[10:25] *** vmlemon_ has joined #kamaelia

[10:27] < vmlemon_> Hi

[10:28] < ms-> hi

[10:30] < Lawouach> heya

[10:40] < vmlemon_> Crikey, the 2nd of November already

[11:03] *** ms- nods

[11:10] *** Lawouach has joined #kamaelia

[11:30] *** Rigolo has joined #kamaelia

[11:30] < Rigolo> good afternoon

[11:49] *** Davbo has joined #kamaelia

[11:54] < vmlemon_> Hi Rigolo and Da

[11:54] < vmlemon_> *Davo

[11:54] < vmlemon_> *Davbo

[11:54] < vmlemon_> (Can't type with my keyboard at an angle :|)

[11:57] < Davbo> :-)

[11:57] < Davbo> Hey vmlemon_, all.

[11:58] < vmlemon_> (Some idiot decided they'd move the keyboard, so that they could squander time on eBay, and never managed to put it back on the desk when I managed to force them off ;))

[12:02] < Rigolo> mmm I solved that problem by getting more computers and keyboards into the house :-)

[12:02] < vmlemon_> Heh

[12:03] < Rigolo> but now I feel like I am responsible for the complete profit of my electricity companay :-)

[12:03] < Rigolo> company also :-

[12:03] < Rigolo> 0

[12:03] < Rigolo> (damm keyboard ... can not type on it anymore)

[12:03] < vmlemon_> I did use a /etc/hosts and Apache2 hack to display a false HTTP Forbidden error a while ago

[12:03] < vmlemon_> although it didn't seem to work as planned

[12:04] < Rigolo> what happend then?

[12:04] < vmlemon_> Ended up needing to disabling it so that the said eBay addict could use the crappy thing, which kinda rendered it moot :|

[12:04] < vmlemon_> Gonna try GreaseMonkey and JavaScript snow

[12:05] < vmlemon_> that should p*ss them off ;)

[12:05] < Rigolo> I feel that a little investment in an extra computer is well worth it :-)

[12:05] < vmlemon_> Kind of annoyingly, the only connectivity we have is either GPRS/EDGE or UMTS, so sharing it with another machine wouldn't work too well

[12:05] < vmlemon_> (It'd be about 20 times slower than it already is)

[12:07] < Rigolo> mm where are you located then? .. and I hope you setup a router with a UMTS modem attached to it .. with squid etc etc behind it to cache as much as possible

[12:08] < vmlemon_> Aah, the coComment plug-in I installed on one of my Drupal sites actually works now

[12:08] < Rigolo> I did it for a computer shop that just opened up ..and the phone company kind of forgot to deliver the DSL

[12:08] < vmlemon_> I thought it didn't

[12:08] < vmlemon_> We're near Harrogate, although we can't afford BT line rental and ADSL at the moment :(

[12:08] < Davbo> what does squid do Rigolo?

[12:09] < vmlemon_> It's a caching proxy server

[12:09] < Davbo> ah

[12:09] < Rigolo> squid is a http cache ... never mind

[12:09] *** vmlemon_ finds WWWOFFLE to be good, although it seems flaky

[12:09] < Rigolo> vmlemon_: but it must be cheaper to get dsl then to use umts ...

[12:09] < vmlemon_> There was another one that I found, although the name escapes me

[12:09] < Rigolo> or are you on some unlimited dataplan?

[12:11] < vmlemon_> 3 give me 1GB across 30 days on a £10 prepaid allowance if I use UMTS, and Orange provide either £1/day or £5/week "unlimited usage" for flaky GPRS/EDGE

[12:11] < vmlemon_> (Also prepaid at the mo)

[12:11] < vmlemon_> 3's offering goes up to 10GB for £25 if memory serves correct

[12:12] *** Davbo wishes T-Mobiles 3G coverage wasn't so good near him

[12:12] < vmlemon_> Heh

[12:13] < Davbo> tempts me so much to get a G1 ;)

[12:13] < Rigolo> and what happens when you go over that limits?

[12:13] < vmlemon_> You can actually *get* T-Mobile 3G?

