2008-07-31 GnuCash IRC logs

00:11:55 *** rogo has left #gnucash
01:15:12 *** zloog has joined #gnucash
01:15:27 <zloog> hi gnucashiers
01:15:30 <zloog> anyone here?
01:17:08 <zloog> Well I'm new to gnucash and accidently ran all my transactions as credits from the expenses:books account. Is there a way I can move those transactions to be pulling $ from my checkings instead?
01:59:01 *** aindilis` has quit IRC
02:37:24 *** nbinont has joined #gnucash
03:27:56 *** warlord-afk is now known as warlord
03:28:55 <warlord> zloog: did you want to move ALL the txns in Ex:Books into your checking account? Or only some of them?
04:06:56 *** merc has joined #gnucash
04:27:06 *** merc has quit IRC
04:38:45 *** warlord has quit IRC
05:04:31 *** andi5 has joined #gnucash
05:04:32 *** gncbot sets mode: +o andi5
05:22:38 *** StocksR has joined #gnucash
05:27:15 *** StocksR has quit IRC
05:29:41 *** warlord has joined #gnucash
05:29:41 *** gncbot sets mode: +o warlord
05:32:27 *** ErKa has joined #gnucash
05:46:02 *** warlord has quit IRC
05:46:06 *** warlord has joined #gnucash
05:46:06 *** gncbot sets mode: +o warlord
06:50:35 *** warlord has quit IRC
07:13:07 *** Rolf1 has left #gnucash
07:13:48 *** Rolf1 has joined #gnucash
07:14:56 *** twunder has joined #gnucash
07:27:00 *** twunder has quit IRC
07:31:06 *** JimRaehl has left #gnucash
07:46:47 *** JimRaehl has joined #gnucash
07:55:43 *** warlord has joined #gnucash
07:55:44 *** gncbot sets mode: +o warlord
07:58:57 *** twunder has joined #gnucash
08:22:01 *** twunder has quit IRC
08:24:32 *** emrec has joined #gnucash
08:24:52 <emrec> (08:20:44 AM) emrec: hi
08:24:52 <emrec> (08:22:33 AM) emrec: I needed a small accounting program for my small business and i found out gnuCash. Im trying to install it to my ubuntu/debian 8.04 and im getting and error mesaage when i configure it as "pkg-config --modversion glib-2.0' returned 2.9.6, but GLIB (2.16.3) was found! If pkg-config was correct, then it is best to remove the old version of GLib ....
08:24:52 <emrec> (08:22:59 AM) emrec: i searched about it but i couldnt find any proper solution
08:24:53 <emrec> (08:23:06 AM) emrec: is there anyone can help?
08:25:55 <warlord> emrec: apt-get install gnucash
08:28:42 *** |gunni| has joined #gnucash
08:28:43 *** _gunni_ has quit IRC
08:29:03 <emrec> warlord: ok im installing it now
08:29:04 <warlord> (unless there's some particular reason you want to compile gnucash manually?)
08:29:18 <emrec> nope
08:29:55 <warlord> did you compile other stuff by hand? like glib?
08:30:01 <emrec> yeap
08:30:06 <emrec> the latest version
08:30:07 <warlord> Oops!
08:30:16 <warlord> Well, you just destroyed your system's integrity.
08:30:25 <emrec> 2.9.6 from their site
08:32:21 <emrec> i couldnt see this note on the web-site "installing compiled gnucash" that's why i tried to do it manually. ANd im also really rookie to use Linux
08:32:29 <emrec> :(
08:33:23 <warlord> yeah. right now I'd recommend that you remove all your manually-compiled stuff. Having it there will REALLY cause lots of problems.
08:34:58 <warlord> (or re-install, if that's an option)
08:38:28 <emrec> i compiled by #make install and how will i remove manually compiled glib? by the way apt-get insatll gnucash has finished without any error
08:38:46 <warlord> You can /try/ "make uninstall"
08:42:59 <emrec> warlord:ok glib has been removed and gnucash is working properly now.. Tnx a lot warlord
08:43:16 <warlord> you're welcome
09:10:09 *** Rolf1 has quit IRC
09:12:37 *** emrec has left #gnucash
09:18:25 *** Rolf1 has joined #gnucash
09:29:38 *** donfede has joined #gnucash
09:40:57 *** twunder has joined #gnucash
09:46:56 *** Rolf1 has quit IRC
09:50:28 *** Rolf1 has joined #gnucash
09:53:39 *** zloog has quit IRC
10:04:38 *** warlord has quit IRC
10:42:03 *** warlord has joined #gnucash
10:42:03 *** gncbot sets mode: +o warlord
11:06:06 *** nbinont has quit IRC
12:26:05 *** andi5 has quit IRC
13:15:30 *** ErKa has quit IRC
14:16:21 *** warlord has quit IRC
14:22:47 *** warlord has joined #gnucash
14:22:47 *** gncbot sets mode: +o warlord
14:23:02 *** warlord is now known as warlord-afk
14:28:57 *** twunder has quit IRC
15:35:19 <andrewsw> congrats boys!
