ignas | yes, yes i know, now imagine doing that 20 times in one day to understand how i felt packaging everything | 00:00 |
---|---|---|
fsufitch | <_< | 00:01 |
fsufitch | Installed /home/fsufitch/cando-trunk/eggs/zope.app.catalog-3.5.1-py2.4.egg | 00:01 |
fsufitch | Searching for zope.mimetype>=1.1dev-r72462 | 00:01 |
fsufitch | Reading http://download.zope.org/zope3.4/zope.mimetype/ | 00:01 |
fsufitch | No local packages or download links found for zope.mimetype>=1.1dev-r72462 | 00:01 |
fsufitch | error: Could not find suitable distribution for Requirement.parse('zope.mimetype>=1.1dev-r72462') | 00:01 |
fsufitch | make: *** [build] Error 1 | 00:01 |
ignas | can you lisppaste5 the whole log | 00:02 |
ignas | for me to see | 00:02 |
fsufitch | ok | 00:02 |
ignas | ok | 00:08 |
ignas | found the bug | 00:08 |
ignas | :/ | 00:08 |
fsufitch | ? | 00:08 |
ignas | apparently no one has released a new schooltool package after i have fixed the dependencies | 00:08 |
fsufitch | . | 00:09 |
fsufitch | that's IT? | 00:09 |
ignas | we can try one more thing though | 00:09 |
ignas | open setup.py | 00:09 |
ignas | and replace 'http://ftp.schooltool.org/schooltool/eggs/3.4' with 'http://ftp.schooltool.org/schooltool/eggs' | 00:09 |
ignas | and try make run again | 00:09 |
fsufitch | ok | 00:10 |
fsufitch | Installed /home/fsufitch/cando-trunk/eggs/zc.table-0.6-py2.4.egg | 00:13 |
fsufitch | error: Installed distribution zc.resourcelibrary 0.6.1 conflicts with requirement zc.resourcelibrary>=0.7dev-r72506 | 00:13 |
fsufitch | make: *** [build] Error 1 | 00:13 |
fsufitch | this seems a bit simpler | 00:13 |
ignas | no, not really :/ | 00:13 |
ignas | schooltool package that was released has wrong dependencies set up | 00:13 |
ignas | ok | 00:13 |
ignas | not wrong | 00:13 |
ignas | but unstable | 00:13 |
ignas | so you are suffering from the unstability of the eggs world | 00:14 |
ignas | but without someone pushing the big red button | 00:14 |
ignas | and tagging+releasing the new schooltool2007 alpha | 00:14 |
ignas | fixing everything is a bit difficult | 00:14 |
fsufitch | ah | 00:14 |
ignas | because old schooltool release relies on the unstable stuff, and is not compatible with the sable stuff | 00:15 |
ignas | which was not so bad while unstable eggs worked | 00:15 |
fsufitch | >.< | 00:16 |
ignas | i can try and release an emergency schooltool egg for now to get you working, but that is still a bad idea, as you might be stuck when someone will release the real thing | 00:17 |
ignas | so - how badly do you need to work on cando | 00:17 |
fsufitch | i don't need to work on it that badly | 00:21 |
fsufitch | and anyway, i could always send a diff of what i've done to aelkner or someone else | 00:22 |
ignas | i mean - i can package it and upload it to ignas.pov.lt/schooltool/, and you will have to replace the http://ftp.schooltool.org/schooltool/releases/2007.0.alpha5/ urls with http://ignas.pov.lt/schooltool/ | 00:23 |
ignas | and get it working on that machine | 00:24 |
fsufitch | hmm | 00:25 |
fsufitch | i'd rather i get it to work in a *proper* way | 00:25 |
ignas | ok | 00:25 |
ignas | i'll push for that happening during our monday meeting | 00:26 |
ignas | or even earlier | 00:26 |
ignas | i'll test cando for working with the updated egg | 00:26 |
ignas | and then write an email to jinty | 00:26 |
fsufitch | allrite | 00:28 |
fsufitch | thx | 00:28 |
*** Ninno has quit IRC | 00:29 | |
ignas | omg, apparently stable cando was getting broken because i had the experimental schooltool release installed on my computer | 01:03 |
fsufitch | hahaha | 01:03 |
ignas | i spent like 30 minutes trying to find out why are setuptools trying to install some old version of zope.html | 01:04 |
fsufitch | hehehe | 01:05 |
fsufitch | nice | 01:05 |
ignas | what is not so nice is that i don't know how to build a proper schooltool release | 01:12 |
fsufitch | O_O | 01:13 |
fsufitch | *gasp* ignas doesn't know something schooltool related | 01:13 |
ignas | yep, i don't do anything related to schooltool releases | 01:15 |
fsufitch | ah | 01:15 |
fsufitch | who does? | 01:15 |
ignas | jinty | 01:15 |
fsufitch | ah | 01:15 |
ignas | i hate setuptools | 01:26 |
fsufitch | heh | 01:26 |
fsufitch | oh btw | 01:26 |
fsufitch | my school has a research lab required to do in senior year | 01:26 |
ignas | yes | 01:26 |
