Semantic Web Interest Group IRC Chat Logs for 2002-07-13

00:54:35 <niq> niq is now known as confused

02:02:56 * confused is away: having a well-earned nervous breakdown

02:39:36 * confused is back (gone 00:36:39)

02:40:39 <eikeon> Short nervous breakdown.

02:50:06 <jordan> welcome to the club

10:40:44 * dajobe lol at earlier rdf/xml ranting

10:41:49 * danb_lap revists log with nick2mugshot CSS sstyle enabled in mozilla chat client, see http://www.hacksrus.com/~ginda/chatzilla/faces.pl

10:42:01 <danb_lap> only hixie shows up with a photo

10:42:53 <danb_lap> If only we had a way of harvesting nick-to-photos, we could auto generate a more comprehensive stylesheet.

10:42:56 <danb_lap> hmm...

10:43:12 <dajobe> I've an rdf/xml file of nicks here -> other stuff

10:43:21 <dajobe> but I'm not sharing, let people do that themeselves

10:43:32 <dajobe> privacy of personal info

10:43:47 <danb_lap> sharing what? foaf:depiction factoids?

10:43:53 <dajobe> if that's what it is

10:43:54 <dajobe> yeah

10:44:30 <dajobe> [OT] just discovered 'mrproject' - gnome project admin tool - can do gantt charts ;)

10:45:09 <dajobe> well maybe not OT, it has an XML format, adn with suitable scraping, who knows what you could get out

10:45:38 <danb_lap> I went looking for a good way of dumping such charts to HTML tables / CSS. Didn't find one.

10:48:41 <dajobe> B:these quads seem to have 5 things + context, +seq (an integer). Nice to see a formal definition of your proposal.

10:48:42 <dc_rdfig> added comment B6

10:49:39 <dajobe> B:these quads seem to have 6 things + context, +seq (an integer), + a series of alternate terminals (what are they for?) . Nice to see a formal definition of your proposal.

10:49:39 <dc_rdfig> added comment B7

10:49:42 <dajobe> B6:""

10:49:42 <dc_rdfig> deleted comment B6

11:13:08 * danb_lap gets mysql running on win2000 laptop, sets about converting PostgreSQL schema

11:20:44 <danb_lap> hmmm, found loads of mysql2postgres migration tools; not many going the other way. interesting to see (though annoying currently!)

13:09:59 <mdupont> anyone awake?

13:37:27 <Seth> yo, mdupont

13:39:04 <mdupont> hey seth

13:39:11 <Seth> wazup?

13:39:24 <mdupont> i have been trying to embed #anchors in the sailor html

13:39:40 <Seth> kewl ... got a page i can play with ?

13:39:49 <mdupont> so that you can click on a object and follow it to the subject

13:39:59 <mdupont> well, I am trying to figure out the uri

13:40:11 <Seth> yep that is one of the next step s

13:40:15 <mdupont>http://localhost:729/appOne.html#sailor?gotoGraph=urgraph

13:40:15 <dc_rdfig> A: http://localhost:729/appOne.html#sailor?gotoGraph=urgraph from mdupont

13:40:23 <mdupont>http://localhost:729/appOne.html?gotoGraph=urgraph#sailor

13:40:24 <dc_rdfig> B: http://localhost:729/appOne.html?gotoGraph=urgraph#sailor from mdupont

13:40:46 <mdupont> A:has the #sailor anchor before the ? parameters

13:40:46 <dc_rdfig> added comment A1

13:40:57 <mdupont> B:has the #sailor afterwards

13:40:57 <dc_rdfig> added comment B1

13:41:04 <mdupont> both dont work :(

13:41:45 <mdupont> <a name="sailor">

13:41:46 <mdupont> <table cellpadding="0" cellspacing="0" border="0" width="100%" bgcolor="#ccffff"><tbody><tr><td valign="top" bgcolor="#ccccff" colspan="2">sailor</td><td bgcolor="#ffff33" >nodeEvent</td><td valign="top"></td></tr><tr><td valign="top" align="right" width="20%" bgcolor="#ccffff">has-a&nbsp;&nbsp</td><td width="80%" bgcolor="#ccccff" align="left">browser</td><td bgcolor="#ccccff"></td></tr></table></tbody></a>

13:42:25 <Seth> mdupont, can you post the html file somewhere on the web ? ... that way i can run it here

13:42:29 <mdupont> I guess I could just put the anchor at the top of the table like <a name="sailor"/>

13:42:45 <mdupont> well I can send you the changes to the html,

13:42:46 <mdupont> hold

13:43:13 <mdupont> # start of the node

13:43:13 <mdupont> production['nodeStart']= lambda: \

13:43:13 <mdupont> '<tr><td valign="top"><a name=\"' + sa('subject') + '\">' + "\n" +\

13:43:13 <mdupont> '<table cellpadding="0" cellspacing="0" border="0" width="100%" bgcolor="#ccffff"><tbody>'

13:43:25 <mdupont> production['nodeEnd']= lambda: '</table></tbody></a>'

13:43:40 <mdupont> in the file html.py

13:45:22 <Seth> how does a name=\"' put a anchor in the page?

13:45:38 <Seth> oh wait ... i see what your doin

13:46:56 <mdupont>http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/introspector/daml/test.rdf?rev=1.2&content-type=application/text

13:46:56 <dc_rdfig> C: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/introspector/daml/test.rdf?rev=1.2&content-type=application/text from mdupont

13:47:07 <mdupont> C:is the latest test quads

13:47:07 <dc_rdfig> added comment C1

13:47:17 <mdupont> C:checked into cvs on sourceforge

13:47:17 <dc_rdfig> added comment C2

13:47:55 <mdupont> <a name="blah">

13:48:14 <mdupont> lets you <a href="#blah">this is a link</a>

13:48:21 <mdupont> to jump to it.

13:50:43 <Seth> probably fastest if you just email me the two files in question: html.py and appOne.html ... send to seth@robustai.net ... can you do that?

13:54:01 <mdupont> sure

13:54:18 <mdupont> but look at the lines I pasted

13:54:27 <mdupont> </table></tbody>

13:54:32 <mdupont> this is backwards,

13:54:41 <mdupont> it should be </tbody></table>

13:54:55 <mdupont> lemme finish a quick test

13:54:59 <mdupont> and I will send it to you

13:55:01 <Seth> was that my error?

13:55:09 <mdupont> i think so. :(

13:55:25 <Seth> not surprising since im dyslexic

13:56:17 <Seth> but what your doing should work ... unless you cant hyperlink to anchors inside tabels .. in which case we schould scrap the tables

13:57:56 <mdupont> i am just testing the code.

13:58:23 <mdupont> dont worry, I am my own form of spelling errors.

13:58:39 <mdupont> i do like the grammer stuff you are doing here.

13:58:44 <mdupont> it looks a bit like xslt

14:02:19 <Seth> have you considered breaking your app out ... use introspector.py and introspector.html .... start where i left off in html.py and appOne.html ...