[12:13] < vmlemon_> We can't get it at all here

[12:13] < vmlemon_> Or Orange, O2 or Vodafone's 3G offerings

[12:13] < Davbo> Yeah the mast is really near by

[12:13] < vmlemon_> strangely

[12:13] < vmlemon_> Yay for half-built networks

[12:13] < vmlemon_> Rigolo: Good question

[12:13] *** Rigolo could get T-mobile 3G .. but they do not have the G1 in .nl yet :-)

[12:13] < Davbo> I can remember when people were protesting the mast, was fun times

[12:14] < vmlemon_> For Orange, nothing (it reverts to 1p/KB until the balance runs out, or I buy another GPRS "add-on" type thing)

[12:14] < Rigolo> but I have my data via the cable ... 12Mbit/1.5Mbit down/up at the moment .. going to 16Mbit/3Mbit soon

[12:14] < vmlemon_> For 3, they send you an SMS when you're getting to a certain amount of quota

[12:15] *** vmlemon_ wonders how much BT line rental is these days

[12:15] < Rigolo> vmlemon_: is that the only option? not some open wifi nearby?

[12:15] < vmlemon_> Last time I checked, the only Wi-Fi APs available belonged to neighbours, and they all had at least WEP or WPA on them

[12:16] < vmlemon_> :(

[12:16] < Davbo> WPA eh? Do you have an nvidia card? ;-)

[12:16] < vmlemon_> Hah

[12:16] < vmlemon_> ATI sadly ;)

[12:16] < Davbo> Hehe

[12:16] < vmlemon_> Although my desktop doesn't have a Wi-Fi card in it

[12:17] < vmlemon_> I did try to use a certain WEP password cracking tool, with no results at all

[12:17] < vmlemon_> even after letting it run for about a week nonstop

[12:18] < vmlemon_> (And getting several 100MB of libpcap-format packet logs)

[12:19] < vmlemon_> Looks like YouTube and last.fm streaming sort of work over UMTS

[12:19] < vmlemon_> (Although the fact that support in Amarok is broken for last.fm, means that it isn't very useful)

[12:22] < Rigolo> btw. broadband from BT is 16 pound/month

[12:23] < vmlemon_> Amazing, it actually worked, after resubmitting my test comment

[12:23] < vmlemon_> Not too convinced about running it though

[12:24] < vmlemon_> (If the experience is going to be that crappy, from my testing)

[12:27] < vmlemon_> Hmm, has anyone else renewed a domain name recently?

[12:33] < Davbo> Couple of months ago, yeah vmlemon_

[12:37] *** vmlemon_ wonders why WHOISing his returns "Renewal request being processed" for "Registration status"

[12:37] < vmlemon_> Unless Nominet are glacially slow

[12:39] *** Davbo shrugs, I control all my domains on my old dreamhost account

[12:40] < vmlemon_> Aah

[12:40] < vmlemon_> Not sure if renewing it through the registrar (Heart Internet here) makes a difference...

[12:42] < Davbo> doubt it, your host will just be connecting through Heart's API i guess

[12:42] < Davbo> to register on your behalf

[12:43] *** Davbo might be wrong

[12:43] < vmlemon_> I used their control panel

[12:43] < vmlemon_> although I happen to use 3rd party DNS servers

[12:43] < Davbo> heh, same here

[12:43] < vmlemon_> (Since I run my sites from a shared Xen partition that I contribute towards)

[12:43] < Davbo> Anyway, I'll be back later. :-)

[12:44] < vmlemon_> ns0.xname.org, ns1.dazjorz.com and ns1.xname.org

[12:44] < vmlemon_> so WHOIS says

[12:44] < vmlemon_> See you

[12:50] < vmlemon_> Ooh, http://trolltech.com/developer/technical-preview-qt-for-s60

[12:50] < vmlemon_> Who would have thought that would have happened?

[12:52] *** vmlemon_ wonders why the "commercial license" download is smaller than the GPL'd one

[12:55] *** elvum has joined #kamaelia

[13:59] *** vmlemon_ sees that the lipsum is still under "Source & SVN" on the Kamaelia homepage...

[16:10] *** Davbo has joined #kamaelia

[17:23] < Davbo> hahaha!! they really shouldn't let me use python

[17:23] < Davbo> it's just too damn easy

[17:24] < Davbo> everyone else fiddling with tokens to handle this file and i just "import csv"

[17:33] < vmlemon_> Heh

[17:53] *** vmlemon_ has joined #kamaelia

[18:38] *** mhrd-home has joined #kamaelia

[18:38] < vmlemon_> Hi mhrd-home

[18:39] < mhrd-home> hi vmlemon_

[18:41] < mhrd-home> ms- : r5487 log notes ... can you point me at which examples demonstrate the discrepancy?

[18:43] < ms-> evening

[18:43] < ms-> Test claims:

[18:43] < ms-> FAIL: If a graphline's "control" inbox is not specified to be wired to a child component in the graphline, then any message (including shutdown messages) flows to the "control" inbox of all children without linkages going to their "control" inbox only.

