| --- Log opened Sun Oct 07 00:00:15 2012 |
| 00:11 | <&McMartin> | One of SVN 1.7's features is to move all .svn files into the root checkout directory and nowhere else. |
| 00:11 | <&McMartin> | However, many Linux distros stallout out their SVN versions before this, in part because the git community takes continued development of SVN as an actual personal insult |
| 00:15 | <&McMartin> | (SVN 1.6 fixed branches, more or less, so that merges and such actually worked the way one would expect) |
| 00:24 | < ToxicFrog> | Sorry, I should have contexted |
| 00:24 | < ToxicFrog> | <McMartin> DVCS justifications all eschew the notion of some kind of master repository, becuase that's all client-server and not DISTRIBUTED!!!! and thus it is useless |
| 00:24 | < ToxicFrog> | <McMartin> So their intended purpose is for people to all be nominally working on the same project, but nobody's sets of historical commits is actually a subset of anyone else's. |
| 00:25 | < ToxicFrog> | The closest I've seen to this is "I find eschewing a central server and just having everyone pull from each other works better, but if you want a central server instead that's ok I guess" |
| 00:26 | <&McMartin> | The official git instruction manual has everyone pulling from each other but not as a clique |
| 00:27 | <&McMartin> | THis is why the first six times I tried to use git (a) I couldn't get it working since I only had one user and (b) I kept wondering aloud why anyone would ever want this. |
| 00:28 | <&McMartin> | Also, the git manuals I was using treated 'push' as some weird advanced option that you'd probably never want to use |
| 00:28 | < ToxicFrog> | I don't think I've ever actually read the official git instruction manual (as opposed to the tutorials and 'git help ____') |
| 00:28 | <&McMartin> | Yeah, it turns out that's for the best. |
| 00:28 | <&McMartin> | git was explicitly designed by people who hate svn and everything it stands for, and this includes the notion of a master repository anywhere |
| 00:28 | < ToxicFrog> | Also, I'm not sure how (a) works. Isn't using git with only one user and no central server just 'git init' and then you're done? |
| 00:28 | <&McMartin> | Or something that could be used as one |
| 00:29 | <&McMartin> | well, that wasn't actually what I wanted, though. |
| 00:29 | <&McMartin> | Also, the official docs don't contemplate *that* either! |
| 00:29 | < ToxicFrog> | o.O |
| 00:29 | <&McMartin> | To get what I want, my workflow is: |
| 00:30 | <&McMartin> | Like, th eonly thing the official docs at the time contemplated was git's initial use case, with the names changed |
| 00:30 | <&McMartin> | And that is the total chaos that is the Linux kernel development model in its umpteen variations. |
| 00:30 | <&McMartin> | Anyway, the McMartin workflow is as follows |
| 00:30 | <&McMartin> | (a) git init |
| 00:31 | <&McMartin> | (b) Do enough work to get a first commit in |
| 00:31 | <&McMartin> | (c) commit it |
| 00:31 | <&McMartin> | (d) scp it over to Iodine, since it's the only machine with an SSH server (or have been on Iodine in the first place) |
| 00:31 | <&McMartin> | (d) do a git clone --bare into my repo collection directory |
| 00:31 | <&McMartin> | (e) delete my stuff from a-d |
| 00:32 | <&McMartin> | (f) CLone off the repo collection with git+ssh to get master/origin set right. |
| 00:32 | <&McMartin> | (g) Comence actual work |
| 00:33 | <&McMartin> | Learning that b and c were necessary required digging into the internals in the git book, IIRC. |
| 00:33 | <&McMartin> | Because all empty repos are indentical and you can't push to one. |
| 00:35 | <&McMartin> | Anyway |
| 00:36 | <&McMartin> | Going back to git-scm.com, I see they've actually completely rewritten the docs, and now the assumption is that you're using git because you want to use github |
| 00:36 | <&McMartin> | THis is noticably more reasonable. |
| 00:38 | < ToxicFrog> | Er |
| 00:39 | < ToxicFrog> | You can most certainly push to an empty repo |
| 00:39 | <&McMartin> | I have never in my life gotten this to work. |
| 00:39 | <&McMartin> | "No parent commit, aborting" |
| 00:39 | <&McMartin> | Also, attempting to clone it gives a warning. |
| 00:39 | < ToxicFrog> | And the "canonical" way to do this for stuff like gitolite is to either create a blank repo on the server, clone it to the client, and work from that; or replace d-f with "create blank repo on the server, git remote add origin foo, git push master' |
