IRC log of swhack on 2002-02-27

Timestamps are in UTC.

00:04:52 [GabeW]

GabeW has quit ("Client Exiting")

00:22:55 [bijan]

Ouch.

00:23:44 [bijan]

Hm. If I go with your interpretation of query...my life becomes vastly easier.

00:29:05 [bijan]

Sean, are you around?

00:29:12 [bijan]

* bijan askes again.

00:29:40 [sbp]

yes, I am

00:29:54 [sbp]

I'm getting bored driving up and down the same old strip

00:30:29 [bijan]

Hmm?

00:30:56 [sbp]

you asked if I was around...

00:31:21 [bijan]

driving up and down?

00:31:27 [bijan]

Anyway, if I have the following:

00:31:28 [bijan]

:bob :parent :jane.

00:31:28 [bijan]

:jane :parent :janeII.

00:31:28 [bijan]

this log:forAll :x, :y. :z.

00:31:28 [bijan]

{:x :parent :y} log:implies {:y :child [:parent :y].}.

00:31:30 [bijan]

{:x :parent :y. :y :parent :z} log:implies {[a :gParent]. [a :gChild].}.

00:31:58 [bijan]

would you expect a filter to generate 6 triples with 4 bnodes?

00:33:02 [sbp]

sorry had to pop out again - the midnight hour is my "busy hour"

00:33:12 [sbp]

heh, that didn't come out right, did it?

00:33:15 [sbp]

anyway...

00:33:16 [bijan]

Heh.

00:33:30 [bijan]

* bijan feels close.

00:33:48 [bijan]

I have some forms for builtins, and now for bnodes.

00:33:57 [bijan]

Lists will be pretty much all that's left.

00:34:04 [bijan]

That and implementing all the builtins :(

00:34:07 [sbp]

O.K., for the first rule we get: :jane :child [ :parent :jane ] . :janeII :child [ :parent :janeII ] .

00:34:25 [bijan]

X = 'document#jane'

00:34:25 [bijan]

Y = 'document#child'

00:34:25 [bijan]

Z = anon_61 ;

00:34:26 [bijan]

X = anon_61

00:34:26 [bijan]

Y = 'document#parent'

00:34:26 [bijan]

Z = 'document#jane' ;

00:34:28 [sbp]

ugh, I should label those bNodes: call them _:x and _:y

00:34:28 [bijan]

X = 'document#janeII'

00:34:30 [bijan]

Y = 'document#child'

00:34:32 [bijan]

Z = anon_62 ;

00:34:34 [bijan]

X = anon_62

00:34:36 [bijan]

Y = 'document#parent'

00:34:38 [bijan]

Z = 'document#janeII'

00:34:40 [bijan]

Ooh, an bnode syntax.

00:34:48 [bijan]

But that's not a fundemental bit.

00:35:07 [bijan]

Of the inference engine.

00:35:43 [sbp]

then we have 2 bNodes for the second bit, surely

00:35:58 [bijan]

Yep.

00:35:59 [bijan]

Ok.

00:35:59 [sbp]

so four bNodes, I agree

00:36:49 [sbp]

and six triples...

00:36:50 [bijan]

now, you think that the first :jane :child anon_61 would be blocked by :jane :child :bob.

00:36:56 [sbp]

no, wait, more than six

00:37:05 [sbp]

nope... six

00:37:17 [sbp]

blocked?

00:37:29 [bijan]

If I had :jane :child :bob in the target context.

00:37:40 [bijan]

And given the principle that we don't add duplicates if we can help it.

00:37:58 [bijan]

Should I nevertheless add a :jane :child anon_61?

00:38:15 [sbp]

you don't have :jane :child :bob in there at all

00:38:20 [bijan]

I know.

00:38:24 [bijan]

*Suppose*.

00:38:27 [sbp]

oh, sorry: if

00:38:28 [sbp]

got it

00:38:29 [bijan]

Contray to the expresssed facts ;)

00:38:36 [bijan]

That in the *target* context...

00:38:52 [bijan]

So, e.g., if I were applying rules and the kb had :jane :child :bob.

00:38:55 [sbp]

I think you should add it

00:39:00 [bijan]

Me too.

00:39:05 [bijan]

<sigh/>

00:39:32 [bijan]

Could you check cwm on this?

00:39:36 [sbp]

I could

00:39:42 [bijan]

Would you?