14:02:37 <Seth> that way you could customize your page and saying grammar for your application

14:03:10 <Seth> i think that you may want to display one node per page, rather than them all on the same page .. might make the hyperlinking more coherant

14:03:31 <mdupont> hmmm.....

14:03:39 <mdupont> welll.....

14:03:52 <mdupont> I think that the sailor needs this in general to be honest.

14:04:08 <mdupont> you need to be able to look at a object as a href

14:04:18 <mdupont> or on a new page

14:04:22 <mdupont> or inside the page.

14:04:28 <mdupont> ?

14:05:10 <Seth> well yes ... big question ... for your app would you prefer looking at one node at a time .. or a whole bunch of nodes at a time ? .... we can do it either way ... which is better for your app?

14:05:17 <mdupont> but I would think about beefing this up a bit to be able to do that.

14:05:32 <mdupont> Well lets say that you want to sail through a RDF file

14:05:45 <mdupont> you want to get an overview of all the things in there

14:05:48 <mdupont> and then focus in.

14:06:02 <mdupont> and be able to pull 1,2,3...n objects ont a page.

14:06:20 <mdupont> like you want to right click on a object and say, load onto this page.

14:06:32 <mdupont> or feed a list of predicates to load automatically

14:07:00 <mdupont> then you want to take a graph and say, lay this out into a GRAPHVIS/VCG/DIA graph

14:07:05 <Seth> right click is hard cross platform here .. but we could put different buttons on each node .. one would isolate it .. another would index it

14:07:14 <mdupont> and make an imagemap

14:07:25 <mdupont> sure, little buttons on the top.

14:07:30 <mdupont> or a handle on the side.

14:07:38 <mdupont> a context menu.

14:07:52 <mdupont> but cross platform, you mean ie5/6 and mozilla right?

14:08:05 <mdupont> we'll that wont be that hard.

14:08:22 <mdupont> there are hiermenus javascript libs for both browsers....

14:08:38 <Seth> yes and linux and apple and pc / ie mozilla ... hard to get right click to work in all platforms

14:09:01 <Seth> better to stick with html + java script

14:09:50 <Seth> so lets decide something to do here and actually accomplish it .. ok?

14:10:20 <mdupont> ok,

14:10:44 <Seth> i suggest a new app/grammar .... that will isolate each node ... and\or we could get anchors working on the hyperlinks .. which is best approach for you?

14:11:05 <mdupont> Hmmm, how is this.

14:11:18 <mdupont> a per node viewer yes.

14:11:40 <mdupont> with the later the option to traverse links relativly and pull in new nodes.

14:11:56 <mdupont> the anchors are not working in the tables at all.

14:12:08 <mdupont> i tried with a brand new document.

14:12:31 <Seth> we may not be able to put anchors inside tables ... i dont know .. maybe someone who knows html better could tell us

14:12:39 <mdupont> so this will have to be rethought a bit.

14:12:45 <mdupont> I think your right.

14:13:05 <mdupont> hmmm... anyway. I will think about it some more.

14:13:37 <Seth> so we scrap the table ...

14:13:41 <mdupont> I am going to get going soon.

14:14:11 <mdupont> I did do some work with something similar two weeks ago

14:14:21 <Seth> did you want to work on it now or think some more?

14:14:32 <mdupont> and the tables are ok, just have to leave out the outter table.

14:14:50 <mdupont> Well, my girlfriend just came home, so I gotta run.

14:15:07 <mdupont> but I will do some html prototyping and get back to you.

14:15:21 <mdupont> I think a table per node is ok.

14:15:27 <Seth> k, see ya ... only thing to remember if you add a grammar is that you must put it in the list of grammars inside sailor

14:15:41 <mdupont> I dont know what you mean by that?

14:15:58 <mdupont> add a grammer?

14:16:16 <Seth> later we change that so that all grammars put in the sailor/grammar directory are automatically used

14:17:16 <mdupont> you mean html.py is a grammer?

14:17:22 <Seth> yes

14:17:48 <Seth> in the constructor for wake in sailor.py you find the code:

14:17:50 <Seth> self.grammars=['semenglish','quads','html','n3','inputForm','rdf']

14:18:34 <mdupont> ok.

14:18:40 <Seth> to add a grammar, just add to that list .. otherwise the grammar will not be visable and not accessable to the saying(grammar) method

14:18:43 <mdupont> so those are the rules used to generate output

14:18:49 <mdupont> not for parsing?

14:19:18 <Seth> these grammar.py files control both the reading and the saying of that grammar

14:19:54 <Seth> html.py is just an output grammar, so not has no reading implementation

14:19:56 <mdupont> ok, so I removed the outter form and table around the nodes

14:20:01 <mdupont> and the anchors are working

14:20:05 <Seth> :)

14:20:08 <mdupont> ok

14:20:21 <mdupont> so I dont think that table or form are needed.

14:20:38 <mdupont> when I find the time I will update this and send you the files.

14:20:51 <mdupont> Do you want to setup a sourceforge project?

14:21:03 <mdupont> I can host sailor on the introspector cvs if you want....

14:21:25 <Seth> ok, just make them new grammars.py and app.html files .. perhaps where there is a 1-1 correspondance the files should be named the same

14:21:49 <Seth> sure, that would work for me

14:22:10 <mdupont> ok, and you have to check the links...

14:22:43 <mdupont> are you planning on making this a commercial app? are you builing a business around it?

14:22:57 <mdupont> robustai, is that a company?

14:23:11 <Seth> i read in your test.rdf quads file .. it looks good ... but you should call it test.quads ... we will be keying on the .quads to know in which grammar to read in the file

14:23:23 <Seth> yep, robustai is my domain

14:23:34 <mdupont> so you want to make a sailor top-level project on the introspector cvs? I will check in the 0.90 ok?

14:23:48 <mdupont> oooppp.... I should have fixed that...

14:23:49 <mdupont> sorry.

14:24:01 <mdupont> ok, nice chatting with you. Gotta run.

14:24:03 <Seth> yes that would work for me

14:24:08 <Seth> see ya

14:24:09 <mdupont> I think this will work out well...

14:24:30 <mdupont> we can get the perl/python interaction going.... coool.....

14:24:49 <mdupont> you use windows alot?

14:24:59 <mdupont> do you have cygwin installed?

14:27:57 <Seth> i use windows (almost exclusively) and no i dont have cygwin installed .. what is it?

14:29:59 <mdupont> www.cygwin.com

14:30:12 <mdupont> it is the full linux emulation layer for windows.

14:30:27 <mdupont> with perl,python,gcc,apache,cvs,... everthing

14:30:55 <mdupont> I have been porting my applications between the two using cygwin.

14:31:07 <mdupont> so you should install cvs and ssh

14:31:12 <Seth> kewl .. guess ill have to try it ... does sailor run in cygwin without modification ?

