Timestamps are in GMT/BST.
[0:20] * blank_ (n=smains@ERR.CYLAB.CMU.EDU) has joined #freenet
[0:53] * MrNaught (n=mrnaught@S01060013103e9ce9.ed.shawcable.net) has joined #freenet
[2:58] * sanity (n=ian@81-178-100-27.dsl.pipex.com) has joined #freenet
[3:00] * Zxcvb (n=msmustdi@12-215-141-137.client.mchsi.com) has joined #freenet
[3:00] <Zxcvb> how dependant is freenet on the root DNS servers?
[3:03] * NullAcht15 (n=NullAcht@dslb-082-083-209-053.pools.arcor-ip.net) has joined #freenet
[3:17] * Zxcvb (n=msmustdi@12-215-141-137.client.mchsi.com) Quit ("Leaving")
[3:34] * NullAcht15 (n=NullAcht@dslb-082-083-209-053.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[3:35] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) Quit (SendQ exceeded)
[4:03] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) has joined #freenet
[4:19] <nextgens> hi
[4:20] * spaetz (n=spaetz@62.48.121.241) has joined #freenet
[5:20] * NullAcht15 (n=NullAcht@dslb-082-083-232-185.pools.arcor-ip.net) has joined #freenet
[5:40] * tav (n=tav@host-212-158-205-5.bulldogdsl.com) Quit (Read error: 104 (Connection reset by peer))
[5:43] * tav (n=tav@host-212-158-205-5.bulldogdsl.com) has joined #freenet
[6:18] * sleon|tuX (n=sleon|tu@e180010214.adsl.alicedsl.de) has joined #freenet
[7:16] * sleon|tuX is now known as h4x0r
[7:18] * h4x0r is now known as h4x0r|cr4k3r
[7:26] * h4x0r|cr4k3r is now known as sleon|tuX
[8:16] <nextgens> sanity: here ? :)
[8:44] * NullAcht15 (n=NullAcht@dslb-082-083-232-185.pools.arcor-ip.net) Quit (Remote closed the connection)
[8:45] * MrNaughty (n=mrnaught@S01060013103e9ce9.ed.shawcable.net) Quit ("\(^_^)/'")
[8:45] * MrNaught (n=mrnaught@S01060013103e9ce9.ed.shawcable.net) Quit (Read error: 104 (Connection reset by peer))
[8:45] * NullAcht15 (n=NullAcht@dslb-082-083-232-185.pools.arcor-ip.net) has joined #freenet
[8:46] * MrNaughty (n=mrnaught@S01060013103e9ce9.ed.shawcable.net) has joined #freenet
[8:50] * FallingBuzzard (n=FallingB@66.151.22.70) has joined #freenet
[8:58] <toad_> nextgens: url?
[8:58] <toad_> url is as it was, cool
[8:59] <nextgens> :|
[8:59] <toad_> okay, we have branches/legacy/{stable,unstable}
[8:59] <nextgens> you could browse it directly with a webbrowser... it will be faster than eclipse ;)
[9:00] * toad_ checks out unstable and stable
[9:00] <toad_> nextgens: have you moved/ported the newSnaps and harvestRefs scripts?
[9:00] <nextgens> not yet
[9:01] <nextgens> but they will be refactored
[9:01] <nextgens> as we can't checkout from SF shells
[9:01] <toad_> why would they run on an sf shell?
[9:01] <nextgens> we will have to build snapshoots on emu and the upload
[9:01] <toad_> they run on dodo at present
[9:01] <toad_> right, that's what the scripts do
[9:01] <toad_> they're in ~ian on dodo
[9:01] <toad_> ~/bin/newSnap.sh and ~/bin/harvestRefs.sh
[9:01] <toad_> called from cron
[9:02] * nextgens saw only the ~/bin/regenwebsite
[9:02] <nextgens> ok
[9:02] <nextgens> I will set it up with post-commit and not cron
[9:02] <toad_> hmmm okay
[9:02] <nextgens> as releases will be tags in subversion ; not as cvs
[9:03] <toad_> does that mean we can use the subversion tag number directly in Version in 0.7?
[9:03] <nextgens> toad_: do you have the mysql root password ?
[9:03] <toad_> well we could but it's for the whole repository isn't it...
[9:03] <toad_> nextgens: i don't think so
[9:03] <nextgens> :-S
[9:04] <nextgens> toad_: yes, for the whole repository
[9:04] <nextgens> so version numbers won't be +1
[9:04] <toad_> okay, why does eclipse not want to build stable?
[9:04] <toad_> for the whole repository is bad, i think maybe we just keep it as is
[9:04] <nextgens> because of ian's .project IMHO ;)
[9:04] <toad_> with manually updated
[9:04] <nextgens> you should add the JRE in your project
[9:05] <toad_> i checked out stable as a project...
[9:05] <nextgens> and paths might be differents
[9:05] <toad_> add where?
[9:05] <toad_> it doesn't have any builders, including the internal one
[9:05] <nextgens> to me, .projects and .classpath should be added to svn::ignore
[9:05] <nextgens> normal behaviour
[9:06] <toad_> how to fix?
[9:06] <nextgens> right click on the project root in the java perspective and add the JRE (tick the bow)
[9:06] <nextgens> bow^wbox
[9:06] <nextgens> and maybe add freenet-ext.jar
[9:06] <toad_> where?
[9:06] * nextgens is getting stable
[9:07] <nextgens> properties
[9:07] <nextgens> Java build path
[9:07] <nextgens> Libraries
[9:07] <toad_> there isn't a java build path
[9:07] <nextgens> and order and Export
[9:07] <toad_> i don't think it knows it's a java project...
[9:08] * toad_ closes and reopens... now it's doing something...
[9:08] <nextgens> in svn browser, right click and select 'co as a project'
[9:09] <toad_> that's what i did... will do again...
[9:09] * spaetz (n=spaetz@62.48.121.241) Quit (Remote closed the connection)
[9:09] <toad_> maybe Check Out As...