00:39:42 [bijan]

:)

00:39:48 [sbp]

I would

00:39:55 [sbp]

just the rule that you pasted in above, yes?

00:40:04 [bijan]

Yes.

00:40:57 [sbp]

:bob :parent :jane .

00:40:58 [sbp]

00:40:58 [sbp]

:jane :child [

00:40:58 [sbp]

:parent :jane ];

00:40:58 [sbp]

:parent :janeII .

00:40:58 [sbp]

00:41:01 [sbp]

:janeII :child [

00:41:02 [sbp]

:parent :janeII ] .

00:41:27 [sbp]

6 triples, two bNodes

00:41:45 [sbp]

not what we expected... ugh

00:41:46 [bijan]

This is with :jane :child :bob?

00:41:58 [sbp]

input:-

00:41:59 [sbp]

@prefix : <#> .

00:41:59 [sbp]

@prefix log: <http://www.w3.org/2000/10/swap/log#> .

00:41:59 [sbp]

:bob :parent :jane.

00:41:59 [sbp]

:jane :parent :janeII.

00:41:59 [sbp]

this log:forAll :x, :y. :z.

00:42:01 [sbp]

{:x :parent :y} log:implies {:y :child [:parent :y].}.

00:42:03 [sbp]

{:x :parent :y. :y :parent :z} log:implies {[a :gParent]. [a :gChild].}.

00:42:09 [sbp]

oh, heh. you borked that

00:42:15 [bijan]

You need a :jane :child :bob.

00:42:19 [bijan]

In the input.

00:42:30 [sbp]

that's better, now it gives:-

00:42:31 [sbp]

[ a :gChild ].

00:42:31 [sbp]

00:42:31 [sbp]

[ a :gParent ].

00:42:32 [sbp]

too

00:42:38 [sbp]

you want to add that triple? O.K.

00:42:43 [bijan]

Wait.

00:42:45 [bijan]

I'm confused.

00:42:54 [sbp]

you had: this log:forAll :x, :y. :z.

00:43:04 [sbp]

so it didn't recognize the :z as being universally quantified

00:43:13 [bijan]

Oops.

00:43:26 [sbp]

I added :jane :child :bob. to the document

00:43:31 [sbp]

and now it gives... [big paste]

00:43:40 [sbp]

[ a :gParent ].

00:43:41 [sbp]

00:43:41 [sbp]

[ a :gChild ].

00:43:41 [sbp]

00:43:41 [sbp]

:bob :parent :jane .

00:43:46 [sbp]

00:43:46 [sbp]

:jane :child [

00:43:46 [sbp]

:parent :jane ],

00:43:47 [sbp]

:bob;

00:43:49 [sbp]

:parent :janeII .

00:43:51 [sbp]

00:43:53 [sbp]

:janeII :child [

00:43:55 [sbp]

:parent :janeII ] .

00:43:57 [sbp]

[EOF]

00:44:25 [bijan]

Ok, so we're right.

00:44:28 [sbp]

yep

00:44:47 [bijan]

Ugh, that complicates things a bit.

00:51:28 [bijan]

Ok, y'know something really cool?

00:51:38 [sbp]

what?

00:51:53 [sbp]

[ a :CoolThingKnownBySean ] .

00:52:01 [bijan]

By compiling n3 to prolog I get an *exact* specification of the semantics of the rules.

00:52:11 [bijan]

In fairly transparent terms.

00:52:24 [sbp]

that's a good point. DanC tried ot express N3 in terms of KIF once before...

00:52:38 [bijan]

Yes, but in prolog, it becomes executable :)

00:52:52 [bijan]

For example:

00:52:54 [bijan]

n3implies(1, id(user, user32, user33), _G412):-

00:52:54 [bijan]

rdf(_G413, 'document#parent', _G415, user),

00:52:54 [bijan]

\+ rdf(_G415, 'document#child', Anon_4, _G412),

00:52:54 [bijan]

\+ rdf(Anon_4, 'document#parent', _G415, _G412),

00:52:54 [bijan]

catch(\+ atom_prefix(Anon_4, anon_),_, fail), %this will *NOT* work! Literals.

00:52:54 [bijan]

write_ln(ho),

00:52:57 [bijan]

logimplies,

00:52:58 [bijan]

rdf_gen_id('anon_', New_anon_1),

00:53:00 [bijan]

rdf_assert(_G415, 'document#child', New_anon_1, _G412),

00:53:02 [bijan]

rdf_assert(New_anon_1, 'document#parent', _G415, _G412).