[18:43] < ms-> survey says:

[18:43] < ms-> https://kamaelia.googlecode.com/svn/branches/private_MPS_ImprovedGraphlineShutdown/Kamaelia/Examples/UsingChassis/Graphline/DemoShutdown.py

[18:43] < ms-> Graphline(

[18:43] < ms-> TO_SHUTDOWN1 = Waiter(),

[18:43] < ms-> TO_SHUTDOWN2 = Waiter(),

[18:43] < ms-> TO_SHUTDOWN3 = Waiter(),

[18:43] < ms-> linkages = {}

[18:43] < ms-> ),

[18:43] < ms-> Wouldn't shutdown unless that test *should* pass

[18:44] < ms-> There's a few examples in here now btw:

[18:44] < ms-> https://kamaelia.googlecode.com/svn/branches/private_MPS_ImprovedGraphlineShutdown/Kamaelia/Examples/UsingChassis/Graphline

[18:44] < mhrd-home> you're saying the test_Graphline.py fails on that item?

[18:44] < ms-> Yep

[18:44] < mhrd-home> hmm, all tests pass for me(!)

[18:44] < ms-> ie the /test/ fails, but what the test says it's testing works

[18:44] < ms-> hehe

[18:44] < ms-> yay, heisenbug

[18:44] < mhrd-home> \o\ /o/ \o\

[18:45] *** mhrd-home double checks he's 'sudo python setup.py install'ed recently

[18:45] < ms-> Just reinstalled and tested

[18:46] < ms-> same failure

[18:46] < ms-> Interesting test harness set though

[18:46] < mhrd-home> heh, just done same and its still passing :-)

[18:46] < ms-> I've made a note to come back to it

[18:46] < ms-> I hate software sometimes

[18:46] < ms-> :-)

[18:46] < mhrd-home> lol

[18:46] < ms-> (will be largely afk this evening btw)

[18:46] < mhrd-home> python version? I'm 2.5.2

[18:46] < mhrd-home> k, np

[18:47] < ms-> ~/code.google/kamaelia/branches/private_MH_ImprovedGraphlineShutdown_Tests/Python/Kamaelia/Chassis> svn update

[18:47] < ms-> At revision 5487.

[18:47] < ms-> ~/code.google/kamaelia/branches/private_MH_ImprovedGraphlineShutdown_Tests/Python/Kamaelia/Chassis> ./test_Graphline.py

[18:47] < ms-> ...F.........F.........

[18:47] < ms-> (two tests marked in code)

[18:47] < Davbo> hehehe, you know it's a good bug when you reach "what version of python you running?"

[18:47] < mhrd-home> yep - saw - thanks for marking

[18:47] < ms-> ~/code.google/kamaelia/branches/private_MPS_ImprovedGraphlineShutdown/Kamaelia> svn update

[18:47] < ms-> At revision 5487.

[18:48] < ms-> ~/code.google/kamaelia/branches/private_MPS_ImprovedGraphlineShutdown/Kamaelia> sudo python setup.py install --force

[18:48] < ms-> ~/code.google/kamaelia/branches/private_MPS_ImprovedGraphlineShutdown/Axon> sudo python setup.py install --force

[18:48] < mhrd-home> ok, just force reinstalled Axon and Kamaelia, both from the branch, both at r5487. still passing :-)

[18:48] < ms-> ~/code.google/kamaelia/branches/private_MH_ImprovedGraphlineShutdown_Tests/Python/Kamaelia/Chassis> ./test_Graphline.py

[18:48] < ms-> ...F.........F.........

[18:48] < ms-> heh

[18:49] < ms-> ~/code.google/kamaelia/branches/private_MH_ImprovedGraphlineShutdown_Tests/Python/Kamaelia/Chassis> python -V

[18:49] < ms-> Python 2.5.1

[18:49] < ms-> """ hehehe, you know it's a good bug when you reach "what version of python you running?"""

[18:49] < ms-> That's one of the more annoying kinds IMO

[18:50] < Davbo> Yeah, /me was being sarcastic...it doesn't come accross on IRC

[18:50] < Davbo> sometimes i forget :-)

[18:50] < ms-> np

[18:50] < mhrd-home> ms- : do you know which assertion its failing at?

[18:50] < ms-> I'll pastebin

[18:50] < mhrd-home> ta

[18:51] < ms-> http://pastebin.com/m7d7ca8df

[18:51] < ms-> I *think* it's not activating components not listed in linkage

[18:51] < ms-> linkages

[18:51] < mhrd-home> k - will investigate

[18:51] < ms-> Which is incorrect behaviour

[18:52] < ms-> Could be a stale reference

[18:52] < mhrd-home> other possibilities are that the test isn't letting enough schedule cycles pass before checking assertions (have not yet found a satisfactory fix to avoid that kind of issue)

[18:52] *** ms- nods

[18:52] < vmlemon_> Anyone tried the BBC content-through-customised Totem thing, yet?

