[06:36] *** Uraeus has joined #kamaelia

[06:45] *** mhrd-afk has joined #kamaelia

[06:45] *** mhrd-afk is now known as mhrd

[06:51] < MS-> morning

[06:52] *** vmlemon_ has joined #kamaelia

[06:53] < mhrd> greetings

[06:53] < mhrd> (back from http://www.ibc.org )

[06:53] < MS-> Ah - how was IBC?

[06:53] < mhrd> *busy*

[06:54] < MS-> Uraeus accused me of being the only person from R&D who wasn't there :)

[06:54] < mhrd> but good

[06:54] < MS-> cool

[06:54] < mhrd> heh

[06:54] < MS-> Gave my presentations at Pycon last weekend

[06:54] < mhrd> ah yes, briefly bumped into him; didn't really have time fora chat though

[06:54] < mhrd> how did it go?

[06:54] < MS-> first: http://www.slideshare.net/kamaelian/practical-concurrent-systems-made-simple-using-kamaelia-presentation/

[06:54] < MS-> second: http://www.slideshare.net/kamaelian/sharing-data-and-services-safely-in-concurrent-systems-using-kamaelia-presentation/

[06:54] < MS-> Went well

[06:55] < MS-> Met Davbo

[06:55] < mhrd> you're branching out from cats into robots now? ;-)

[06:55] < mhrd> oh?

[06:55] < MS-> Yep

[06:55] < MS-> Well, I've wanted to put robots in for some time but thought my artistic style was lacking somewhat

[06:55] < vmlemon_> RoboCat? ;)

[06:56] < MS-> Robots were a very good idea - more people "got" it - including someone who listened last year and *didn't* get it last year

[06:56] < mhrd> cool

[06:56] < MS-> Yep, Davbo had registered for pycon - it's a cheap conference :)

[06:56] < mhrd> Birmingham?

[06:56] < MS-> Yep

[06:57] < mhrd> lots of people?

[06:57] < MS-> 250 or so

[06:57] < mhrd> cool - good number

[06:57] < MS-> Sponsored by George Wright as well

[06:57] < MS-> Next year Europython is organised by us (the pycon uk lot)

[06:57] < vmlemon_> mhrd: The cats speak now ;)

[06:57] < MS-> which means the numbers will be up to twice as many

[06:58] < MS-> (given a low level of overlap between the two)

[06:58] < MS-> Also, next year doesn't clash with IBC or anything (end of June)

[06:59] < mhrd> vmlemon_ : heh - that the speak'n'write or something new new?

[07:00] < mhrd> ( apologies for not loitering much recently - busy busy :) )

[07:00] < MS-> Asking "lots of people" about IBC is a misnomer I guess - what were you there for? Helping with Nigel's stuff ?

[07:01] < MS-> vmlemon_: You'll note that the cat doesn't have speech bubbles - it has thought bubbles

[07:01] < mhrd> was assisting with demoing the high-frame-rate work (John Zubry*%^"£, Richard Salmon, Mike Armstrong, Steve Jolly)

[07:01] < MS-> Ahh, I see

[07:01] < MS-> Well, I don't, this is text, but I get what you mean \o/

[07:01] < MS-> :)

[07:01] < vmlemon_> Damn

[07:02] < mhrd> - evangelising the fact that if we're considering going to even higher resolutions in teh future then we *must* increase frame-rate too otherwise portrayal of motion will become worse and worse - eg. sports coverage becomes nigh-on impossible to shoot

[07:03] < mhrd> then yesterday (post IBC) in central Amsterdam with Nigel Earnshaw for the Axmedis project final presentations to the EU commissioners (its an EU Framework collaborative project)

[07:03] < MS-> There's something wrong with that statement that bugs me (sport). Not sure what it is, since I know the people looking at it know **far** more about it than me.