16:19:10 *** Ozaed has joined #gnucash
17:02:52 *** rog1 has joined #gnucash
18:34:58 *** cortana has joined #gnucash
18:36:01 *** _gunni_ has joined #gnucash
18:44:35 *** rog1 has left #gnucash
18:45:07 *** rog1 has joined #gnucash
18:45:14 *** |gunni| has quit IRC
19:11:15 *** Cypha has joined #gnucash
19:14:55 <Cypha> Hi all, in love with GnuCash. I purchase alot fo items from megastores like Amazon which just show as amazon on my statement. Its easy to lookup in my email what the purchase was and then assign it to the correct account - However I would like to sau eaxtly what it was e.g. Niko D40 Camera and not just assign to the Hobby account. Where would you put this infromation?
19:15:35 <Cypha> Pls excuse the bad spelling - New logitech keyboard!#
19:18:23 <jsled> The transaction description or split memo.
19:18:54 <jsled> There's only one description for the transaction, but each split (and thus each account) has a separate memo.
19:19:27 <jsled> I generally find the split to be most appropriate … especially if one amazon order has individual line items that want to go into different Expense accounts.
19:21:42 <Cypha> Thanks @jsled, Im gonna go into GnuCash now and check it out!
19:26:39 <Cypha> Ok, I understand what you mean. I remember the split transactions from the tutorial. However, that is really handy when I have an Amazon order consisting of different items from different accounts. I but alot of gadgets from Amazon for example. Instead of creating accounts for photography, GPS, Gaming etc. I would rather just file them under the electronics account and record somewhere what each transaction specifically is. Are you saying I should edit the desc
19:27:08 <jsled> yeah, that sounds reasonable.
19:27:42 <jsled> I have a catch-all "Expenses:Electronics" account, but if, say, a subset of the order is homebrewing stuff, then I'll mark it "Expesnes:Homebrewing".
19:27:52 <jsled> And record in the split descriptions what the actual items are.
19:28:17 <jsled> Really, it's "put as much effort into recording as much detail as you care about reporting on". :)
19:28:40 <jsled> If you don't care to break things out later by expense account/hobby, then just dump it all into "Expenses" and be done with it.
19:28:52 <Cypha> ok, Im getting it.. I guess I should be afraid to create more sub-accounts if I need that detail
19:29:03 <Cypha> I mean shouldnt!
19:29:42 <Cypha> Thanks jsled, that makes total sense. I will create sub-accounts that I care to report on - the rest can be lumped together
19:29:53 <jsled> right.
19:30:13 <jsled> And you can always move things around (either way, really) later … though note that there are no batch operations on transactions.
19:30:46 <jsled> For instance, you can't go through a generic Expenses account list and easily select a subset of 20 transactions to move into a "Expenses:Specific" account.
19:31:12 <Cypha> ahh ok.
19:31:16 <jsled> But … if you have an Expenses:Specific account, you can – as a side-effect of deleting that account – re-parent the transactions in Expenses.
19:31:40 <Cypha> ok, fully understand.
19:32:56 <Cypha> I really appreciate your help. Coming to grips with my finances has been an eye opening experience to say the least.
20:13:44 *** unfair has joined #gnucash
20:14:32 <unfair> how can i get a price quote for a canadian security listed on the tse? if i put tse as the type in the security editor, yahoo will not find a match
20:15:29 <unfair> i think because gnucash is using for example tse:cef.a as the search term, for which yahoo finance returns nothing, but if i search just cef.a, it will return the proper quote
20:15:46 *** twunder has joined #gnucash
20:16:03 <unfair> the problem is the gnucash requires a security 'type' and then uses that as the prefix for the search, which screws the whole thing up
20:16:44 <jsled> unfair: try using the 'gnc-fq-*' apps (which gnucash calls) to try to localize the issue.