| 00:40 | < ToxicFrog> | Attempting to clone it emits a warning but 'git push' from the clone works just fine. |
| 00:40 | <&McMartin> | I have never had it work |
| 00:40 | < ToxicFrog> | Attempting to push to a freshly-created bare repo also works fine, although you need to specify the branch name the first time |
| 00:40 | <&McMartin> | Perhaps it is because the remotes were set wrong or |
| 00:40 | < ToxicFrog> | (otherwise it will say "no refs in common or none specified; try specifying a branch to push") |
| 00:40 | <&McMartin> | Ah, or that. |
| 00:41 | <&McMartin> | Yeah, and then I'd specify master and it would give the "no parent commit, aborting" or whatnot |
| 00:41 | < ToxicFrog> | (since the default is "push all branches in common and nothing else") |
| 00:41 | < ToxicFrog> | Yeah, that works fine for me and always has |
| 00:41 | < ToxicFrog> | ben@thoth:/tmp/b$ git push c |
| 00:41 | < ToxicFrog> | No refs in common and none specified; doing nothing. |
| 00:41 | < ToxicFrog> | ben@thoth:/tmp/b$ git push c master |
| 00:41 | < ToxicFrog> | * [new branch] master -> master |
| 00:42 | <&McMartin> | What is that "c"? |
| 00:42 | <&McMartin> | When I say "git push" doesn't work I mean "git push" the command does not work |
| 00:42 | <&McMartin> | As in, just those two words. |
| 00:42 | <&McMartin> | Which *does* work in my workflow. |
| 00:42 | < ToxicFrog> | c is the result of: |
| 00:42 | < ToxicFrog> | mkdir /tmp/c |
| 00:42 | < ToxicFrog> | (cd /tmp/c && git init --bare) |
| 00:43 | < ToxicFrog> | git remote add c /tmp/c |
| 00:43 | <&McMartin> | OK, so, at some point I'm getting "origin" set as my default in this workflow. |
| 00:44 | <&McMartin> | So that I never have to type it except on Windows |
| 00:44 | <&McMartin> | ANd when in windows in the bare case, it would suggest "git push origin master" and then fail there too. |
| 00:44 | < ToxicFrog> | Pretty sure the default for push is just the first [remote] entry in .git/config |
| 00:45 | < ToxicFrog> | If you create a repo with 'git clone' (even if it's a clone of an empty repo), this is automatically set up to be the repo you cloned from and is named "origin" |
| 00:45 | < ToxicFrog> | Oh wait, no, sorry |
| 00:45 | < ToxicFrog> | I am wrong |
| 00:45 | < ToxicFrog> | It's determined by the remote= entry for the branch you're on |
| 00:47 | < ToxicFrog> | (also, looking at gitolite again - the procedure for adding a repo to it is: add entry for that repo to the config file, with whatever access controls you want; commit and push the config file; now there is an empty bare repo on the server for you to clone) |
| 00:47 | <&McMartin> | Yeah, and on github a "new" repo has a sample README in it. |
| 00:47 | <&McMartin> | So you clone that and go to town |
| 00:48 | < ToxicFrog> | Pretty sure on gitolite a new repo has nothing - no commits, no files |
| 00:48 | < ToxicFrog> | It's been a while since I actually used it because I keep everything on github now, though~ |
| 00:48 | <&McMartin> | Yes, it's pretty clear that my issue was that certain config values within .git/ were wrong in my workflow |
| 00:49 | <&McMartin> | But getting something that gave me everything that running svn+ssh on Iodine did was a ridiculous amount of experimentation an headbutting down brick walls |
| 00:49 | | * simon_ is happy he isn't the only one hoping to deal as little as possible with the remote repository settings. |
| 00:49 | <&Derakon> | See, my introduction to DVCS was creating a Bitbucket account. |
| 00:49 | <&Derakon> | So I created the account, cloned it to my local machine, and was off and running. |
| 00:50 | <&McMartin> | Yes, but that's not what I wanted it for |
| 00:50 | <&McMartin> | I didn't want to *publish* it |
| 00:50 | <&Derakon> | ...I tell a lie, my introduction was Jetblade, which I started locally and then pushed to Google Code later. |
| 00:50 | < simon_> | usually, to avoid having to add remote origins manually, I create new repos on the far end and clone them in order to inherit the settings. |
| 00:50 | <&McMartin> | I wanted a backup with revision information that I could use to keep development up to date on multiple target platforms. |