14:31:33 <mdupont> yes I think so...

14:31:42 <mdupont> I tried sailor at work,

14:31:54 <mdupont> and it must have been under cygwin...

14:32:00 <Seth> hmmm ...wonder why danbri had problems with it ... oh well

14:32:02 <mdupont> no, i installed from scratch.

14:32:07 <mdupont> no

14:32:22 <mdupont> i installed python native.

14:32:38 <mdupont> but I will try it out under cygwin... later on next week if I find a minute.

14:32:51 <mdupont> but for ssh and cvs cygwin is perfect.

14:32:56 <Seth> so do you end up with two separate file systems ?

14:33:13 <mdupont> no.

14:33:27 <mdupont> the cygwin are native win32 apps with a cygwin.dll

14:33:38 <mdupont> you can call grep and sort from dos

14:33:54 <mdupont> the kernel is still ntos

14:34:13 <mdupont> just the gcc is ported to win32, and bash as well

14:34:25 <mdupont>http://www.cygwin.com/setup.exe

14:34:25 <dc_rdfig> D: http://www.cygwin.com/setup.exe from mdupont

14:34:34 <Seth> which shell do you use?

14:34:37 <mdupont> D:The Cygwin setup exe for win32

14:34:37 <dc_rdfig> added comment D1

14:34:44 <mdupont> I use CMD and bash

14:34:53 <mdupont> depends on how I feel

14:35:03 <mdupont> and i use emacs most of the time, under windows and linux

14:35:13 <mdupont> so... I really gotta go.

14:35:37 <Seth> ok ... so really ... see ya later .. thanks for working with me on this :)

14:35:43 <mdupont> kewl.

14:35:55 <mdupont> I think that we are on a similar wavelength.

14:36:03 <mdupont> this could get interesting...

14:36:05 <Seth> :)

14:36:07 <mdupont> see ya

14:36:12 <Seth> yep

14:47:10 <Seth> anyone want to discuss the idea that we can have free software that where you not have a licence to use that software as a component of a commercial system without negoitiating with the creators of the softare ?

15:10:21 <danb_lap> not really!

15:11:22 <Seth> meaning that you dont ... kewl ... but my question was addressed to anyone

15:12:17 <danb_lap> it doesn't sound very 'free', more like 'cheap'

15:13:13 <danb_lap> GPL'ing it, and allowing commercial folk to get in touch if they want it under a different license... that makes sense. See mysql.com, I think.

15:13:17 <Seth> actually , your prob right, free is the wrong term here ... probably 'open' would be better as in http://opencontent.org/opl.shtml

15:14:18 <Seth> i dont understand the idea of 'a separate licence' how do you exclude sale and still make it free ?

15:15:45 <Seth> look my intent here is clear .. im just tring to find the best expression of that intent legally ... its open people can use it and distribute it for free, but if they charge money for it, then i want to negoitiate a royalty ... hwat licence expresses that ?

15:16:28 <danb_lap> you don't exclude sale. You exclude people selling it on under non-GPL license...

15:16:36 * danb_lap doesn't know a license that does quite what you want, sorry

15:18:19 <Seth> i dont undestand your point, the licence i am looking for does exclude selling it. Of course i want to encourage people to sell it, i just want my cut when they do.

15:20:23 <danb_lap> I meant, on the model I sketched, you couldn't exlude sale, but could put conditions on how was sold

15:20:39 <danb_lap> good luck anyway!

15:21:42 <Seth> k

15:29:16 <xover> Uhm, Seth, with GPL people _can_ sell it, but nobody would as they _must_ give the source and so nobody would buy it. For anyone to be able to sell it they would need to negotiate other licensing terms with you. Terms that might include giving you a cut.

15:29:34 <xover> Isn't that what you're after?

15:32:14 <Seth> xover, but the source will always be open

15:32:50 <Seth> in that I will be disclosing it, so that is not possible with this senario

15:33:26 <xover> Hmmm. One of us is not understanding the other, I think.

15:34:03 <xover> You want to release something to the community for free; source and modification rights, etc. Right?

15:34:15 <Seth> right

15:34:46 <xover> You also want to allow people to take that code and sell it, you just want to make sure that if they do you would get royalties?

15:34:56 <Seth> yes.

15:35:14 <danbri> seth, see http://www.mysql.com/support/arrangements.html for an example

15:35:24 <xover> Ok. SO you release as GPL. "The community" gets all the freedoms you want them to have etc.

15:35:41 <xover> A company comes along and wants to sell it as a commercial app...

15:36:04 <xover> ...since the license is GPL they can do that, but then they have to give everyone their source...

15:36:31 <xover> ...meaning few or none would buy it, making that impracticable or just plain uneconomical.

15:36:40 <Seth> oh i see your point.

15:36:52 <xover> IOW, GPL wouldn't legally prevent anyone from selling it, but it would make it stupid to try it.

15:37:17 <xover> Sooo, they need to get in touch with you and negotiate alternate licensing terms.

15:37:38 <xover> These terms can be _anything_ you want them to be, and affects the public GPLed code not at all.

15:37:59 <xover> As danbri sez, this is more or less what MySQL AB is doing.

15:38:01 <Seth> but what if the company doesnt care about disclosing their soruce ? .. say they are doing same as me . Your making too many assumptions about how the market place may evolve.

15:38:48 <xover> Well, what's then the difference between "the community" getting their free(dom) code from the company and them getting it from you?

15:38:50 <danbri> then its a pyramid scheme... eventually somewhere down the line, money's changing hands and theres a business model...

15:39:07 <xover> They get the freedoms either way.

15:39:42 <Seth> woops .. i cant follow that last point.

15:39:48 <xover> The point being, this situation is an assumption about the market, but one I think that is unlikely to change.

15:41:55 <Seth> well im talking about the assumption that seeing the code is going to be the prefered way to operate in the future. protecting your investment by not disclosing the code is not going to be a viable alternative.

15:43:01 <Seth> but yes, i can see how your model works, were we to make your assumption.

15:43:51 <xover> There is no way you can have your cake and eat it too. If you allow someone to modify and redistribute you cannot prevent them from also selling the service of packaging or providing custom modifications. Then they aren't charging for your code, they're charging for services rendered.

15:46:09 <Seth> good point. if i allow modify and redistribute, then i must accept that modification as either the same package or is a new package. if new, then i have nothing to say about it, and in that case i have just provided ideas. if same, then i get my cut.

15:46:41 <Seth> so each module is considered as a separate thing.

15:48:01 <xover> Hmmm. What I was thinking of is: I need OpenSP to have feture foo. I hire someone to take the GPL OpenSP source and add feature foo. The programmer didn't sell the OpenSP code; s/he sold the _service_ of modifying it.

15:48:11 <Seth> if someone whats to steal the thing by non sustatantive modifications, well let them have it. This race is to the swift and to those who cooperate and dont try to screw others ... i think we will be able to tell which people belong to which group and decide how to interact with them accordingly.

