code logs -> 2008 -> Sun, 09 Nov 2008< code.20081108.log - code.20081110.log >
--- Log opened Sun Nov 09 00:00:14 2008
00:13 You're now known as TheWatcher[T-2]
00:14 You're now known as TheWatcher[zZzZ]
00:26 AnnoDomini [~farkoff@Nightstar-29799.neoplus.adsl.tpnet.pl] has quit [Quit: I am Magellan-R-MGE. I see your group is missing a Hygiene officer.]
01:51
<@Derakon>
...right, I forgot about that whole "SSL requires a signing authority" thing.
01:51
<@Derakon>
Unless you want to fake it and have every browser scream that you aren't trusted, of course.
01:52
< Vornotron>
Eh.
01:52
< Vornotron>
SSL isn't that important, I mean
01:52
< Vornotron>
Unless you're selling stuff.
01:52
<@Derakon>
It is if I'm going to have "real" accounts for Race for the White House.
01:52
< Vornotron>
No it's not.
01:52
< Vornotron>
Der, how does nickserv get your password?
01:53
<@Derakon>
My impression was that nickserv was not remotely secure, which is why that particular password is not especially strong.
01:53
<@ToxicFrog>
I note that "well, this other system we use is insecure" is not an excuse for continuing to build insecure systems.
01:53
< Vornotron>
Well, no.
01:53
<@ToxicFrog>
And it's not.
01:53
<@ToxicFrog>
Hell, I can read the passwords in cleartext right out of the database.
01:53
<@Derakon>
I suppose I could encrypt things client-side manually.
01:53
<@ToxicFrog>
With strings.
01:53
<@Derakon>
Heh.
01:54
< Vornotron>
But I don't any reason to make it absurdly secure, I mean really, it's just a game.
01:54
<@Derakon>
What I really want is a nice Perl-based system that'll do all this for me automatically~
01:54
<@Derakon>
On the one hand, yes, Vorn, your point is valid. On the other hand, it's also an exercise.
01:54
<@ToxicFrog>
Personally, I'd like a "weak https" that does crypto but not authentication.
01:55
<@Derakon>
But for the moment, I think I'll just replace the account_id column in the Player table with a "player_name" column.
01:55
<@ToxicFrog>
IE, it doesn't guarantee the identity of either end (and doesn't require a signer), but it does guarantee that the traffic can't be sniffed.
01:55
<@Derakon>
Yeah.
02:00
<@gnolam>
Firefox's "HTTPS with a self-signed certificate is worse than unencrypted! It's the DEVIL!" stance is just one more reason why I have no faith in that particular project anymore.
02:01
<@ToxicFrog>
...
02:07
<@gnolam>
(Other reasons include the obnoxious so-called "awesome bar" and the general trend towards a) bloat and b) obstinate refusal to let you change settings)
02:08
<@Derakon>
I actually rather like the "awesome bar", though I do agree that they should have let you use the old behavior if you prefer it.
02:12
<@ToxicFrog>
"awesome bar"?
02:14
<@gnolam>
That's what they call breaking the address paradigm by having the location bar search through titles as well as URLs, in your entire bookmark list and page history.
02:15
<@gnolam>
And they purposely disabled the about:config entry that let you disable that horrible behavior.
02:16
<@Derakon>
Gnolam: note that this also lets you find URLs by whatever happens to be unique about them, which means a lot less typing to get to e.g. deeply-nested subdomains.
02:16
<@Derakon>
If I want to get to http://www.foo.com/bar/baz/quux and there's also bar/baz/flarghle and bar/baz/fnordlingrads, I can just type "foo quux" and get what I want.
02:17
<@gnolam>
And if you want to go to, say Google News when you also visit, say, Slashdot, you're fucked.
02:17
<@Derakon>
Er...
02:17
<@gnolam>
Because Slashdot has "news for nerds, stuff that matters" in its title.
02:17
<@Derakon>
Or you could type "news go".
