[12:09] <HiddenWolf> any server admin here?
[12:09] <HiddenWolf> GPG error on nl.archive.ubuntu.com
[12:09] <HiddenWolf> W: GPG error: http://nl.archive.ubuntu.com breezy Release: Unknown error executing gpgv
[12:10] <HiddenWolf> W: GPG error: http://nl.archive.ubuntu.com breezy-updates Release: Unknown error executing gpgv
[12:12] <Znarl> HiddenWolf : Only recently?
[12:12] <HiddenWolf> Znarl, yes
[12:13] <crimsun> wait 30 minutes and retry
[12:13] <HiddenWolf> crimsun, what changed?
[12:13] <crimsun> more than likely, it's syncing
[12:14] <crimsun> dunno about 'breezy'
[12:14] <HiddenWolf> Znarl, isn't it still a redirect to archive.u.c?
[12:16] <Znarl> nl has recently changed to a mirror.
[12:23] <Brunellus> stupid question.  how do I send a bug report using the bug reporting tool?  it seems to want to use sendmail;  how can I tell it to send the mail through my SMTP server?
[12:26] <Znarl> HiddenWolf : OK, I've pointed nl back at the master archive site.
[12:27] <HiddenWolf> Znarl, I didn't ask you to, was just alerting you to that error. :)
[12:34] <Znarl> HiddenWolf : Ok, I'll change it back.
[12:36] <HiddenWolf> Znarl, I didn't ask you to do that either. :)
[12:36] <HiddenWolf> Znarl, do what is right. I have no clue what's causing this.
[12:37] <Znarl> HiddenWolf : Bug me again if it continues to happen in an hour, ok?
[12:39] <HiddenWolf> Znarl, nog gonna happen. It's 12:30 at night.
[12:39] <HiddenWolf> Znarl, I'm tucking in.
[12:39] <HiddenWolf> g/t
[12:40] <HiddenWolf> Night all.
[01:18] <psusi> anyone know why aio_read is creating another thread instead of telling the kernel to do the IO asynchronously like it should?
[01:22] <crimsun> how do you know that kernel thread isn't doing it?
[01:22] <psusi> because when I step over the aio_read call in gdb it says a thread was created
[01:23] <crimsun> oh, you're not doing kdb stuff?
[01:24] <psusi> nope... I'm trying to patch dd to use aio to keep the pipeline full when doing copies with O_DIRECT
[01:24] <crimsun> you might want to ask in #kernelnewbies on irc.oftc.net, then.
[01:25] <psusi> hrm.... that's a channel for newbie kernel hackers?
[01:25] <crimsun> the channel is misleading.
[01:25] <crimsun> channel _name_
[01:25] <psusi> hrm.... thanks
[01:42] <psusi> hrm.. if I install the debug libc package... how can I direct a program under gdb to use it instead of the stripped one?
[01:52] <ispiked> psusi: it should automatically be able to load the symbols. at least with my experience that's how it's worked.
[02:19] <beezly> umm, i think i accidentally uploaded a package to upload.ubuntu.com earlier - that shouldn't upset anything/anyone should it?
[02:19] <crimsun> no
[02:20] <crimsun> it'll just be rejected silently unless your key is in the uploader keyring
[02:20] <beezly> excellent :) - although the world is going to miss out on my crappy remote apt util! (no great loss I assure you!)
[03:59] <psusi> when I do a tags-search in emacs, it finds things with that text even in comments... is there a way to limit it to things names of functions?
[04:17] <Burgundavia> how did the osnews thread on Flight 1 devolve to images of Trolltech shafting KDE?
[04:18] <crimsun> every thread on osnews degenerates </offtopic>
[04:18] <Burgundavia> sad, just sad
[04:19] <zakame> huhu
[04:19] <Lathiat> hahaha
[04:20] <Burgundavia> ok this one cracks me up the most
[04:20] <Burgundavia> http://www.osnews.com/permalink.php?news_id=12721&comment_id=62436
[04:20] <Burgundavia> freedesktop.org is a giant Trolltech conspiracy. I knew it!
[04:21] <Lathiat> bahahah
[04:41] <desrt> man
[04:41] <desrt> i don't understand why people take screenshots of dapper
[04:41] <desrt> it's like "wow it looks exactly like the screenshots for breezy awesome!"
[04:47] <wasabi> wonder if there is a way to make all toolbars use small icons
[04:48] <wasabi> oops wrong window
[04:57] <wasabi> woh
[04:57] <wasabi> my ibook just woke up.
[04:57] <wasabi> i let it fall asleep on accedent, and it just woke
[04:57] <wasabi> it's never done that in linux before
[05:05] <Burgundavia> why does everybody and their dog write a media player? Are there not enough in the world alrady?
[05:06] <wasabi> So I was just thinking. Samba sucks.
[05:06] <wasabi> I mean, by default.
[05:07] <wasabi> WIsh: by default you should be able to install it and it should function just like a WIndows box.
[05:07] <wasabi> Should auto hook into pam, auto generate smbpasswd, make sure it's in sync with /etc/passwd.
[05:07] <wasabi> No shares enabled by default, of course, but the basic config should be in place.
[05:08] <Burgundavia> then produce the patches to do that
[05:08] <wasabi> Thinking about it. :0
[05:09] <mpt> Burgundavia, there are no good ones, which motivates many people enough to write another one, but not enough to write a good one :-)
[05:10] <Burgundavia> mpt, yes
[05:10] <Burgundavia> the reality with media players is that no perfect one exists to satisfy all people
[05:15] <jbailey> wasabi: The password synchronisation is a bitch.  I don't think CIFS tosses passwords around in the same format shadow stores them in.
[05:15] <jbailey> wasabi: The right solution is Keberos on every machine. =)
[05:16] <wasabi> No, you're right, It doesn't.
[05:16] <wasabi> It's not possible to fix perfectly.
[05:17] <wasabi> But a default samba install should at least inform the user to change his Ubuntu password to activate his Windows access, and automatically set up PAM to do so.
[05:17] <wasabi> Basically it should assure he doesn't have to maintain two passwords.
[05:17] <jbailey> Synchronisation is teh suck.
[05:18] <wasabi> No way around it, it doesn't use the same hash format.
[05:18] <wasabi> and it's one way.
[05:18] <wasabi> For an office, yeah, Kerberos should be used.
[05:18] <wasabi> For a home? Probably not.
[05:18] <jbailey> Right, thus my assertion that Kerberos by default everywhere. =)
[05:18] <wasabi> You tell MS that.
[05:19] <jbailey> AD uses Kerberos internally.
[05:19] <wasabi> AD doesn't run on my mom's computer.
