[00:53] *** bcarlyon|laptop has joined #kamaelia

[06:27] *** jlei_ has joined #kamaelia

[07:44] *** j_baker_ has joined #kamaelia

[08:11] < ms-> morning

[08:11] < jlei_> morning, ms

[08:12] < Lawouach> morning

[09:06] *** vmlemon_ has joined #kamaelia

[09:26] *** Chong- has joined #kamaelia

[09:26] < Chong-> Morning, all.

[09:46] < vmlemon_> Hi

[09:54] *** Uraeus has joined #kamaelia

[10:20] < Chong-> Morning, vmlemon_

[10:38] *** vmlemon_ has joined #kamaelia

[11:02] *** vmlemon__ has joined #kamaelia

[11:26] *** vmlemon__ has just found http://www.wackypackages.org/bestof/jollymean.html

[11:35] *** vmlemon_ has joined #kamaelia

[11:39] *** Uraeus has joined #kamaelia

[12:03] *** Davbo has joined #kamaelia

[12:18] < Davbo> Good idea's ms- :-)

[12:19] *** vmlemon_ wonders if the license at http://paster.dazjorz.com/?a=rawpaste&p=3077 could be considered as "full Open Source", seeing as you can't actually redistribute code provided under it (although you can modify it)

[12:28] *** ms- reads

[12:28] < ms-> Davbo: Glad they're useful

[12:29] < ms-> "Distribution of the Package by You, whether in source or binary form is not permitted."

[12:29] < ms-> That means it's not open source

[12:29] < Davbo> Writing a wishlist of things i want now

[12:29] < ms-> Cool

[12:29] < Davbo> if you could add what you want too sometime, that'd be great

[12:30] < Davbo> we should have a lot to shoot for

[12:30] < ms-> Will do

[12:30] < ms-> Having a long list ideas is however a good thing. It allows you to go "I actually really want this" for prioritising stuff

[12:30] < Davbo> extending to more useless but cool features might be a hobby of mine after GSoC :-)

[12:31] < ms-> :)

[12:31] < ms-> That's also the sort of thing that other people could be interested in doing

[12:31] < ms-> You'll probably want to look at what j_baker is doing with packaging btw

[12:31] < ms-> since what he's doing is really nice

[12:31] < Davbo> Oh yeah I have

[12:32] < Davbo> he said it would break my svn folder so I haven't downloaded anything yet lol

[12:32] *** ms- nods

[12:32] < Davbo> interested though

[12:32] < ms-> you know that you can just copy your entire svn folder by doing:

[12:32] < ms-> cp -R my_current_svn svn_to_test_jbakers_stuff_so_I_dont_lose_my_stuff

[12:33] < ms-> If you didn't, you do now :)

[12:34] < Davbo> Well that wasn't the problem really (and yes i'm aware of cp lol) It was more the fact that if it doesn't really know what to do with .svn maybe it's not ready yet.

[12:34] < Davbo> I'd sooner let him test it and then when he's bragging about how good it is I'll take a look ;-)

[12:36] < ms-> :)

[12:36] *** vmlemon_ finds the http://www.jumpstartforwireless.net/downloads.htm page slightly odd, seeing as none of the CVS links work, and there aren't actually any tarballs for download, despite what the page claims

[12:37] < Davbo> I need another paint tool which has a RGB colour selector, The Gimp's isn't what i'm looking for really..

[12:37] < Davbo> trying to think of the best algorithm for it

[12:38] *** vmlemon_ doubts Photoshop Express would be any use...

[12:38] < Davbo> heh

[12:39] < ms-> kolourpaint ?

[12:39] < vmlemon_> From what I can gather, it's web-based, slow, and needs a load of other junk downloading before it works...

[12:39] < vmlemon_> (Or at least that's the impression I get from some)

[12:39] < Davbo> danke ms-

[12:39] < Davbo> ah nah, that's not what i'm looking for either hmm

[12:39] < ms-> vmlemon_: re photoshop express Works "OK" under linux. It's definitely slow.