[9:10] <nextgens> ok, bugtracker is set up
[9:11] <nextgens> though it needs a DB to be configured :/
[9:11] <nextgens> sanity: could you create a mysql DB for it when you will be back ?
[9:11] <toad_> okay
[9:11] <nextgens> arf
[9:11] <toad_> if i do check out as... then i can make it work properly
[9:11] <nextgens> sanity: forget about it :)
[9:12] <nextgens> there were no password :<
[9:15] * nextgens is creating the database
[9:17] * toad_ tries to get this to build the Right Way... I suspect that isn't possible with eclipse...
[9:18] <toad_> (with ant integration, i mean)
[9:19] <nextgens> I cant's help :$
[9:24] <nextgens> toad_: you should get a mail with the mantis's admin password
[9:25] <nextgens> or at least how to reset it
[9:25] <toad_> it works
[9:25] <nextgens> yes
[9:25] <toad_> mantis works?
[9:25] <nextgens> of course it does ;)
[9:25] <toad_> eclipse/ant integration on stable also works :)
[9:25] <nextgens> but still, I find mantis ugly :-P
[9:26] <nextgens> and even worst, mantis and SVN passwords won't be in sync. for now :<
[9:26] <nextgens> toad_: create a non priviledged account for you'r dayly use ;)
[9:27] * nextgens is setting up the https proxy
[9:28] <toad_> well that's normal... separate tool, separate password
[9:28] <toad_> many people who use mantis won't have svn access
[9:31] <sanity> nextgens: the "signup for a new account" link just returns to the login page
[9:32] <sanity> (on mantis)
[9:32] <nextgens> normal, administratively denied for now
[9:33] <sanity> ok, well, can you create an account for me then? or give me the pwd
[9:33] <nextgens> toad_: could you fw the mail please ? I don't have the password :$
[9:33] <toad_> to where?
[9:33] <toad_> last time i sent you an email it got lost
[9:34] * nextgens has resetted it and set toad's mail as the administrator's one
[9:34] <nextgens> toad_: to sanity
[9:34] <nextgens> toad_: mine should work too
[9:34] <nextgens> cc me if you'd like to test
[9:34] <toad_> sent it (it doesn't work btw)
[9:34] <nextgens> toad_: ?
[9:34] <nextgens> what doesn't work ?
[9:35] <nextgens> the password ?
[9:35] * nextgens is crossing finger when restarting apache2
[9:35] <sanity> toad: can you send it unencrypted please?
[9:35] * sanity doesn't do gnupg these days
[9:36] <sanity> nextgens: incidentally, i'm not sure if toad mentioned it, but emu is shared between freenet project inc, and cematics ltd, my consulting company
[9:37] <sanity> nextgens: please be careful not to touch anything related to cematics if you can avoid it
[9:37] <sanity> nextgens: on that subject, i may see if i can set up a separate project on mantis for cematics bugtracking
[9:37] <nextgens> i don't ... apache2 is for https
[9:37] <toad_> sanity: why?
[9:37] <sanity> toad_: why what?
[9:38] <nextgens> sanity: or even a separate mantis ;)
[9:38] <toad_> i can't send you any confidential mails about chinese militants any more :)
[9:38] <sanity> that may be an option, yes
[9:38] <sanity> toad_: the chinese bugged your keyboard long ago
[9:38] <nextgens> :-D
[9:38] <toad_> :)
[9:38] <toad_> seriously, encryption is good practice - why do you object to it?
[9:39] <sanity> i don't object to others doing it, just when they force me to
[9:39] <toad_> umm, by sending you encrypted mails?
[9:40] * nextgens is restarting apache2
[9:40] <sanity> toad: exactly
[9:41] <nextgens> raah
[9:41] <toad_> nextgens: you done the auto-snapshot script yet?
[9:41] <nextgens> not yet
[9:41] <toad_> sanity: translated, that means "I forgot my passphrase"
[9:41] <nextgens> and the svn is unaviable for now
[9:41] <nextgens> toad_: erf
[9:41] <toad_> nextgens: i just committed to it
[9:41] <sanity> toad: no, it means that gnupg is a PITA to use on a mac
[9:41] <toad_> AHHH!
[9:42] <sanity> and i am getting too old for this paranoia lark
[9:42] <toad_> so THAT'S it!
[9:42] <toad_> well I'm glad you don't code :)
[9:42] <sanity> and all our computers are bugged anyway so whats the point?
[9:42] <sanity> i write plenty of code, thank you very much
[9:42] <toad_> on freenet :)
[9:42] <sanity> including the framework upon which you are building freenet 0.7 lest i remind you :-)
[9:43] <nextgens> huhuhu
[9:43] <sanity> toad: remember dijjer?
[9:43] <toad_> which is not designed for paranoia, and therefore has had to have it retrofitted :)
[9:43] <sanity> dijjer is pretty secure actually
[9:43] <toad_> anyway i sent you the passwd
[9:43] <sanity> it incorporates two forms of integrity checking
[9:43] <sanity> cool, thx
[9:44] <toad_> sanity: how do you integrity check stuff cached from the web btw? i've always wondered...
[9:44] <toad_> do most servers support HEAD _and give you a Content-MD5_ ?
[9:44] <sanity> toad: you didn't send me the password, you just sent me a reset url
[9:44] <toad_> yes
[9:44] <toad_> because that's all that nextgens sent me, and it didn't work
[9:45] <sanity> probably because nextgens had already used it, and it is one-time only...?
[9:46] <nextgens> :-S
[9:46] <nextgens> ok, I'll set up accounts :<
[9:46] <sanity> nextgens: how hard would it be to set up a separate mantis on bugs.cematics.com with a separate db?
[9:46] <nextgens> sanity: easier than to manage one mantis with 2 projects IMHO
[9:47] <nextgens> but both will require different credentials for people
[9:47] <sanity> nextgens: of course
[9:47] <nextgens> not a big deal if it's not freenet related
[9:47] <sanity> nextgens: can you give me a quick HOWTO, since i don't really know where everything is...
[9:48] <nextgens> sanity: I finish this one and then set up the other ... fine for you ?