[05:19] <jbailey> I don't know what MS does for single sign on in home networks, though.
[05:19] <wasabi> They don't.
[05:19] <wasabi> They just pass hashes.
[05:19] <wasabi> And then prompt if it's wrong.
[05:20] <Burgundavia> or they let you in without a password
[05:20] <wasabi> Only on 9*. ;)
[05:20] <wasabi> Or if you enable guest.
[05:20] <jbailey> But we *could* do it on home Ubuntu networks.  Opportunistically trust other machines on the local network when someone with root privs says "trust that machine"
[05:20] <jbailey> And use that as an opporutnity to send principals back and forth.
[05:20] <wasabi> But we "don't need to."
[05:20] <Burgundavia> wasabi, you cannot disable guest on XP Home machiens
[05:20] <wasabi> =)
[05:21] <wasabi> http://www.microsoft.com/windowsxp/using/setup/learnmore/tips/goingsolo.mspx
[05:21] <jbailey> wasabi: Sure we do.  I find it annoying that I can't trivially just browse the files on the machines in my network without setting up an account, etc.
[05:21] <wasabi> I agree. We can work on that. ;)
[05:21] <wasabi> But for the purposes of getting my dad sharing files with Ubuntu in 2 months.
[05:21] <wasabi> Hehe.
[05:22] <jbailey> Dude, anything like you're thinking of is unlikely to be dapper material.
[05:22] <Burgundavia> jbailey, is there a sane fix to the browse problem?
[05:22] <jbailey> Having samba suddenly twiddle pam is a bit of an invasive change.
[05:22] <wasabi> I agree. It would need to be thought out better than what I just blurted out on IRC. ;)
[05:23] <wasabi> Maybe a samba-workgroup-config package or something.
[05:23] <wasabi> Anyways, I want it to be easy for Mortals.
[05:23] <jbailey> Burgundavia: I haven't though it through, but I'm imagining something like a directory.  When you install the machine it could zeroconf or whatever and say "look, there's a network here.  Trust it?"  And then ask for a password to attach to the other machines or something like that.
[05:23] <wasabi> Having some sort of better pluggable PAM infrastructure would be useful though.
[05:24] <wasabi> I am so tired of hand editing these things.
[05:24] <jbailey> Similar to registering a windows or netware machine into their respective security setups.
[05:24] <jbailey> The ideal bit would be somehow making it version changes to the authentication database so you didn't need a single server.
[05:24] <wasabi> jbailey: I like that idea... but it raises a lot of questions. ;)
[05:24] <jbailey> Good.  Any idea worth having should. =)
[05:25] <wasabi> In a home I still see each computer of more of an untrusting island.
[05:25] <wasabi> Where any trust between them is always initiated with sceptism, etc.
[05:25] <wasabi> People bringing in comps for a lan party, peopel driving by, people with work PCs
[05:26] <jbailey> Right.  Have to assume this whole network has real ipv4 addresses and is sitting on the public 'net
[05:26] <wasabi> Each system having it's own user base, and then, on first auth, attempting to "link" two accounts between systems.
[05:26] <wasabi> First time you browse to a PC, it asks you for the user name on that PC, which you type in, and then it asks you if you want to "link" these names, or something.
[05:27] <wasabi> Basically just shares some token.
[05:27] <wasabi> Would let you bring in a work PC, link it to a home PCs local account, for the purposes of local PC access.
[05:27] <wasabi> Obviously it would have no meaning on the work network.
[05:27] <wasabi> But isn't this just like remembering passwords? :)
[05:27] <jbailey> Right, but that machine should understand that there are multiple contexts.  tie that in with network manager or something.
[05:28] <wasabi> Or the entire idea could just be snipped in the butt, remember passwords, but offer to update changed passwords on other systems.
[05:29] <jbailey> The problems I want to solve are: 1) If I have files on my wife's machines, I should be able to get them.  2) If my wife has a file on her machine that she wants me to have, I should be able to get it without emailing it through our mail server 6000 km away
[05:29] <wasabi> Why can't you just create an account on your wifes PC?
[05:29] <wasabi> And access that file through that account from yours?
[05:29] <jbailey> I could, but then I have to remember the password for it.
[05:29] <wasabi> That's basically all we're talking about.
[05:29] <jbailey> And that I created it.
[05:29] <Burgundavia> wasabi, too much overhead
[05:29] <wasabi> No you don't.
[05:29] <wasabi> You'd have to remember that password just as much as you'd have to remember some mythical token we just talked about.
[05:30] <wasabi> For any system to function you have to a) create a security token on each system involved and store with it information that the other system knows.
[05:30] <jbailey> It's almost like I want a zeroconf AFS. =)
[05:30] <wasabi> Creating a user with a password fulfils that.
[05:31] <wasabi> Perhaps it just needs a better interface.
[05:31] <wasabi> Perhaps, for instance, the first time you try to access a machine with smb:// that doesn't have a remembered password, you are asked to a) auth or b) create a new remote account
[05:32] <Burgundavia> but that means you need to root account on that box
[05:32] <wasabi> Yup.
[05:32] <wasabi> Obviouslyl.
[05:32] <Burgundavia> because I sure has hell don't want somebody random creating accounts on my box
[05:32] <wasabi> Exactly.
[05:32] <Burgundavia> that is way too much overhead
[05:32] <wasabi> You don't want that in any system.
[05:32] <wasabi> What?
[05:32] <wasabi> Who else do you propose a system to allow access?
[05:32] <wasabi> AI?
[05:32] <wasabi> At some point a human has to step in and say Bob is allowed to access these files.
[05:34] <Burgundavia> how does OS X handle filesharing in a windows network?
[05:35] <wasabi> By default, like we do.
[05:35] <wasabi> Prompts for password. Password gets saved.
[05:35] <Burgundavia> can you see shares by default
[05:35] <Burgundavia> ?
[05:35] <wasabi> To share, it uses samba, but preconfigures it to allow local users access.
[05:35] <wasabi> And does track password changes.
[05:35] <wasabi> Doesn't share any shares by default.
[05:35] <wasabi> But once you turn it on, you can browse (nothing)
[05:36] <wasabi> actually maybe Home is shared
[05:56] <mahangu> im just downloading the first dapper image
[05:56] <mahangu> how can I get on the testing team?
[06:01] <Burgundavia> mahangu, the laptop testing team?
[06:08] <Burgundavia> minghu1, did you mean the laptop testing team?
[06:09] <wasabi> jbailey: don't suppose you know much about kerberos eh? You talk about it a lot. ;)
[06:09] <wasabi> I've got a horrid problem I'm trying to debug.
