| ojwb | are there any public stats for comparative usage for the various ubuntu releases? | 01:00 |
|---|---|---|
| mwhudson | i doubt it | 01:02 |
| mwhudson | there aren't any public stats for the usage of ubuntu at all, afaik and the private ones are pretty much guesses i think :) | 01:02 |
| ojwb | just package downloads normalised would be handy | 01:04 |
| ojwb | I'm curious is there's much point backporting packages to hardy now, for example | 01:04 |
| ajmitch | ojwb: it's a bit hard to get even download stats, given the number of mirrors around | 01:07 |
| ajmitch | I don't know if the 'official' mirrors supply stats back to canonical or not & even then you can have further caching | 01:07 |
| ajmitch | but yeah, I doubt there's a lot of point backporting to hardy | 01:08 |
| ojwb | yeah, though for my purposes "quite a few people" vs "practically nobody" is useful | 01:08 |
| ojwb | i gave up on dapper some time ago | 01:08 |
| ajmitch | hardy should be EOL on desktop support soon | 01:09 |
| ojwb | yes, though this is xapian which is widely used on server | 01:09 |
| ojwb | +s | 01:09 |
| ajmitch | how hard is it to backport for hardy as well? | 01:09 |
| ojwb | it already supports it, but there are a few special cases which are for hardy only now | 02:20 |
| * ojwb wonders if there are PPA stats | 02:20 | |
| ojwb | I can't see an obvious link in the interface | 02:20 |
| ojwb | be a nice guide for cases like this | 02:20 |
| ajmitch | http://blog.launchpad.net/cool-new-stuff/tracking-ppa-download-statistics | 02:21 |
| ojwb | that was quick | 02:21 |
| ojwb | hmm the linked to blog is frustrating | 02:23 |
| ojwb | tells you how to manipulate a ppa object | 02:23 |
| ojwb | but not how to create it | 02:23 |
| ojwb | which seems to be the hard part... | 02:23 |
| * ojwb bookmarks it anywya | 02:25 | |
| ajmitch | I've managed to get a ppa record & the build records by a bit of hacking around | 02:35 |
| ojwb | yeah, don't really have time for that this week - off to the UK tomorrow | 02:35 |
| ajmitch | e.g. lp.people['ajmitch'].getPPAByName(name='ppa') gets the first ppa I have (and probably the only one) | 02:36 |
| ajmitch | hopefully it's a nice, uneventful flight :) | 02:37 |
| * ojwb wonders what an lp is | 02:37 | |
| ojwb | hopefully | 02:37 |
| ajmitch | from launchpadlib.launchpad import Launchpad | 02:37 |
| ajmitch | lp = Launchpad.login_anonymously('staging') | 02:38 |
| ajmitch | then with the ppa, I cheated & used getPublishedBinaries()[0].getDownloadCount() to get something to show :) | 02:40 |
| ajmitch | I think there's a better way than looking at wach published binary, but I haven't used this | 02:40 |
| ojwb | aha | 02:41 |
| ojwb | from launchpadlib.launchpad import Launchpad - needs more mentions of launchpad I feel | 02:41 |
| ajmitch | probably | 02:41 |
| ojwb | hmm, 0 | 02:41 |
| ojwb | nobody loves your first binary | 02:42 |
| ajmitch | it's old, probably predates the stats | 02:42 |
| ajmitch | & I was using it only for build testing in the PPA | 02:42 |
| ojwb | ah | 02:44 |
| ojwb | 25! | 02:44 |
| ajmitch | you're getting somewhere with it? | 02:45 |
| ojwb | ish | 02:45 |
| ajmitch | looks like you can get stats per day | 02:46 |
| ojwb | for b in lp.people['xapian-backports'].getPPAByName(name='ppa').getPublishedBinaries(): print "%d\t%s %s" % (b.getDownloadCount(), b.binary_package_name, b.binary_package_version) | 02:55 |
| ojwb | doesn't work for the xapian-1.2 ppa for some reason | 02:57 |
| ajmitch | ojwb: what do you get? | 03:08 |
| ojwb | oh, it works sometimes | 03:09 |
| ojwb | other times: TypeError: 'int' object is unsubscriptable | 03:09 |
| ojwb | i guess that's flakey server or something | 03:09 |
| ojwb | me speak english good | 03:09 |
| ojwb | seems to work the second time in quick succession pretty reliably | 03:10 |
| ojwb | I guess the first may pull data into cache | 03:10 |
| ojwb | looks like hardy is about 10% of lucid | 03:13 |
| ojwb | where lucid is 100 ish | 03:13 |
| ojwb | varies a lot by the subpackage though | 03:13 |
| ajmitch | might be useful to graph it & see how many of the hardy downloads are recent comared to lucid | 03:14 |