20:17:53 <unfair> ok ill look into it, this is on windows btw
20:22:52 <jsled> unfair: So, if I use Type=TSE, {Security = "testing", symbol = "cfe.a", get Quotes, multiple, "Canada"}, it reports a last price of 13.26…
20:25:18 <unfair> yeah, that doesn't work for me, i can get an amex quote though, using multiple, usa
20:26:06 <jsled> So, if you run `gnc-fq-dump canada cef.a`, what do you get?
20:30:07 <unfair> it returns info for cef.a but it says it is missing a date
20:30:58 <jsled> that's ... weird.
20:32:25 <unfair> yeah, getting a quote for usa is fine though
20:32:29 <unfair> it returns ibm fine
20:38:28 *** twunder has quit IRC
21:05:32 *** unfair has quit IRC
21:13:07 *** rog1 has quit IRC
21:17:25 *** twunder has joined #gnucash
21:28:26 *** cortana has quit IRC
22:06:55 <Rolf1> jsled: are you still working on the reporting stuff?
22:07:12 <Rolf1> I am just browsing some messages to gnucash-devel from October of last year
22:07:31 <jsled> nope. I haven't done anything since the fall, and I doubt I will.
22:07:52 <Rolf1> lack of interest?
22:09:36 <jsled> For a long time now, yeah.
22:12:12 *** rog1 has joined #gnucash
22:13:25 <Rolf1> why is that?
22:13:41 <Rolf1> what are you interested in getting moved forward in gnucash?
22:14:23 <jsled> Oh ... well, I've basically decided the current code is more trouble than it's worth for what I want out of gnucash.
22:17:41 <jsled> I might have said this before while you were in-channel, but I keep thinking back to the assertion (or "Fact") from Robert Glass: Fact 19: Modification of reused code is particularly error-prone. If more than 20 to 25 percent of a component is to revised, it is more efficient and effective to rewrite it from scratch.
22:17:51 <jsled> I'd say gnucash easily qualifies.
22:18:34 <jsled> startup, options, reporting, the register and the ledger are all things that everyone would agree need to be re-written.
22:19:29 <jsled> I'd suggest that all those roughly familiar with gnucash internals could be persuaded that it would be better for even the *core* of the engine to be done differently – in late-2008 – than how QOF works.
22:21:20 <jsled> And, moreover, I'm spoiled by working in managed languages eveyrwhere else, so working in C is utter fail.
22:22:59 <andrewsw> sounds like you're advocating a total rewrite of the whole thing...
22:23:11 <jsled> Plus, I'm pretty convinced that web apps – even personal ones – are the future.
22:23:58 <jsled> Plus, "every successful large system can be shown to have grown from a successful smaller system". </paraphrase>
22:24:15 <jsled> andrewsw: well, a smaller new app, to start.
22:24:23 <andrewsw> yes.
22:24:26 <jsled> But, yeah.
22:24:44 <jsled> Which is so damn frustrating, because there's man-years of value there.
22:24:51 <andrewsw> yes.
22:25:13 <Rolf1> restart from scratch sounds interesting
22:25:29 <Rolf1> But I don't agree with the "web apps is the future"
22:25:30 <jsled> But I just don't see the right way to massage it into greatness. But I can see the path that grows from a kernel, pulling in pieces of the existing source as … good suggestions.
22:25:48 <Rolf1> I choose not work with sql-ledger for a reason
22:25:56 <andrewsw> maybe not from scratch, but from the existing data file structure?
22:25:57 <jsled> (or, add QIF import to the "well known needs rewriting" list)
22:26:11 <jsled> andrewsw: Well, clearly. :)
22:26:18 <jsled> And the same core model, in fact.
22:26:35 <Rolf1> jsled: what do you want out of gnucash?
22:26:58 <Rolf1> what language do you suggest?
22:27:08 <jsled> http://paste.lisp.org/display/64513
22:27:52 <jsled> Rolf1: Oh, not too much. A wicked-efficient way to enter receipts (I'm now 5 months behind because it's dreadful to do it, but that's not all gnucash)
22:28:07 <jsled> bank statement import with reconcilation.
22:28:18 <jsled> Uh ... basic expense reporting.
22:28:48 <Rolf1> IOW, business not personal use as focus
22:28:54 <Rolf1> same here
22:29:08 <jsled> er... no, not business at all.