[06:11] <minghu1> Burgundavia: You probably need to ask whoever you intended to ask again.  Sorry for the nick collision :-)
[06:12] <Burgundavia> minghu1, ah, ok
[06:12] <Burgundavia> for some reason I read your two nicks as the same
[06:12] <Burgundavia> the mind is funny sometimes
[06:15] <Kinnison> night all
[06:19] <mahangu> hi, did anyone see what I wrote before?
[06:20] <mahangu> (I don't wanna ask again, sorry got d/c)
[06:41] <Burgundavia> mahangu, did you want to join the laptop testing team?
[06:43] <mahangu_> Burgundavia, yeah, i run ubuntu on a thinkpad t42
[06:43] <Burgundavia> mahangu, ok, https://wiki.ubuntu.com/LaptopTestingTeam should get you started
[06:43] <Burgundavia> the key thing is that you test EVERY feature of the laptop and file bugs for everything that doesn't work
[06:44] <Burgundavia> also install all Flight cds so feedback can come during the development cycle
[06:44] <Burgundavia> there is also a laptop-testing list that you should probably subscribe to
[06:53] <mahangu_> Burgundavia, ill do that
[06:53] <mahangu_> :)
[06:54] <Burgundavia> mahangu, thanks
[07:03] <mahangu_> Burgundavia, for running the devel images
[07:03] <mahangu_> i probably shouldn't use my primary work laptop, correct?
[07:04] <Burgundavia> mahangu_, you can dual boot
[07:08] <mahangu_> Burgundavia, yeah that's true
[07:08] <mahangu_> it means ill have to resize the partition im currently running breezy on though
[07:09] <Burgundavia> not hard, the installer can do it
[07:14] <mahangu_> Burgundavia, really? without messing my data?
[07:14] <mahangu_> i was told to boot xp and use partition magic
[07:14] <Burgundavia> yes
[07:15] <Burgundavia> no to the second part
[07:15] <Burgundavia> ajmitch, you getting another one from canonical
[07:16] <ajmitch> not that I've heard
[07:16] <Burgundavia> ok
[07:16] <ajmitch> I'd be quite surprised if I did
[07:17] <ajmitch> I've been checking out what to replace it with
[07:17] <mahangu_> ajmitch, you borked your lappy?
[07:17] <ajmitch> no, it was stolen
[07:17] <mahangu_> whoa
[07:17] <mahangu_> :S
[07:18] <mahangu_> from where?
[07:18] <ajmitch> from UBZ, at the hotel
[07:18] <mahangu_> UBZ>
[07:18] <mahangu_> ?
[07:18] <ajmitch> UbuntuBelowZero, the developer summit we just had
[07:19] <mahangu_> ah
[07:56] <zakame> hi all! :D
[07:56] <zakame> just to make sure, but is build-depending on libgamin-dev preferable to build-depending on libfam-dev?
[10:00] <Burgundavia> bon soir, neuralis 
[10:01] <neuralis> Burgundavia, g'day, sir
[10:02] <Burgundavia> have you started in on the server testing framework yet?
[10:02] <neuralis> Burgundavia, nope, the specs are in mdz's queue
[10:02] <Burgundavia> ah
[10:03] <neuralis> Burgundavia, i want to make sure he's happy with them, after which i'll also mail laptop-list re: using the backend for laptops 
[10:03] <Burgundavia> excellent
[10:03] <neuralis> Burgundavia, and i've mailed malc re: laptop certification, without response so far
[10:03] <Burgundavia> have you spoken with mjg59 regarding what specifically he needs from the reports?
[10:04] <Burgundavia> and if you have, is that recorded somewhere?
[10:05] <neuralis> no -- i want to address any concerns mdz has *first*, and then fan out and talk to the rest of the people
[10:05] <Burgundavia> ok
[10:06] <neuralis> no worries, it's on my list, it'll get done
[10:22] <mojo85> I suggest the 'Add Application' app and 'Sumit Hardware' app should use icons of the new naming scheme (Tango project)
[10:37] <pef> hello
[10:47] <MagnusR> mag runes
[10:47] <MagnusR> foobar
[10:47] <MagnusR> Wrong window, sorry!
[10:48] <neuralis> JaneW, we need to make ubuntu plushies. cute, stuffed animals that i can give to people. anything in the works? :)
[10:56] <Lathiat> neuralis: we could have little plush dragons ;)
[10:58] <neuralis> Lathiat, +1. but i'd settle for a duck, if it were dapper enough.
[01:20] <\sh> infinity / lamont-away: could one of you two check where the last upload of armagetron is hiding? latest uploaded version is 0.2.7.0-1.1ubuntu1 and it didn't appear in the buildlogs
[01:46] <\sh> elmo: please sync dnprogs from unstable, dropping ubuntu changes ok
[01:48] <jbailey> wasabi: I've done two smallish kerberos deployments (about a dozen servers in each one.  One with about a hundred users, one with three users and a whole bunch of service principals)
[01:48] <jbailey> wasabi: I have to Oreilly Kerberos book here if you have specific questions.
[03:11] <ptlo> neuralis_ hi :)
[03:12] <neuralis_> ptlo, hey
[03:50] <slomo_> BenC: the ppc kernel build failure didn't disappear with the new binutils :(
[03:51] <BenC> slomo_: assuming I can keep kernel-package updated to the latest ppc/powerpc changes, it will build in -4.4
[03:51] <BenC> I disabled HMT and NUMA for ppc64 to get that building
[03:51] <slomo_> BenC: ok, thanks... when can we expect -4.4?
[03:52] <BenC> tomorrow
[03:52] <BenC> -rc2 was tagged today, so seems like a good time (ppc64->powerpc merge is done now too, so it helps a lot)
[03:53] <slomo_> hehe, maybe i can finally try the newest bcm43xx driver then ;) seems like they made great progress in the last days
[03:58] <HiddenWolf> BenC, anything cool in -4.4?
[03:59] <BenC> hiddenwolf: devmem patch from fedora, enabled ACPI_BLACKLIST=2000, should build on sparc and ia64 now
[03:59] <BenC> lots of little changes aswell
[04:00] <HiddenWolf> BenC, sparc and ia64 are on the road to being officially supported?
[04:01] <BenC> no idea, but I know that fabbione is pushing to get sparc to be a first class citizen
[04:01] <fabbione> HiddenWolf: not yet no
[04:01] <HiddenWolf> I'm amazed that interest for IA64 is so consistent.
[04:01] <fabbione> there is nobody other than lamont giving love to ia64
[04:03] <fabbione> HiddenWolf: who is asking for it?
[04:04] <HiddenWolf> fabbione, ah, no. not specifically.