[12:40] < vmlemon_> Interesting

[12:40] *** vmlemon_ has never tried it

[12:40] < ms-> Davbo: https://kamaelia.svn.sourceforge.net/svnroot/kamaelia/trunk/Sketches/MPS/Systems/Paint/Paint.py ?

[12:40] *** vmlemon_ meant to try Xara LX out again, to see if they've improved it since last time he looked

[12:40] < ms-> vmlemon_: By coincidence I'd been looking at it (photoshop express_ earlier today

[12:42] < ms-> Davbo: That's a distant ancestor of the whiteboard

[12:44] < Davbo> damnit don't have Tkinter

[12:44] < Davbo> 1 moment :-)

[12:45] < Davbo> ooo I was thinking about doing it that way

[12:45] < Davbo> what you think to that ms-?

[12:47] < ms-> It's one way.

[12:47] < ms-> It'd probably nice nicer to explore pygame based widgets though

[12:47] < ms-> Either writing a trivial simple one

[12:48] < ms-> Or nabbing some in

[12:48] < ms-> Simply because tkinter is a real pain to use on Macs

[12:49] < vmlemon_> If PS Express is slow, I'd hate to see how "well" something like a web-based version of Pro/Desktop or Blender would work ;)

[12:49] *** Davbo has joined #kamaelia

[12:49] < Davbo> sorry about that

[12:50] < vmlemon_> Aww, it looks like development has stalled since 2006 on Xara LX...

[12:50] < vmlemon_> It looked promising, too

[12:50] < Davbo> that the 3d tool?

[12:50] < vmlemon_> It was a drawing/painting app

[12:51] < vmlemon_> Apparently, they never actually open sourced the core drawing engine...

[12:52] < ms-> That doesn't surprise me tbh

[12:54] < vmlemon_> Looks like there's still some activity on the mailing lists, although the main project site seems dormant

[12:56] < vmlemon_> (Most of it being speculation on whether or not the project is still alive, or trying to build a Cairo-based backend for it)

[12:56] < vmlemon_> http://www.xaraxtreme.org/maillists/archive/dev/dev_042008/msg00070.html

[13:02] < ms-> I see

[13:03] < ms-> My take would be that the latter option is better

[13:03] < vmlemon_> http://www.xaraxtreme.org/maillists/archive/dev/dev_022007/msg00098.html

[13:03] < ms-> Since then they'd no longer be dependent on the team that's letting it be dead

[13:03] < vmlemon_> Seems to be what's happening right now, or at least where the banter is

[13:04] < vmlemon_> It'd be a shame for it to vanish off the face of the Earth completely, though, even if it didn't get much interest either from contributors or from Xara themselves

[13:04] < ms-> inkscape is annoying. It's really nice, but it's export is just incompatible with other things

[13:04] < ms-> (svg export this is)

[13:04] < vmlemon_> I like Inkscape, but I find that it does "non-standard" things with SVG, so that it doesn't render correctly in other SVG viewers

[13:05] < vmlemon_> (Konqueror's/KSVG? comes to mind)

[13:06] < vmlemon_> Hmm, "PS One of the obvious 'disastrous' scenarios for CDraw, and indeed the whole project, is if were to be ported back to Windows and used to compete against our commercial versions, so jeopardising the revenue and jobs of the staff working on the product..."

[13:07] < ms-> Yeah. It's SVG breaks hard in firefox too

[13:09] < vmlemon_> I haven't used it for a while, but I tended to export stuff as PNG, and keep a copy of the original Inkscape-extended SVG and resources alongside it

[13:10] *** vmlemon_ has noticed that KOffice does strange/bad things to OpenOffice's ODF files, too

[13:11] *** ms- has merged PygameText branch

[13:12] < vmlemon_> (I even posted to the OO.o bug tracker ages ago with sample files, and they didn't know what to make of it )

[13:12] < vmlemon_> Probably should have pestered the KOffice folks, though

[13:15] < ms-> It probably is KOffice's problem to be honest

[13:17] < vmlemon_> It looks like it handled formulae objects in a different way that was incompatible with OpenOffice, and it also stripped certain metadata from files