[9:48] <sanity> if you don't mind, that would be great :-)
[9:48] <nextgens> or do you want to do it yourself ?
[9:48] <toad_> does anyone here have any actual experience working for an ISP? jrandom is convinced that ISPs routinely run expensive surveillance on their users...
[9:48] <nextgens> toad_: no :)
[9:49] <toad_> nextgens: tell me when you have implemented snapshot auto-rebuild
[9:49] <toad_> then i will announce 5106 and 60274
[9:49] <sanity> nextgens: well, i would prefer if you could do it since you grok how things are set up, but since its not freenet-related i wouldn't expect you to
[9:49] <nextgens> ok
[9:49] <toad_> alternatively i could do it
[9:49] <nextgens> as you wish
[9:49] <toad_> or i could get on with hacking 0.7 :)
[9:50] <nextgens> toad_: hack .7 ;)
[9:50] <sanity> toad: i think 0.7 needs to be your priority
[9:50] <toad_> right
[9:50] <toad_> absolutely agreed
[9:50] <nextgens> I can't do it for you whereas I know how to do that
[9:51] * nextgens commited a small thing on PacketSender this morning
[10:02] <nextgens> sanity: toad_: you've got mail
[10:03] <nextgens> https is now aviable
[10:05] * nextgens is now on the post-commit stuff
[10:07] <nextgens> it would be good to have some backups somewhere ...
[10:07] <nextgens> especially svn and mysql
[10:08] <sanity> nextgens: your wish is my command :- http://www.bytemark.co.uk/support/tech/backupspace.html
[10:09] <nextgens> :)
[10:10] <sanity> nextgens: i think emu's username is @emu@
[10:10] <sanity> um, "emu"
[10:10] <sanity> nextgens: what is your email addr?
[10:10] <danderson> toad_: does the repo look satisfactory?
[10:11] <nextgens> nextgens AT laposte DOT net
[10:11] <danderson> note you also have a snapshot of the last CVS HEAD (everything and anything, including CVSROOT) in tags/cvs-archive
[10:11] <sanity> nextgens: i just fowarded emu's original root password in case you need it to access the backup server
[10:11] <danderson> nextgens: can you make that read-only?
[10:13] <nextgens> sanity: thanks
[10:13] <nextgens> danderson: sure
[10:14] <nextgens> done
[10:14] <danderson> just to avoid me having to perform more surgery on the repo because of an accidental commit ;)
[10:15] <nextgens> sanity: you need to create the project in mantis ...
[10:15] <nextgens> I let you do it
[10:21] <sanity> nextgens: which project? freenet?
[10:21] <sanity> ok
[10:21] <nextgens> yes
[10:22] <sanity> hmm, looks like we could have a separate project for cematics, and just make it non-public...?
[10:22] <sanity> nextgens: are you sure having a completely separate mantis deployment is the better option?
[10:23] <nextgens> I don't mind
[10:23] <nextgens> sanity: how many users will your project have ?
[10:23] <sanity> ok, lets try it the easy way, if that doesn't work, we will try something fancier
[10:23] <sanity> nextgens: 3 or 4
[10:23] <nextgens> sanity: if it's non public you will have to create accounts ...
[10:23] <nextgens> ok
[10:24] <nextgens> so, it might be easier like that
[10:27] <toad_> i don't have an admin password
[10:27] <toad_> bbiab
[10:28] <toad_> nextgens: you will have to run harvestRefs in a cronjob even if you don't run newSnap in a cronjob
[10:28] <nextgens> toad_: you account (toad) has admin privileges
[10:28] <toad_> nextgens: seeya
[10:28] <toad_> nextgens: ahhh okay
[10:28] <toad_> will be back in an hour, and then set password on toad
[10:28] <nextgens> ok, not sure I'll still be here :/
[10:36] <sanity> nextgens: i am trying to make bugs.cematics.com point to the same place as bugs.freenetproject.org - but its not working :-/
[10:36] <sanity> nextgens: i have set bugs.cematics.com as an alternative virtual server name for bugs.freenetproject.org - but it doesn't seem to work, any ideas?
[10:37] <nextgens> yes
[10:37] * Ash-Fox (i=UNKNOWN@ecl42.neoplus.adsl.tpnet.pl) has joined #FreeNET
[10:37] <nextgens> let me fix it
[10:38] <nextgens> sanity: btw, do you need php for cematics.com ?
[10:39] <sanity> nextgens: not right now, but it would be useful to have the option - why?
[10:39] <nextgens> because it's not enabled ATM
[10:39] <nextgens> sanity: https only or both for bugs ?
[10:40] <sanity> nextgens: both i think
[10:40] <nextgens> ok, what certificate should I use ?
[10:40] <nextgens> could you generate one ?
[10:40] <sanity> nextgens: its not a big deal, but when mantis sends out emails with links, they are using http rather than https
[10:41] <nextgens> :-S
[10:41] <nextgens> not good
[10:41] <sanity> nextgens: i'm not sure how? can you?
[10:41] <nextgens> as root : apache2-ssl-certificate
[10:41] <sanity> nextgens: are you using webmin to administer the web servers?
[10:42] <nextgens> sanity: NO!
[10:42] <sanity> nextgens: where do i put it
[10:42] <nextgens> default location I'll move it
[10:42] <nextgens> sanity: webmin shouldn't be used IMHO
[10:42] <nextgens> nor accessible from outside ;)
[10:42] <sanity> nextgens: why not?
[10:43] <nextgens> because it's only 'blinding' the administrator about real configuration
[10:43] <nextgens> ie, if I modify a file and put an option it doesn't understand, you won't see it
[10:43] <sanity> ssl thing done
[10:44] <nextgens> sanity: accessible from the outside world ? because it's far more efficient to bruteforce root password with SSL than with ssh ;)
[10:44] <sanity> nextgens: perhaps, however it is extremely useful in practice for those who don't have the time to muck around with config files
[10:44] <nextgens> and even, root login should be denied
[10:45] <nextgens> sanity: they should delegate work ;) but shouldn't install such a tool :)
[10:45] <sanity> that presumes there is someone to delegate to - are you volunteering to be our full-time sysadmin?