[04:04] <fabbione> BenC: i am getting an ia64 soon
[04:04] <HiddenWolf> fabbione, but afaik IA64 isn't called itanic for no reason. :)
[04:04] <fabbione> HiddenWolf: you mean Titanic?
[04:04] <fabbione> ;)
[04:04] <BenC> with all the support behind x86-64, I don't think ia64 is viable anymore
[04:06] <HiddenWolf> BenC, anandtech had an article on why IA64 was the most promising architecture, beating Sparc and PPC and X86. 
[04:06] <HiddenWolf> BenC, that was on technical merit tho, not viability.
[04:06] <BenC> hidden: it's a marketing failure though
[04:06] <HiddenWolf> BenC, Both a marketing and a sales failure, really.
[04:07] <fabbione> HiddenWolf: well..  sparc is still used a lot more compared to ia64
[04:08] <BenC> sparc64 would be a lot better if 64-bit code actually ran decent compared to 32-bit code
[04:08] <BenC> as it is, 64-bit is reserved for when you have to have that much addressing
[04:08] <BenC> otherwise it's a performance hit
[04:10] <HiddenWolf> fabbione, anand's point was more: It'll scale better to future performance.
[04:12] <fabbione> HiddenWolf: sounds like they did send a lot of ia64 to these people writing the review
[04:12] <BenC> fabbione: uploading new kernel-package now
[04:13] <fabbione> BenC: cool
[04:13] <fabbione> BenC: is it going to build on sparc?
[04:13] <fabbione> ah meh
[04:13] <fabbione> never mind
[04:13] <doko> nice, dapper_probs lists the whole archive as uninstallable
[04:13] <HiddenWolf> doko, why?!
[04:13] <fabbione> BenC: i read "new kernel source"
[04:14] <fabbione> HiddenWolf: ia64 is taking over archive.ubuntu.com!
[04:14] <HiddenWolf> fabbione, cool. :P
[04:14] <fabbione> impressive
[04:14] <fabbione> out of 6 arches
[04:14] <doko> HiddenWolf: tell me
[04:14] <fabbione> sparc is the one that can still install more than all the others :D
[04:15] <BenC> fabbione: have you tried compiling -4 on sparc yet?
[04:15] <fabbione> BenC: no
[04:15] <BenC> I should hook up my e3k in here just temporarily tonight
[04:15] <fabbione> i did let the buildd catch up
[04:16] <HiddenWolf> doko, have you upped OOo2 final to Dapper yet?
[04:17] <doko> HiddenWolf: go for it if you want ;-P
[04:18] <HiddenWolf> doko, ehm...
[04:25] <Kamion> doko: uninstallables> please rebuild python against openssl 0.9.8
[04:25] <Kamion> python2.4 that is
[04:25] <doko> ahh, ok ...
[04:25] <Kamion> we need to get openssl097 into the archive though, so that this isn't so destructive
[04:26] <Kamion> elmo: should openssl097 still be listed as broken?
[04:27] <Kamion> elmo: I think probably not - if you could remove it from josie's broken list and do an auto-sync, the uninstallables list would appreciate it :-)
[04:35] <Kamion> elmo: in fact it probably won't autosync because it's modified - I'll merge
[04:43] <Kamion> doko: ok, I've uploaded a merged openssl097; the world should look somewhat better once that makes it into the archive
[04:43] <Kamion> we should still get rid of libssl0.9.7 linkage where possible
[04:48] <BenC> it's bad when you have to get a dolly to move a computer from one side of the room to the other
[04:49] <mahangu> BenC, eh?
[04:50] <ogra> BenC, thats not bad ... bad is if you have to get a team *and* a dolly to move it 
[04:51] <mahangu> if I have a development idea (which i have half implemented), who can I talk to?
[04:51] <BenC> worst ever for me was that the dolly wasn't strong enough to hold the computer
[04:51] <BenC> took six people, and the computer was never actually lift off the ground, we had to lean it into a trailer to transport it
[04:52] <HiddenWolf> mahangu, just talk here. :)
[04:52] <mahangu> ok
[04:52] <mahangu> well, i haven't written a primer or anything
[04:52] <mahangu> but im working on this live support system for distros
[04:52] <mahangu> basically, i noticed that many new users dont know how to get on IRC
[04:52] <BenC> mahangu: you must pass the six tests of Ubuntu, and climb to the highest peak in Africa :)
[04:52] <mahangu> so I figured, why not write a nice front end for it?
[04:53] <mahangu> im new to perl and linux, but i did a hack job, and got a working prototype
[04:53] <doko> Kamion: uploaded python2.4 as well to link against 0.9.8
[04:53] <mahangu> it's basically a bot that interfaces between a thin GTK client and a support channel
[04:54] <Kamion> mahangu: the biggest problem with that sort of thing is generally setting up an appropriate support channel; #ubuntu is too busy and basically inappropriate for that sort of thing
[04:54] <mahangu> GTK client connects to IRC network --> spews data in channel (this is basically username, contact email and help request)
[04:54] <mahangu> Kamion, I'm aware of that :) just getting there
[04:54] <Kamion> doko: thanks
[04:54] <mahangu> so basically, there will have to be a seperate channel for it
[04:54] <mahangu> like #distroname-support
[04:54] <mahangu> so far, the folks at Taprobane (www.taprobane.org) , a local GNU / Linux distro seem to like the idea
[04:54] <mahangu> and im working on it
[04:55] <mahangu> i havent implemented a client yet
[04:55] <mahangu> but the bot so far
[04:55] <mahangu> 1) scans the support channel for activity
[04:55] <mahangu> if there is activity, channels request there
[04:55] <mahangu> 2) if no activity sends to support-mailing-list
[04:55] <mahangu> any thoughts?
[04:56] <mahangu> i was just thinking while handing out Ubuntu CDs the other day that none of the people I was giving them to would even know what IRC was
[04:56] <Kamion> I'm wondering if it's complementary to https://launchpad.net/support, or if the two fill basically the same use case
[04:56] <mahangu> so their support was minimal
[04:57] <Kamion> well, let's say https://launchpad.net/distros/ubuntu/+tickets, better URL
[04:57] <mahangu> Kamion, yeah, id say it'l work really well with that
[04:57] <mahangu> again, i have some pretty sparse code on the Taprobane CVS
[04:57] <mahangu> anyone think it's worth a shot?
[04:58] <mahangu> im not really an experienced coder, and have never been to CS school
[04:58] <mahangu> but im quick to learn and am willing to dedicate a big portion of my time to see this through
[04:58] <Kamion> my main concern is that IRC is actually kind of a poor medium for real customer support
[04:59] <Kamion> are you sure you want to use it as the backend?