[13:29] *** ms- added beginnings of notes on changelog to http://edit.kamaelia.org/ReleaseJune2008

[13:35] < Davbo> back later

[13:36] *** ms- wonders if http://edit.kamaelia.org/Components/pydoc/Kamaelia.Video.PixFormatConversion.html can be used trivially (or extended) to support conversion of pygame surfaces to YUV surfaces suitable for stuffing into the diac encoder

[14:02] *** vmlemon__ has joined #kamaelia

[14:05] *** vmlemon__ is now known as vmlemon_

[14:06] *** basicdemokbot has joined #kamaelia

[14:06] *** basicdemokbot has parted #kamaelia

[14:22] *** orphans has joined #kamaelia

[14:22] < vmlemon_> Interesting, http://www.riscosopen.co.uk/content/

[14:23] < ms-> Surprising too

[14:23] < ms-> I wouldn't've expected to see that

[14:23] < vmlemon_> Open Source RISC OS...

[14:24] < ms-> Would be interesting to see that on a GP2X :)

[14:24] < ms-> (since that's ARM based)

[14:26] < vmlemon_> Not sure what the terms of use is on it, or which version of RISC OS it's based on, but it looks promising

[14:26] < orphans> ms-, saw you merged JL_PyGameText - cheers :)

[14:27] < vmlemon_> Hah, "Anyway, new companies are always welcome at least if they are not called "RISC OS Closed Ltd.""

[14:29] < ms-> orphans: :)

[14:30] < ms-> vmlemon_: Again, I'm not sure that it's open source - it looks like it fails the OSI definition due to a possible snafu regarding commercial/non-commercial

[14:30] < ms-> cf http://www.riscosopen.co.uk/content/documents/licences

[14:30] < vmlemon_> They have tarballs, by the looks of things

[14:31] < Chong-> ms-: hi, one question,

[14:31] < orphans> ms-, accepted/rejected list all look cool too - I know a bit more what I need to put in/not put in now too :)

[14:31] < vmlemon_> Dunno if it's best described as "Sources Available"/"Transparent Development"/"Shared Source"/"Commercial Open Development" or whatever the buzzword of the nanosecond is for that sort of thing

[14:31] < Chong-> At present, the bondLength of all particles are fixed according to the definition of physical laws.

[14:31] *** orphans stops bugging ms- now and gets on with his weekend :)

[14:31] < Chong-> For a particle system with the same size, we can calculate a good value for bondLength; but for a particle system with different sizes, it could become a problem.

[14:32] < Chong-> I'm thinking about letting bondLength(A,B)=A.radius+B.radius+aValue. I have not find a good way to do it until now. Do you have any suggestion about it?

[14:32] < ms-> Play and see what works

[14:33] < vmlemon_> Looks like the sort of thing QNX are trying to do at the moment...

[14:33] < Chong-> orphans: now, I am embarrassed to ask ms- questions :-)

[14:34] < vmlemon_> ((Mystical hand-waving) "Look! Shiiiny Oopen Code!", even though it isn't really)

[14:34] < vmlemon_> ;)

[14:34] < ms-> vmlemon_: Castle's license specifically prohibit's specific uses

[14:34] < ms-> that means it's definitely not open source

[14:34] < vmlemon_> Ugh

[14:35] < ms-> section 2.4 prevents OEM'ing

[14:35] < ms-> in a hardware product

[14:35] *** vmlemon_ wonders if we'd need to wait for Castle/RISC OS Open to go bust, and then a million more years for it to enter the public domain, before it could ever be considered as "open source"...

[14:38] < ms-> Chong-: You can ask, however don't expect an answer, and if one isn't forthcoming, re-ask on the list (which is generally a good idea with *any* open source project)

[14:39] < ms-> not expecting an answer doesn't mean to say you won't get one

[14:39] < ms-> Just that it might not come from the person you expect

[14:39] < ms-> Also, just asking me misses the fact that this is a collaborative project and everyone's opinions are valid

[14:39] < ms-> (which isn't the way all projects are run, I admit)

[14:40] < Chong-> ms-: got it. I would post it on the list later. :-)

[14:43] < ms-> I answered your question though

[14:43] < ms-> """I'm thinking about letting bondLength(A,B)=A.radius+B.radius+aValue. I have not find a good way to do it until now. Do you have any suggestion about it?