[10:45] <sanity> :--)
[10:47] <sanity> nextgens: FYI - bugs.cematics.com not working yet
[10:48] * greycat (i=rfc1413@wooledge.org) has joined #freenet
[10:49] <nextgens> gnuh ?
[10:50] <nextgens> working but with the wrong certificate here
[10:50] <sanity> nextgens: ah, works with https, but not http
[10:51] <nextgens> sanity: you made a mistake with the certificate :
[10:51] <nextgens> it should be issued for bugs.cematics.com ; not emu ;)
[10:52] <nextgens> the certificate is the one you have provided but with wrong FQDN
[10:52] <sanity> nextgens: oh
[10:53] <sanity> i can do it again...
[10:53] <nextgens> ok,beep me when done
[10:54] <sanity> done
[10:54] <nextgens> sanity: be carefull to give rights to your users only to your projects :
[10:54] <nextgens> atm, they are manager on all projects
[10:55] * nextgens doesn't trust your users :?
[10:55] <nextgens> ;)
[10:55] <sanity> its by brother and my girlfriend :-)
[10:55] <sanity> s/by/my
[10:56] <nextgens> ok, certificate in due place
[10:56] <nextgens> apache2 has been reloaded
[10:56] <nextgens> working :)
[10:56] <sanity> cool :-)
[10:56] <sanity> hmmm
[10:57] <sanity> can we make http://bugs.cematics.com/ redirect to https://bugs.cematics.com/ - ditto for bugs.freenetproject.org
[10:57] <sanity> ?
[10:57] <nextgens> yes
[10:57] <nextgens> I will
[10:58] <nextgens> it will be simpler than an other virtual host
[10:58] <sanity> ok, cool
[10:58] <nextgens> and I'll do the same with bugs.freenet
[11:09] <nextgens> ok, almost finished
[11:09] <nextgens> raaah, directories aren't working yet :|
[11:26] <sanity> wow, accessing mantis is now *really* slow
[11:26] <sanity> page just sits there loading for over 20 seconds
[11:27] <nextgens> I know, I'm on it
[11:27] <nextgens> it's related to the denial of access of subdirectorys
[11:31] * FallingBuzzard (n=FallingB@66.151.22.70) Quit (Read error: 104 (Connection reset by peer))
[11:31] * FallingBuzzard (n=FallingB@66.151.22.70) has joined #freenet
[11:44] <nextgens> sanity: is it faster now ?
[11:44] <nextgens> there still be the directory problem
[11:45] <nextgens> but it should be usable now
[11:45] <sanity> just tried it, still very slow - been loading for over 10 seconds
[11:46] <nextgens> not here :<
[11:46] <nextgens> hard to debug
[11:46] <sanity> this is https://bugs.freenetproject.org/main_page.php
[11:46] <sanity> same problem with bugs.cematics.com :-(
[11:46] <nextgens> loading imediately here
[11:47] <nextgens> maybe because of my :@ proxy
[11:47] <nextgens> could anyone else test ?
[11:49] <sanity> looks like it is trying to load http://127.0.0.2/css/default.css
[11:50] <nextgens> is it better now ?
[11:53] <sanity> nope, still trying to load from 127.0.0.2
[11:55] <nextgens> :-S
[12:01] <nextgens> should be ok now
[12:01] <nextgens> sanity: please try again
[12:01] <sanity> that works
[12:02] <nextgens> :)
[12:02] <nextgens> and the redirect thing too ?
[12:10] <nextgens> toad_: back ?
[12:11] <toad_> nextgens: hi
[12:11] <toad_> nextgens: what do you need?
[12:12] <nextgens> your opinion on how to transfert tags to SF
[12:12] <nextgens> tags^wsnapshoots
[12:12] <nextgens> should we build them on emu or dodo ?
[12:12] <nextgens> whith which JVM ?
[12:12] <nextgens> should we use rsync or cat |bzip2 |ssh ?
[12:13] <toad_> build them on emu
[12:13] <toad_> dodo is going away
[12:13] <toad_> use rsync, or whatever is easiest
[12:13] <nextgens> ok, is the JVM version important ?
[12:13] <toad_> seednodes we will have seednodes.ref, .bz2, .gz and .zip
[12:13] <toad_> so may want to do the recompression on sf
[12:13] <toad_> no, as long as the target is 1.4
[12:14] <nextgens> is any jvm already installed ?
[12:15] <toad_> dont know
[12:17] * sleon_ (n=sleon|tu@e180009089.adsl.alicedsl.de) has joined #freenet
[12:22] <nextgens> :@ there are .exe in the repository !
[12:22] * sleon|tuX (n=sleon|tu@e180010214.adsl.alicedsl.de) Quit (Read error: 110 (Connection timed out))
[12:23] * sleon_ is now known as sleon|tuX
[12:26] <toad_> hi
[12:36] * nextgens is rtfming svnlook
[12:40] <nextgens> seeya tomorrow
[12:40] <toad_> thanks
[12:40] <nextgens> the post-commit script isn't ready though :$
[12:41] <sanity> http://bugs.cematics.com/ redirect doesn't work yet
[12:42] <nextgens> fixed, a typo :$
[12:43] * blank_ (n=smains@ERR.CYLAB.CMU.EDU) Quit (K-lined)
[12:44] <sanity> cool, works :-)
[13:27] * Code-Chris (n=kicken18@80-193-62-89.cable.ubr03.gill.blueyonder.co.uk) has joined #freenet
[13:32] * smains (i=[U2FsdGV@tor/session/x-8649b7978a352ff2) Quit (SendQ exceeded)
[13:36] <toad_> SSK@iNCJJBnFIQSIS5CrV-8DytbkjaEPAgM,UqQevqaamFe212HBkCVsGQ/VAVA/2//
[13:36] <toad_> the top item
[13:36] <toad_> http://web.media.mit.edu/~fviegas/survey/blog/results.htm
[14:05] <toad_> how do i send a gpg-signed email from the command line?