[04:59] <mahangu> Kamion, isn't it a stable means of communication? rather than doing all that handshaking on our own?
[04:59] <Kamion> it's very vulnerable to noise, inappropriate comments, etc.
[04:59] <mahangu> Kamion, yes
[04:59] <mahangu> i have a solution (I think)
[04:59] <Kamion> and you have to sit online for ever to get a response, which is pretty awkward if you have a pay-for line
[05:00] <mahangu> Kamion, yeah that's why it sends unanswered requests to the mailing list
[05:00] <mahangu> the client will tell the user
[05:00] <Kamion> it's better if responses can come back to you asynchronously and you can log in to find if somebody's answered
[05:00] <mahangu> "Sorry looks like no one is aorund, I've sent your request to the mailing lit"
[05:00] <mahangu> Kamion, hmm, that would be neat
[05:00] <Kamion> why not just do everything through a mailing list then? it's confusing to have two means of communication
[05:01] <mahangu> Kamion, oh right
[05:01] <mahangu> so a mailing list front end
[05:01] <mahangu> that's a lot of text manip though
[05:01] <Kamion> or, as I say, the Launchpad support tracker - you have the asynchronous communication there
[05:01] <mahangu> stripping mail signatures etc?
[05:01] <mahangu> Kamion, that is web based, right?
[05:01] <Kamion> and if you talked to the folks on #launchpad they might be able to expose an XML-RPC interface for you
[05:01] <Kamion> right
[05:02] <mahangu> i was thinking of a GTK application
[05:02] <mahangu> of course
[05:02] <mahangu> yes, i understand what you are saying
[05:02] <mahangu> what is the (average) turn around time for a launchpad reply?
[05:02] <Kamion> I'm not saying you shouldn't pursue an IRC-backed option, just suggesting that you might not want to get too hung up on IRC for this, that's all :)
[05:02] <Kamion> no idea - it's never actually been advertised anywhere so the people using it are the people who happen to have come across it
[05:02] <Kamion> (so far)
[05:02] <mahangu> yeah I understand
[05:03] <mahangu> Kamion, where can I find one or more devs with a little free time who might like to run through some ideas with me?
[05:03] <mahangu> Kamion, http://cvs.taprobane.org/viewcvs.py/taprobane/hermes/hermes.pl?rev=1.7&view=log
[05:03] <Kamion> for launchpad, or in general?
[05:03] <mahangu> Kamion, well in general
[05:05] <Kamion> hmm, I guess here's as good as any for IRC, or try the ubuntu-devel@lists.ubuntu.com mailing list; it's always better to find somewhere specific if possible though
[05:06] <mahangu> Kamion, thanks, ill try
[05:06] <Kamion> (I'm mostly doing other things at the moment though, I'm afraid)
[05:07] <mahangu> np
[05:31] <mahangu> back
[05:35] <\sh> elmo: please sync grip from unstable, ubuntu changes can be dropped thx
[07:47] <desrt> so what's the story on 2.6.15?
[07:48] <wasabi> Once upon a time, a long time ago, there was a kernel.
[07:49] <\sh> elmo: please sync k3b-i18n from unstable, overriding ubuntu changes ok
[07:57] <desrt> no.  seriously
[07:57] <desrt> there are a few bugs that i'm waiting to see if the new kernel will fry
[07:58] <HiddenWolf> desrt, it's in, but unstable.
[07:58] <desrt> so why doesn't it show up on apt-get dist-upgrade?
[08:00] <\sh> elmo: please sync kdirstat from unstable, overriding ubuntu changes ok
[08:01] <fabbione> desrt: because PPC is still FTBFS on the buildd
[08:01] <fabbione> desrt: it should be fixed with tomorrow
[08:01] <fabbione> 's upload
[08:01] <desrt> ah.  excellent
[08:01] <desrt> can you do me a favour, fabio?
[08:01] <fabbione> possibly
[08:01] <desrt> view this document :)
[08:01] <desrt> http://desrt.mcmaster.ca/random/new-key.asc
[08:01] <desrt> also: explain why there are 2 'b's in your name :)
[08:02] <fabbione> the second B stands for Big
[08:03] <fabbione> like Big Fabio -> fabbione in italian
[08:03] <desrt> ah.  i see.
[08:03] <tseng> desrt: he has a tshirt that says Ride The Italian Stalion
[08:03] <fabbione> no
[08:03] <tseng> there is no modesty to be had here
[08:03] <fabbione> it says: "Italian Stallion" with Ferrari's font
[08:03] <desrt> i should count my blessings that i didn't see it :)
[08:03] <tseng> haha
[08:03] <jbailey> fabbione: So what should elmo's shirt say?
[08:04] <jbailey> What's the elmo doll called in Italy? =)
[08:04] <fabbione> jbailey: "I ride katie?"
[08:04] <fabbione> dunno..
[08:04] <fabbione> i never seen that dolls in italy
[08:04] <desrt> jbailey; i shall be in montreal in january
[08:04] <jbailey> desrt: Why are you coming here in the coldest month of the year? =)
[08:04] <desrt> jbailey; it's called CUSEC
[08:05] <desrt> canadian university software engineering conference
[08:05] <desrt> at concordia/mcgill
[08:06] <desrt> a lot of my friends are going and they didn't have a very difficult time convicing me to come along
[08:06] <fabbione> desrt: dude..
[08:06] <fabbione> your new key is sick
[08:06] <desrt> fabbione; i love it :)
[08:06] <fabbione> 3360R ???
[08:06] <jbailey> desrt: Ah, nice.  I should poke my head in there.
[08:06] <fabbione> make it 4096 and get over it
[08:06] <desrt> jbailey; you certainly should
[08:06] <desrt> fabbione; 4096 is for posers :)
[08:07] <fabbione> yeah right
[08:07] <fabbione> you are jalous that my "key" is bigger than your
[08:07] <desrt> i wanted more security without doing the cliche thing of choosing the biggest possible key
[08:07] <desrt> so i put in 3333
[08:07] <desrt> and it rounded up to 3360
[08:07] <jbailey> 3333 rounded to 3360?
[08:07] <jbailey> This isn't on an old style Pentium, is it? =)
[08:07] <desrt> by gpg logic, yes
[08:08] <desrt> old-style pentium 4 :)
[08:08] <desrt> fabbione; so i guess you have a 4096bit rsa key?