15:48:58 <xover> Exactly! Let evolution sort out the cheap knock-offs.

15:49:29 * danbri thought that was a cheap crack at Evolution (the Outlook knock-off) :)

15:49:39 <danbri> (assumes it wasn't...)

15:49:58 <xover> AH, but Evolution _is_ evolution at work. :-)

15:50:26 * danbri grins

15:50:31 <Seth> xover, i agree ... survival of the fittest .. and being uncoooperative will be a disadvantage in this new economics

15:51:56 <Seth> xover, are you now agreeing with my basic proposal of : open source, but non commercial without negoiting royalty?

15:52:49 <xover> You asked for a license to cover needs A, B, and C. I suggested GPL would do that. Not sure I "disagreed" with anything in there?

15:53:36 <Seth> But GPL says you can sell it without paying royalties, doesn't it?

15:54:16 <xover> Yes, but it also says you _must_ give up source if you do, so it is nearly impossible to do that.

15:54:37 <xover> For the simple reason that if they implement any desireable features, you can just get their source and add it to the free version.

15:54:56 <xover> There is no value-add they could possibly make in pure code that they could convince anyone to pay for.

15:55:25 <Seth> yes, we covered that .... you assume that not giving up the source is a viable way of doing business in such a community.

15:55:26 <xover> The only thing they can sell are _services_ surrounding it. Help people install it, help people develop new features, etc.

15:56:14 <xover> I think I'm missing your point.

15:56:25 * danb_lap ponders a new name for rubyrdf (seems unfair taking such a generic name, given my tools are somewhat kooky and non-production-grade)

15:56:25 <Seth> so your are kind of tricking them with this assumption .. well they can just get around that ... it muddies the water and makes the logic too complex.

15:57:18 <xover> Hmm. I still don't see the problem. WHat is the scenario you are worried about? (assuming you have released as GPL)

15:57:27 <Seth> xover, i agree with you if your assumption holds water ... but i claim that it does not.

15:57:44 <xover> Which assumption?

15:58:29 <Seth> that people will always protect their investment in software by hiding the source.

15:59:58 <Seth> im changeing that to: protect your investment in software by always collecting royalties when others sell it.

15:59:59 <xover> No, that's not my assumption.

16:00:24 <xover> But isn't that what you want to do?

16:00:42 <Seth> well, then remove that assumption and tell me how a GPL will give me what i want

16:01:05 <Seth> .. dont forget im a little slow with binary logic

16:01:22 <xover> Ok. You publish Foo as GPL.

16:01:28 <Seth> ok

16:01:35 <xover> The community uses it, feeds you patches, all is well.

16:01:41 <Seth> ok

16:01:45 <xover> I come along and decide I want to make money off it.

16:01:53 <Seth> kewl

16:02:21 <xover> So I hack up a fancy new Frobnicator for it, label it "Xover's Great Frobnicated Foo", and offer it up for sale.

16:02:58 <Seth> does Frobnicator have any of my modules within it?

16:03:09 <xover> Since the code is GPL, I _must_ release my source. You get the source for the Frobnicator, add it to plain old Foo, and the two products are exactly equal again.

16:03:40 <xover> Frobnicator _is_ your source, only with 30 extra lines in one of the source files (or something like that).

16:04:22 <Seth> well under GPL it is free, you can do with it what you want, even sell it, then i cannot come to you and ask for a royalty

16:04:37 <Seth> i want to ask for royalty

16:04:42 <Seth> so i cannot do it that way

16:04:46 <xover> My assertion, or assumption, is that nobody will actually want to pay for that. Why would they pay for "Xover's Frobincator" when they can get the _exact_ same code for _free_ from you?

16:05:29 <xover> In other words, if _I_ want to be able to make money from that, I have to go to you and negotiate a _new_ license. One that does _not_ require me to release the source back to the free version.

16:05:41 <Seth> maybe you forgot that i *want* you to sell it and make money on it.

16:05:51 <Seth> i just want my cut

16:06:06 <xover> Yes, and the above prevents that, how?

16:07:02 <Seth> i fail to see why you would have to come to me to negoitate a new licence, if it is gpl, you already have your licence.

16:07:49 <Seth> your senario only happens if you try to hid that fact by hiding the code.

16:08:04 <xover> No.

16:08:22 <xover> GPL lets me sell your code. Period.

16:08:36 <Seth> well what forces you to come to me to negoitate a new licence ?

16:08:46 <xover> Nothing.

16:09:08 <Seth> so why would anybody ever pay me a royalty?

16:09:18 <xover> _Unless_ I want to do something which the GPL won't let me do.

16:09:39 <Seth> like forinstance?

16:10:08 <xover> Closing up the source, adding a "killer feature", and selling it as a commercial product.

16:10:53 <xover> (or linking _non_-GPL code with it; code from an existing, commercial, system)

16:11:45 <xover> My base premise here is that you _want_ "the community" to get the code for free, and be allowed to modify and redistribute it. Did I get that wrong?

16:12:07 <xover> IOW, someone selling it is an _exception_ from the norm.

16:12:29 <Seth> well gpl forcs you not to close the source, but does allow your to add killer feater and sell ... it comes down to the point of the assumption that closing source is viable in the marketplace .. and that closing source will prevent me from knowing that you are just selling my system without paying me for it. i still dont see your point.

16:12:59 <Seth> no your got that ritht .. i want community to modify and redistribute.

16:13:25 <xover> It allows me to add killer feature and sell, but it also requires me to give you back the source for my killer feature.

16:13:40 <Seth> so?

16:13:56 <xover> That way, my version doesn't have anything that your version does. But my version costs money, which your version doesn't.

16:14:13 <xover> Why would anyone pay for mine when they can get yours for free?

16:14:16 <Seth> maybe you dont care about disclosing the source of your killer featere. your whole argument keeps hinging on that assumption.

16:14:29 <xover> No.

16:15:23 <xover> Take your source three and make an _exact_ copy of it. Burn each of the two on their own CD. Label one "Free" and one "$50". Place them in a store and see which version people will get.

16:15:49 <xover> Thetwo are identical. Why would anyone pay for what they can just as easily get for free?

16:15:57 <Seth> well coulr you state again how you remove that assumption and still force a company to negoitate my cut ?

16:17:13 <xover> Maybe Apple comes along and wants your Foo in Mac OS X...

16:17:16 <Seth> xover, i conceed that last point .. but it is irrelivant to what i am saying. that senario would not happen, obviously .. well actually it does happen all the time, i see stuff in the computer store i can get free on the internet.

16:17:53 <xover> Apple doesn't relase stuff as GPL. They release stuff under the Apple Public License or somesuch.

16:17:57 <Seth> well with gpl, apple can do that. with my 'cant sell withth giving seth his cut' does not allow apple to do that.

16:18:39 <xover> To enable Apple to use Foo they would have to negotiate a license from you that lets them release their own version under AFPL.