02:18
<@Derakon>
I admit that the rollout was pretty badly handled, but the feature actually is useful.
02:18
<@gnolam>
=> more typing.
02:18
<@Derakon>
It just takes some getting used to.
02:18
<@gnolam>
No. It is an awful feature. The URL bar is for URLs.
02:18
<@Derakon>
It also tends to do a good job of sorting things.
02:19
<@Derakon>
The URL bar is for finding URLs for sites that you visit regularly.
02:19
<@gnolam>
If you want to search through page titles in your history, /search the fucking history/.
02:19
<@Derakon>
You shouldn't have to start typing at an arbitrary point (the beginning) just to get them.
02:23
<@gnolam>
Ctrl+H, type, enter.
02:23
<@ToxicFrog>
I note that Opera has had this feature for a long time, and it's sufficiently useful and unobtrusive that I didn't even notice until I checked for it just now.
02:24
<@ToxicFrog>
I'm not sure what your point is about, say, slashdot vs google.
02:24
<@ToxicFrog>
If you type in "news" without this feature, you get nothing, or news.com, or something.
02:24
<@Derakon>
He's saying that if he types "news", which is the beginning of news.google.com, Slashdot also shows, because it has "news" in its page title.
02:24
<@ToxicFrog>
If you type in "slashdot.org" or "news.google.com" with this feature, it behaves just like a normal URL bar.
02:25
<@gnolam>
If you type in the entire URL, yes.
02:25
<@gnolam>
Autocompletion: good. Mindless matching of *STRING* in both URLs and titles: bad.
02:25
<@McMartin>
If you're blindly typing "ne*down arrow*enter*" without seeing what you're selecting, you're probably going to end up getting a random individual article anyway.
02:26
<@ToxicFrog>
What McM said.
02:26
<@Derakon>
You're clearly convinced that you're right, so I don't think we're accomplishing anything here.
02:26
<@gnolam>
Having an about:config entry and then deliberately removing it: unforgivable.
02:26
<@McMartin>
It's not even autocompletion!
02:26
<@McMartin>
You have to deliberately select a menu item.
02:26
<@ToxicFrog>
Not being able to turn this off is stupid and wrong as a matter of principle, but none of the arguments you're making about the feature as a feature make sense.
02:26
<@gnolam>
McMartin: it isn't?
02:27
<@McMartin>
It's not.
02:27
<@McMartin>
Autocompletion, when wrong, requires you to delete things.
02:27
<@McMartin>
This is emphatically not the case here.
02:27
<@gnolam>
So the autocompletion feature in, say, OpenOffice isn't autocompletion?
02:27
<@McMartin>
I have to delete shit in OO.o all the fucking time.
02:28
<@gnolam>
I can't think of anything that autocompletes /and autoaccepts/ the completion.
02:28
<@McMartin>
Reversion of capital letters, mainly
02:28
<@ToxicFrog>
(incidentally, if you are using Opera, you're looking for User Prefs -> Address bar content search to turn it off~)
02:29
<@McMartin>
I admit you could if you wanted fold that under a different thing, like AutoFormat
02:29
<@McMartin>
However, those autocompleters offer one possibility, and hitting "enter" enforces it.
02:29
<@ToxicFrog>
I generally consider that to be autoreplace, not autocomplete.
02:29
<@ToxicFrog>
With autocomplete, it's "is this the rest of what you were typing?"
02:29
<@McMartin>
In FF, at least, hitting enter without selecting from a menu sends what you've typed as a request.
02:30
<@ToxicFrog>
With autoreplace, it's "oh, that was wrong, I'll just go back and fix it for you, no need to thank m- what are you doing with that chainsaw AUGH NO"
02:30
<@McMartin>
While typing, say, "news.google.com*enter*" does not send me to news.google.com/article/etc
02:30
<@McMartin>
Also, because space is autocomplete acceptance, which is extremely annoying when you *wanted* to say "Bob and Jane were wed last week", not "Bob and Jane were Wednesday last week"
02:31
<@ToxicFrog>
This is why, if leaving autocomplete on, I remap it to tab.