[08:09] <fabbione> desrt: 2 of them
[08:09] <fabbione> + a standard 1024 DSA
[08:09] <desrt> fabbione; you deserve the second 'b'
[08:09] <\sh> elmo: please sync gnome-themes-extras, overriding ubuntu changes ok...thx
[08:10] <fabbione> desrt: i know i do :)
[08:12] <desrt> an interesting dos attack would be to create a new strong set
[08:12] <desrt> just crank out 100s of thousands of keys and have them all sign each other
[08:12] <desrt> then upload the resulting mess to some poor keyserver
[08:13] <fabbione> desrt: you would only notice a suddenly increase of keys on the keyservers
[08:13] <fabbione> it won't affect the strongset
[08:13] <desrt> fabbione; yes it would
[08:13] <fabbione> nopr
[08:13] <fabbione> nope
[08:13] <fabbione> only if one of the key is crosssigned with the strongset
[08:13] <desrt> because my new group of keys would be the new strongset
[08:13] <fabbione> nope
[08:13] <fabbione> the strong set is not detected
[08:13] <desrt> "the strongset" is defined as the largest set
[08:14] <fabbione> the starting point of the strong set is known :)
[08:14] <fabbione> this is a limitation in basically all the tools that do key stats
[08:14] <Kamion> desrt: it's only a DoS attack on people who don't hardcode the strong set. :)
[08:14] <desrt> the only way to fix that would be to say that the strong set is the set with my key in it
[08:14] <desrt> :)
[08:14] <Kamion> so more a Denial of Theory attack
[08:14] <fabbione> Kamion: exactly
[08:15] <desrt> there exists, i understand, another set of size 98
[08:15] <desrt> someone needs to cosign those guys in :)
[08:15] <fabbione> desrt: see.. the point is that prople like you tend to disappear very fast from the net after an attack like that
[08:16] <fabbione> desrt: there are over 2M keys on the keyserver and afaik the strongset is ~200K
[08:16] <fabbione> so there is more or less a ratio to 1:10
[08:16] <desrt> that's weird
[08:16] <fabbione> no it's not
[08:17] <fabbione> most of the keys in there are single keys
[08:17] <desrt> it means that only 1 in 10 keys are effectively signed
[08:17] <fabbione> that never signed or being signed
[08:17] <desrt> so weird
[08:17] <desrt> anyway... have you noticed with rsa keys
[08:17] <fabbione> i don't find it weird
[08:17] <desrt> some keyservers think that they have different fingerprints
[08:18] <fabbione> a lot of people upload keys only becuase they have been told to reading the gpg howto
[08:18] <fabbione> and gave away the key one minute later
[08:18] <desrt> some of the web-based interfaces seem to use md5 checksums for rsa keys
[08:18] <fabbione> desrt: yeah.. old format
[08:18] <desrt> kinnison bought me a lovely purple book in montreal
[08:19] <desrt> i've written my key fingerprint inside of the cover in sha1 format
[08:19] <desrt> i really hope i don't have to erase it :p
[08:19] <fabbione> eheh
[08:20] <fabbione> http://keyserver.kjsl.com/~jharris/ka/2005-11-13/top50table.html
[08:20] <fabbione> desrt: ^^
[08:20] <fabbione> just about in the middle
[08:20] <desrt> Fabio M. Di Nitto
[08:21] <desrt> mako is quite high on the list, hm?
[08:21] <fabbione> yeah
[08:21] <Treenaks> whee, I'm 849
[08:21] <fabbione> but mako also travels N times as much as i used to
[08:21] <desrt> i wonder how i can determine my rank
[08:21] <desrt> my msd is a rather pathetic 5.something
[08:22] <desrt> actually
[08:22] <desrt> interesting
[08:22] <desrt> if you sign my key with a key that has MSD x
[08:22] <desrt> then my key automatically has a MSD of at most x+1
[08:22] <fabbione> desrt: right
[08:22] <desrt> cool
[08:22] <desrt> sign me up!
[08:22] <fabbione> so in the next calculation you will jump up to death
[08:22] <desrt> death?
[08:22] <desrt> you are 3.8199
[08:23] <desrt> so i will probably be very slightly less than 4.8199
[08:23] <fabbione> i am 25, 271, 751
[08:23] <sivang> desrt: what kind of book is it?
[08:23] <fabbione> and you got signed by all of them
[08:23] <desrt> sivang; just a notebook
[08:23] <desrt> er
[08:23] <desrt> did you just upload to the keyserver?
[08:23] <fabbione> desrt: look at your email
[08:23] <sivang> desrt: ah , I thought some interesting technical book :)
[08:23] <fabbione> + you need to sign me too
[08:23] <sivang> fabbione: keysining over IRC ? ;-)
[08:23] <desrt> ah
[08:23] <desrt> excellent
[08:23] <fabbione> sivang: no
[08:24] <fabbione> i had his documents and keys already
[08:24] <desrt> gpg:         new signatures: 3
[08:24] <desrt> gpg: sending key AFAA6FF6 to hkp server keyserver.ubuntu.com
[08:24] <desrt> thanks
[08:24] <fabbione> desrt: you need to sign my keys
[08:24] <desrt> just let me verify that the keys that just signed me are the keys that i signed with my old key :)
[08:24] <desrt> wait
[08:24] <fabbione> oh btw
[08:24] <desrt> did you accept my signatures?
[08:24] <fabbione> your script is broken
[08:24] <desrt> or ignore them?
[08:25] <fabbione> no i couldn't
[08:25] <desrt> it's scott's script :)
[08:25] <fabbione> gpg was complaning about the mail format being broken
[08:25] <fabbione> and didn't import
[08:25] <desrt> odd
[08:25] <fabbione> i use scott script
[08:25] <sivang> desrt: use Kinnison's , they are great :)
[08:25] <fabbione> that was not scott's script output
[08:25] <desrt> daniel's are so overkill :)
[08:25] <desrt> weird.
[08:25] <desrt> i made some very minor modifications to it
[08:26] <fabbione> well.. fix them
[08:26] <desrt> like having gpg ask me the passphrase each time instead of the perl script doing it
[08:26] <desrt> and if i say something other than '2' it asks me "are you sure?"
[08:26] <desrt> nothing to do with the format of the mail
[08:26] <fabbione> well the output mail was somehow corrupted
[08:26] <fabbione> and i couldn't import them
[08:27] <desrt> and i signed your key before the mods :)
[08:27] <desrt> hmmmmm
[08:27] <desrt> this is a problem
[08:27] <desrt> because i am no longer sure what keys i signed
[08:27] <desrt> lemme go through my things and see if i can find the piece of paper you gave me
[08:30] <fabbione> desrt: btw... with my signature, you will be 4 hops from Linus :)
[08:30] <desrt> linus sucks.  i want rms!
[08:30] <desrt> :)
[08:31] <\sh> hmmm...