00:53:18 [bijan]

Ignore the head, it's just infrastructure.

00:53:26 [bijan]

The first line checks that the premise holds.

00:53:46 [bijan]

2-3 check that the conclusion doesn't

00:53:56 [bijan]

2-5

00:54:03 [bijan]

6 is debugging :)

00:54:11 [bijan]

the logimplies is just a marker.

00:54:18 [bijan]

The last three lines add the triples.

00:54:32 [sbp]

excellent

00:54:40 [bijan]

But it's also pretty logical.

00:54:56 [bijan]

(the catch is wart. atom_prefix throws exceptions...)

00:55:17 [sbp]

what's the thing about literals for?

00:55:26 [bijan]

Oh, because of atom_prefix

00:55:29 [bijan]

It will work *now*.

00:55:35 [bijan]

Since I'm catching the exceptions.

00:55:39 [sbp]

* sbp eats some lovely garlic bread meanwhile

00:55:45 [sbp]

ah, got it

00:55:54 [bijan]

atom_prefix(literal(foo), X) will hork and exception rather than failing.

00:56:27 [bijan]

Alas, they arent' working :)

00:56:49 [bijan]

Duh.

00:56:56 [bijan]

Becasue I want an exception to == true :)

00:57:12 [bijan]

there we go.

00:57:48 [bijan]

Ugh.

00:58:00 [bijan]

Well, it will be interesting to see what optimization opportunities there are.

00:58:05 [bijan]

On real N3 code.

00:58:14 [bijan]

Right now I just query the crap out of everything.

00:58:26 [sbp]

yeah, CWM has a real problem with that

00:58:49 [bijan]

I still don't have good performance testers.

00:59:01 [bijan]

Oh, and between each application, i recompile *all* the rules.

00:59:22 [bijan]

As compiling gets more compilicated, and iterations get larger, it's going to make sense not to do that.

00:59:31 [bijan]

I.e., only compile new rules.

01:01:42 [sbp]

sbp has changed the topic to: Coming soon: "Swhack - the movie"

01:02:17 [sbp]

we could get Lucy Lawless to play xena

01:03:04 [bijan]

Ooo, scopt of the negations is wrong wrong wrong.

01:03:33 [sbp]

negations? in N3?

01:04:17 [sbp]

heh... just wait 'till you have to implement log:notIncludes. You're going to explode in a kind of critical frenzy. I can't wait

01:04:30 [bijan]

I'm NOT looking forward to it.

01:04:48 [sbp]

CWM's implementation of it *really* confuses me. It makes no sense

01:05:17 [bijan]

Well, I don't see that I need to implement all the log built-ins.

01:05:22 [bijan]

They seem sorta shady to me.

01:05:37 [sbp]

yep, but they're the meat on the N3 bone

01:05:59 [bijan]

Eh.

01:06:07 [bijan]

At SOME point someone else has to join in and help :)

01:06:13 [bijan]

Mustn't they?

01:06:23 [sbp]

well, most of the things that are done with N3 require builtins...

01:06:26 [sbp]

someone else?!

01:06:35 [bijan]

Sure, but not *all* the log built-ins

01:06:44 [sbp]

not just all, but more!

01:06:56 [bijan]

Er...

01:06:58 [sbp]

what's the point of making a CWM clone if you can't extend it :-)

01:07:00 [bijan]

* bijan doesn't parse that.

01:07:02 [bijan]

Oh sure!

01:07:06 [sbp]

that's why I added the crypto stuff to Eep

01:07:17 [bijan]

But *I* don't have to implement *all* the log built-ins.

01:07:22 [sbp]

sure you do

01:07:26 [bijan]

Nope.

01:07:32 [bijan]

Part of my point is to have a proof of concept.

01:07:33 [sbp]

otherwise, it's not much of a clone, is it? A sub-clone, perhaps :-)

01:07:51 [bijan]

And mostly on the inference engine.

01:08:04 [bijan]

I've still not seen a supposedly tough n3 file.

01:08:09 [bijan]

I.e., one that cwm grinds on forever.

01:08:17 [sbp]

er... the pool stuff is goo

01:08:25 [bijan]

Hmm.

01:08:43 [sbp]

[but it requires built-ins]

01:08:52 [bijan]

Sure.