02:31
<@McMartin>
Which works great until you fire up the spreadsheet
02:31
<@ToxicFrog>
Which I have yet to do, so~
02:32
<@McMartin>
Anyway, neither of these things is a mutating combo box; it's closer to MSVC's Intellisense stuff than anything else, and that requires explicit enabling gestures that are Not Part Of Standard Interaction.
02:32
<@McMartin>
Ergo, I say it is not autocomplete, and cannot understand why additional options therein is a flaw of any kind.
02:32
<@gnolam>
What TF said. Autoreplacement != Autocompletion.
02:32
<@ToxicFrog>
What Opera and FF do is neither, though.
02:32
<@McMartin>
Other than "my hardwired reflexes for blind menu navigation no longer behave as I expect"
02:33
<@ToxicFrog>
You type stuff in, it doesn't mess with it but you can use up/down to switch between what you typed and guessed alternatives and completions.
02:36
<@McMartin>
I'm still a little unclear on the interaction pattern this breaks.
02:36
<@McMartin>
"blind menu navigation" is, after all, intentionally a strawman
02:36
<@gnolam>
Autocompletion of something you actually type VS autocompletion of what you type + possibly unrelated metadata.
02:37
<@ToxicFrog>
Except it is not autocompletion
02:37
<@McMartin>
And that's not an interaction pattern.
02:37
<@ToxicFrog>
If you type in "news" and press enter, that's what it tries to load; it doesn't randomly pick the first search result and go to that instead.
02:37 * gnolam headdesks.
02:37
<@ToxicFrog>
I can think of two legitimate complaints against this feature, but neither of them seem to be what you're railing against
02:38
<@ToxicFrog>
Well, no, three
02:38
<@ToxicFrog>
To wit:
02:38
<@McMartin>
I guess it might be embarrassing if your entire porn browsing history came up while typing in girlscouts.org or whatnot
02:38
<@McMartin>
(four!)
02:38
<@ToxicFrog>
(1) I can no longer type in "n e down enter" without seeing what I'm doing and end up at news.google.com (a behaviour that isn't guaranteed with the old behaviour, either)
02:39
<@ToxicFrog>
(2) searching my entire history is a noticeable performance hit for a feature I never use (actually true on one of my machines)
02:39
<@ToxicFrog>
(3) the search ordering is stupid, so that typing in "news" gets me CNN articles I viewed four years ago and "news.google.com" is result #53 in the list
02:39
<@Derakon>
(Which it mostly isn't)
02:40
<@ToxicFrog>
And (4) is as McM said.
02:40
<@McMartin>
(4) is an entirely different user error made more blatant~
02:40
<@gnolam>
And again, it doesn't have to autoaccept to qualify as autocompletion.
02:41
<@McMartin>
That doesn't mean it still isn't.
02:41
<@McMartin>
Unless every combobox widget everywhere is "autocompletion".
02:41
<@McMartin>
Or unless bookmarks themselves are.
02:41
<@ToxicFrog>
And you still haven't made any complaints apart from "it's different and I don't like it". Why don't you like it?
02:41
<@McMartin>
You still aren't clear as to what exactly is wrong with the behavior.
02:42
<@McMartin>
Yeah. There's a not-very-hidden assumption here that there is only one possible thing to put beneath an address bar while typing in it, and this is laughable on its face.
02:44
<@Derakon>
Oooh, fun bug. http://blogs.zdnet.com/Burnette/?p=680
02:44
<@McMartin>
"it's giving the wrong answer" implies a right answer, and that's at best debatable.
02:45
<@ToxicFrog>
Also raises the question of what you consider "wrong"; if "wrong" is "my favorite site starting with 'news' isn't the first result when I type that in", see "not guaranteed by a plain address bar either"
02:47
<@Derakon>
Incidentally, as a web developer, I dearly love being able to type "local <foo>" to go to any page on my local site without having to type out "localhost/bar/baz/quux/foo".