[14:13] <toad_> how do i send a MIME message from the command line?
[14:13] <toad_> i can make a detached signature no problem...
[14:28] <toad_> done..
[14:28] <toad_> announced the migration to SVN, and disabled everyone's email address
[14:28] <toad_> err
[14:28] <toad_> disabled everyone's CVS access
[14:28] <toad_> except me and ian
[14:34] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[14:36] <danderson> toad_: do you have an archive link for the mail?
[14:36] <toad_> hmmm?
[14:36] <danderson> 20:31 <@toad_> announced the migration to SVN, and disabled everyone's email address
[14:36] <danderson> is there an archive for the ML somewhere?
[14:36] <toad_> yes
[14:37] <toad_> http://article.gmane.org/gmane.network.freenet.devel/16825
[14:37] <danderson> thanks.
[14:37] <toad_> grrr, the mail archives are still broken
[14:37] <toad_> probably sanity's fault :|
[14:37] <danderson> nope, working here
[14:38] <toad_> i suppose gmail MAY work... but we really ought to have our own, gmail isn't exactly ideal
[14:38] <toad_> they don't even have a global search!
[14:38] <toad_> the data really isn't reusable
[14:38] <danderson> oh, right
[14:45] <toad_> okay, what next?
[14:45] <toad_> should i create a branch to put my publish/subscribe work onto?
[14:45] <toad_> it's been a long time since i made a commit...
[14:45] <toad_> to 0.7
[14:45] <danderson> well, it depends
[14:45] <danderson> do you want the publish/subscribe to be a branch ? :)
[14:46] <toad_> :|
[14:46] <toad_> well not in the long term
[14:46] <danderson> short/long term is no problem
[14:46] <danderson> right now, should it be a feature branch rather than break trunk?
[14:46] <toad_> it probably makes sense to have it in SVN in the short term
[14:46] <toad_> and then merge it
[14:46] <danderson> ie. does the work on publish/subscribe involve fairly long work that will break trunk?
[14:46] <danderson> if so, you're wanting a feature branch.
[14:46] <toad_> yes
[14:47] <danderson> svn cp -m "Branching trunk for work on the publish-subscribe architecture." https://emu.freenetproject.org/svn/trunk/ https://emu.freenetproject.org/svn/branches/publish-subscribe
[14:48] <danderson> then, if you have a trunk working copy, you can tell svn to switch over to the branch:
[14:48] <danderson> svn switch https://emu.freenetproject.org/svn/branches/publish-subscribe
[14:48] <danderson> to be issues from the root of the trunk working copy
[14:48] <toad_> well, i'll probably branch and then diff with a local copy, and bring the changes in, then commit them
[14:48] <danderson> yeah.
[14:49] <danderson> just mentionning svn switch, because at that stage, both the trunk and branch are identical, so switching will save a lot of time/bandwidth, by just switching the metadata over to the branch.
[14:49] <danderson> so that commits go to the branch instead of trunk.
[14:49] <toad_> ps i love instant diff :)
[14:50] <danderson> (this also works if trunk and the branch are different; it'll only send the diffs between the two over the network. That's how I can get the 1.3.x release branch for svn with only a few dozen diffs, rather than recheck out a whole working copy)
[14:50] <danderson> heh, yeah
[14:50] <toad_> okay, i have a checked-out trunk
[14:51] <toad_> rename/move probably isn't what i want
[14:51] <danderson> ?
[14:51] <toad_> subclipse
[14:51] <danderson> oh.
[14:51] <danderson> toad_: perhaps create the branch using the svn commandline program.
[14:52] <toad_> doesn't look like i can do it from within subclipse
[14:52] <toad_> so yeah
[14:52] <danderson> I gave the commands above, it'll be much easier.
[14:52] <danderson> I don't know subclipse at all, so I couldn't really help you out there.
[14:53] <toad_> svn switch https://emu.freenetproject.org/svn/branches/publish-subscribe
[14:53] <toad_> D .project
[14:53] <toad_> D .externalToolBuilders
[14:53] <toad_> D .classpath
[14:53] <toad_> D LICENSE.Mantissa
[14:53] <toad_> what's with all the D's?
[14:55] <toad_> danderson: what do the D's mean?
[14:55] <danderson> D means deleted.
[14:55] <toad_> why would any files be deleted?
[14:55] <danderson> meaning those files don't exist in the branch.
[14:55] <danderson> which is weird.
[14:55] <toad_> eclipse artifact maybe, i'll check it out from scratch
[14:56] <danderson> I'll try checking out trunk and switching here, to see what gives.
[14:56] <toad_> is it really easy to merge the branch back into trunk?
[14:57] <toad_> hrrm
[14:57] <toad_> i think i may have branched from too high up :)
[14:57] <danderson> toad_: it is described in the svn handbook. I believe there are 3 commands to issue to do it properly.
[14:57] <toad_> yep, i did
[14:57] <toad_> should've been trunk/freenet :|
[14:58] <danderson> toad_: that doesn't matter
[14:58] <danderson> toad_: any files you don't touch in the branch take up zero extra space
[14:58] <danderson> and usually it's good to branch the whole of trunk
[14:58] <toad_> it explains why the switch didn't work
[14:58] <toad_> it is?
[14:58] <danderson> so that for example, you could change stuff in the website dir of the branch
[14:58] <toad_> ahhh
[14:59] <danderson> and have the website come up to date with branch changes when you merge back
[14:59] <danderson> stuff like that
[14:59] <toad_> so may as well keep it as-is?
[14:59] <danderson> yeah.
[15:00] <toad_> want to check out only the freenet/ subdir though, for now
[15:00] <danderson> yeah, of course.
[15:00] <danderson> but the fact that the whole of trunk is branched is nice if you ever need it
[15:00] <toad_> yeah
[15:01] <toad_> given the svn zero-space-cost niceness
[15:01] <danderson> exactly
[15:01] <danderson> branching any directory is just as cheap as branching any other
[15:01] <danderson> it takes up exactly 1 new node in the SVN filesystem, which takes negligible disk space.