01:09:05 [bijan]

I'll, perhaps, implement all the builtins necessary to run soemthing liek that.

01:09:31 [sbp]

in http://www.w3.org/2001/08swws67/

01:09:39 [sbp]

the poolGame* stuff

01:09:45 [bijan]

If I had a good example of an N3 program that took CWM 1/2 hour to run on a fast machine, I'd focus on getting it running.

01:10:22 [bijan]

Good god.

01:10:26 [bijan]

3 files?

01:10:32 [sbp]

yeah?

01:10:45 [sbp]

what were you expecting?

01:11:01 [bijan]

I dotn' know, I just want to make sure I understand it.

01:11:06 [sbp]

oh!

01:11:10 [sbp]

rules.n3!

01:11:15 [bijan]

Hmm/

01:11:23 [sbp]

that doesn't have any builtins, but when you apply it to data, it'll grind for quite a while

01:11:32 [bijan]

Pointer?

01:11:48 [bijan]

Ah. you think on that?

01:11:57 [sbp]

http://infomesh.net/2001/05/rdflint/rules.n3

01:12:11 [bijan]

Yes, got it.

01:13:25 [sbp]

try it on http://www.w3.org/2001/03/earl/0.95.n3

01:13:35 [sbp]

I'll run it locally, to see how long it takes...

01:14:35 [sbp]

* sbp sets it off

01:15:00 [newhak]

newhak (djflsdkjfl@ACA5AC4E.ipt.aol.com) has joined #swhack

01:15:02 [sbp]

it's getting files from the Web too... you should add that functionality (if you haven't got it already)

01:15:06 [sbp]

Hi there newhak

01:15:11 [bijan]

Argh. No I don't

01:15:27 [sbp]

this channel is publically logged etc., to make an off-log comment, prefix your junk with "# "

01:15:30 [Morbus]

hey there newhak.

01:15:34 [Morbus]

got any good 0 day?

01:15:48 [bijan]

Which is getting files from the web?

01:15:54 [sbp]

CWM

01:15:58 [newhak]

hi

01:16:07 [sbp]

ooh, CWM barfed:-

01:16:09 [sbp]

RuntimeError: ('Should only see first and rest in list mode', ((1, 'file:/home/#0_work'), (0, 'http://www.w3.org/1999/02/22-rdf-syntax-ns#type'), (0, 'http://infomesh.net/2001/05/rdflint/rules.n3#_g35'), (0, 'http://www.w3.org/2000/01/rdf-schema#Resource')))

01:16:16 [bijan]

I mean, which file is making it get stuff from the web?

01:16:26 [sbp]

file? it's from command line...

01:16:29 [bijan]

eek, if cwm can't run them ...

01:16:38 [sbp]

CWM used to be able to

01:16:41 [newhak]

newhak has left #swhack

01:16:50 [bijan]

I want to load up a KB

01:16:51 [Morbus]

damn

01:16:52 [bijan]

Think.

01:16:54 [bijan]

Time

01:16:55 [Morbus]

i really wanted 0 day.

01:17:13 [bijan]

If possible, solely from local data.

01:17:24 [bijan]

I don't want to test my network latency or bandwidth :)

01:18:25 [bijan]

Why would rules.n3 and the eral thing fetch stuff fromthe net?

01:18:26 [sbp]

Tim's validation thing can also take a while, but that uses mega-builtins

01:18:37 [sbp]

rules.n3 doesn't take anything from the Web

01:18:48 [bijan]

* bijan is lost as why you said it would :)

01:19:23 [sbp]

I didn't say that it would. I meant that I just put the URIS in the command line, and CWM slurped the files from the Web. It can also take them using log:semantics

01:19:32 [bijan]

Ah, yes.

01:19:39 [bijan]

That's not a benchmarking feature though.

01:19:43 [bijan]

it is a feature I need to add.

01:19:43 [sbp]

nope

01:19:54 [bijan]

It's not?

01:19:58 [bijan]

Whew, then.

01:21:43 [bijan]

Oh bugger, the earl things has triple quoted stuff.

01:22:01 [bijan]

Definitely somethign to add soon :)

01:22:40 [sbp]

it's not a benchmarking feature, that is... it *is* a feature you need to add :-)

01:22:45 [sbp]

* sbp is running more CWM tests

01:22:54 [sbp]

ugh, it barfed again

01:23:12 [bijan]

Heheheh.