03:00 * Vornotron got pissed off at Safari when it updated
03:01
< Vornotron>
It now puts unspeakablevorn.livejournal.com at the top of the "uns" list, when I never visit it intentionally.
03:02
< Vornotron>
(I want unspeakablevorn.livejournal.com/friends)
03:02
<@Derakon>
In FF3 you could probably type "friend" and get that link on the top.
03:36 Attilla [~The.Attil@Nightstar-9469.cdif.cable.ntl.com] has quit [Quit: <Insert Humorous and/or serious exit message here>]
04:23 gnolam [lenin@79.136.60.ns-4387] has quit [Quit: Morning has broken]
04:30
<@Derakon>
Hrm...it is a bit irritating that I have to make 9 DB queries to get the current game state.
04:33
< Vornotron>
You need a lot of queries for things like that
04:33
<@Derakon>
Ahh, Perl.
04:33
<@Derakon>
$sql .= join(',', map("?", keys %$states));
04:34
<@Derakon>
And then map({$sth->bind_param($i++, $_)} keys %$states);
04:35
<@Derakon>
(I'm not certain that does what I want...I'm just writing the function ATM, and I'll debug it once it's done. But those two lines should make a string "?,?,?,?,?" etc. with one '?' for each state, and then bind the state IDs in for each '?' in the $sql string)
04:59 * Derakon writes a Perl->JSON serializer.
05:04
<@jerith>
Are you serialising perl code to JSON?
05:04
<@Derakon>
I'm serializing Perl data objects.
05:04
<@Derakon>
Perl code would be a much trickier prospect.
05:04
<@jerith>
Ah. That makes slightly more sense.
05:23
< Vornotron>
Der: you'd think that sort of thing would already be in cpan
05:24
<@Derakon>
Vorn: very likely, but it took less time to write than it would have to find, download, install, and learn to use~
05:24
<@Derakon>
I only have three datatypes to work with here: scalar, array, and hash.
05:43
<@Derakon>
"Error: There is no game with the name [object HTMLDivElement]"
05:58 Serah [~Z@Nightstar-5401.atm2-0-1041217.0xc329e232.boanxx12.customer.tele.dk] has quit [Quit: Be right back, got some smiting and righteous justice to attend to.]
05:59
<@Derakon>
Okay...could someone please take a look at http://derakon.dyndns.org/~chriswei/wh/map2.html and help me figure out why the line "document.getElementById('joinMapName').value" is returning "[object HTMLDivElement]" every time?
05:59
<@Derakon>
It's an input element, not a div. And it definitely has a unique ID.
07:03 Derakon is now known as Derakon[AFK]
07:39 AnnoDomini [~farkoff@Nightstar-29799.neoplus.adsl.tpnet.pl] has joined #Code
07:39 mode/#code [+o AnnoDomini] by ChanServ
08:00 Vornotron is now known as Vornicus-Latens
09:53 You're now known as TheWatcher
10:52 Vornotron [~vorn@Admin.Nightstar.Net] has joined #code
10:53 Vornicus-Latens [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout]
11:48 Attilla [~The.Attil@Nightstar-9469.cdif.cable.ntl.com] has joined #code
11:48 mode/#code [+o Attilla] by ChanServ
12:28 Vornucopia [~vorn@Admin.Nightstar.Net] has joined #code
12:30 Vornotron [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout]
13:24 Vornotron [~vorn@Admin.Nightstar.Net] has joined #code
13:25 Vornucopia [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout]
13:58 gnolam [lenin@79.136.60.ns-4387] has joined #Code
13:58 mode/#code [+o gnolam] by ChanServ
16:46 Vornotron is now known as Vornicus
17:05 Derakon[AFK] is now known as Derakon
17:20
<@ToxicFrog>
I now see what gnolam meant about FF3 and self-signed certificates.