[07:04] < mhrd> s/shoot/shoot such that it is watchable/

[07:04] < MS-> Same comment applies

[07:04] < MS-> esp the "not sure what it is"

[07:05] < mhrd> if you find words to elucidate, get back to me :)

[07:05] < MS-> but know that the people looking at it know *far* more about it than me :)

[07:05] < MS-> So if I said anything to them, I'd spend the next 3 weeks being told chapter and verse why I was wrong :-D

[07:06] < mhrd> we had a good demo reel that showed the effects (needs a 100 fps HD display though :-) )

[07:06] *** MS- nods

[07:06] < mhrd> so it ain't going to get uploaded to youtube any time soon :)

[07:06] < MS-> As I say, chapter and verse :)

[07:07] < MS-> Also, just because I have a feeling that it's wrong somehow doesn't make me right, and I'm more than willing to accept they're right :)

[07:07] < MS-> Anyone gone AAAAAAAARRRRRRRRRRRRRRRRRRRRRRRRR yet btw?

[07:07] < MS-> International speak like a pirate day and all that?

[07:07] < MS-> Or have you not been scarfed yet?

[07:08] *** mhrd sighs :-)

[07:08] < mhrd> on the train today - bicycle at the shop for fault finding/repair, so got an early train and have avoided that

[07:09] < vmlemon_> Yay for error dialogue boxes that state the obvious... "This document is trying to put Acrobat in full screen mode, which takes over your screen."

[07:09] < MS-> mhrd: :)

[07:09] < mhrd> oh cripes - I've got 11 days of expenses to sort

[07:09] < MS-> enjoy

[07:09] < vmlemon_> (Erm, isn't that the point of Full-Screen Mode?)

[07:11] *** mhrd hunts for coffee

[07:12] < MS-> vmlemon_: You'd rather hope so

[07:13] < aa_> [] do not display this warning again

[07:14] < aa_> "trendy UI"

[07:14] < vmlemon_> Hah

[07:15] < vmlemon_> Hmm, don't they call it Adobe Reader now? So that technically invalidates it...

[07:16] < vmlemon_> Given that it only displays PDFs, it was pretty pointless changing the name, in my view

[07:17] < aa_> they are nice with that product to be honest

[07:17] < MS-> I prefer kpdf myself

[07:17] < MS-> :)

[07:18] < aa_> they give aaway free redistribution licenses that allow people to embed the installer in 3rd part installers

[07:18] < MS-> For most purposes it seems quicker

[07:19] < vmlemon_> I find that AR renders better for most documents

[07:20] < vmlemon_> (some of which don't seem to give anything other than blank pages in kpdf, unless that's solved)

[07:22] < MS-> I've not had any problems with the version of kpdf here - I know it used to have problems but it seems OK

[07:22] < Lawouach_> morning

[07:22] < vmlemon_> Hi Lawouach

[07:23] < MS-> version string: KPDF 0.5.7 ( Using KDE 3.5.7 "release 72.9" openSUSE 10.3)

[07:23] < MS-> Lawouach_: Morning

[07:24] < vmlemon_> KPDF 0.5.8 here on KDE 3.5.8

[07:25] < vmlemon_> Still, it's certainly fast

[07:35] < vmlemon_> Cat and mouse game!

[07:36] < MS-> :-)

[07:37] < MS-> It's a nice example as well - very explicit controller and thing to control, followed by a very explicit thing handling the model

[08:06] *** MS- gets coffee

[08:06] < MS-> biab

[08:39] *** vmlemon_ has joined #kamaelia

[10:35] *** Davbo has joined #kamaelia

[14:35] < MS-> Yay, sourceforge is well and truly bust

[14:35] < MS-> http://kamaelia.sourceforge.net/Home

[14:36] < MS-> Traceback (most recent call last): File "/home/groups/k/ka/kamaelia/cgi-bin/Wiki/wiki", line 24, in ? import pprint File "/usr/lib64/python2.4/pprint.py", line 39, in ? from cStringIO import StringIO as _StringIO ImportError: /usr/lib64/python2.4/lib-dynload/cStringIO.so: failed to map segment from shared object: Cannot allocate memory

[14:36] < MS-> I have no polite words

[14:36] < vmlemon_> Ooh

[14:36] < vmlemon_> Way to go!

[14:37] < MS-> At least I've managed to make it such that if you click on website for http://sourceforge.net/projects/kamaelia

[14:37] < MS-> that it points at the server we DO control

[14:37] < MS-> But that doesn't fix all the links pointing at sourceforge's "wonderful" hosting

[14:38] < MS-> Can't fix things on the server either handily:

[14:38] < MS-> ~/Incoming> telnet shell.sf.net 22

[14:38] < MS-> Trying 66.35.250.208...