01:23:42 [sbp]

the debug information certainly isn't very helpful, either

01:24:36 [bijan]

heh.

01:24:44 [bijan]

Right now I have no good error messages.

01:24:50 [bijan]

I hope eventually to add some.

01:25:02 [bijan]

But being able to inspect and query the dbs is pretty helpful.

01:25:24 [bijan]

Weird.

01:27:02 [bijan]

Oops. lists.

01:27:05 [bijan]

no can do.

01:27:51 [sbp]

argh, it fucked up again?! what the smeg is wrong with it? it's like it can't handle any damn lists at all

01:28:11 [Morbus]

where the hell is AaronSw?

01:28:13 [bijan]

Well, I *know* I can't :)

01:28:55 [sbp]

* sbp tries an old version of the CWM thing

01:30:55 [sbp]

Pff:-

01:30:57 [sbp]

$ u timenow; python cwm/cwm.py http://infomesh.net/2001/05/rdflint/rules.n3 http://www.w3.org/2001/03/earl/0.95.n3 --think > earl`u timenow`.n3; u timenow

01:31:00 [sbp]

20020227-013010

01:31:02 [sbp]

20020227-013029

01:31:04 [sbp]

works almost instantly, given that it has to get the files off of the Web

01:31:10 [sbp]

it's so much better than the new CWM. The new one seems quite broken

01:31:13 [bijan]

So not a good test.

01:31:38 [bijan]

Heh. /me is proud to have a CWM that is almost merely as broken as teh current CWM!

01:32:12 [sbp]

it *is* a good test, it's just that the latest CWM is buggy

01:32:22 [sbp]

well, it's not a good test in that it runs fast...

01:32:25 [bijan]

Yes.

01:32:28 [sbp]

heh, let's try the pool stuff

01:32:30 [bijan]

Sorry, that's all I meant.

01:33:39 [sbp]

cool

01:33:58 [sbp]

* sbp tries feeding it all of the pool stuff, and getting it to filter, using the *-filter* file

01:34:54 [Morbus]

Morbus has quit ("http://www.disobey.com/")

01:35:06 [sbp]

a minute and a half-ish

01:35:07 [sbp]

$ u timenow; python cwm/cwm.py http://www.w3.org/2001/08swws67/poolGame-kb.n3 http://www.w3.org/2001/08swws67/poolGame.n3 --think --filter=http://www.w3.org/2001/08swws67/poolGame-filter.n3 > pool`u timenow`.n3; u timenow

01:35:08 [sbp]

20020227-013328

01:35:08 [sbp]

20020227-013452

01:35:23 [bijan]

I can't read your timestamps :)

01:35:38 [bijan]

130 some seconds?

01:36:04 [sbp]

HH:MM:SS

01:36:20 [sbp]

1 minute, 34 seconds

01:36:29 [sbp]

90-ish seconds...

01:36:38 [sbp]

that's pretty efficient, actually

01:36:42 [bijan]

That's a reasonable test.

01:36:46 [bijan]

No, you mean that's pretty quick.

01:36:56 [bijan]

We have no idea if it's efficient.

01:37:03 [sbp]

no, I mean efficient compared to the new CWM

01:37:08 [bijan]

Ah!

01:37:10 [bijan]

Yes. well.

01:37:13 [sbp]

which would take a year if it could do it at all! :-)

01:37:18 [bijan]

New CWM sucks. :)

01:37:29 [sbp]

good ol' 1.82

01:43:05 [AaronSw]

what is he on now?

01:43:08 [AaronSw]

* AaronSw waves

01:43:58 [AaronSw]

Morbus, I took the day off.

01:45:30 [AaronSw]

* AaronSw plays "Spanish Lady" again, in preparation for the RDF Core singalong

01:46:08 [sbp]

ooh, very neat: http://www.w3.org/2000/10/swap/pim/usPlace2LatLong.n3

01:47:28 [sbp]

erk: 80-ish seconds

01:48:58 [AaronSw]

lol: http://www.ftrain.com/robot_exclusion_protocol.html

01:49:24 [AaronSw]

.google aaron

01:49:24 [sbp]

been there, done that. Stop reading our logs! they're private

01:49:25 [xena]

aaron: http://www.rall.com

01:49:33 [sbp]

heh. that's baffling, that

01:50:10 [sbp]

aha: "These terms only appear in links pointing to this page: aaron" - http://www.google.com/search?q=cache:krhpLrSaHI8C:www.rall.com/+Aaron&hl=en

01:50:18 [sbp]

that is unbelievably weird

01:50:28 [AaronSw]

.google link:www.rall.com aaron

01:50:29 [xena]

no results found.