17:21
<@ToxicFrog>
ME: I would like to configure my router now. Opera?
17:21
<@Derakon>
You can improve the behavior a bit (to where the hidden "I know what I'm doing" div is always unhidden and the cert is retrieved automatically) with some fooling in about:config.
17:21
<@ToxicFrog>
OPERA: Sorry, I'm wedged on a buggy sshfs mount and you won't realize this for another five minutes.
17:21
<@ToxicFrog>
ME: Ok. Firefox?
17:21
<@ToxicFrog>
FIREFOX: YOU ARE ENTERING THE ASSHOLE OF THE GREAT SATAN. ABANDON EVERY HOPE, YE THAT ENTER.
17:23
<@ToxicFrog>
After getting it to connect, it then worked for a few pages and then stopped working.
17:23
<@ToxicFrog>
Complaining that the certificate has the same serial number as, er, the certificate.
17:23
<@Derakon>
That sounds unrelated.
17:23
<@Derakon>
...oh.
17:23
<@ToxicFrog>
And that the certificate authority should issue a new one.
17:28
<@Derakon>
Ahh, yes. Set browser.xul.error_pages.expert_bad_cert to true and browser.ssl_override_behaviour to 2.
17:29
<@Derakon>
The former unhides the div; the latter prefetches the cert.
17:29
<@Derakon>
So you go from four clicks to two.
17:29
<@Derakon>
(I'm not saying this is ideal, but it's a heck of a lot better than the default)
17:41 ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has quit [Ping Timeout]
17:56 ToxicFrog [~ToxicFrog@Nightstar-24851.acanac.net] has joined #code
17:56 mode/#code [+o ToxicFrog] by ChanServ
18:00 * Consul sets off to find a Linux proggy that can decode Nikon raw image format...
18:02
<@Consul>
It would be great if the GIMP could read them directly...
18:03
<@Derakon>
There might be a decoder for ImageMagick to work with them, though I didn't get it with my compile.
18:05
<@Consul>
There appears to be a GIMP plugin called ufraw which can do the trick. I'll see if I can give it a go.
18:05
<@Consul>
And as luck would have it, it's available in the Fedora repos.
18:07
<@gnolam>
Derakon: the standard command line tool for it is dcraw.
18:10
<@Consul>
I'm hoping that the 12-bits-per-pixel dynamic range of the raw format over jpeg's 8-bits will help out in the jewelry photography stuff I'm doing.
18:10
<@Consul>
That, and I'd like to shoot in a lossless format.
18:10 ToxicFrog [~ToxicFrog@Nightstar-24851.acanac.net] has quit [Operation timed out]
18:10
<@Consul>
I should clarify: that's 12 bits per pixel per color channel.
18:11
<@Consul>
These photos are also destined for CMYK printing.
18:13
<@Consul>
The good news is, I should be able to turn the raws right over to our graphic designer and printer lady.
18:13
<@Consul>
So I need raw capability only to check the photos and sort out the good ones.
18:27 ToxicFrog [~ToxicFrog@Nightstar-25829.acanac.net] has joined #code
18:27 mode/#code [+o ToxicFrog] by ChanServ
18:29
<@Derakon>
Right! I can successfully create a game and then tell the API to join a game...now I have to write the Javascript code to interpret all the game data.
18:47 crem_ [~moo@Nightstar-28703.adsl.mgts.by] has quit [Ping Timeout]
18:57
<@gnolam>
Well, NEF isn't 100% lossless. But still.
18:58
<@gnolam>
(I consider post processing cheating, so I almost only shoot in JPEG, heh)
18:59
<@Derakon>
Cameras are demonstrably not as accurate as the human eye is.
18:59
<@Derakon>
So some degree of post-processing can actually make the image more accurate (so long as you trust the person doing the postprocessing).
19:01
<@ToxicFrog>
JPEG isn't lossless either.