[14:43] < ms-> (15:37:53) ms-: Play and see what works"""

[14:44] < ms-> Besides, the system already handles the case you appear worried about

[14:44] < ms-> https://kamaelia.svn.sourceforge.net/svnroot/kamaelia/trunk/Code/Python/Kamaelia/Kamaelia/Visualisation/ER/ERLaws.py

[14:45] < ms-> ~/kamaelia/trunk/Code/Python/Kamaelia/Kamaelia/Visualisation/ER> grep -i bondlength ERLaws.py |grep -v def

[14:45] < ms-> entity_entity = SimpleLaws( bondLength = relationBondLength,

[14:45] < ms-> maxRepelRadius = 2.3 * relationBondLength,

[14:45] < ms-> relation_relation = SimpleLaws( bondLength = relationBondLength,

[14:45] < ms-> entity_attribute = SimpleLaws( bondLength = _COMPONENT_RADIUS*2,

[14:45] < ms-> entity_relation = SimpleLaws( bondLength = _COMPONENT_RADIUS*3,

[14:45] < Chong-> ms-: I see. Sorry, I had thought you said that to others :-)

[14:46] < ms-> Reading your question in more detail, the problem lies in the question:

[14:46] < ms-> "For a particle system with the same size, we can calculate a good value for bondLength; but for a particle system with different sizes, it could become a problem."

[14:46] < ms-> Specifically:

[14:46] < ms-> "it could become a problem"

[14:46] < ms-> that means you don't *know* it's a problem

[14:47] < ms-> You need to determine that first

[14:47] < ms-> Since, actually, we do already have the case you're worried about (it's fairly common - I believe the AxonLaws does the same) it would suggest it isn't a problem

[14:48] < ms-> and worth not worrying about

[14:48] < Chong-> I know it's a problem. for two particles which are very big, they could overlay each other;

[14:48] < ms-> In which case, that's NOT the same thing as what you said

[14:48] < Chong-> for two particles which are very small, the distance between them would be too far.

[14:49] < ms-> You aren't being precise in what you're talking about

[14:49] < ms-> which is the problem

[14:50] < ms-> What you meant to say, I guess

[14:50] < ms-> is more like

[14:50] < ms-> Normally a particle class creates objects which are the same size

[14:50] < Chong-> English is not my first language ;)

[14:51] < ms-> Then be precise in your own language first, then translate

[14:51] < ms-> What you meant to say was

[14:51] < ms-> presumably something like

[14:52] < Chong-> I'll. :-)

[14:52] < ms-> At present, a particle class creates objects which are the same size.

[14:52] < ms-> This means that having the same bond length for these makes sense.

[14:52] < ms-> I am creating a particle class which renders objects which are different sizes.

[14:53] < ms-> This means that the bondlength may need to be different

[14:53] < ms-> It may need to be different for different particles in the same class

[14:53] < Chong-> correct. Your guess is just what I had thought.

[14:53] < ms-> "I would like to know how to proceed"

[14:54] < ms-> "what is your suggestion"

[14:54] < ms-> My suggestion would be : play, see what works

[14:54] < ms-> My second suggestion : consider resizing particles to a standard size *unless* they have been clicked on

[14:54] < ms-> And allow a click on particle to be rendered full size.

[14:55] < ms-> However, that still falls under the heading of "play, see what works"

[14:55] < ms-> So the answer remains the same

[14:55] *** ms- goes back to what he was doing

[14:56] < Chong-> ms-: thanks for your advice. Enjoy your time.