| 00:50 | < ToxicFrog> | Yeah, my introduction was just "create repo on server, work on server, eventually make a 'master' bare repo and clone that to the laptop for work there" |
| 00:51 | < ToxicFrog> | Then when I started doing more interesting things I installed gitolite (I think it was gitorious then?) and used that. |
| 00:51 | < simon_> | when I sometimes create a local repo and I eventually want it on github, I have to spend time lookup up the docs. |
| 00:52 | < simon_> | I'm using the GitHub free-for-students private repo option. I'd use BitBucket instead, but it would invalidate the Octopussy sticker on the lid of my laptop. |
| 00:54 | < gnolam> | ? |
| 00:56 | < simon_> | as in, I have no real preference for either of them, so I'm just using GitHub because all my stuff is already there. :) |
| 00:59 | < gnolam> | "it would invalidate the Octopussy sticker on the lid of my laptop" |
| 01:00 | < rms> | <McMartin> (f) CLone off the repo collection with git+ssh to get master/origin set right. # my work flow: git init; ssh git@d1 mkrepo X; git add .; git commit -m 'init'; git remote add origin git@d1:X; git push origin master; ### now I can just git add->commit->pull->push as needed |
| 01:00 | < rms> | (mkrepo is a script that mkdir's, cd's then git init --bare's |
| 01:01 | < rms> | <McMartin> "No parent commit, aborting" <-- you have to add a branch to commit to. |
| 01:04 | | You're now known as TheWatcher[T-2] |
| 01:06 | < simon_> | gnolam, I would not upset any deity that has both the temper of a feline and the affluence of an octopus. |
| 01:07 | <&McMartin> | rms: Yes, as noted, I had to work out my workflow by guess and check against documentation that actively hated everything about the workflow I wanted to make |
| 01:08 | < gnolam> | I... still don't get it. |
| 01:08 | < simon_> | "The work of G?del showed that the notion of computation is essentially unique." -- what does unique mean, here? I would understand if it said universal, in that there is only one kind of computation, but is this what unique also refers to? |
| 01:08 | < rms> | Odd |
| 01:08 | <&McMartin> | It should not be tremendously surprising that the result is like the mess you get when grandma writes you saying "never mind, I solved my email problem" |
| 01:08 | < simon_> | gnolam, I'm just being silly. |
| 01:08 | <&McMartin> | That said, the list of commands you give is not exactly intuitive either =P |
| 01:09 | < rms> | No, but it's mostly the same stuff github walks you through |
| 01:10 | <&McMartin> | Which I wasn't trying to use! |
| 01:10 | <&McMartin> | I just wanted a local backup! |
| 01:10 | <&McMartin> | The docs did not contemplate this |
| 01:10 | | You're now known as TheWatcher[zZzZ] |
| 01:10 | <&McMartin> | Also, ISTR the github docs were "push this button, win" |
| 01:10 | <&McMartin> | WHich they were when I was using Github, which was not the problem spec |
| 01:11 | <&McMartin> | A 3/4 complete draft of The Git Book was pretty much what got me a working workflow, in the end. |
| 01:13 | < rms> | Oh, your complaint is the initial setup stuff, yeah I went through the same shit, finally gave up, got github account then a few months later figgured how to do it on my own. |
| 01:17 | < ToxicFrog> | For future reference, the minimal easy setup without github or gitolite or anything, as best I can tell, is: |
| 01:17 | < ToxicFrog> | $ ssh server GIT_DIR=/path/to/foo.git git init --bare |
| 01:18 | < ToxicFrog> | $ git clone server:/path/to/foo.git |
| 01:18 | < ToxicFrog> | $ cd foo; hack hack hack; git add; git commit; git push origin master |
| 01:37 | <&McMartin> | And speaking of, I think I've got my terrain generator extracted from Sable |
| 01:37 | <&McMartin> | It's so bad ;_; |
| 01:45 | | * gnolam hands McMartin a Power Glove. |
| 02:08 | | * Derakon mutters at Python's cProfile module. |
| 02:08 | <&Derakon> | The only way to enter it is by handing it a string to exec(). |
| 02:09 | <&Derakon> | Which makes it impossible to use if you have any context you need to provide. |
| 02:10 | <&Derakon> | E.g. if I have a large GUI program, then traditional profiling is useless because most of the program is spent in an event loop...but I want to profile a single function makeDungeonLevel(self, depth, width, height). |
| 02:10 | <&Derakon> | Of course these are not available to exec(). |
| 02:14 | <&Derakon> | ...ah, an undocumented function allows you to supply context! runctx() |