01:51:50 [AaronSw]

* AaronSw reports this using the quality_form

01:53:33 [bitsko]

bitsko (~ken@kmacleod.static.iaxs.net) has joined #swhack

01:53:51 [sbp]

* sbp waves

01:53:56 [bijan]

Ken returns.

01:54:28 [bitsko]

hey. know anybody who has an RSS generating toolkit (more than just one website) that'd be willing to put out an update that includes admin:generatorAgent?

01:54:40 [bijan]

Rael?

01:54:51 [bitsko]

heh, I was just about to say that!

01:55:04 [bitsko]

he also has privs for Perl's XML::RSS

01:55:04 [bijan]

You mean something used by more than one *person*?

01:55:13 [bijan]

Or a big old site?

01:55:59 [bitsko]

generatorAgent is particularly for the software used to generate the RSS, different person that the website owner

01:56:38 [bijan]

right, but an aggregator could collect 0.91 files, and genreate rss1.0

01:56:44 [bijan]

Aaron! Arron might do that.

01:56:48 [bijan]

he's a tool guy.

01:57:25 [sbp]

you'll never get his attention that way.

01:57:30 [sbp]

You need to say, "Aaron, someone using a namespace with a hash on the end!"

01:57:37 [bitsko]

heh

01:58:00 [AaronSw]

sorry, i'm reading the logs, but i hear lines with "aaron" in them and no periods with characters following them

01:58:05 [bitsko]

btw, I just chatted with Jeff Barr and explained the idea and he's cool with it

01:58:14 [AaronSw]

and my name is not Arron

01:58:25 [bijan]

Sorry, a simple arror on my part.

01:58:42 [sbp]

heh. You make me sic

01:59:13 [bijan]

This is a comma typo banter on swhack?

01:59:25 [bitsko]

Aaron: I need some RSS 1.0 toolkit authors to incorporate admin:generatorAgent

01:59:45 [AaronSw]

oh yeah, i sort of started that mess like that didn't i...

02:00:52 [bitsko]

* bitsko holds breath: this may be the first widely used community-driven extension to RSS (of any flavour)

02:01:06 [bijan]

Oookay, I've got a working and perhaps feasible target for compiling bnoded rules.

02:01:24 [bitsko]

bnoded?

02:01:29 [AaronSw]

sbp, have you met the unix time command?

02:01:34 [bitsko]

* bitsko wonders what Bijan is cooking...

02:01:35 [sbp]

he means "bNoded"

02:01:41 [AaronSw]

$ time python somelongfile.py

02:01:46 [sbp]

unix time: that sucks

02:01:54 [bijan]

{[: a :child]} log:implies {[ a :parent]}

02:02:01 [bijan]

Ken: CWMclone!

02:02:10 [sbp]

wow! the new CWM had been going for 20 minutes

02:02:14 [sbp]

* sbp stops it

02:02:18 [bijan]

See: http://www.unc.edu/~bparsia/sw/cwmclone/cwmclone.html

02:02:38 [bijan]

Hmm. To which I've just added some grovely stuff about sean.

02:02:55 [sbp]

ooh, grovely stuff!

02:03:14 [AaronSw]

why does unix time suck?

02:03:37 [bijan]

* bijan is reminded of the excellent "Long Painful History of Time"

02:03:38 [bijan]

On the web.

02:03:46 [sbp]

* sbp would slave for Bijan if he hadn't mentioned Eep crapping out :-)

02:03:50 [bitsko]

* bitsko wonders if he means Unix time(3) or time(1)

02:03:56 [bijan]

No you wouldn't :)

02:03:57 [bitsko]

or is that time(2)?

02:04:00 [AaronSw]

.google "Long Painful History of Time"

02:04:00 [xena]

"Long Painful History of Time": http://www.stud.ifi.uio.no/~enag/lugm-time.html

02:04:07 [sbp]

ah: http://www.stud.ifi.uio.no/~enag/lugm-time.html

02:04:08 [AaronSw]

i mean the one that times programs for you

02:04:10 [sbp]

Pff

02:04:14 [AaronSw]

instead of this crazy u timenow stuff

02:04:25 [AaronSw]

Ooh, Naggum.no!