[14:59] < ms-> I'm going through the IRCClient branch right now. I'm working on the release

[14:59] < ms-> The AIM branch & pygame branch have been merged

[14:59] < ms-> Then I'll move onto continue splitting out code from my branch

[15:22] *** vmlemon_ has joined #kamaelia

[15:29] *** ms- kills the prefab that confused chong, replaces all of that prefab with 2 demos, shifts support code to Kamaelia.Support.Protocol.IRC

[15:29] < ms-> leaves Kamaelia.Protocol.IRC.IRCClient in much better shape

[15:43] < Chong-> ms-: cool. it seems that next release is coming very soon with your hard work :)

[15:54] < ms-> Well, maybe

[15:54] < ms-> I'm not going to release broken code

[15:54] < ms-> But there is a ton of stuff in it

[15:54] < ms-> http://edit.kamaelia.org/ReleaseJune2008

[15:54] < ms-> cf  Current Changelog notes (since last release)

[15:58] *** ms- notes that Sourceforge has stopped forwarding mails regarding changes

[15:58] < ms-> :-(

[16:10] *** ms- sends mail regarding merge notes/branch improvements regarding that branch

[16:14] < ms-> Branch merged

[16:14] < Chong-> From the Changelog, a lot of useful stuff have been added. That's awesome.

[16:15] < Chong-> Another negative point of Sourceforge pushes us toward goole group :-)

[16:16] < ms-> No, that's nothing to do with the mailing lists

[16:16] < ms-> It's everything to do with integration of their systems

[16:16] < ms-> The backup is clear though:

[16:16] < ms-> http://cia.vc/stats/project/kamaelia/.rss?ver=2&medium=xhtml&limit=100

[16:16] < ms-> Which is useful

[16:17] < Chong-> ms-: Do you mean the commit message?

[16:18] < ms-> The commits, yes

[16:19] < ms-> OK, new branch time

[16:19] < ms-> merge of Kamaelia-Logger

[16:20] < ms-> Hm.

[16:21] < ms-> Actually that requires the TCP mods to be merged first

[16:21] *** ms- creates private_MPS_TCPServerImprovments instead

[16:22] < Chong-> ;)

[16:23] < Chong-> I see. if google group could give us enough space, we can also move svn server there.

[16:23] < ms-> Um, no

[16:23] < Chong-> ms-: What's happened?:-)

[16:23] < ms-> ?

[16:24] < ms-> (17:28:12) Chong-: I see. if google group could give us enough space, we can also move svn server there.

[16:24] < ms-> (17:28:22) ms-: Um, no

[16:24] < ms-> I was just saying "no"

[16:24] < Chong-> I see. :-)

[16:25] < Chong-> So, you think sourceforge's is better than google's?

[16:26] < ms-> No, it's not enough of a benefit at this instant to warrant the &^*^ing about it would involve

[16:26] < ms-> Besides, SF backs up content around the world

[16:26] < ms-> But the key thing is the benefit vs effrot

[16:27] < Chong-> I see. Especially the new release is around the corner.

[16:27] < Chong-> I will go out for a while. See you later.

[16:28] *** Chong- has parted #kamaelia

[16:28] < vmlemon_> Any more progress on the shiny, new, hype-infused site?

[16:31] < ms-> Priorities - get code merged first :-)

[16:31] < ms-> But sortof, yes

[16:32] < ms-> The fact it's all on edit.kamaelia.org makes a huge difference in practical terms :-)

[16:32] < ms-> http://edit.kamaelia.org/Developers/ has been tidied up a touch

[16:33] < ms-> I'm also looking to take a version of Lawouach's new intro which he wrote for his site and improve the Kamaelia one somewhat: http://edit.kamaelia.org/SylvainIntroduction

[16:34] < ms-> http://edit.kamaelia.org/Home is also now more up to date

[16:34] < ms-> Getting the content in place for changing the shape of the site

[16:34] < ms-> http://edit.kamaelia.org/Developers/Projects/KamaeliaPublish

[16:34] < ms-> http://edit.kamaelia.org/Developers/Projects/KamaeliaJam

[16:34] < ms-> http://edit.kamaelia.org/Developers/Projects/KamaeliaPaint

[16:35] < ms-> are all a part of that as well

[16:35] < ms-> as is /KamaeliaGrey /KamaeliaLogger

[16:35] < ms-> etc

[16:35] < ms-> Constant forward progress.

