked21:46
freemangordonkerio: no21:46
keriono what?21:47
freemangordon<kerio> yeah, usb hostmode as pali made it requires bq24200_charger.ko which is borked21:48
keriowell, it *is* borked!21:49
freemangordonUSB hostmode (gui) does not need it21:49
freemangordonnoone forces you to use Pali's version21:49
freemangordonthere is h-e-n gui on repos21:49
keriothat's why i said "usb hostmode as pali made it"21:49
kerioand that's why i mentioned it as we were talking about which patches to include in the cssu kernel21:49
kerioand that's also why i'm still using h-e-n and not usbmode21:50
kerio.sh21:50
freemangordonwell , those bqxxxx modules for sure won;t make it in cssu until proven stable. though on the other hand thay are disabled by default, so will see21:50
*** macmaN has quit IRC22:26
*** macmaN has joined #maemo-ssu22:29
*** arcean_ has quit IRC22:35
*** arcean has joined #maemo-ssu22:36
*** Free-MG has joined #maemo-ssu22:54
*** macmaN has quit IRC23:00
*** macmaN has joined #maemo-ssu23:03
*** MrPingu has quit IRC23:06
*** andre__ has quit IRC23:20
*** Free-MG has quit IRC23:21
*** NIN101 has quit IRC23:52
*** nox- has joined #maemo-ssu23:59
*** lizardo has quit IRC23:59

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!