16:19:10 <xover> IOW, the code is still open ("Open Source"[tm], even), but they had to negotiate royalties with you to do it.

16:19:18 <Seth> well yes, ok i see that, because apple is not disclosing their source.

16:19:24 <xover> Yes they are!

16:19:39 <Seth> but me thinks that is only one segment of the market here.

16:19:46 <xover> The whole underpinnings of Mac OS X are Open Source ("Darwin").

16:20:18 <xover> The rest of the "segment" are closing their source to "protect their investment"! But you insisted that I not consider that in my argument!

16:20:59 <Seth> is mac os gpl?

16:21:06 <xover> No!

16:21:22 <Seth> what's the salient diff between mac os and gpl ?

16:21:33 <xover> Mac OS X is proprietary code on top of an Open Source underpinning.

16:21:49 <xover> The Open Source code is not GPL, but a custom Open Source licence.

16:21:57 <Seth> and the proprietary code is not open?

16:22:02 <xover> No.

16:22:11 <Seth> it is open?

16:22:19 <xover> No, it is not open. :-)

16:22:30 <xover> (sorry about that)

16:23:09 <Seth> can others than apple sell mac os ?

16:23:27 <xover> Not Mac OS, no, but they can sell "Darwin".

16:23:44 <xover> "Darwin" is the Open Source underpinnings for Mac OS X.

16:24:02 <xover> It runs on Macs but also on PCs (x86).

16:24:14 <xover> "Darwin" started out as pretty much jusy FreeBSD.

16:24:51 <xover> Apple explicitly does not want any GPLed code in there because of it's so-called "viral" properties.

16:25:28 <Seth> so your saying that mac os could not embedd a sailor agent in their proprietary extension because then such a embbedding would desgtroy appel's business model and so that they would come to me and negoitate a licence ... is that your point?

16:25:48 <xover> Yes! :-)

16:27:04 <xover> Nobody can close the source without paying you a fee. And anyone not wanting to use GPL, even if it's still open source, will also have to pay you a fee.

16:27:09 <Seth> fine ... then gpl would work for me in this kind of senario ... im saying that is just too tricky for me ... i want to be clearer ... i dont view gpl as a kind of religion that i must incant

16:27:45 <xover> The GPL is about as pragmatic as you can get! The FSS is pure dogma, but the GPL is just legalese!

16:27:50 <xover> s/FSS/FSF/

16:28:30 <xover> Legalese is complicated and meaningless, but them's the breaks. The world is (unfortunately) like that and no license can get around that.

16:28:46 <Seth> and i dont believe that in the future that kind of senario will happen that much .. i envision a future where all code is open and viewable ... that is what i want ... i just want there to also be a straight forward way to make money by publishing good systems that people will want to sell

16:29:26 <xover> You are making a mistake in your assumptions here, I think. ...

16:29:57 <xover> ... If you can get the product for free you won't pay for it. BUT yuo may want to pay for a value-add service. ...

16:30:13 <xover> ... such as packaging, support, custom modifications, etc. ...

16:30:46 <Seth> well the assumption is that it is viable to assembel components of free stuff put it together and sell it as a package ... hide it or dont ... most of the market will buy it instead of assembeling theri own form the free components

16:30:53 <xover> ... and I don't think there is any loophole in the current Intellectual Property laws that will _both_ let you release it for free _and_ prevent someone from providing such services.

16:32:54 <xover> assembly: Yes, but then you aren't selling the _code_, you are selling the _service_ of packaging it. And you cannot prevent that without also preventing "the community" from using it (fully).

16:33:24 <xover> If you want to prevent packagers like that (without paying a fee) you must also limit what "the community" are allowed to do with it.

16:33:32 <xover> IOW, you;re requirements are incompatible.

16:34:57 <Seth> my rstriction is on copying the code. dont forget im not using gpl.

16:35:30 <xover> Well then the community also are not allowed to copy the code.

16:35:44 <Seth> you can copy the code all your want, you cannot sell a copy of the code.

16:36:17 <xover> Hey Seth? I'll give you $50 bucks if you download Sailor and install it on my machine, ok?

16:36:33 <xover> Did any of us violate your licens?

16:37:11 <xover> (there should probably have been "-marks around that thought experiment above) :-)

16:37:31 <Seth> shucks, though i just made $50

16:37:38 <xover> :-)

16:38:17 <xover> The point is, if you allow me to make a copy for myself, you can't prevent me from charging people money to make that same copy for them.

16:39:26 <Seth> good question though .. im torn between saying that your labor involved in installing is your own and i dont fairly get a cut of it ... and for the $50 that would be fair. Now if you charged $50,000 for the same labor, you would be cheeting yourself but then i would want my cut.

16:39:53 <xover> Yes, but current law does not allow you to make that distinction.

16:40:03 <Seth> i mean you would be cheeting your client

16:40:24 <Seth> and your business model would fail because of it.

16:40:24 <xover> (I assume you understand that "IANAL" and you should consult a real lawyer if you... blah blah)?

16:40:39 <Seth> yep

16:41:25 <Seth> im not trying to prevent people from adding extra services to sailor and charging whaever the market will bear .. and im not thinking that i need a cut of that.

16:42:08 <Seth> where extra services = packageing ... well yep that i guess is a problem .. dont know how to fix that one

16:43:35 <xover> I can sell my labour at whatever price I think I can get people to pay for it. You can never get a cut of that, no matter what.

16:43:49 <Seth> agree

16:44:28 <Seth> but in the marketplace you actually cannot sell it for whatever large sum you want, because nobody will buy it

16:44:54 <Seth> me thinks that is beside the point here

16:45:53 <Seth> it is packageing as labor that breaks my licence here ... i just need to distinguish between added value and packaging the same value

16:45:55 <xover> Linux + Sendmail + Smap (all OpenSOurce) netted a local consultant company 16 hours at $300 and hours plus travel expenses.

16:46:35 <Seth> so?

16:46:55 <xover> SO the amount charged can be pretty damn steep if the labour is sufficiently valuable to you.

16:47:14 <Seth> yes i agree .. like i said .. im not trying to get a cut of that.

16:47:57 <xover> That leaves people modifying or adding to the source and selling the result. Right?

16:48:21 <Seth> if Borg Gates puts sailor in his operating system and sells his operating system .. that is just his packageing of my system and i want my cut.

16:48:34 <xover> No!

16:48:49 <Seth> right on the modifying .. but we go module by module

16:49:23 <Seth> so you say bill will claim he is just adding packageing labor ?

16:49:29 <xover> If Sailor is GPL, anything it touches is in serious danger of also becoming GPL (the "viral" clause).

16:49:41 <xover> Thus BorgOS would also end up GPL.

16:49:54 <xover> I suspect he would find that a wee bit difficult to sell. :-)

16:51:24 <xover> Ok. Say Microsoft packages Sailor with the next version of Windows.

16:51:27 <Seth> hmmm ...so your saying this viral clause has actually been effective in collecting money form large companies ?