[08:31] <desrt> (i'm already 3 hops from rms)
[08:31] <\sh> is collecting "gpg signatures" something like collecting autographs from VIPs?
[08:31] <desrt> absolutely :)
[08:31] <fabbione> \sh: gpg is more fun
[08:32] <sivang> \sh: judging by the current discussion, seems so :)
[08:32] <desrt> with gpg you can publish your signature collection to the world to brag
[08:32] <desrt> without fear of someone stealing the signatures
[08:32] <sivang> desrt: how come ?
[08:33] <desrt> fabbione; so why 3 keys?
[08:33] <fabbione> desrt: 1 (old) debian 2 (new) debian 3) canonical/ubuntu
[08:33] <\sh> hmmm...then I'm proud to have actually a signature of kinnison and lamot...which gives me a real hard on ... but linus or rms..who wants them? ,)
[08:33] <fabbione> desrt: but i still use all 3 of them
[08:33] <\sh> lamont even
[08:33] <desrt> Kinnison; !
[08:34] <desrt> kinnison and keybuk said they would sign my keys, accepted my fingerprint and have not yet done so!
[08:34] <fabbione> kids...
[08:34] <fabbione> desrt: gpg keysign is not "push me to sign your key"
[08:34] <fabbione> it happens when it happens
[08:34] <fabbione> it's not like sex that needs to finish asap ;)
[08:34] <desrt> you sound very zen :)
[08:35] <fabbione> i have learned that
[08:35] <fabbione> it took me a year to get my key signed by Martin Schulze
[08:35] <desrt> who is that?
[08:35] <fabbione> just because he kept forgetting
[08:35] <fabbione> but he did
[08:35] <\sh> fabbione: "needs to finish asap"? 
[08:35] <fabbione> sex is an obsolete and ancient form of fum
[08:35] <fabbione> fun
[08:35] <fabbione> it's much better to play PS2
[08:35] <fabbione> ehehhe
[08:36] <desrt> your wife agrees?
[08:36] <fabbione> ok i am off to spend sometime with my wife
[08:36] <desrt> hah!
[08:36] <fabbione> desrt: sometimes ;)
[08:36] <desrt> enjoy the playstation
[08:36] <fabbione> thanks
[08:36] <fabbione> \sh: than you are screwed
[08:36] <\sh> which reminds me, that this is real OT 
[08:37] <fabbione> \sh: on a sunday evening?
[08:37] <fabbione> OT?
[08:37] <fabbione> ok
[08:37] <\sh> fabbione: i'm not screwed
[08:37] <fabbione> is there anybody in here that wants to talk about Ubuntu Devel? ;)
[08:37] <desrt> i think i've thought about a way to hax0r Kinnison's scripts
[08:37] <fabbione> desrt: you are welcome to try with one of my keys
[08:38] <fabbione> but please don't flood me with mails
[08:38] <fabbione> just use the one with the less amount of UID
[08:38] <desrt> no... i mean getting daniel to sign a false uid
[08:38] <fabbione> you can't
[08:38] <desrt> he puts the fingerprints of keys that he trusts
[08:38] <desrt> and reruns his script every now and again so that if a new UID got added it sends an email there
[08:38] <sivang> desrt: I have my key signed by daniel, but still hadn't had chance to sign his ...:-/ It's been SO hectic in work since I returned from UBZ, so much backlog
[08:38] <fabbione> yeah
[08:39] <desrt> so just add a new uid with a valid email and a fake name
[08:39] <fabbione> desrt: nope..
[08:39] <fabbione> that doesn't work
[08:39] <desrt> his script checks the name?
[08:39] <fabbione> becuase the script checks if the UID has been signed by the master key
[08:39] <desrt> yes
[08:39] <fabbione> you would need to trick a lot the key to do so
[08:40] <desrt> i'm saying adding a fake uid to my own key
[08:40] <desrt> so i have 2 uids
[08:40] <desrt> Ryan Lortie <desrt@desrt.ca>
[08:40] <desrt> Jesus H. Christ <jesus@desrt.ca>
[08:40] <fabbione> even so, the fake uid is still binded to your primary key
[08:40] <desrt> i bet daniel's script would sign the 2nd uid
[08:40] <fabbione> even if you kill Ryan Lortie
[08:40] <fabbione> you can still see it in the key
[08:40] <desrt> wrong
[08:40] <fabbione> desrt: yes it will sign
[08:40] <fabbione> desrt: you are wrong
[08:40] <desrt> since i know p and q i can remove ryan lortie from the key
[08:40] <fabbione> you can revoke the uid
[08:40] <desrt> i can generate a new key without 'ryan lortie' in it that has the same fingerprint
[08:40] <fabbione> you can't delete it forever
[08:41] <fabbione> desrt: good luck :)
[08:41] <fabbione> let me know when you have done ;)
[08:41] <desrt> i don't want to do it
[08:41] <fabbione> gotta go
[08:41] <desrt> i'd totally fuck up my key in the process :)
[08:41] <fabbione> later
[08:41] <desrt> ciao!
[08:42] <\sh> fabbione: hf
[08:43] <Kamion> fabbione: oh, for the sake of some actual relevant conversation :-), I've implemented most of the stuff you need for whatever that server rescue mode thing was
[08:44] <sivang> Kamion: is it GUI in any sense?
[08:44] <Kamion> no
[08:44] <Kamion> if you want that, use a live CD
[08:44] <Kamion> rescue mode is not designed for prettiness, it's designed for "oh my god my system is hosed help me"
[08:46] <loogaroo> hi
[08:48] <sivang> Kamion: ah I see. Me and Corey discussed that it would be nice to have a GUI rescue mode for a "desktopish" installation, for the sake of "regular users" (tm)
[08:50] <sivang> can anyone comment if I'm better off using glade-gnome-2 for doing a GUI app in PyGTK or gazpacho?
[08:52] <slomo_> sivang: whatever appeals more to you... both create valid glade files so...
[08:52] <sivang> slomo_: ah nice, I was afraid there were some incompatibilities that might have prevented use of one over the other
[08:55] <\sh> but gazpacho is sometimes segfaulting...when it comes to glade glade files
[08:55] <sivang> \sh: what's glade glade files?
[08:55] <\sh> sivang: glade-2 .glade files
[08:56] <sivang> \sh: aren't we always using glade-2 .glade files? (or is gazpacho producing glade-1 files?)
[08:57] <\sh> sivang: well..actually we're using xml files (which are .glade files) but gazpacho sometimes doesn't work as expected..it crashed e.g. when I wanted to edit gajims glade file with it..