[15:01] <danderson> only changes start adding weight to the branch.
[15:02] <toad_> this is because you already have to have an object describing the whole dir snapshot somewhere?
[15:02] <danderson> yeah, it's due to the way the Subversion filesystem works internally.
[15:02] <danderson> it represents the history as a tree of "noderevs", each representing one node of the filesystem at one specific revision
[15:03] <danderson> the nice thing is, with branching (among other ops), the new node that the branch adds doesn't need to duplicate contents
[15:03] <danderson> it just says "my contents is node-rev XYZ"
[15:03] <danderson> where node-rev XYZ is trunk at revision N
[15:04] <danderson> if you see what I mean.
[15:04] <toad_> yeah, i think so
[15:04] <danderson> the only extra property you need then, is that any node already in the filesystem is immutable
[15:05] <danderson> if any operation tries to edit a node-rev, the filesystem silently duplicates it into a mutable node
[15:05] <danderson> which can be edited, and becomes immutable when the commit is finalized
[15:06] <danderson> that way, extra data appears in the FS only when changes are recorded; and old states are guaranteed to never be changed (at least not through the library API)
[15:07] <danderson> anyhow, the internals are a little complex (but well documented if you care enough to find out), but the bottom line is that copying is extremely cheap
[15:07] <danderson> making branching for features and release stabilization easy.
[15:12] <sanity> this debate with jrandom is a waste of time
[15:12] <sanity> he is basically doing the whole "its theoretically possible to break it, therefore it is useless"
[15:13] <sanity> a bit ironic given that it is far easier to compromise I2P in the same way
[15:13] <danderson> aah, the classic energy beast.
[15:13] <sanity> energy beast?
[15:14] <danderson> feed it arguments and thoughts, and it will consume them and excrete mindless and unjustified denial.
[15:14] <danderson> every good project has a few :-)
[15:15] <sanity> are you accusing me of that, or him?
[15:15] <danderson> sanity: him, according to what you're saying
[15:15] <danderson> I'm assuming you're explaining why in practice the system remains quite usable and has interesting applications
[15:16] <danderson> and the replies all start with "yes, but it is broken by design, because it can theoretically be completely broken down."
[15:16] <toad_> well, the argument is that it can be easily broken down
[15:16] <toad_> but yes
[15:17] <danderson> any system, given adequate conditions, can be easily broken down imho.
[15:17] <danderson> it's the old "foolprof software vs. fools" problem
[15:18] <sanity> well, his argument that through nationally deployed traffic analysis, freenet 0.7 users may be identifiable
[15:18] <sanity> our argument is that this might very well be true, but that is a hell of a lot more difficult to do than to compromise any of today's options for anonymous exchange of information
[15:19] <sanity> should people be denied something better just because it isn't perfect?
[15:20] <sanity> is the website updating set up yet?
[15:20] <danderson> sanity: exactly.
[15:20] <sanity> well, i am bored with the debate, so if you would like to take over, please feel free :-)
[15:20] <danderson> if we waited for perfection, we'd still be punching cards for programs :-)
[15:20] <danderson> sanity: I don't think it is. nextgens was setting it up but had to go.
[15:21] * sanity is getting a hp-16c calculator :-)
[15:21] <sanity> its a rpn programmers calculator from 1982
[15:21] <sanity> i thought it would be nice to have a calculator on my desk, and then decided to try to find something interesting :-)
[15:37] <toad_> whacky
[15:40] <sanity> i like the simplicity of that type of thing - makes a nice break from the complexity of today's computers
[15:59] * Romster (i=[U2FsdGV@tor/session/x-6731c7c8ffa724e7) Quit (Remote closed the connection)
[16:06] <toad_> Subject: Not such a theoretical risk,
[16:06] <toad_> but darknet still useful was Re: [Tech] Re: [i2p] I2P conspiracy
[16:06] <toad_> theories flamewar
[16:06] * kingMac (n=King@151.56.219.160) has joined #freenet
[16:07] <toad_> sanity: please read that message, it is from the flamewar, but i think it is useful
[16:07] <toad_> you don't have to read the rest of the flamewar :)
[16:07] <kingMac> hi all
[16:08] <kingMac> Can i ask for frost question?
[16:08] <toad_> kingMac: hi
[16:08] <toad_> you can ask...
[16:08] <kingMac> thank you
[16:08] * sleon|tuX (n=sleon|tu@e180009089.adsl.alicedsl.de) Quit ("Ex-Chat")
[16:09] <kingMac> I have a iMac
[16:09] * kingMac (n=King@151.56.219.160) has left #freenet
[16:10] <sanity> reading now
[16:12] <sanity> i don't know, i think the argument against jrandom is simple : a better solution today is better than a perfect solution at some vaguely defined point in the future
[16:13] <toad_> i don't think he expects to ever have a perfect solution
[16:13] <toad_> certainly i2p will never be a perfect solution
[16:13] <sanity> you are overcomplicating the issue by engaging him on the specifics
[16:13] <sanity> rather, you should challenge him on whether the Chinese are better off with HTTP proxies, or whether they are better of with Freenet 0.7, imperfect as it might be
[16:14] <toad_> he's arguing that a perfect solution is impossible and we shouldn't give a false sense of security because the moment politicians start caring about it, it can be trivially neutralized... my response is that that's not true; although you probably CAN detect freenet on the internet no matter what we do, unless you are prepared to go from a moderate totalitarian state to a really nasty one, you can't prevent sneakernet+wifi etc transp
[16:15] <sanity> he is providing a perfect example of the "ivory tower anonymity" crowd, if it isn't perfect, they aren't interested, regardless of what people in China, Iran, and elsewhere actually need
[16:15] <toad_> well, it's an unfortunate fact that darknet freenet probably is detectable; ISPs will soon be required in the EU to do traffic analysis to log voip calls, so they can as easily do the same for the rest, and if the govt provides a box to analyse the logs in real time... they can find us
[16:16] <toad_> sanity: not quite. he thinks they need walkie-talkies
[16:16] <toad_> sanity: because a scalable anonymity system will never be perfect
[16:16] <sanity> it will be straight-forward enough to obscure Freenet traffic if/when it becomes necessary
[16:16] <toad_> you think so?