| 02:18 | | iospace is now known as boundaryspace |
| 03:17 | | Attilla [Obsolete@Nightstar-7d877edc.as43234.net] has quit [Ping timeout: 121 seconds] |
| 05:39 | <&Derakon> | So, problem I have with Pyrel right now. |
| 05:39 | <&Derakon> | Each Thing is Pyrel has a stats database -- Things include creatures, items, terrain, etc. |
| 05:39 | <&Derakon> | Obviously different kinds of Things will store different information in their stats. |
| 05:39 | <&Derakon> | Now, the stats on a creature include stuff like strength, hitpoints, speed, stealth, etc. |
| 05:40 | <&Derakon> | And items can include those same stats, which is meant to indicate that they provide a bonus to that stat when they are equipped. |
| 05:40 | <&Derakon> | I'm working on trying to figure out how to properly transfer stats from an item to a creature when the item is equipped. |
| 05:40 | <&Derakon> | The stats "database" is a mapping of strings to lists of stat modifiers. |
| 05:40 | <&McMartin> | Each stat on a creature should have a base/current. |
| 05:40 | | boundaryspace is now known as iospace |
| 05:41 | <&McMartin> | A la Might And Magic. |
| 05:41 | <&Derakon> | Each stat modifier has a tier; the tiers are Inherent > Permanent > Temporary > Derived. |
| 05:41 | <&McMartin> | I'd probably procedurally work out all but "Inherent" based on inventory/currently active status effects |
| 05:41 | <&McMartin> | And do so each time to make sure I never missed a spot |
| 05:42 | <&Derakon> | The idea here is that Inherent represents fundamental stats; Permanent represents stat mods of indefinite duration; Temporary represents temporary buffs; Derived represents modifiers that come from other stats. |
| 05:42 | <~Vornicus> | WHat's in each of those? Where would a Helm Of Intellect go? A Tome Of Learning? |
| 05:42 | <&Derakon> | McM: give me a bit... |
| 05:42 | <&Derakon> | The idea with the tiers is to handle the issue where a given stat modifier depends on some other stat. |
| 05:42 | <&Derakon> | E.g. if a modifier boosts your STR based on your magic device skill. |
| 05:43 | <&McMartin> | That sounds like it has the potential to loop |
| 05:43 | <&Derakon> | Each modifier is only allowed to refer to modifiers of a strictly lesser tier, to prevent circular dependencies. |
| 05:43 | <&Derakon> | Hence the tier system. |
| 05:44 | <~Vornicus> | It was actually pretty easy in Diablo to have a piece of equipment that you were wearing but if you took it off you could no longer put it on. |
| 05:46 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Operation timed out] |
| 05:46 | <~Vornicus> | herp |
| 05:47 | <&McMartin> | Something about this design still feels suspiciously off to me |
| 05:47 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 05:47 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 05:47 | <&McMartin> | Something about this design still feels suspiciously off to me |
| 05:47 | <&Derakon> | Dammit, what was my last line? |
| 05:48 | <&McMartin> | "hence the tier system" |
| 05:48 | <&Derakon> | Uh. |
| 05:48 | <&Derakon> | That was awhile ago. |
| 05:48 | <&Derakon> | Pastebin time. |
| 05:48 | <&McMartin> | I'm not at full mental power right now |
| 05:48 | <&Derakon> | http://pastebin.com/96CUSri6 |
| 05:48 | <&McMartin> | But Temporary and Derived don't feel to me like they can sensibly coexist in a tier system. They're going to intrinsically want to get in each other's way. |
| 05:49 | <&Derakon> | Can you think of a specific example? I'm not seeing a conflict. |
| 05:49 | <&Derakon> | Though note that Derived tier is intended for modifiers that are based on other stats. |
| 05:49 | <&McMartin> | Right |
| 05:49 | <&McMartin> | So, say I've got some device that boosts Intellect based on my Use Widget Skill |
| 05:50 | <&Derakon> | Right. |
| 05:50 | <&McMartin> | Which is, itself, based on DEX. |
| 05:50 | <&McMartin> | I might want more intellect, and want to get it via a DEX boost. |
| 05:50 | <&McMartin> | That seems to bounce both directions in the tier; either my DEX boost can't hit Use Widget, or Intellect can't notice Use Widget |
| 05:50 | <&McMartin> | Er, both directions across tiers |
| 05:51 | <&Derakon> | Hmm. |
| 05:52 | <&Derakon> | So the item in question has a Permanent-tier modifier that says "give +INT based on Widget skill". |