19:02
<@Derakon>
Yyyyep.
19:04
<@gnolam>
Nobody claimed it was.
19:12 Vornotron [~vorn@Admin.Nightstar.Net] has joined #code
19:14 Vornicus [~vorn@ServicesOp.Nightstar.Net] has quit [Ping Timeout]
19:14
<@gnolam>
Wait...
19:14
<@gnolam>
You said this was for CMYK printing? And you are going to use the GIMP?
19:35
<@Consul>
Read my sentence again.
19:35
<@Consul>
I only need a way to preview and puck out the best images.
19:35
<@Consul>
As it turns out, Ufraw does this fine standalone.
19:35
<@Consul>
Tnen the raws themselves get shuffled off to the lady doing the catalog design.
19:36
<@Consul>
err, pick out
19:36
<@gnolam>
Ahh.
19:37
<@Consul>
Yeah, turns out that GIMP still can't do anything beyond 8 bits per channel.
19:39 * Consul waits a half-hour for his honey-water to cool down so he can get the mead going.
19:47
<@gnolam>
Mead. \o/
19:48
<@Consul>
It's a basic recipe, takes about four or so months to mature, 3 to be decent, 2 to be drinkable.
19:48
< GeekSoldier>
I take it that the process in similar to that for beer, except without grains and hops?
19:48
<@gnolam>
It's a shame my pet beekeeper had a hive collapse.
19:48
<@Consul>
GeekSoldier: Actually, no, beer is a very different process that requires boiling. The mead only needs to be brought up to enough heat to dissolve the honey.
19:49
< GeekSoldier>
hmm. I brew beer at home, but I've never done mead. is taste primary dependant on honey quality and yeast-type?
19:49
<@Consul>
It's as complex as making wine.
19:49
<@Consul>
So many factors go into it.
19:50
<@Consul>
But yes, quality honey will make the biggest difference, I'll bet.
19:50
<@gnolam>
I agree.
19:50
<@Consul>
I buy stuff from the store. I'm not experienced enough to be a snob. :-)
19:50
<@gnolam>
But unlike traditional beer, even traditional meads are often spiced.
19:50
<@Consul>
I made a pineapple melomel (a fruit mead, basically) that was pretty darned good. Took a year of aging, though.
19:51
< GeekSoldier>
interesting. I guess that's what I like about beer; you could get a decent batch in about a month.
19:52
<@Consul>
Well, it is wine, when it comes down to it.
19:52
<@gnolam>
But making beer yourself is easy. Making /good/ beer yourself is hard. :)
19:53
<@gnolam>
But most people manage a drinkable mead the first time.
19:53
<@Consul>
The key there is to not be stupid like I was and try making stouts the first time out.
19:53
<@Consul>
Start with a light or an amber.
19:53
< GeekSoldier>
I made a decent pilsner my first batch.
19:54
<@Consul>
Myself, I really like wheat beers.
19:54
<@Consul>
So if I try beer again, that's what I might try next.
19:54
<@Consul>
My other problem is, I only have enough gear to make 1 gallon batches.
19:54
<@Consul>
Which is fine by me, really.
19:54
<@Consul>
But the recipes don't always scale well.
19:55
<@Consul>
And everything is sold in the shops assuming five-gallon batches.
19:55
< GeekSoldier>
yikes. I'm limited to 2 gallon batches at the moment, and I go through those quickly.
19:55 * gnolam has enough to make 3 x 25 l of wine/beer/mead at any given time. ^-^
19:56
<@gnolam>
Brewing equipment is surprisingly cheap.
19:56
< GeekSoldier>
I'm going to purchase a couple 5 gallon buckets soon.
19:57
<@Consul>
My biggest problem is space.
19:57
<@Consul>
I don't have the room to store five gallons of beer in bottles. Also, the Grolsch-style bottles I like are pricey. I have 24 of them.