16:51:52 <xover> Microsoft and Apple won't touch GPL with a ten foot pole because the GPL looks like it's unbeatable!

16:52:02 <xover> (never tested in court though!)

16:52:20 <xover> Say Microsoft packages Sailor with the next version of Windows.

16:52:41 <xover> If Sailor is GPL, they must also provide the source for it.

16:53:03 <Seth> but that is just the opposite effect than i want ... what i want is for ms to package sailor with the os and give me my cut .. not that they will just rewrite it and give me nothing.

16:55:16 <xover> First of all, they can't "rewrite" it or modify it etc. and get around the license that way. They must _reimplement_ it. From scratch. In a clean-room. Without looking at the Sailor source.

16:55:46 <Seth> whatever ... they certaily have the resources to do that

16:55:54 <xover> If they so much as load the Sailor home page you have grounds enough to sue them for Copyright infringement (MS uses this tactic itself).

16:55:55 * danb_lap_ wanders past

16:56:06 <danb_lap_> you two still going!

16:56:18 * danb_lap_ wonders what the conclusion was

16:56:45 <xover> But if you assume they have the resources to do that, why would they ever be interested in giving you a cut? Why would they ever be interested in using Sailor?

16:56:48 <Seth> not yet arrived at .. im getting good feedback here form xover .. for which i am very gratful

16:57:03 <danb_lap_> :)

16:58:10 <Seth> again, yes if that is the kinhd of company they run, then the will not ... in that world, i loose ... my assumption is that is not the only possible world

16:59:05 <xover> Yes. So lets go back to my little, small-time, company "A. Fawlty SOftware Co.". :-)

16:59:25 <Seth> ok ...

16:59:34 <xover> What exactly does Sailor do, BTW? :-)

17:00:25 <Seth> ... see http://robustai.net/sailor/ .... its a software agent that reads and writes on the semantic web and is a personal information tool

17:00:53 <xover> Ok, say I have a PIM ala. Evolution/Outlook or something like that.

17:01:07 <Seth> k

17:01:43 <xover> I have some Open Source parts (the HTTP, SMTP, etc. bits say) and some Propriatry parts (the Exchange connector, the Access DB file acces code, etc.)

17:02:30 <xover> To enable this I use a custom Open SOurce license -- Not GPL -- for my Open Source parts. (otherwise likning the proprietary stuff against it would make that code GPL too).

17:02:48 <xover> Now I want to add Sailor into this product.

17:03:01 <xover> The product retails for ~$200.

17:03:22 <Seth> ok

17:03:44 <xover> If Sailor is GPL, I can't add it as is to my code because then all of it would end up as GPL and my business model evaporates.

17:03:56 <xover> So what do I do?

17:04:37 <xover> Well, I /could/ forget about Sailor and implement the features I want from scratch.

17:04:53 <xover> But I'm not Microsoft. I don't have the resources to do that.

17:05:16 <xover> Hmmm. How can I solve this quandry?

17:05:21 <Seth> so you negoitiate :) .... your senario works in this case.

17:05:49 <xover> The same goes for something that is semi-open source, or completely proprietary.

17:06:33 <Seth> and your saying that i would never get the ms account in any case?

17:06:46 <xover> Pretty much, yeah.

17:07:25 <xover> The Windows 2000 TCP stack is heisted from NetBSD, BTW.

17:07:41 <Seth> and ...

17:08:01 <xover> Nobody got paid for that because the license let anyone do whatever they wanted with the code.

17:08:42 <xover> GPL code OTOH and so anathema at Microsoft that you'd better not even think of going near it on your time off or you risk reprisals.

17:08:43 <Seth> so bill would have rewritten it from scratch but gpl just gave it to him

17:09:12 <xover> No, GPL would have stopped him. The BSD License (Open Source, but not GPL) let him get away with it.

17:09:21 <Seth> oh

17:09:34 <danb_lap_>http://rdflib.net/cvs/rdflib/syntax/parser.py

17:09:35 <dc_rdfig> E: http://rdflib.net/cvs/rdflib/syntax/parser.py from danb_lap_

17:09:56 <danb_lap_> E:Python RDFLib RDF/XML parser

17:09:56 <dc_rdfig> added comment E1

17:10:37 <Seth> i use that :)

17:10:50 <xover> The GPL can basically be summed up as: "These are the freedoms we want _everyone_ to have. The only freedom you do not get, is the freedom to take away the freedoms of _others_"

17:11:24 <xover> Te corollary is, that if you want to curtail someone else's freedoms you have to negotiate an alternate license.

17:11:59 * danb_lap_ wonders whether this might be even more on-topic for #gnu...?

17:12:06 <Seth> and so you rely on the greed of the big corportations to want to lock up their stuff to prevent them from using your stuff.

17:12:30 <xover> Yes.

17:13:04 <Seth> im thinking that if you say ... look come to me and talk money and you get what you want, they will be more prone to do buisiness with you.

17:13:15 <xover> Yes.

17:13:52 <Seth> so it dont fragment the world into feee by guys .. and greedy guys ... it's clear in both domains

17:13:52 <xover> SO your README says: "Alternatelicenses are avilable, everyone else gets the GPL (see file LICENSE for details)"

17:14:38 <Seth> hmmmm ... that last one is starting to look very attractive :)

17:14:58 <xover> BTW, both the FSF and Eric Raymond have written very lengthy treatises on this.

17:15:41 <Seth> i know .. im just now getting into this area because its first time in a real long while that i got something to give and or sell

17:16:16 <xover> You need to decide whether what you want is to "share" or to "make money".

17:16:33 <xover> If you decide on "share" then any money you make is probably accidental and pleasant surprise.

17:16:44 <xover> And coversely for the other option.

17:16:58 <Seth> i want to do both ... i dont accept binary logic as viable here

17:17:17 <xover> It's not binary; it's a question of dominant attribute.

17:17:57 <Seth> my dominant attitude is: share ... but my better half keeps going for the bucks

17:18:14 <bijan> Anyone have favored RDF tutorial material?

17:19:17 <mdupont> hey there

17:19:34 <Seth> anyway .. thanks for your help, xover, ... am leaning twards the "Alternatelicenses are avilable, everyone else gets the GPL " case

17:19:47 <Seth> hi mdupont

17:19:51 <xover> Glad to hear it, and glad I could be of some help!

17:19:57 <mdupont> Hey Seth

17:20:04 <mdupont> Just had a nice nap

17:20:30 <Seth> been discussing licencing ... so let me ask you a question .. is your stuff all gpl ?

17:22:32 <mdupont> yes

17:23:01 <mdupont> I am looking into what if a program dumps XML and how that can be also treated as linking as well

17:23:32 <mdupont> It is important that people just dont dump to some other format and then say "Oh its not GPL any more"

17:24:14 <xover> Is it an "API" or a mere "data format"?