| ojwb | this was for the 1.2.4 release, which was december I think | 03:15 |
| * mwhudson_ gets 403s from nz.archive.ubuntu.com ... whut | 12:32 | |
| ibeardslee | morning | 20:27 |
| ojwb | morning | 20:49 |
| hads | Mornin | 21:04 |
| ajmitch | morning | 21:38 |
| Atamira | mornin | 21:41 |
| ibeardslee | sigh | 21:43 |
| Atamira | whats the matter ibeardslee ? | 21:43 |
| ibeardslee | politicians | 21:43 |
| Atamira | heh. they're all terrible | 21:44 |
| Atamira | according to forbes, we're the 8th in the world for a corrupt govt...with iceland and finland | 21:44 |
| ibeardslee | the problems is they can make choices about what the right thing to do is | 21:45 |
| timClicks | i'm pretty confused that things have come to this | 21:48 |
| timClicks | especially b/c officials and politicians have been told of the negative consequences | 21:48 |
| Atamira | oops. cant find the article..i might be wrong on that | 21:48 |
| timClicks | Atamira, I think you'll find we're the least corrupt country | 21:49 |
| ibeardslee | it seems (without getting as crass as I really could) that they are starting to line us up to bend over for the TPPA | 21:49 |
| timClicks | from Transparency International | 21:49 |
| Atamira | we're mild compared to some of the others in the world thats true | 21:49 |
| Atamira | we were debating who was the worse | 21:49 |
| Atamira | the sth africans at work reckon africa is the most corrupt | 21:50 |
| timClicks | I know that there people within MED & MFAT that are concerned | 21:50 |
| Atamira | technically its somalia..which i guess is close | 21:50 |
| timClicks | but, Wellington's very smal | 21:50 |
| timClicks | and leaks can be career destroying | 21:50 |
| ibeardslee | not soverignty destroying? | 21:51 |
| timClicks | this is a trend that has been going on for a long time | 21:53 |
| timClicks | many trade agreements are taking place internationally | 21:53 |
| timClicks | actually, WIPO (World IP Organisation) agreements didn't really suit the USA | 21:54 |
| timClicks | so they created TRIPS | 21:54 |
| timClicks | I don | 21:54 |
| ojwb | no, you tim | 21:54 |
| timClicks | I don't have too much time this morning to go over the whole thing | 21:55 |
| timClicks | But, in terms of sovereignty.. I wouldn't fight the battle on that front | 21:55 |
| timClicks | Politicians are advised by economists | 21:55 |
| timClicks | who are advised by economic models | 21:55 |
| timClicks | that give way to empirical data | 21:56 |
| timClicks | Therefore, if the right data can get to the policy advisers, then the legislation will change | 21:56 |
| timClicks | There's increasing evidence that strict IP laws protect the businesses that know how and are able to utilise those monopolies | 21:58 |
| timClicks | But, they tend to make a country worse off | 21:59 |
| timClicks | For example, take copyright | 21:59 |
| timClicks | the argument goes that we need a 100+ year monopoly in order to incentivise creative works | 22:00 |
| timClicks | however, I think that confuses the incentive to create a work from the incentive to distribute it | 22:00 |
| timClicks | distribution costs are extremely low for digital content | 22:01 |
| timClicks | therefore, they're mostly irrelevant | 22:01 |
| timClicks | which means, we should only provide the term of the copyright that would incentivise the *creation* of the work | 22:02 |
| timClicks | that's the hard bit - getting a new novel written | 22:02 |
| timClicks | once it's written, it's easy to reproduce | 22:02 |
| timClicks | we need to create a system that does enough to get creators rewarded for the first bit. Once that's done, what matters is that consumers can access a work at the lowest cost possible (which increases consumer surplus, e.g. economic win) | 22:03 |
| ojwb | and there's a lot of work which is based on out of copyright stuff, so out of copyright likely has economic benefits | 22:09 |
| ojwb | not for the original author, but for the economy, probably | 22:09 |
| ojwb | endless sherlock holmes inspired stuff, for example | 22:10 |
| snail | morning all | 23:08 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!