[16:16] <toad_> say we make it look like VOIP
[16:16] <sanity> until it is necessary, there is little point in inconveniencing our users
[16:16] <toad_> why would I have 10 permanent VOIP connections to the same 10 peers, 24x7?
[16:16] <sanity> not VoIP, what about bittorrent?
[16:17] <sanity> these vague things are impractical to enforce in law
[16:17] <toad_> we could maybe make it look like bittorrent, yes... assuming that they like bittorrent
[16:17] <sanity> will they make it illegal to connect to more than one other peer at a time?
[16:17] <toad_> i doubt that they'll be too keen on trackerless bittorrent
[16:18] <sanity> but again, don't fall into the trap of trying to persuade anyone that Freenet is impossible to attack. the important thing is that it is far harder to attack than what is available to people now.
[16:18] <toad_> and once you get beyond completely local traffic analysis to say ISP level, you can easily blow the bittorrent defence... or just by traffic quantities, perhaps... but doing it GLOBALLY per ISP might well be expensive, I'd expect it to be essentially local unless you spend a lot of money
[16:18] <toad_> sanity: well, that is a fact for sure
[16:19] <sanity> ok, so why are you trying to argue a point you can't win, when the argument you should be having is easy to win?
[16:20] <toad_> sanity: well, does it matter that freenet is harder to attack than what they have now, when if freenet was attacked on a political level, it would necessarily fall?
[16:20] <sanity> stop trying to persuade jrandom that Freenet is impossible to compromise, or can be made impossible to compromise - you will lose that argument. start arguing that Freenet is more difficult to compromise than any other option that is available to people
[16:20] <toad_> i never claimed it was impossible
[16:20] <toad_> not to anyone
[16:20] <sanity> if sufficient resources are thrown at Freenet, yes, it will fail to continue to disguise who is part of the Freenet network
[16:20] <toad_> certainly not to jrandom
[16:21] <sanity> but you AREN'T arguing the important point, which is that Freenet is BETTER than ANY OTHER AVAILABLE OPTION
[16:21] <toad_> well you have made that point very clearly on several occasions
[16:21] <sanity> to the extent that jrandom distracts you from point that out, he is winning the argument
[16:21] * Eol (n=Eol@tor/session/x-7a3213f7bb455c66) has joined #Freenet
[16:21] <toad_> oh and i did, in response to that mail i cc'd you on
[16:21] <sanity> a/point/pointing
[16:22] <sanity> well, perhaps, but you continue to try to argue the point you can't really win - and that means that it appears to observers of your debate that you are losing. you are alllowing him to dictate the battleground.
[16:23] <toad_> And in the meantime, they have a system which costs $10M to block (any
[16:23] <toad_> time you have to involve politicians you get into big money, for a
[16:23] <toad_> start), instead of $10K. This is bad how...?
[16:23] <toad_> isn't that basically your point?
[16:24] <sanity> my point is that freenet 0.7 is a significant improvement over what is available today, and that alone makes it a worthwhile endeavour - even if it isn't perfect
[16:25] <toad_> because it advances the state of the art, which will eventually produce something that is better still
[16:25] <sanity> exactly
[16:25] <sanity> take some time to look back over your debate with jrandom
[16:25] <toad_> whereas he is making the orthogonal argument that _anything_ you build will be easily detectable and therefore worthless
[16:26] <sanity> you will see that he has continually challenged you, not to prove that freenet 0.7 is better than other things, but that it meets some unspecified standard - and you fall for the debate every time
[16:26] <toad_> unspecified standard is "freenet will always be easy to detect"
[16:26] <sanity> "easily" is a relative term. it is far far easier to detect the tools that people are using today (including I2P) than it will be to detect Freenet 0.7
[16:26] <toad_> for some arbitrary value of "easy"
[16:27] <sanity> exactly, he is forcing you to accept some kind of absolute measurement of effectiveness as the standard, but the real standard is relative to the next best option
[16:28] <sanity> if freenet 0.7 is better than the next best option, and it is far better, then it is a worthwhile endeavour
[16:28] <sanity> so, challenge him to describe a better option
[16:28] <sanity> and if he dares to, tear it apart
[16:29] * FallingBuzzard (n=FallingB@66.151.22.70) has left #freenet
[16:29] <toad_> he has repeatedly refused to offer a better option
[16:29] <toad_> well, lets see... has he?
[16:29] <sanity> then ask yourself, what exactly is the point being debated?
[16:29] <toad_> he said something about radios, i think...
[16:30] <sanity> he has suggested that a multitude of crap solutions are better, because it will be harder to attack each of them individually than one single system
[16:30] <toad_> well ultimately his point is that it is futile to give the chinese a scalable anonymous p2p, because it will get noticed and get blocked
[16:30] <toad_> well, they're not solving the same problem
[16:31] <danderson> in that case, it is also futile to plug a computer into the internet, because it'll only go and broadcast stuff for people to intercept.
[16:31] * Eol just notes who cares if its easily detectable ... if you want to play the absolute game, then hyptothetically everybody body uses freenet, all traffic is freenet, so its freenet traffic isn't interesting ... (jumps back into lurk mode)
[16:31] <toad_> LOL
[16:32] <toad_> sanity: i think there's a difference of goals too
[16:32] <toad_> our ideology is get freedom of speech to the people
[16:32] <toad_> in bulk
[16:33] <toad_> what he is saying is that's impossible, so help really small groups
[16:33] <toad_> ("militants")
[16:38] <toad_> yep, here's the key
[16:38] <toad_> > Not everything has to scale to be useful. Helping 50 people is better
[16:38] <toad_> > than helping none. And as I said in other emails, scaling in this case
[16:38] <toad_> > may in fact be counterproductive, as it changes the economics and
[16:38] <toad_> > invites more powerful attacks.