22:29:38 <jsled> Oh, basic multi-user, too. My finacee and I need to see shared/joint expenses.
22:30:09 *** twunder has quit IRC
22:30:41 * andrewsw is out of practice with gnucash
22:30:53 <Rolf1> jsled: what does the python thing do?
22:31:03 <Rolf1> I'm too tired to understand and test
22:31:26 <jsled> Oh, that's a django model definition. From that is generated a DB schema, object model, accessors, DB query interface, &c.
22:31:28 <andrewsw> can the python bindings be used to rewrite portions of the app, gradually replacing the C code one chunk at a time?
22:31:35 <jsled> Oh, and admin interface.
22:31:48 <andrewsw> or is that just rewriting the same problems in a new language?
22:31:48 <jsled> andrewsw: See Fact 19 above. :)
22:32:09 <andrewsw> yeah
22:32:32 <andrewsw> of the core devs, are you alone in this or has this come up before?
22:32:53 <Rolf1> jsled: Fact 19 does not speak against rewriting gradually
22:33:01 <Rolf1> I even think it supports that
22:33:23 <Rolf1> There is no use throwing everything away and hoping to have something to replace it in due time
22:33:40 <jsled> andrewsw: I don't know if I'd consider myself a core dev. I've not really contributed any new code in a couple of years.
22:33:45 <Rolf1> gradual replacement is a very good thing IMHO
22:33:59 <jsled> I moderate the mailing lists. I sometimes look at bugs. I might read the mailing list, but usually don't.
22:34:23 <Rolf1> I've also come to realize that what I want out of gnucash will require quite a bit of external help
22:34:31 <Rolf1> or a miracle
22:34:39 <jsled> Rolf1: nope, it's pretty explicit. If you have to change more than a small fraction of the thing, you're better off starting from scratch.
22:34:47 <jsled> I can't reason it effectively, but my experience bears that out.
22:35:03 <Rolf1> jsled: but it still depends on perspective
22:35:10 <andrewsw> well there's an issue right there. The two *lead* devs aren't really core devs anymore and so who is producing code?
22:35:12 <Rolf1> the question is if the thing is one big box
22:35:18 <Rolf1> or whether you have modules
22:35:26 <Rolf1> with well-defined connectors in and out
22:35:44 <Rolf1> if you have modules you're lucky and good to go with gradual replacement
22:35:48 <jsled> There's this funny thing about programming. There's so much implicit in the code … often naturally so (I mean: not bad programming, just an inherent feature of how software is written) … code has significant inertia.
22:35:57 <Rolf1> one module at a time (complete and 100%)
22:36:00 <jsled> s/is written/is expressed/
22:36:27 <andrewsw> I know andi5 does a lot and Charles Day as well, Cstim, a bit, but the two guys with the most in depth knowledge aren't really doing much anymore. (not criticism, just observation)
22:36:30 <jsled> andrewsw: right. :) andi5. Charles Day. Phil Longstaff.
22:36:56 <jsled> No, not criticism. warlord admits it too.
22:37:10 <jsled> Yeah, it'
22:37:22 <Rolf1> jsled: are you serious with thinking about the rewrite?
22:37:28 <andrewsw> You guys have a *presence*...
22:37:34 <jsled> It's a frustrating place for the project. So much potential. So much inertia. So hard to get stuff done in the code base.
22:37:47 <Rolf1> same thing for the user
22:37:53 <jsled> Rolf1: one thing that gnucash does have is a reasonable internal structure … mostly.
22:37:54 <Rolf1> At least as much frustration
22:37:58 * jsled will expound in a bit.
22:38:03 <Rolf1> ok
22:39:22 <Rolf1> I'll get some sleep now
22:39:29 <Rolf1> but I'll check the backlog tomorrow
22:39:36 <Rolf1> looking forward to your ideas
22:42:42 *** Rolf1 has quit IRC
22:44:25 *** nawab has quit IRC
22:44:31 *** nawab has joined #gnucash
22:46:54 <jsled> yeah, they could be used to do that.
22:47:06 <jsled> But … it's a slog.
22:47:21 <jsled> One of the problem with gnucash is the fact that most of it's in C, but a non-trivial part is in scheme.
22:47:49 <jsled> Unfortunately, there's a pretty serious "impedance mismatch" between those two languages.
22:48:05 <jsled> That mismatch is not *as much* between C and python, but it's still there.