fsufitch | and we're supposed to start thinking up of projects | 01:27 |
ignas | oh | 01:27 |
fsufitch | and one of my ideas was to do a study on which ubuntu interface would be the most friendly to windows-users switching to ubuntu | 01:27 |
ignas | "ubuntu interface" ? | 01:27 |
ignas | as in xubuntu kubuntu ubuntu ? | 01:27 |
fsufitch | no, those are subdistros | 01:28 |
fsufitch | although i could compare DEs | 01:28 |
fsufitch | i mean like the whole look+feel | 01:28 |
fsufitch | like, is the default GNOME really the best/the most intuitive? | 01:28 |
fsufitch | i mean, it isn't for me | 01:28 |
fsufitch | which is why i'm using kde with a really eccentric setup | 01:28 |
ignas | :D | 01:29 |
fsufitch | i don't even have a proper taskbar | 01:29 |
fsufitch | it's hidden away in a corner | 01:29 |
fsufitch | crying | 01:29 |
fsufitch | waiting for me to press the expand button | 01:29 |
fsufitch | but i never do | 01:29 |
fsufitch | <_< | 01:29 |
fsufitch | yeah | 01:29 |
fsufitch | well, i could compare the default setups of GNOME, KDE and XFCE | 01:30 |
fsufitch | although all of them SUCK in my opinion | 01:30 |
fsufitch | the kde one most | 01:30 |
fsufitch | rly space inefficient | 01:30 |
fsufitch | and too windows-ish XD | 01:30 |
fsufitch | but what i wanted to ask is | 01:30 |
fsufitch | do you think this project is a good idea? | 01:30 |
ignas | well - designing an intuitive interface and performing a lot of user tests would be kind of interesting | 01:31 |
ignas | if you can find enough users | 01:31 |
ignas | with simmilar level of computer literacy | 01:31 |
ignas | and if you are capable of customizing the UI to cover a lot of ground | 01:32 |
ignas | at least you would get to know how user testing is done | 01:32 |
fsufitch | i doubt i can do too many users while keeping the testing environment sane | 01:32 |
fsufitch | and within my research grant limits | 01:32 |
fsufitch | because i was thinking to get a bunch of ppl simple computers (laptops preferably) with linux on them | 01:33 |
fsufitch | and letting them use for a month or so | 01:33 |
fsufitch | with checkups on progress | 01:33 |
fsufitch | then a new set of people | 01:33 |
fsufitch | with improved setups | 01:33 |
ignas | an interesting approach, though i would not bet on it to be suitable for designing a universal user friendly UI | 01:34 |
ignas | from the scientific point of view | 01:34 |
ignas | everything depends on the goals of the project | 01:35 |
fsufitch | hmm | 01:36 |
fsufitch | i could always fall back on my other project idea of researching fuzzy logic in machine learning | 01:36 |
ignas | ??? | 01:36 |
ignas | not sure what you said makes too much sense | 01:37 |
fsufitch | exactly | 01:37 |
ignas | i mean fuzzy logic is a very very broad thing | 01:37 |
ignas | so you'd have to concentrate on something like hmm - reinforcement learning | 01:37 |
ignas | or expert system kind of things | 01:37 |
fsufitch | hmm | 01:38 |
ignas | but the former is heavy on maths and the later is old ;) | 01:38 |
fsufitch | <_< | 01:38 |
ignas | fuzzy logic most of the time means that "your machine can perform in different ways with the same input" | 01:38 |
ignas | well it can mean that the output is defined to some degree of certainty as well | 01:39 |
ignas | in the first case - a lot of algorithms can be used to train such "machines" | 01:39 |
ignas | reinforcement learning being one i have heard/read about most | 01:39 |
ignas | in the second case - i'd find it difficult to connect the learning part with the fuzzy part ;) | 01:40 |
ignas | (btw even though reinforcement learning is heavy on maths the concept itself is awesome) | 01:40 |
ignas | IMHO at least | 01:40 |
ignas | as for the first study subject - the "best/most intuitive" is a very subjective measurment, so you would have to define it in some way to be able to measure it | 01:42 |
ignas | as objectively as possible | 01:42 |
fsufitch | umm | 01:42 |
fsufitch | i did a small research project on fuzzy logic in 10th grade | 01:42 |
fsufitch | and it's defined as the opposite of binary logic | 01:43 |
fsufitch | like | 01:43 |
fsufitch | 1 DOES equal 2 | 01:43 |
fsufitch | 1 equals 3 as well | 01:43 |
fsufitch | but to a lesser degree than 1 ==2 | 01:43 |
fsufitch | so | 01:43 |
fsufitch | (1==2)<(1==3) | 01:43 |
fsufitch | but 1==2 is undefined | 01:43 |
fsufitch | and so is 1==3 | 01:43 |