17:24:31 <Seth> i think were talking different gpl here .. i was talking about the GPL licence for free software

17:25:56 <mdupont> errrrr the GNU public license

17:26:33 <Seth> what's the exact url of the liceence that xover has been recommending ?

17:26:34 <mdupont> I am talking about people using you software, and linking to it via XML and making a derivitive work

17:26:56 <mdupont> and then saying "Oh, I dont have to aggree to the GPL, because..... its not linking"

17:27:28 <Seth>http://www.gnu.org/licenses/gpl.html

17:27:29 <dc_rdfig> F: http://www.gnu.org/licenses/gpl.html from Seth

17:27:29 <bijan> Folks, there are established fora for licence discussions.

17:27:41 <mdupont> yup

17:27:42 <bijan> With people who have worked through this stuff before.

17:27:44 <danb_lap_> #gnu comes to mind

17:28:06 <mdupont> the best place is gnu.misc.discuss and fs-licensing mailling list

17:29:52 <danb_lap_> well that killed the mood!

17:30:02 <danb_lap_> what shall we talk about now? :)

17:30:07 <mdupont> I have been looking into a new recursive end user license called RGPL

17:30:18 <mdupont> Recursive GPL

17:30:33 <Seth> F:| GNU General Public License

17:30:33 <dc_rdfig> titled item F

17:30:34 <mdupont> that covers the usage of the XML data as linking.

17:31:12 <mdupont> Is there anyone who used the RDF perl libraries?

17:32:04 <danb_lap_> which ones? there are a few now...

17:32:11 <Seth> danb and bijan, look the people i wanted to discuss this with are all here .. the bandwidth of this channel will justify this kind of discussion .. were it to become much more lengthy we will take it off channel .. but for now please extend a bit of tolerance to what other people want to discuss here .. ok?

17:32:29 <mdupont>http://sourceforge.net/cvs/?group_id=19878

17:32:31 <dc_rdfig> G: http://sourceforge.net/cvs/?group_id=19878 from mdupont

17:32:33 <bijan> No.

17:32:43 <mdupont> G:| Introspector CVS acces page

17:32:43 <dc_rdfig> titled item G

17:32:44 <bijan> Especiallya s you could have started up a scratch group for it.

17:32:45 <mdupont> G:| Introspector CVS access page

17:32:45 <dc_rdfig> titled item G

17:33:08 <mdupont> Most people rightfully hate licensing disussions by laymen

17:33:24 <Seth> im gonna ignore bijan's intolorance

17:33:26 <mdupont> the BuggerOff license it the right thing for ending the discussion

17:34:02 <bijan> Fine

17:34:06 <xover> bijan: Great idea. Pity you didn't come up with it a little sooner....

17:34:25 <danb_lap_> I'm in two minds. People shouldn't live in fear of being deemed offtopic. But discussions of GPL etc can rathole and retread old ground. Setting up a quick side-channel eg #sailorgpl is cheap and cheerful

17:34:44 <danb_lap_> ...and means those who skim the chatlogs for RDF stuff get what they came for.

17:35:11 <xover> "fear of"?!? Just a polute pointer to buggar off to a private channel would have been fine. No "fear" necessary.

17:35:16 <Seth> you forget that were i to have done that i would not have pulled xover into the conversation

17:35:19 <xover> s/polute/polite/

17:35:21 <mdupont> danb_lap_: I aggree.... Seth why dont you setup a temp channel, or we goto sailor and have this talk?

17:35:56 <Seth> fine, anyone who wants to continue to discuss licencing is welcomed into #sailor

17:35:59 <xover> Seth: We could have started here and moved out-of-band when it looked like it would drag out.

17:36:01 <danb_lap_> I mean, I don't want people to feel over inhibited. But then again, we're supposed to be RDFish. Temp channels are easy.

17:36:14 <danb_lap_> thanks guys, appreciated!

18:02:29 <Seth> danb must feel very proud .. he has this channel all to himself

18:04:43 <larsbot> and Seth must feel very badly treated :)

18:04:54 <Seth> yep :(

18:07:25 <danb_lap_> aww, c'mon

18:08:26 <danb_lap_> oh hey, Guha's back

18:10:24 <Guha_> hi danbri

18:11:38 <danb_lap_> hi there

18:17:49 <mdupont>http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/introspector/sailor/

18:17:50 <dc_rdfig> H: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/introspector/sailor/ from mdupont

18:18:18 <mdupont> H:| The HTML access to the Sailor source code in the introspector cvs repository on sourceforge

18:18:19 <dc_rdfig> titled item H

20:00:01 <Seth> why does the link to the lates logs at http://rdfig.xmlhack.com/ still go to http://ilrt.org/discovery/chatlogs/rdfig/2002-07-10.html ?

20:05:53 <dajobe> it does not

20:07:20 <Seth> fine ... it still redirects to the url above instead of the actual logs

20:07:33 <Seth> actual latest log

20:07:34 <dajobe> it does not

20:07:50 <Seth> it does on my client

20:07:56 <dajobe> so you are broke

20:08:14 * jordan is broke

20:09:22 <Seth> just tried it in mozilla and in ie .. same redirect to 7/10 logs

20:09:34 <Seth> its now 7/13

20:09:49 <dajobe> It's my software, I know what it does, it does always go to today and it does now.

20:10:04 <Seth> dajobe, did you try it right now?

20:10:09 <dajobe> of course

20:10:45 <Seth> well i wonder why its performing differently on my system two differenct browsers and on jordan's too

20:10:47 * xover wishes he was equally certain his software had no bugs... :-)

20:11:35 <dajobe> telnet ilrt.org 80

20:11:35 <dajobe> Trying 137.222.16.8...

20:11:35 <dajobe> Connected to ilrt.org.

20:11:35 <dajobe> Escape character is '^]'.

20:11:35 <dajobe> GET /discovery/chatlogs/rdfig/latest HTTP/1.0

20:11:35 <dajobe> HTTP/1.1 301 Moved

20:11:37 <dajobe> Date: Sat, 13 Jul 2002 20:11:29 GMT

20:11:39 <dajobe> Location: http://ilrt.org/discovery/chatlogs/rdfig/2002-07-13.html

20:11:45 <dajobe> anything else is *your* problem

20:12:23 * Seth is baffeled

20:12:33 * dajobe gives up

20:12:58 <Seth> jordan, did you get the same result that i do ?

20:13:15 <xover> Are you sure that wasn't just local cache issues Seth?

20:13:24 <xover> (worked fine all along for me)

20:13:52 <Seth> well i reloaded the page ...

20:14:44 <Seth> hmmmm ...now it works magically

20:15:20 <Seth> but not in mozilla

20:16:18 * Seth kills mozilla trys again

20:17:45 <Seth> well mozilla must have some wierd caching prob here

20:20:55 <xover> Hmmm. Well... "301" is "Moved Permanently". Mozilla may be deciding that the redirect really is permanent and pointing to whatever it was the first time you followed it.