22:48:52 <jsled> The language idioms aren't quite as far apart, but a) they still are, and b) any managed vs. non-managed split has problems.
22:49:52 <jsled> As for "rewriting the problems" … no, it doesn't need to be a bug-for-bug rewrite. And most of the internal interfaces – where they exist – are sound.
22:50:52 <jsled> A good example is the QIF importer. It would be a mistake to naïvely rewrite the (scheme) QIF importer into C function-by-function, but instead to adapt it into C, keeping the rough overall structure and maybe even literally keeping the same utility functions and regexps and whatnot …
22:51:02 <jsled> But, adapting it into what the "generic importer" wants.
22:51:46 <jsled> The [report] options stuff, OTOH, is *so* tied to dynamic expression and evaluation of options in a really lisp-y idiom.
22:52:34 <andrewsw> IMO the extremem lispyness of the report options is a huge barrier.
22:52:54 <jsled> That's a case where I'd say "screw it. Existing open and saved reports are going to be lost" Getting the backward compatability right means the thing never gets delivered; getting improved code with a bit of user pain is more important.
22:53:07 <andrewsw> yup.
22:53:30 <jsled> But even in the C, things are not all roses.
22:54:03 <jsled> I don't think – generally – that a fine-grained, implicit data-modification model (a-la QOF) is the way to do things.
22:54:11 <jsled> I think it's magical and brittle.
22:54:34 <jsled> Too much of the abstract business logic is tied directly to UI reactors.
22:55:28 <jsled> The register – while it does have a good MVC separation – just didn't track gtk, and is thus really broken.
22:56:13 <jsled> The areas in which people made major improvements aren't the areas in which the register is implemented … so it can't leverage those improvements.
22:59:05 <jsled> And that's not to blame the improvers. It's just the way software works, to a large part.
23:02:27 <jsled> Rolf1: as for the gradual replacement … I do think it's possible. And were the ratio of okay:rewrite different, I'd advocate it. But I think the scales are tipped in this case.
23:02:56 <jsled> [[[
23:02:57 <jsled> # A complex system cannot be "made" to work. It either works or it doesn't.
23:02:57 <jsled> # A simple system, designed from scratch, sometimes works.
23:02:57 <jsled> # Some complex systems actually work.
23:02:57 <jsled> # A complex system that works is invariably found to have evolved from a simple system that works.
23:02:57 <jsled> # A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over, beginning with a working simple system.
23:03:02 <jsled> ]]] - http://en.wikipedia.org/wiki/Systemantics
23:07:31 <jsled> I mean, Systemantics is a bit silly and a bit strange, but these aren't isolated conclusions. In large parts, Agile methods and open source are rooted in the same paradigm: build the simplest possible thing. Get it in front of people. Iterate to the next level. Keep a feedback loop that includes: customer feedback, automated testing, code debt.
23:09:00 <jsled> So, yeah, that python is from a little experiment. It only has a few man-hours of effort behind it, over the last few months. If it gets somewhere good, I'll broadcast that and solicit contributions.
23:09:15 <andrewsw> I didn't want to be the first to mention agile, but since I'm in an agile class at the moment, it was sure starting tolook like smelly code.
23:09:17 <jsled> But I'm a big believer in releasing projects much closer to "1.0" than "0.0.1".
23:10:10 <jsled> Well, especially at the core, it's an old project. There are conventions and libraries for things now that didn't exist at the time.
23:10:33 <jsled> QOF is fucking awesome when you realize when it was born.
23:10:58 <jsled> Compare it to the LINQ stuff that Microsoft's doing.
23:11:27 <jsled> But now ... those things are *language features".
23:11:31 <jsled> s/"/*/
23:11:35 <jsled> so, why re-implement them.
23:11:48 <jsled> And why burden yourself with that implementation.
23:12:34 <jsled> Of course, even simpler is memory management and string handling in C.
23:12:47 <jsled> Or object orientation in glib/gnome.
23:13:03 <jsled> Life's too short to write your own fucking vtables.
23:14:18 <jsled> Anyways … new PS3 games/demos out tonight. Less talk more rock.
23:14:20 <jsled> 'night!
23:14:41 <andrewsw> goodnight jsled. thanks for the ...
23:14:45 <andrewsw> stuff
23:14:47 <jsled> heh
23:16:47 <andrewsw> got it!
23:16:50 <andrewsw> oops..
23:17:00 <jsled> hmm?
23:17:58 <andrewsw> wrong window.