fsufitch | it's all relative | 01:43 |
ignas | well - in AI fuzzy logic is used to define these other things | 01:44 |
ignas | when it is related to Expert systems | 01:44 |
ignas | you have this huge program that can tell you what the weather is going to be tomorrow | 01:44 |
ignas | but instead of saying "rain = True" | 01:45 |
ignas | it says "rain = 90% probability" | 01:45 |
fsufitch | yeah | 01:45 |
ignas | the other application is (at least in reinforcement learning) | 01:45 |
fsufitch | which SO many people love doing | 01:45 |
fsufitch | "i'm probably going to finish the project tomorrow" | 01:45 |
ignas | you are trying to control a small car on a track | 01:45 |
ignas | you have inputs like velocity, distance to the next turn etc | 01:46 |
ignas | but your car instead of always turning right and accelerating | 01:46 |
ignas | has some chance to just accelerate | 01:46 |
ignas | for example | 01:46 |
ignas | and what your learning algorithm is "modifying" | 01:46 |
ignas | is the probabilities that the car will do something in some situation | 01:46 |
fsufitch | yeah | 01:47 |
ignas | so after some time your car is doing the right thing 99% of the time | 01:47 |
ignas | and experimenting the 1% of the time | 01:47 |
ignas | just to see if there is a better way | 01:47 |
fsufitch | yeah, that's what i was thinking | 01:47 |
fsufitch | my 10th grade project touched on that | 01:47 |
ignas | it's called "reinforcement learning" | 01:47 |
fsufitch | did you know that some air conditioning systems use that stuff? | 01:47 |
fsufitch | so that you don't go from hot->cold->hot->cold... | 01:48 |
fsufitch | ah | 01:48 |
fsufitch | so then i missed other types of learning | 01:48 |
ignas | the learning algorithm | 01:48 |
ignas | expert systems by the way can be set up manually, your teapot that turns off automatically is a "fuzzy expert system" ;) | 01:49 |
ignas | with 1 input - the temperature | 01:49 |
ignas | and 1 output "turning off" | 01:49 |
fsufitch | heh | 01:50 |
ignas | fsufitch: http://www.cs.ualberta.ca/~sutton/book/ebook/the-book.html | 02:06 |
fsufitch | ooooooooooooooh | 02:07 |
fsufitch | interesting! | 02:07 |
* fsufitch bookmarks that | 02:07 | |
*** Aiste has joined #schooltool | 02:14 | |
*** ignas has quit IRC | 02:51 | |
*** Aiste has quit IRC | 04:06 | |
*** Ninno has joined #schooltool | 10:43 | |
*** Aiste has joined #schooltool | 11:36 | |
*** didymo has quit IRC | 12:37 | |
*** th1a_ has quit IRC | 14:36 | |
*** th1a has quit IRC | 14:36 | |
*** jfroche has quit IRC | 14:36 | |
*** fsufitch has quit IRC | 14:36 | |
*** Lumiere has quit IRC | 14:36 | |
*** SteveA has quit IRC | 14:36 | |
*** mheath has quit IRC | 14:36 | |
*** lisppaste5 has quit IRC | 14:36 | |
*** Fujitsu has quit IRC | 14:36 | |
*** JohnnyST has quit IRC | 14:36 | |
*** SteveA has joined #schooltool | 14:38 | |
*** Lumiere has joined #schooltool | 14:38 | |
*** fsufitch has joined #schooltool | 14:38 | |
*** th1a has joined #schooltool | 14:38 | |
*** th1a_ has joined #schooltool | 14:38 | |
*** lisppaste5 has joined #schooltool | 14:38 | |
*** Fujitsu has joined #schooltool | 14:38 | |
*** jfroche has joined #schooltool | 14:38 | |
*** mheath has joined #schooltool | 14:38 | |
*** JohnnyST has joined #schooltool | 14:38 | |
*** fsufitch has quit IRC | 14:39 | |
*** Lumiere has quit IRC | 15:21 | |
*** SteveA has quit IRC | 15:21 | |
*** th1a_ has quit IRC | 15:21 | |
*** th1a has quit IRC | 15:21 | |
*** jfroche has quit IRC | 15:22 | |
*** mheath has quit IRC | 15:22 | |
*** Fujitsu has quit IRC | 15:22 | |
*** JohnnyST has quit IRC | 15:22 | |
*** lisppaste5 has quit IRC | 15:22 | |
*** SteveA has joined #schooltool | 15:24 | |
*** Lumiere has joined #schooltool | 15:24 | |
*** th1a has joined #schooltool | 15:24 | |
*** th1a_ has joined #schooltool | 15:24 | |
*** lisppaste5 has joined #schooltool | 15:24 | |
*** Fujitsu has joined #schooltool | 15:24 | |
*** jfroche has joined #schooltool | 15:24 | |
*** mheath has joined #schooltool | 15:24 | |
*** JohnnyST has joined #schooltool | 15:24 | |
*** mheath has quit IRC | 15:24 | |
*** mheath has joined #schooltool | 15:24 | |
*** lisppaste5 has quit IRC | 17:50 | |
*** lisppaste5 has joined #schooltool | 17:58 | |
*** fsufitch has joined #schooltool | 19:03 | |
*** didymo has joined #schooltool | 22:52 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!