[14:38] < MS-> telnet: connect to address 66.35.250.208: Connection refused

[14:38] *** vmlemon_ has joined #kamaelia

[14:40] < vmlemon_> Aren't they still performing "critical" maintenance on it?

[14:41] < vmlemon_> (AKA screwing it up even more) ;)

[14:43] < MS-> Yes they're breaking things

[14:43] < MS-> For no apparent good reason

[14:44] < MS-> I don't mind the idea that they're changing the maintenance of the website from shell to web

[14:44] < MS-> But disabling shell, breaking the webserver, BEFORE putting in place the alternative is a real pain.

[14:44] < vmlemon_> Classy

[14:45] < MS-> Just marks the end of the road with SF for me

[14:45] < MS-> We've got our own webserver, even if I view it as at best temporary (due to what's happening where the box is)

[14:45] < MS-> we're using different mailing lists and different source control site

[14:46] < MS-> And so breaking the webserver that most links go to becomes more or less irrelevant IMO

[14:46] < vmlemon_> Google Code and Groups?

[14:46] < MS-> Yep

[14:46] < MS-> Aside from annouce lists etc

[14:47] < vmlemon_> I can't see those breaking soon

[14:48] < MS-> Dunno. They're getting rather proficient at breaking their services

[14:48] < mhrd> MS- : related note: where are the doc builds supposed to be placed now?

[14:49] < MS-> That's an interesting question - they may as well end up on the server

[14:49] < MS-> the same machine rather :)

[14:49] < mhrd> the upload has been borked for 3 days in a row now, so I'm assuming sf.net have pulled the shell plug now

[14:49] < vmlemon_> The ex-fish box?

[14:49] < MS-> Somewhere inside here I guess: /srv/www/sites/edit.kamaelia.org/hotdocs/

[14:50] < mhrd> k

[14:50] < MS-> This is one: /srv/www/sites/edit.kamaelia.org/hotdocs/Docs/Axon

[14:50] < MS-> This is the other /srv/www/sites/edit.kamaelia.org/hotdocs/Components/pydoc

[14:51] < MS-> You're happy to change/fix that? Cool :-)

[14:51] *** mhrd on it

[14:51] *** vmlemon_ wonders what would happen to Kamaelia and the wiki, if things took a turn for the worst and MS- or mhrd left the BBC (not saying that will happen, though)...

[14:51] < MS-> yay :)

[14:51] < MS-> much appreicated

[14:52] < MS-> vmlemon_: Lawouach and I have a little plan regarding that

[14:52] *** mhrd defers vmlemon_'s question to MS

[14:52] < MS-> I can't get paid for GSOC since the tax situation for the BBC is *really* complicated

[14:52] < MS-> As in more pain than anyone would ever want

[14:52] < MS-> Lawouach however can be paid for all the students on our behalf

[14:53] < MS-> So, the idea is if he's willing to work through the pain (which could take 6-8 weeks...)

[14:53] < MS-> *if*

[14:53] < MS-> then any "excess" could be used to pay upfront for a seperate hosting server

[14:54] < MS-> Since I personally own the domain kamaelia.org, we can happily then shift everything over to that and let it have a life of its own

[14:54] < MS-> It would also mean that non-BBC people could have access to it as well

[14:55] < MS-> This is kinda necessary since the location where edit.kamaelia.org is currently hosted is in a machine room on a site that the BBC has to vacate at some point in the next 6-18 months or so

[14:55] < vmlemon_> Ugh

[14:55] < MS-> It also deals with the "people leaving the BBC" issue (though that's not likely)

[14:56] < MS-> It might sound like jumping through hoops for no reason BUT it actually protects the project from lots and lots of very annoying BBC politics

[14:56] < MS-> So is a good thing overall

[14:56] < MS-> I've never liked the idea of hosting the project on BBC servers for the simple reason that you raised "what happens when X and Y decide it's time to move on"

[14:57] < MS-> Hosting it *off* BBC servers makes it clear that it's out there to stay

[14:57] < MS-> :)

[14:57] < MS-> I'm actually planning on doing website things this weekend

[15:01] *** mhrd tests

[15:01] < MS-> mhrd: I just realised - if you argue that most of the concepts in Kamaelia are relatively old, but being brought to a new audience, then it's "introducing a technology or approach to a new sector" - which is the textbook definition of innovation

[15:01] < MS-> mhrd: Don't worry about breaking the site btw :)

[15:01] < mhrd> MS-: data backed up? (in case I do something silly) ?