| 05:52 | <&Derakon> | And Widget skill has a Derived-tier modifier that says "Give +skill based on DEX". |
| 05:52 | <&Derakon> | Yeah, that wouldn't work entirely as designed. |
| 05:52 | <&McMartin> | Now I wander into the Haste Device, which would normally be a Temporary-tier boost to DEX for 12 turns. |
| 05:53 | <&McMartin> | I should see that reflected in my Intellect, too. |
| 05:53 | <&Derakon> | (But you could declare that Widget skill's inherent nature is DEX-based) |
| 05:53 | <&Derakon> | You can't have everything affect everything, though. |
| 05:53 | <&Derakon> | That way madness and circular dependencies lies. |
| 05:53 | <&McMartin> | I suppose that's true. |
| 05:53 | <&McMartin> | Maybe my objection is to the naming of the tiers~ |
| 05:53 | <&Derakon> | Given the fundamental proposition that you can have stat mods based on other stats, there must be a hierarchy of some kind. |
| 05:54 | <&McMartin> | Right |
| 05:54 | <&McMartin> | Hm |
| 05:54 | <&McMartin> | Yeah, I think the usual way I see this is "all modifiers that are a fraction of a stat are a fraction of *the base* stat" |
| 05:54 | <&McMartin> | Unless you're playing Morrowind, in which case, MAX_INT in all stats at level 1 thanks to potion abuse ahoy! |
| 05:54 | <&Derakon> | So you have Inherent tier and Everything Else tier. |
| 05:54 | < ToxicFrog> | The approach most games take to this is "no circular dependencies of any kind", usually by dividing up the stats, not the bonuses |
| 05:55 | <&McMartin> | True, actually. |
| 05:55 | <&Derakon> | TF: howso? |
| 05:55 | < ToxicFrog> | So you have your core stats (str dex con pow) and your derived stats (hp mp acc atk resist speed) |
| 05:55 | <&McMartin> | Well, take GURPS. Or SPECIAL. |
| 05:55 | <&McMartin> | Skills are effectively derived stats, but they can't affect your core stats. |
| 05:55 | < ToxicFrog> | Increases to core stats are reflected in derived stats; the reverse is not true. |
| 05:55 | <&McMartin> | Advantages/Feats/Perks can affect anything, but you can take any given perk and put it into the hierarchy. |
| 05:56 | <&Derakon> | Mm... |
| 05:56 | < ToxicFrog> | Bonuses can be "add 50% of your POW to your ATK" but you'll never see the reverse, or "add 50% of your POW to your STR" |
| 05:56 | <&Derakon> | If possible I'd really rather not set up an encoded hierarchy of stats in the game. |
| 05:56 | <~Vornicus> | Most of the systems I know don't get a situation where yoou get the add 50% of your POW to your STR |
| 05:56 | <&Derakon> | In part because I'm trying to stay generic here. |
| 05:56 | <&Derakon> | (Recognizing that this makes life harder) |
| 05:56 | <&McMartin> | Hm |
| 05:56 | <~Vornicus> | I have seen "if you have x POW, add y to your STR" |
| 05:56 | < ToxicFrog> | Why is that worse than setting up a heirarchy of bonus types? |
| 05:57 | <&McMartin> | Well, do you want it to be elegant or just generic~ |
| 05:57 | <&Derakon> | TF: because it presupposes what the stats are. |
| 05:57 | <&Derakon> | And how they relate to each other. |
| 05:57 | <&McMartin> | Because one way of being generic is "apply each modifier exactly once, in an arbitrary but consistent order" |
| 05:57 | <&McMartin> | Say, order of donning them. |
| 05:57 | <&Derakon> | You can't come in and say "I'm going to create a new stat Gunsmithing and work it into the system cleanly". |
| 05:57 | <&McMartin> | s/donning them/application/ |
| 05:57 | <&Derakon> | (The Angband community is all about the game mods, frankly) |
| 05:58 | < ToxicFrog> | Not so; you can, for example, classify stats as primary/secondary/tertiary without specifying what those stats are, just as you're doing right now with bonuses. |
| 05:59 | <&Derakon> | Ehhh...I still don't like it. |
| 05:59 | <&Derakon> | In no small part because I want to be able to say "your magic device skill can affect your core stats" and things like that. |
| 05:59 | <&Derakon> | It seems like an interesting mechanic. |
| 05:59 | <&Derakon> | And I'd like to allow it and other such creative systems. |
| 06:01 | <&Derakon> | McM's example wherein DEX, INT, and Use Widget skill are intertwined can be mitigated to an extent, though it still won't be perfect. |
| 06:01 | <&Derakon> | 1) say that Use Widget skill is fundamentally based on DEX -- then its DEX-based mod is Inherent-tier (and doesn't get any external DEX bonuses, which is less ideal). Boosting your Inherent-tier DEX improves Use Widget improves INT. |