[16:40] < ms-> :)

[16:43] *** vmlemon__ has joined #kamaelia

[16:43] *** vmlemon__ is now known as vmlemon_

[16:51] *** ms- unifies SimpleServer & ServerCore's core code

[16:55] < ms-> based on j_baker's arguments, I've decided to go with option 2 in this http://groups.google.com/group/kamaelia/msg/f8c0ed45365ad15b?

[16:55] < ms-> Sorry, option 3

[16:55] < ms-> (essentially)

[16:56] < ms-> the idea being it then supports "putting back in place" existing protocol handlers relatively easily.

[16:56] < ms-> However, with a minor change to deal with the reason why I was going to do 2

[16:57] *** bcarlyon|laptop has joined #kamaelia

[17:05] *** vmlemon_ has joined #kamaelia

[17:15] *** vmlemon__ has joined #kamaelia

[17:15] *** vmlemon__ is now known as vmlemon_

[18:17] < ms-> Great, my new branch is misspelt :)

[18:26] *** Chong- has joined #kamaelia

[19:54] *** j_baker_ ponders trying out his packaging system on cygwin...

[19:55] < j_baker_> It should theoretically work.

[19:55] < ms-> Probably, yes

[19:56] < ms-> Did you see my note about wsgiref ?

[19:56] < j_baker_> Yes. That will be worked on tomorrow.

[19:57] < j_baker_> If I get a chance today, I'm going to try the system out on the Spam Assistant for practical reasons.

[19:57] *** j_baker_ gets a lot of spam

[19:58] < j_baker_> It's just gotten ridiculous the sheer amount of it all. Even with Apple Mail's spam tool.

[19:58] < ms-> cool

[19:59] < ms-> I'll upload my current phrases.txt file then

[20:00] < ms-> done

[20:00] < ms-> stats:

[20:00] < ms-> michaels@r44116:~/Pop3/SPAMSTORE$ grep X-OriginalArrivalTime: *|awk '{print $2,$3,$4}'|sort|grep 2008|uniq -c

[20:00] < ms-> 157 02 Jun 2008

[20:00] < ms-> 409 03 Jun 2008

[20:00] < ms-> 411 04 Jun 2008

[20:00] < ms-> 456 05 Jun 2008

[20:00] < ms-> 458 06 Jun 2008

[20:00] < ms-> 426 07 Jun 2008

[20:00] < ms-> 365 08 Jun 2008

[20:01] < ms-> Number of spams caught by it so far, per day

[20:06] < ms-> Branch ready to merge: private_MPS_TCPServerImprovments

[20:06] < ms-> Includes (for example) ServerCore

[20:07] < ms-> (mail posted to list)

[20:19] < j_baker_> Impressive.

[20:19] < j_baker_> (the spams caught that is)

[20:25] < ms-> Yep - all it does is remember decisions I take

[20:26] < ms-> I have to manually note them in phrases.txt, but often I take a long subject line and split it in two

[20:26] < ms-> in case it's being generated by a markov chain

[20:26] < ms-> eg that sentance "in case it's being generated by a markov chain"

[20:26] < ms-> could get split into

[20:26] < ms-> "in case it's being generated"

[20:26] < ms-> "generated by a markov chain"

[20:27] < ms-> And generally I find that assists in pre-empting spams

[20:27] < ms-> It's not perfect, but its *alot* quicker than I was doing

[20:32] *** vmlemon has joined #kamaelia

[20:53] < ms-> Branch private_MPS_KamaeliaLogger/ ready for merge

[21:23] < ms-> OK, what else is on my branch ...

[21:30] < vmlemon_> o.O http://tech.slashdot.org/comments.pl?sid=577355&cid=23700501

[21:39] *** bcarlyon|laptop has joined #kamaelia

[21:47] *** vmlemon_ has joined #kamaelia

[22:24] < ms-> Branch private_MPS_STM ready for merge

[23:51] < ms-> Branch private_MPS_Tools2AppsConsolidation also ready for merge, though I haven't written the merge mail yet