20:21:19 <dajobe> hmm

20:21:27 <dajobe> ok, I'll 302 it

20:22:01 <dajobe> done

20:22:53 <xover> Well, still works here so at least that didn't break anything. :-)

20:23:23 <Seth> still doent work here with mozilla 1.1a

20:23:54 <dajobe> probably your ISP and some stupid cache

20:24:39 <Seth> well it worked in ie after i reloaded the page, but not in mozilla

20:25:07 <deltab> tried shift-reload?

20:26:07 <Seth> just did .. same result

20:34:31 * xover queries Bugzilla...

20:34:43 <xover> Hmmm. Perhaps this is relevant: http://bugzilla.mozilla.org/show_bug.cgi?id=142252

20:35:01 <dajobe> I've been using moz only, so WFM

20:35:39 <xover> Yeah, but thing apparently depends on Moz's cache behaviour is to "Verify cached pages: Never".

20:36:18 <xover> It defaults to "Auto" IIRC, but Seth may have changed it.

20:36:57 <dajobe> I see I should avoid looking at Seth's software till he decides a license

20:37:41 <xover> Yes, always good to be certain of the license terms before you "taint" yourself with looking at code. :-)

20:37:44 <dajobe> yes

20:38:03 * xover stays well clear of any information provided on *.microsoft.com...

20:38:15 <dajobe> that goes for open/free as well as closed/proprietary/...

20:38:37 <dajobe> yes, so the links in the chump today might be bad for some people to look at

20:38:57 <dajobe> arg, bare URLS

20:38:58 <dajobe> C:

20:38:58 <dc_rdfig>http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/introspector/daml/test.rdf?rev=1.2&content-type=application/text

20:38:59 <dc_rdfig> (1:mdupont) is the latest test quads

20:39:00 <dc_rdfig> (2:mdupont) checked into cvs on sourceforge

20:39:04 <dajobe> C:|please title me

20:39:04 <dc_rdfig> titled item C

20:39:19 <dajobe> D:|cygwin setup for win32

20:39:19 <dc_rdfig> titled item D

20:39:24 * xover looks fearfully over his shoulder for danbri... :-)

20:39:28 <dajobe> what's that got to do with rdf...

20:39:51 <dajobe> dc_rdfig: view 1

20:39:51 <dc_rdfig> H: The HTML access to the Sailor source code in the introspector cvs repository on sourceforge (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/introspector/sailor/)

20:40:01 <dajobe> H:license not yet clear, Seth still thinking about it

20:40:01 <dc_rdfig> added comment H1

20:40:41 <jordan> re earlier comment: "and on jordan's too" ? I just meant that I was broke.

20:41:59 <dajobe> surely A: and B: are rather usless too

20:42:12 <dajobe> nevermind

20:42:21 <xover> bare URLs: I looked into setting up chump the other day, and the first thing I did was to add a prefix to URLs for them to be recognized.

20:42:37 <dajobe> that's what #swhack does

20:44:44 <xover> BTW, does anyone know if logger's source is avilable?

20:44:50 <dajobe> lol

20:44:57 <dajobe> yes, since I wrote it

20:45:03 <xover> :-)

20:45:11 <dajobe> .google irc chat logger

20:45:12 <datum> irc chat logger: http://ilrt.org/discovery/chatlogs/rdfig/help.html

20:45:16 <dajobe> darn

20:45:23 <dajobe> close

20:46:08 <xover> I wasn't able to find it from either of the RDFIG log pages or the ILRT main page.

20:46:09 <dajobe> see http://cvs.ilrt.org/cvsweb/redland/logger/

20:46:25 <dajobe> or the w3c version at http://dev.w3.org/cvsweb/2001/sw/logger/

20:46:52 <dajobe> and it's GPLed

20:47:02 <xover> Ooooo. In Perl even? /me *likes*! :-)

20:47:09 <dajobe> of course

20:47:19 <dajobe> yeah, pointers are missing, sorry

20:53:25 <dajobe> ok, added some

20:55:31 <xover> Ah. Much better! :-)

20:57:10 <sbp> <dajobe> that's what #swhack does

20:57:10 <sbp> ?

20:57:22 <dajobe> add a prefix to lines to prevent accidently chumping

20:57:24 <sbp> oh, chumpy, yeah

20:57:25 <dajobe> # isn't it?

20:57:35 <dajobe> or is that [off]?

20:57:35 <sbp> chumpy's been down for ages and ages

20:57:58 <sbp> Aaron made the chumping syntax "@ URI"

20:58:05 <dajobe> aarons blogspace server still down

20:58:06 <sbp> the [off] thing is for loggy

20:58:09 <dajobe> yeah

20:58:10 <sbp> yep

20:58:29 <dajobe> I wanted to make [off] so annoying to type, people wouldn't use it, but would keep things OT and in public ;)

20:58:32 <sbp> I'm surprised that he didn't just move the chump bot to his other machine, and he's not around now... oh well

20:58:37 <sbp> heh!

20:58:38 <dajobe> OT=on topic I meant

20:58:41 <xover> But "#" would make more sense then "[off]", or even "//" for the C++ fans...

20:59:01 <sbp> we used # for a long time, until Aaron couldn't be bothered to keep maintaining his own tree

20:59:04 <sbp> (so he says)

20:59:22 <deltab> it was annoying when code was pasted

20:59:47 <sbp> yeah...

20:59:49 <dajobe> exactly

20:59:53 <dajobe> that's why we aren't changing it here

20:59:55 <dajobe> QED

20:59:58 <sbp> we could have used "!" though

21:00:11 <sbp> ! works quite nicely, IMO

21:00:13 <niq> howbout a user-command to change a bot's magic char on the fly?

21:00:25 <xover> Strong connotations of "not", yes, I like it.

21:01:31 <xover> Per user is probably too much overhead, but easier configurability to avoid having to maintain separate code trees is probably a good idea.

21:02:59 <niq> could be a privileged command, but say I want to cut&paste shell or perl using #, I politely request a privileged person to switch it to ";" (say)

21:03:35 <dajobe> I think this is all over-engineering :) we get by OK mostly

21:03:53 <xover> Heh heh, or make it recognize "[[[" and disable it until it's properly closed? :-)

21:04:04 <niq> but it's b***** annoying to be chumped without asking for it

21:04:18 <niq> [[[

21:04:28 <niq> who has the privilege to close that?

21:04:38 <sbp> ]]]

21:04:38 <sbp> yes, I think "@ URI" would be a welcome change to the chump's syntax. call edd!

21:04:50 <dajobe> it's a bad idea to rely on the state of bots, I think - things can be missed

21:05:06 <dajobe> how much should the bot keep before moaning about [[[

21:05:07 <dajobe> etc.

21:05:37 <niq> yeah, better to use per-line char(s)

21:05:42 * xover is shocked that anyone thought he was serious!

21:06:21 * niq is shocked that xover thought he was ... nevermind