| 06:02 | <&Derakon> | 2) Base Use Widget skill primarily off of race/class/level, with a rather secondary DEX bonus; this is what Angband does. |
| 06:02 | <&Derakon> | Then boosting DEX has no effect on your INT, but being a different class or gaining levels would have a noticeable impact. |
| 06:02 | <~Vornicus> | this device is clearly a Curta |
| 06:03 | <&McMartin> | Fizba Wizh? |
| 06:03 | <&Derakon> | It sounds like the arguments are basically "you can have an elegant system without edge cases, or you can have an inelegant system that can do lots of stuff, but you can't have both." |
| 06:04 | <~Vornicus> | http://en.wikipedia.org/wiki/Curta_calculator a handheld mechanical calculator. |
| 06:04 | | * Derakon googles. "Interesting." |
| 06:05 | <&Derakon> | Heh, the backers tried to screw him out of the company, but neglected to obtain the patent rights. Whoops! |
| 06:23 | | * Derakon posts about the problem to the forums: http://angband.oook.cz/forum/showpost.php?p=73667&postcount=49 |
| 06:23 | <&Derakon> | And now shortly to go to bed. |
| 06:23 | <&Derakon> | (Feel free to respond here if you have any ideas; I'll check my logs in the morning) |
| 06:33 | | Derakon is now known as Derakon[AFK] |
| 07:00 | | Syloq_Home [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Connection closed] |
| 08:00 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [[NS] Quit: Well, most things get better when I kick them!] |
| 08:41 | | Vash [Vash@Nightstar-3ba4108e.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!] |
| 09:34 | <&jerith> | Comment of the day, from some code I'm debugging: |
| 09:34 | <&jerith> | api.sandbox_kill() # it's a harsh world |
| 09:50 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 10:03 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 10:04 | | mode/#code [+o himi] by ChanServ |
| 10:35 | | You're now known as TheWatcher |
| 12:31 | | Attilla [Obsolete@Nightstar-7d877edc.as43234.net] has joined #code |
| 12:55 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 13:08 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 13:08 | | mode/#code [+o himi] by ChanServ |
| 13:37 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 13:50 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 13:50 | | mode/#code [+o himi] by ChanServ |
| 14:48 | < AnnoDomini> | Much of the time, when interacting with various devices, I feel as though I don't have the actual correct skill for operating them, but rely on Use Electronic Device (Int) to Activate Blindly. |
| 15:08 | | Syloq_Home [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 15:46 | < gnolam> | AnnoDomini: :) |
| 17:02 | < AnnoDomini> | Hmm. Is it possible to run a bootable USB in a virtual machine (vbox, here) to get a functional BackTrack Linux with possibility of using monitor mode of a wifi adapter of the host system? |
| 17:08 | < rms> | Yes and maybe |
| 17:08 | < rms> | 1) you can do some complicated mojo to make vbox use a real disk instead of a file |
| 17:09 | < rms> | 2) Bridge the wifi adapter to a tap interface made by the VM (might not work) |
| 17:09 | < rms> | 2b) Instead, if you have a USB adapter, just give the VM direct access to the USB wifi adapter |
| 17:10 | < AnnoDomini> | I *do* have a USB adapter! |
| 17:10 | < Namegduf> | Okay, in that case, yes. |
| 17:10 | < Namegduf> | Boot the OS with the complex mojo for booting off a physical partition |
| 17:11 | < Namegduf> | Then use USB passthrough to pass the USB wifi adapter through |
| 17:11 | < Namegduf> | It's going to be a pain |
| 17:12 | < Namegduf> | And I've not done physical partition booting off a USB thing before. |
| 17:12 | < Namegduf> | But I assume it to be possible. |
| 17:12 | < Namegduf> | I'm using a VM booted off a regular internal drive's real partition right now. |
| 17:12 | < Namegduf> | It's part of my standard setup. |
| 17:13 | < Namegduf> | I've used USB passthrough for a bluetooth adapter before. |
| 17:13 | < AnnoDomini> | Better! I have an ISO file with BackTrack on file. |
| 17:14 | < Namegduf> | Okay, awesome. Just install it into a regular VM |
| 17:14 | < Namegduf> | Then USB passthrough |
| 17:14 | < Namegduf> | No weird fiddling with anything outside the standard UI necessary |