[18:52] < ms-> vmlemon_: I use the "wrong" distro

[18:53] *** Davbo also uses the "wrong" distro

[18:53] < ms-> i'm unsupported by my own org

[18:53] < ms-> *sob*

[18:53] < vmlemon_> Heh

[18:53] < ms-> *whinge*

[18:53] < Davbo> I did want to check it out too

[18:53] < ms-> :)

[18:53] < ms-> Davbo: Try a live cd in a virtual box

[18:53] < Davbo> can probably get it working tbh

[18:53] < mhrd-home> vmlemon_ : sems to work well ... though they don't have some of the particular programmes I was hoping would be there. great to seeit though

[18:53] < vmlemon_> You could just recompile it for your choice of superior distribution ;0

[18:53] < vmlemon_> ;)

[18:55] < ms-> mhrd-home: Indeed it is great to see

[18:55] < ms-> You seen the reactions on the bbc internet blog?

[18:55] < ms-> http://www.bbc.co.uk/blogs/bbcinternet/2008/10/research_on_bbc_content_for_gn.html

[18:56] < ms-> no pleasing some people

[18:56] < mhrd-home> will i want to stab my eyes out?

[18:56] < mhrd-home> that'll be a "yes" then :-)

[18:56] < ms-> yes, definitely

[18:56] < Davbo> I got as far as "GNU/Linux" then closed it

[18:57] < mhrd-home> """20:19 ms- I *think* it's not activating components not listed in linkage .... which is incorrect behaviour"""

[18:57] < mhrd-home> I swear, ages ago, it was decided that was desirable behaviour?

[18:59] < mhrd-home> ...however I think, right now, I'd agree that it doesn't *feel* right

[19:06] *** vmlemon_ has joined #kamaelia

[19:10] < mhrd-home> """20:19 ms- I *think* it's not activating components not listed in linkage""" ... doesn't look like that is the problem (added tests to verify, r5488)

[19:12] < mhrd-home> ms- : (whever, not necessarily today) can you confirm there are no local mods hiding in your checkout of private_MPS_ImprovedGraphlineShutdown and private_MH_ImprovedGraphlineShutdown_Tests ? (I will do same)

[19:16] < mhrd-home> ha - ok, I'm getting the errors too now! looks like its my fault :-)

[19:17] < mhrd-home> ah - I commented out the yield on line 323 of Graphline.py but didn't check it in ... apologies

[19:18] < mhrd-home> ok, checked in fix (r5489)

[19:19] < mhrd-home> ms- : thanks for spotting

[19:21] *** mhrd-home switches to doing some data analysis

[19:41] < ms-> test passes here now too

[19:43] < mhrd-home> cool

[19:44] < ms-> cheers :-)

[19:44] < mhrd-home> np

[19:44] < ms-> In which case, I'll merge it - if we get regressions, we can review

[19:44] < ms-> We're more likely to hit problems if it's merged after all

[19:44] < mhrd-home> I'm happy with that

[19:44] < ms-> cool

[19:46] < mhrd-home> btw: I know there was a lengthy discussion before I reviewed then modified lots. They've been strangely silent since :-)

[19:46] < ms-> :)

[19:48] < Davbo> Python is awesome.

[19:48] < ms-> If I had to pick a potential long term replacement for python, it'd be reia

[19:48] < ms-> http://wiki.reia-lang.org/wiki/Reia_Programming_Language

[19:49] < Davbo> making your own compare function and then providing it to the sort function..

[19:49] < Davbo> genius.

[19:49] < mhrd-home> Davbo: seen that pattern in other languages, though I suspect it involves less typing and declarations in python :-)

[19:50] < ms-> mhrd-home: I was just wondering what language didn't do that...

[19:50] < ms-> ... and then started remembering a few :)

[19:50] < ms-> Shows the sorts of languages I've been using for a while :)

[19:50] < ms-> Mind you, perl has one operator that no other language has which is fun

[19:51] < ms-> The "spaceship" operator

[19:51] < ms-> < =>

[19:51] < Davbo> it just makes the whole thing so transparent and obvious

[19:51] < Davbo> especially since i'm comparing a list of lists in quite a particular way

[19:52] < ms-> Finally, found the examples page I was looking for : http://wiki.reia-lang.org/wiki/Quickstart

[19:53] < mhrd-home> anyone remember how to escape a string for use as a GET variable value in php?