23:18:21 <andrewsw> I just figured out why my silly swing buttons were losing their names when I added actions.
23:18:32 <jsled> swing?
23:18:37 <jsled> Sorry to hear it. :)
23:19:00 <andrewsw> it's okay.
23:19:22 <andrewsw> part of this agile class... we've written a webserver and now are writing browser to talk to it.
23:19:25 <andrewsw> kinda fun.
23:19:26 <jsled> Have you done swing (or other UI stuff) before?
23:19:32 <andrewsw> nope.
23:19:34 <jsled> Oh, that is good.
23:19:46 <jsled> I hope you're using some HTML rendering library?
23:20:16 <andrewsw> for the browser? yes. JEditorPane, if you set the content type, will do that for us.
23:20:36 <jsled> Oh, cool. How deep into HTTP did you get on the web server part?
23:20:57 <andrewsw> It's just a nice little app to practice Observer and other behavior patterns
23:20:59 <andrewsw> um..
23:21:02 <andrewsw> not too deep.
23:21:18 <andrewsw> Very simple server that handles a file backend or memory stored pages.
23:21:21 <andrewsw> simple queries
23:21:32 <andrewsw> but has lots of room to extend
23:21:40 <jsled> Any delving into the various HTTP status codes? Cache-control headers?
23:21:50 <jsled> Did you need to read RFC2616?
23:21:58 <andrewsw> everything is all interfaced up nicely with inverted dependencies and other OOP stuff
23:22:02 <andrewsw> nope.
23:22:10 <andrewsw> we kick out 200, 404, and that's about it.
23:22:13 <jsled> Ah.
23:22:33 <andrewsw> total of about 10 man hours of learning TDD and some nice OOP principles
23:22:36 <jsled> So ... some sort of nod to Dependency Injection/IOC?
23:22:59 <andrewsw> ummm.. I think so. What's IOC?
23:23:09 <jsled> "Inversion of Control"
23:23:15 <andrewsw> okay, sure.
23:23:24 <jsled> Heh. The idea that a component has it'
23:23:56 <jsled> it's collaborators given to it, rather than knowing what they are explicitly.
23:24:03 <andrewsw> yes
23:24:39 <andrewsw> our "server" used to know about and depend on a whole slew of stuff (filereaders, request parsers etc), but now it just has two interfaces that are implemented by another package that contains all that stuff.
23:24:50 <jsled> So, your web server would be configured with some implementation of PersistentPageResolver … right.
23:25:02 <andrewsw> Sure
23:25:25 <jsled> (where PersistentPageResolve might have an impl. that went to the filesystem … or another that layered that with a cache … or a completely fake one for doing reasonable unit testing.)
23:26:09 <andrewsw> the server only handles the network portions, all the rest is handled externally, and that external system returns a "Response" opbject that contains the http protocol stuff and the text of the result.
23:26:16 <andrewsw> and yes, what you said exactly.
23:26:44 <andrewsw> we were supposed to implement a Factory, and I think we did, sort of.
23:27:13 <andrewsw> it's a Factory that builds up the PageResolver object.
23:27:25 <andrewsw> regardless, all very fun stuff, and team programming to boot!
23:27:42 <jsled> What're you using for revision control? SVN, I think you said before?
23:28:07 <andrewsw> We had three teams and when everyone passed a certain point in the test suite, we all passed around the server portion of the system and had to plug in our PageResovler stuff.
23:28:11 <andrewsw> yes SVN
23:28:25 <andrewsw> though we could have used whatever
23:29:06 <jsled> Hmm. So everyone did both halves, then you handed one half off "to the left"?
23:29:13 <andrewsw> yup.
23:29:25 <andrewsw> very interesting to see the assumptions we made versus those of others.
23:29:38 <jsled> I bet! Yeah, that's a good way to do it.
23:29:42 <andrewsw> oh, and you had to make your test suite run on their server...
23:30:08 <andrewsw> so our tests were based on the assumption that we could run arbitrary numbers of servers on any ports we wanted...
23:30:14 <andrewsw> the server we got was a singleton.
23:30:18 <jsled> heh.
23:31:10 <andrewsw> I tried to file a bug report with the other team. They were not amused.
23:31:17 <jsled> hee hee.
23:31:52 <jsled> Alright. I'm afk for real this time.
23:31:53 <jsled> ttyl.
23:32:02 <andrewsw> ciao, thanks for chatting.