[15:02] < MS-> Just doing that :)

[15:02] < Lawouach> which reminds me that I have to fill the various forms to get the required papers

[15:02] *** Lawouach hates that

[15:02] < mhrd> say "when", then I'll restart test

[15:02] < mhrd> heh: http://www.austinthirdgen.org/upload/piechart.jpg

[15:02] < MS-> heh

[15:03] < MS-> OK, done :)

[15:03] < mhrd> rev 5012 ?

[15:03] < MS-> Sounds about right

[15:04] < MS-> Oh no

[15:04] < MS-> repository of rev 5351

[15:04] < mhrd> must still be using sf.net

[15:04] < MS-> Higher than I thought :)

[15:04] < vmlemon_> I think that was probably a TSiF question, even if I didn't mean it to sound that way

[15:04] < MS-> TSiF ?

[15:05] < vmlemon_> "The Sky is Falling"

[15:05] < MS-> Ahh, I see

[15:05] < MS-> Didn't take it that way

[15:05] < MS-> It's a valid Q

[15:05] < MS-> And is something planned for :)

[15:06] < vmlemon_> Was just curious...

[15:06] < MS-> If I thought curiosity was a bad thing, I probably wouldn't choose a cat as a logo :)

[15:07] *** vmlemon_ has joined #kamaelia

[15:09] < vmlemon_> kamlogbot: dance

[15:09] Reply: does the macarena

[15:10] < mhrd> ok, rerunning Axon Docs with svn co from code.google.com

[15:11] < MS-> cool

[15:12] < mhrd> MS- : you backed up to /tmp/backed_up_axon_pydocs ? same name as the build script tries to use :-)

[15:12] *** mhrd will rename

[15:13] < MS-> Nope - I backed up in 2 other locations

[15:13] < MS-> well, 3 actually

[15:13] < mhrd> k

[15:13] < MS-> :)

[15:14] < mhrd> ah, it probably just moved the old docs, which had wrong permissions or something, then when trying to delete them on a subsequent build ... :-)

[15:14] *** MS- nods

[15:14] < mhrd> yep

[15:15] < mhrd> ok, thats new axon docs done

[15:15] *** mhrd tries kamaelia docs

[15:21] < mhrd> ok, looks successful

[15:24] *** vmlemon_ has joined #kamaelia

[15:28] *** MS- looks

[15:28] < MS-> Indeed it does :)

[15:28] < MS-> Much appreciated

[15:29] < mhrd> np

[15:29] < mhrd> relatively straightforward

[15:29] *** mhrd targets next train home

[17:03] *** vmlemon_ has joined #kamaelia

[17:46] < vmlemon_> Hmm, I wonder how that works... "However, under NDA, we are willing to share data about your network with you - in exchange for some data on your network we do not have"

[17:53] < MS-> hmm?

[17:57] < vmlemon_> Aha, just read the next slide of the Akamai presentation, and it relates to them giving peering network operators TCP flow information and statistics, in exchange for them getting NetFlow/sFlow data

[17:57] < vmlemon_> Still, sounds weird to me

[18:01] < vmlemon_> (Essentially something like "We've got your data, but you can't have it until you give us more")

[18:04] < MS-> Some people have broken networks, but don't realise it

[18:04] < MS-> They assume that their network works fine, and someone like Akamai can actually tell them how it really is

[18:05] < MS-> There was a particular european network operator a few years back that was certainly large enough to know better that had completely misconfigured ingress/egress routing set up on one of their major routers

[18:05] < vmlemon_> Sounds fun

[18:06] < MS-> Caused their actual throughput to be significantly lower (several multiples) than it should be

[18:06] < vmlemon_> I'd hate to think of the problems it caused

[18:13] *** vmlemon_ assumes that network was orders of magnitude "noisier" with all the bad BGP or/and OSPF traffic...

[18:14] < vmlemon_> Probably caused some nice loops

[18:22] *** Mgdmpd has joined #kamaelia

[18:24] *** Mgdmpd is now known as vmlemon_

[20:46] *** vmlemon_ has joined #kamaelia