19:57
< GeekSoldier>
Me too. fortunately, I have a pantry area that is beneath the stairs and stays a pretty good temperature.
19:57
< GeekSoldier>
I saved a bunch of those from when I was in Germany.
19:59
<@gnolam>
I should weasel out the recipe Patya used on the choir's last tour. That ale was awesome.
19:59
<@gnolam>
Packing priorities: http://www.lysator.liu.se/~gnolam/RAG/STORK2008/DSC_0028.JPG
20:01
< GeekSoldier>
sweet.
20:02
<@gnolam>
Best of all: the place we were staying at - a school - had strict rules prohibiting any alcohol in the building. So we /had/ to drink it all on the bus on the way up! ;D
20:03
< Vornotron>
Heeeee.
20:03
<@jerith>
This network does not yet have a #beer.
20:04
< Vornotron>
Obviously you must remedy this.
20:04
<@Consul>
How about #homebrewing ?
20:04
<@jerith>
The network now has a #beer.
20:37 crem [~moo@Nightstar-28703.adsl.mgts.by] has joined #code
21:05 crem [~moo@Nightstar-28703.adsl.mgts.by] has quit [Quit: Ep0xa 1.2v]
21:23 AnnoDomini [~farkoff@Nightstar-29799.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
21:28 crem [~moo@Nightstar-28703.adsl.mgts.by] has joined #code
21:30 AnnoDomini [~farkoff@Nightstar-7076.neoplus.adsl.tpnet.pl] has joined #Code
21:30 mode/#code [+o AnnoDomini] by ChanServ
22:13 Serah [~Z@Nightstar-5401.atm2-0-1041217.0xc329e232.boanxx12.customer.tele.dk] has joined #Code
22:13 mode/#code [+o Serah] by ChanServ
22:51
<@Consul>
Does anyone know how to get Open Office Calc to create a function from a scatterplot curve?
22:51
<@ToxicFrog>
Not a clue, sorry!
22:52
<@gnolam>
I'd start looking in Scilab or something similar instead.
22:52
<@Consul>
I don't have time.
22:52
<@Consul>
This is for a physics lab.
22:52
<@gnolam>
I've started hating OpenOffice with a /passion/.
22:52
<@Consul>
I just need to extrapolate the curve at one point.
22:53
<@gnolam>
Do it manually (least squares etc)?
22:53
<@Consul>
I don't know how.
22:53
<@Consul>
I haven't gotten that far in linear algebra yet.
22:54
<@Consul>
Or whatever the math is.
22:54
<@Consul>
I'm also lazy.
22:57
<@Consul>
OO.o is apparently incapable of curve-fitting.
22:57
<@Consul>
Big fat freaking surprise.
22:58
<@gnolam>
Yeah, it's linear algebra.
22:58 * gnolam chalks up another FAIL for OpenOffice.
22:58
<@Consul>
Okay, it says there's an exponential regression available, but I can't find it.
22:58
<@Consul>
It always keeps picking linear.
23:00
<@Consul>
Blargh.
23:01
<@Consul>
I can't get it to switch to anything other than linear.
23:03 AnnoDomini [~farkoff@Nightstar-7076.neoplus.adsl.tpnet.pl] has quit [Quit: "General! Strike me down, then have every last man, woman and child put to the sword. There's no point going on anymore, not now that it's all gone..." "At once, sire."]
23:25 You're now known as TheWatcher[T-2]
23:32 You're now known as TheWatcher[zZzZ]
23:37 C_tiger [~c_wyz@Nightstar-8414.hsd1.ca.comcast.net] has quit [Killed (NickServ (GHOST command used by C_tiger_))]
23:37 C_tiger [~c_wyz@Nightstar-8414.hsd1.ca.comcast.net] has joined #code
23:37 mode/#code [+o C_tiger] by ChanServ
--- Log closed Mon Nov 10 00:00:25 2008
code logs -> 2008 -> Sun, 09 Nov 2008< code.20081108.log - code.20081110.log >