| 17:15 | < Namegduf> | USB passthrough can be done while running with Devices->USB Devices->(device to pass through) |
| 17:15 | < Namegduf> | But can be a bit... picky about when and what it will passthrough, because if the host is using it it gets unhappy |
| 17:16 | < Namegduf> | You can also set them up in the VM's settings, and that (plus maybe starting the VM with the device unplugged, then plugging it in) might work more reliably |
| 17:16 | < Namegduf> | That'd be my first troubleshooting step if it refused to passthrough the simple way. |
| 17:17 | | * AnnoDomini starts installing. |
| 17:28 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 17:33 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code |
| 17:41 | | * AnnoDomini hopes this stupid BT5R2 will install correctly on a 5GB vdrive, even if it nominally requires 11. |
| 17:42 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 17:42 | | mode/#code [+o himi] by ChanServ |
| 17:52 | < AnnoDomini> | Okay, no, screw it. Imma run this as a Live CD with a tiny virtual drive to keep notes on. |
| 17:56 | | * AnnoDomini goes download the upgade to vbox. |
| 17:57 | | iospace is now known as io\PACKERS |
| 18:00 | | * gnolam arghls. |
| 18:02 | | rms is now known as Vasi |
| 18:02 | < Vasi> | gnolam: ? |
| 18:02 | < gnolam> | Currently: people who don't document their function arguments. |
| 18:03 | < gnolam> | Also: people who pick some existing word for their project without creatively misspelling it. |
| 18:03 | | * gnolam glares at Bullet Physics. |
| 18:03 | < Vasi> | Heh |
| 18:03 | | * Vasi is so guilty of the first one |
| 18:12 | <@TheWatcher> | gnolam: damn right on the first one. |
| 18:12 | | * TheWatcher goes to Great Pains to avoid that |
| 18:14 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 18:17 | < gnolam> | The second is ok iff it's a) something that doesn't need documentation or b) a word that has absolutely nothing to do with the project in question. |
| 18:17 | < gnolam> | But "bullet physics"? Hooray, false positive search results. :P |
| 18:17 | < gnolam> | Still not as bad as fscking "Maxima", but still. |
| 18:27 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 18:27 | | mode/#code [+o himi] by ChanServ |
| 19:33 | | mode/#code [+ao ToxicFrog ToxicFrog] by ChanServ |
| 19:53 | | Derakon[AFK] is now known as Derakon |
| 20:05 | < Vasi> | <ender> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND |
| 20:05 | < Vasi> | <ender> 13212 portage 1 0 16128 5520 1420 D 9999 0.2 5124095h ld |
| 20:07 | <@Tamber> | o.0 |
| 20:21 | < froztbyte> | there's a few problems there |
| 20:23 | <&ToxicFrog> | |
| 20:40 | < AnnoDomini> | Namegduf: I'm getting "No USB Devices Connected". |
| 20:40 | < Namegduf> | AnnoDomini: Where? |
| 20:41 | < AnnoDomini> | Devices->USB Devices->No USB Devices Connected. |
| 20:41 | < AnnoDomini> | I'm also getting: http://i.imgur.com/sRhb3.png |
| 20:41 | < AnnoDomini> | When I go to settings. |
| 20:52 | < AnnoDomini> | Any ideas? |
| 21:00 | | * AnnoDomini follows some instructions from the internets, goes reboot. |
| 21:06 | < AnnoDomini> | That cleared it up. |
| 21:09 | < AnnoDomini> | It works! |
| 21:09 | < AnnoDomini> | Thank you for your help, Namegduf. |
| 21:10 | < Namegduf> | Sorry. |
| 21:10 | < Namegduf> | I was pulled away by something and then missed this channel lighting up. |
| 21:10 | < Namegduf> | I'm glad you were able to find a guide. |
| 21:12 | <&ToxicFrog> | death to openoffice |
| 21:12 | <&ToxicFrog> | death to powerpoint |
| 21:15 | < Namegduf> | Go Google Presentations |
| 21:16 | < Namegduf> | (Actually, they've got some bad niggles, too.) |
| 21:16 | < Namegduf> | OO Impress is awful, though |
| 21:23 | | Attilla_ [Obsolete@Nightstar-b60121d0.as43234.net] has joined #code |
| 21:25 | | Attilla [Obsolete@Nightstar-7d877edc.as43234.net] has quit [Ping timeout: 121 seconds] |
| 21:39 | <&ToxicFrog> | Namegduf: how well does Google Presentation handle poster-sized stuff, and importing from PPT/ODP |
| 21:39 | <&ToxicFrog> | ? |
| 21:42 | | io\PACKERS is now known as io\d4ea |
| 21:43 | | io\d4ea is now known as io\dis4ea |
| 21:46 | | Vasi is now known as rms |
| 21:48 | < Namegduf> | I don't know. It's not a drawing tool. |
| 21:52 | <&ToxicFrog> | Well, what I have here is a PowerPoint template, a set of dimensions, and a requirement to make a conference poster |
| 22:00 | < Namegduf> | Oh dear. |