[16:38] <toad_> it's the "help the militants" myth
[16:39] <Eol> traffic is interesting though then....none might actually be better than few as they will have to rely on proven secure methods
[16:43] * greycat (i=rfc1413@wooledge.org) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[16:55] <sanity> toad: he argues that freenet will be easier to challenge on a large scale, than on a small scale - because a small scale won't provide sufficient motivation to impose traffic analysis
[16:56] <toad_> sanity: therefore...?
[16:57] <sanity> its mistaken
[16:57] <sanity> ...because, people who have something to say will say it on the most secure medium available to them
[16:57] <sanity> right now, those mediums are crap
[16:59] <sanity> he thinks that Freenet shouldn't be described as a "globally scalable darknet", because if it does scale sufficiently, it will cause a government to impose the kind of traffic analysis that can break it
[16:59] <toad_> ... therefore it is not a darknet any more
[16:59] <sanity> the fact that he uses the term "hyped" reveals the underlying motivation, which is that he is jealous of our publicity ;-)
[17:00] <toad_> :)
[17:00] <sanity> well, it is still a darknet, according to the definition of darknets - so he is wrong
[17:00] <toad_> what definition?
[17:00] <toad_> i thought darknet just meant a network to share illegal filez on according to the original microsoft usage
[17:01] <sanity> there isn't a concrete one, but basically the idea is that you only connect to people you trust
[17:01] <toad_> so darknet = f2f?
[17:01] <sanity> IMHO, yes = darknet = f2f
[17:05] <sanity> bue essentially i think jrandom's problem is that he recognises that freenet 0.7 will kick I2P's ass, and so he is very keen to poke holes in it - even though those very same holes are much easier to poke in I2P
[17:06] <toad_> we'll see
[17:06] <toad_> i hope so, (on technical as well as selfish grounds!)
[17:07] <toad_> the main hole to poke, either way, is that the justification for freenet or i2p in the west is dubious; it exists, but it's by no means clear cut and whether it is proportionate is iffy
[17:08] <toad_> but if you consider the Rest, and the plausible state of the west in the not so distant future, it becomes pretty clear cut
[17:08] <toad_> but since I2P isn't usable in either case... jrandom has a problem
[17:11] <toad_> do we want a user-contributed patch for OS/X support? yes, i think so
[17:11] <toad_> sanity: you could build a library to put in freenet-ext.jar...
[17:11] <toad_> when i've applied this it will be trivial to build one
[17:12] <sanity> toad: send me clear instructions and i will
[17:12] <sanity> there are reasons in the west now, look at diebold
[17:12] <sanity> ...or decss
[17:13] <toad_> well, you CAN still get them
[17:13] <toad_> the question is, will you be able to get them if they get to Google?
[17:13] <toad_> it's very precarious
[17:13] <sanity> i think its going to get much worse in the UK soon, our current government is showing a significantly authoritarian streak, and they are likely to indulge it given recent terrorist activities
[17:14] <toad_> (this is about as big a reason as any for having a good in-freenet search engine...)
[17:14] <toad_> well, i'm pretty annoyed about the glorification offence
[17:14] <toad_> but BT's moves to block pedo sites are equally disturbing
[17:14] <sanity> the real-time functionality we are planning will facilitate a good search-engine
[17:14] <toad_> how long until EVERYTHING illegal is blocked?
[17:15] <toad_> sanity: true, although i think it's possible (just a bit slower) without it
[17:20] * danderson (n=dave@natulte.net) Quit ("leaving")
[17:24] * danderson (n=dave@natulte.net) has joined #freenet
[17:33] <toad_> is mkdir -p a standard POSIX option, or is it a GNU extension?
[17:34] <toad_> will mkdir provided by shells on whacky systems like macintoshes have it?
[17:34] * danderson (n=dave@natulte.net) Quit ("leaving")
[17:34] * danderson (n=dave@natulte.net) has joined #freenet
[17:38] * Elly (i=elly@ool-457854e8.dyn.optonline.net) Quit (Read error: 104 (Connection reset by peer))
[17:51] <toad_> sanity: please check out Contriba
[17:51] <toad_> sanity: please check out Contrib
[17:51] <toad_> and go to the NativeBigInteger directory
[17:51] <toad_> and run build.sh
[17:51] <toad_> on a macintosh
[17:51] <sanity> tomorrow, its too late now
[17:51] <toad_> that will generate a .jnilib library
[17:51] <toad_> okay, i will email you
[17:53] * fridim (n=fridim@62-74-118-80.kaptech.net) has joined #freenet
[17:53] <fridim> hi
[17:53] <toad_> hi fridim
[17:56] <toad_> bbl
[17:56] * toad_ (i=toad@pdpc/supporter/active/toad-with-underline) Quit (Remote closed the connection)
[18:19] * fridim (n=fridim@62-74-118-80.kaptech.net) has left #freenet
[18:20] * Elly (i=elly@ool-457854e8.dyn.optonline.net) has joined #freenet
[18:58] * Ash-Fox (i=UNKNOWN@ecl42.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[18:58] * Ash-San (i=UNKNOWN@ecc19.neoplus.adsl.tpnet.pl) has joined #FreeNET
[19:24] * Code-Chris (n=kicken18@80-193-62-89.cable.ubr03.gill.blueyonder.co.uk) Quit (Read error: 110 (Connection timed out))
[19:29] * sanity (n=ian@81-178-100-27.dsl.pipex.com) Quit ()
[19:33] * NullAcht15 (n=NullAcht@dslb-082-083-232-185.pools.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
[19:34] * Hory (n=Miranda@82.78.27.85) Quit ("http://CyberLore.net - Recommendations on the best games, freeware and websites.")
[22:21] * Eol (n=Eol@tor/session/x-7a3213f7bb455c66) Quit ("Leaving")
[22:40] * Markaci (n=Markaci@wikipedia/markaci) Quit (Remote closed the connection)
[22:45] * Markaci (n=Markaci@wikipedia/markaci) has joined #freenet
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.