[08:58] <sivang> \sh: I see thanks for the tip. I'm already pretty used to working with glade-2 so I'll default to that :)
[08:58] <Kamion> sivang: that use case is fulfilled by the live CD
[08:59] <Kamion> we don't have resources to help out with the graphical installer at the moment, unfortunately
[09:00] <sivang> Kamion: I understand. Not a problem, good the know the use case is taken care of.
[09:08] <\sh> hmmm....when fabbione is only 4 hops from linus...
[09:09] <\sh> i'm only 3 hops from linus ,(
[09:09] <\sh> ,)
[09:13] <sivang> \sh: then if you sign my key I'm 5 hopes from linux?
[09:14] <sivang> err, s/linux/linus/
[09:14] <slomo_> BenC: is -4.4 including the softmac patch? you can get it from here ftp://ftp.berlios.de/pub/bcm43xx/
[09:15] <BenC> I just synced the latest svn today, saw the softmac log entry
[09:19] <slomo_> the stuff in svn uses softmac but the patch from that ftp is still needed afaik
[10:37] <\sh> killall -FIX  buildd 
[11:03] <Kinnison> desrt: so impatient
[11:04] <desrt> SIGN MY KEYS, BITCHES!!!

[11:06] <desrt> btw.  did you read my hypothetical attack against your keysign scripts?
[11:07] <Kinnison> no, where's that?
[11:07] <desrt> basically, it goes like this:
[11:07] <desrt> i add a new uid to my key "Jesus H. Christ <jesus@desrt.ca>" and upload it to the keyservers
[11:08] <desrt> then a few months later you re-run your script
[11:08] <desrt> (implied: i can receive email at jesus@desrt.ca)
[11:08] <Kinnison> Nice try
[11:08] <desrt> hm.  fair enough :)
[11:10] <\sh> I didn't know that  Jesus had a middle name
[11:11] <\sh> ,-)
[11:11] <slomo> daniels: will you update the whole driver stuff to 7.0 for dapper? or will we stay with 6.8?
[11:12] <Kamion> slomo: it's mostly updated already but FTBFS due to a glibc bug, which is being fixed
[11:13] <slomo> Kamion: ok, thanks... btw, the new binutils seem to have a bug... or cairo has... debian bts #340073
[11:14] <Kamion> slomo: thanks, RC bugs from Debian are already automatically imported into bugzilla for us
[11:14] <Kamion> (I saw that bug in my Ubuntu bugs folder a few minutes ago, so I know it worked ...)
[11:15] <slomo> ok ;) but it's a bug in binutils imho... only shows up in cairo and they worked around it
[11:19] <Kamion> ok, I'm not the binutils maintainer so you're talking to the wrong guy ...
[11:20] <elmo> slomo: no, it's not a bug in binutils
[11:20] <elmo> it a) doesn't only show up in cario, and b) is a genuine bug in cario and the other places it's shown up (e.g. glibc)
[11:21] <slomo> elmo: ok... thanks for clarifying... so an INT_ prefix is dropped by binutils? or what was the mistake?
[11:21] <Kinnison> desrt: Hmm, the xref wants to sign your key
[11:21] <Kinnison> desrt: I think I'll let it, this time :-)
[11:24] <Kinnison> desrt: chocks away
[11:26] <desrt> thanks
[11:26] <desrt> i do think i've signed your key... but it's not on the servers
[11:27] <Kinnison> I have a bunch of key stuff to merge
[11:27] <desrt> gotcha.
[11:27] <Kinnison> 33 mails
[11:28] <Kinnison> mostly it's because I put it there a long time ago thinking it was a good idea, and then didn't want to deal with revoking the uid etc
[11:29] <desrt> ah.  i see.
[11:31] <desrt> do you know if there is a program that is better at manipulating ascii-armor than gpg?
[11:31] <Kinnison> manipulating in what sense?
[11:31] <desrt> like something that will show you what information a particular sniplet contains and let you add/remove bits to it
[11:33] <desrt> :)
[11:33] <desrt> wanna hear something weird?
[11:33] <desrt> a bunch of the keyservers out there refuse to call my key 'afaa6ff6' and call it some other thing instead
[11:34] <desrt> they use md5 as the hash, i think?
[11:34] <Kinnison> seems unlikely
[11:37] <\sh> Kinnison: somehow evolution don't want to play nice..now i have to copy'n'paste your mail into vi and save the file..
[11:37] <\sh> Could not parse S/MIME message
[11:37] <\sh> gpg: armor header: Version: GnuPG v1.4.1 (GNU/Linux)
[11:37] <desrt> http://wwwkeys.ch.pgp.net:11371/pks/lookup?op=index&search=desrt
[11:37] <desrt> observe
[11:37] <Kinnison> \sh: odd, evo on breezy seems fine with them
[11:37] <desrt> pub  3360/4BC2F68B 2005/11/05 Ryan Lortie <desrt@desrt.ca>
[11:37] <Kinnison> \sh: then again, I tend to paste them into gpg --import anyway
[11:38] <\sh> Kinnison: evo on dapper doesn't like it
[11:39] <Kamion> desrt: bug handling keys of your wacky key length, maybe? (random uninformed guess)
[11:40] <desrt> Kamion; that's one of my guesses too except that somewhere i saw that it was using an md5 for my key fingerprint
[11:40] <desrt> (or md5, i assume... it was 128bits)
[11:41] <Kinnison> desrt: It may be a hiccough in RSA handling
[11:41] <Kinnison> I.E. the keyserver may assume an RSA key is not v4
[11:41] <desrt> nod
[11:41] <Kinnison> v3 keys had md5-looking fingerprints
[11:41] <Kamion> desrt: first, get it straight - are you wondering about the keyid, or the fingerprint?
[11:41] <desrt> Kamion; both
[11:42] <desrt> Kamion; the keyid, i think, is incorrect by virtue of coming from a fingerprint done with a different hash
[11:42] <desrt> http://wwwkeys.ch.pgp.net:11371/pks/lookup?op=vindex&search=desrt%40desrt.ca&fingerprint=on&exact=on
[11:42] <desrt>      Key fingerprint = FA 2C B4 95 EC 16 6F 0C  0E CD AA FB E9 A6 38 A1
[11:42] <desrt> weird, eh?
[11:43] <desrt> does the proper sha fingerprint for the DSA key but does (i guess md5?) for the RSA
[11:43] <Kamion> yes, as Kinnison says, RSA v3 key fingerprints were MD5, RSA v4 fingerprints are SHA-1
[11:44] <desrt> i wonder, then... if someone got my rsa key with a v3 version of pgp might they try to encrypt to me using my signing key? :)
[11:44] <Kamion> but you should really ask the keyserver admins - I doubt many people here can do more than guess