| 22:01 | < Namegduf> | How complicated would the template be to replicate in an actual poster-making tool? |
| 22:08 | <&ToxicFrog> | Quite trivial if I could actually extract the images from it. |
| 22:08 | <&ToxicFrog> | Which I cannot seem to do. |
| 22:12 | < Namegduf> | Sounds like it's print screen time |
| 22:12 | <&ToxicFrog> | OH wait got it |
| 22:12 | <&ToxicFrog> | Export to pdf, run pdfimages on it |
| 22:13 | <&ToxicFrog> | Ok, what would you recommend as the least shitty poster-making tool available? |
| 22:13 | <&ToxicFrog> | Bearing in mind that this poster will be mostly text (informational or source code) with occasional diagrams |
| 22:15 | <&ToxicFrog> | oh wait |
| 22:15 | <&ToxicFrog> | there's a "poster" LaTeX documentclass |
| 22:16 | <~Vornicus> | Apparently all is solved |
| 22:17 | <&ToxicFrog> | ...there are in fact like six |
| 22:17 | < gnolam> | Wait, a conference poster in /PowerPoint/? o_O |
| 22:20 | <@TheWatcher> | At the least, I'd be using Publisher. |
| 22:20 | <@TheWatcher> | Powerpoint is just bwuh |
| 22:21 | < rms> | gnolam: if there's /anything/ I've ever learned from dealing with users is that they will almost always use the worst hammer for the screw. |
| 22:22 | <@Tamber> | Sounds like my Engineering teacher. |
| 22:22 | <@Tamber> | Everything was done in Powerpoint; because that's the hammer he knew how to attack screws with~ |
| 22:24 | <&ToxicFrog> | "Publisher" here is...? |
| 22:24 | < rms> | A program named Publisher |
| 22:25 | < rms> | <zb3> http://office.microsoft.com/en-us/publisher/ - Publisher 2010 - Professional newsletter, brochure, postcards ... |
| 22:26 | <&ToxicFrog> | Aah |
| 22:27 | <&ToxicFrog> | What is the closest Linux equivalent? |
| 22:28 | < rms> | Something that handles *TeX I would assume |
| 22:28 | < rms> | I don't think there's an OO mirror to Publisher |
| 22:29 | <@TheWatcher> | Scribus, I think. |
| 22:32 | <&ToxicFrog> | Hmm |
| 22:32 | <&ToxicFrog> | Ok, I think I'll try out both \documentclass{baposter} and scribus |
| 22:34 | <@himi> | As far as I know Scribus knows nothing about TeX |
| 22:34 | <&ToxicFrog> | Yeah, I don't expect it to |
| 22:35 | <@himi> | It'd be good for a poster from scratch - it's a full-on desktop publishing tool |
| 22:36 | <&ToxicFrog> | This isn't quite "from scratch", but the template should be fairly easy to reconstruct |
| 22:36 | <&ToxicFrog> | Images in upper left and right, title centered between them, horizontal divider, poster content, footer containing org/author/contact information white on blue. |
| 22:39 | <@himi> | Yeah, something that simple isn't an issue, though it'll probably take a bit of fiddling to get it to behave the way you expect |
| 22:40 | <@himi> | Scribus is nice, but the model it uses took me a few tries to get used to |
| 22:40 | <@himi> | Anyway, I'm off to work |
| 22:42 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Connection closed] |
| 22:55 | < gnolam> | Hmm. I wonder if the values from this gamepad really are that crap or if SDL is messing up somehow. |
| 23:01 | | Vash [Vash@Nightstar-3ba4108e.wlfrct.sbcglobal.net] has joined #code |
| 23:01 | | mode/#code [+o Vash] by ChanServ |
| 23:02 | < rms> | gnolam: on Linux? |
| 23:03 | < rms> | xxd /dev/input/js0 # something like this |
| 23:04 | < gnolam> | Windows. |
| 23:05 | < gnolam> | And it appears it's the gamepad. |
| 23:05 | | * ToxicFrog pokes Scribus with a stick |
| 23:05 | < gnolam> | The centering is... less than perfect. |
| 23:11 | < gnolam> | Still. I've created a simple RC helicopter simulator in ~3 hours without previous experience with the physics library. So this was just a minor and quite fixable setback in a very productive day. :) |
| 23:12 | <@TheWatcher> | \o/ |
| 23:20 | < gnolam> | (One of those cases when "how hard could that be to implement, really?" turned out to be an appropriate question to ask) |
| 23:33 | <@TheWatcher> | Oh dear. I have another biologist on my C course. |
| 23:34 | < gnolam> | Biologists can't code, I assume? |
| 23:40 | <~Vornicus> | Biologists I'm aware of use python and perl. Very badly. |
| 23:41 | <@TheWatcher> | Biologists are responsible for some of the most hideous Abominations I've ever seen |
| 23:43 | <@TheWatcher> | Hopefully by the time I'm done with this one, he'll be decent enough |
| --- Log closed Mon Oct 08 00:00:30 2012 |