[19:53] < ms-> I'm pleased to say not. PHP is the new COBOL

[19:54] < ms-> sorry, that's not helpful

[19:54] < Davbo> can't remember i'm afraid mhrd-home

[19:55] < Davbo> the new COBOL? hehe

[19:56] *** vmlemon_ has joined #kamaelia

[19:56] < mhrd-home> np

[19:56] < mhrd-home> jus thought I'd ask whilst googling for the solution :-)

[19:56] < mhrd-home> aha urlencode()

[19:57] < ms-> looks like you found it before me

[20:09] < vmlemon_> Heh, that's really puzzling, 3 considers 229.80MB (or thereabouts) of a 1GB usage allowance to be a "low balance"...

[20:09] *** vmlemon_ would say that 10-50MB was "low"

[20:12] < ms-> mhrd-home: Is your test suite ready to merge as well?

[20:12] < ms-> it does pass after all

[20:12] < mhrd-home> yes

[20:12] < ms-> cool

[20:12] < ms-> will do that whilst I'm here

[20:12] < mhrd-home> cool, ta

[20:13] < mhrd-home> it has, imho, broad coverage of most of the claimed functionality. but as you say, most important thing is that it passes :-)

[20:13] < ms-> indeed

[20:13] < ms-> It's a good merge point

[20:19] < ms-> mhrd-home: That test suite for graphline is interesting. I may look at how we can use that to generalise it.

[20:20] < mhrd-home> I experimented a bit - fleshed out some stuff that had been done in teh pipeline test suite.

[20:20] < mhrd-home> made life easier in places, but still a bit hacky in places too :)

[20:20] *** ms- nods

[20:21] < ms-> Well, that's always the way when you're figuring out what works :)

[20:37] *** Davbo wonders what will be better for K-nearest neighbor. Taxicab/Manhattan metric of Euclidean

[20:37] < Davbo> s/of/or

[20:38] < Davbo> googled it and it looks like a few other people have wondered it too :-/

[21:12] *** ms- has merged three branches this evening

[21:15] *** mhrd-home gives ms- a pat on the head for being so productive :-)

[21:16] < ms-> heh

[21:16] < ms-> merging is easy of course - easier than writing the code in the first place :)

[21:18] < vmlemon_> Cool, thanks for getting a Less Hacky version of the FileAppender merged in

[21:18] < ms-> vmlemon_: Thanks for prompting it going in :)

[21:18] < ms-> Will be useful :)

[21:20] < vmlemon_> Does it still exhibit the weird glitch where data will be written in the wrong order, if a Console reader and writer are chained together with it?

[21:20] < vmlemon_> Out of interest

[21:20] < vmlemon_> (Along with a TCP server, if I remember correctly)

[21:21] < vmlemon_> Although it's been a while since I last touched the MSN server project I was working on that utilised those components in that way

[21:23] < ms-> I'm not sure what you experienced there - I think I'd need to see a test case to test that

[21:23] < vmlemon_> (Just reminded me that I needed to check some old scripts that I wrote into a Google Code repository I set up a while back for something)

[21:50] < vmlemon_> Haven't ran it in ages, but the code I've just put at http://code.google.com/p/i9os/source/browse/trunk/Subprojects/Kudasai/core.py produces the said problem, if I remember correctly

[22:09] *** vmlemon___ has joined #kamaelia

[22:11] *** vmlemon___ is now known as vmlemon_

[22:27] < ms-> vmlemon_: Name was based on me asking around for names of shrubberies, and being told about "Camellia". Liking it and mis-spelling it (delberately)

[22:27] < vmlemon_> Aah

[22:27] < vmlemon_> That's the one

[22:27] < ms-> I didn't hear about Camellia Sinesis until some time later

[22:27] < ms-> I'd like to retcon that and say it was based on tea, since tea at Kingswood is very important

[22:27] < ms-> but it wasn't, sadly

[22:28] < vmlemon_> Heh

[22:28] < ms-> But it is a nice co-incidence :)

[22:28] < mhrd-home> yay, tea!

[22:28] < ms-> mhrd-home: precisely :)

[22:29] < vmlemon_> Tea-powered snakes! ;)

[22:33] *** mhrd-home -> bedtime

[22:33] *** mhrd-home has parted #kamaelia

[22:33] < vmlemon_> Good idea here too, got a really nasty headache, and that pesky toothache's came back :(

[23:28] < vmlemon_> kamlogbot: log

[23:29] < vmlemon_> kamlogbot: info

[23:29] < vmlemon_> kamlogbot: logfile

[23:29] Reply: /srv/www/sites/edit.kamaelia.org/docs/logs/kamaelia2008-11-02_log.html