02:04:36 [AaronSw]

I know him.

02:04:40 [bijan]

Personally?

02:04:51 [AaronSw]

No.

02:05:02 [AaronSw]

I should say that I know of him.

02:05:13 [AaronSw]

bitsko, so what exactly do I add to my generator?

02:05:13 [bijan]

Hmm. Actually, sean, you could do worse than try to implement builtins. As a prolog learning task.

02:05:21 [bitsko]

I knew (of) him before he shot SGML in the back ;)

02:06:06 [Morbus]

Morbus (~Morbus@s108.terminal3.totalnetnh.net) has joined #swhack

02:06:14 [bitsko]

Naggum was a huge fan/supporter of SGML when it should have been doing RDF things, but HyTime didn't come soon enough before he bailed.

02:06:18 [jeremiah]

hello

02:06:22 [AaronSw]

hey j

02:06:29 [jeremiah]

AaronSw: I was looking for the plex docs today

02:06:34 [jeremiah]

and I found a newer version

02:06:35 [jeremiah]

it seems

02:06:37 [bijan]

Er...

02:06:39 [bitsko]

* bitsko bugs out for dinner. brb.

02:06:40 [jeremiah]

check the old site, I think they've been republished

02:06:47 [AaronSw]

huh? of what? where?

02:06:49 [bijan]

How tdoes that translate into shooting it in the back?

02:06:54 [bitsko]

Aaron: just add admin:generatorAgent as described

02:07:00 [AaronSw]

described where?

02:07:18 [AaronSw]

BLURB:RDF Core Official Song

02:07:20 [chumpster]

A: RDF Core Official Song from AaronSw

02:07:39 [AaronSw]

A::Golden Triples (trad.)

02:07:40 [chumpster]

commented item A

02:07:45 [AaronSw]

A::[Lyrics from Pat Hayes]

02:07:47 [chumpster]

commented item A

02:07:52 [jeremiah]

one sec

02:07:52 [jeremiah]

chord docs

02:07:52 [jeremiah]

not plex

02:07:52 [jeremiah]

sorry

02:08:03 [bitsko]

http://groups.yahoo.com/group/rss-dev/files/Modules/Proposed/mod_admin.html

02:08:07 [AaronSw]

A::[Chords of Spanish Lady|http://ils.student.utwente.nl/main/r/reilly_paddy/spanish_lady.crd]

02:08:08 [chumpster]

commented item A

02:08:21 [AaronSw]

A::[MP3 from Paddy and the Pale Boys|http://www.thepaleboys.com/Paddy%20and%20the%20Pale%20Boys%20-%20Spanish%20Lady.mp3]

02:08:22 [chumpster]

commented item A

02:08:39 [AaronSw]

jeremiah, yeah, i got them to post a WIP draft

02:08:47 [AaronSw]

but nothing so exciting has changed

02:09:06 [Morbus]

hey AaronSw. good to see you.

02:09:16 [jeremiah]

ok

02:09:17 [AaronSw]

Hi there. I took the day off as it seemed you noticed.

02:09:29 [Morbus]

yes... i was wondering where your rhetoric was.

02:09:34 [AaronSw]

Feeling pretty down this morning, but some Ken MacLeod (the other one) cheered me up.

02:09:42 [AaronSw]

Ken (the this one) changes affiliations weekly it seems.

02:10:35 [Morbus]

i'm working through some reviisons on part 5

02:10:36 [sbp]

* sbp likes coding builtins

02:10:38 [sbp]

just give me the lowdown, a couple of simple examples, and set me to my task :-)

02:10:40 [sbp]

* sbp has a borky connection

02:10:46 [Morbus]

sbp: open a file, for me.

02:11:17 [sbp]

which one?

02:11:24 [bijan]

it shoudl be pretty easy.

02:11:48 [bijan]

Once I have the details figured out :)

02:11:56 [AaronSw]

"First she washed them, then she dried them, over a fire of amber coals."

02:12:01 [bijan]

built_in(literal(S), 'http://www.w3.org/2000/10/swap/string#startsWith', literal(O), Context) :-

02:12:03 [bijan]

atom_concat(O, _, S).

02:12:14 [bijan]

The head will be pretty much boiler plate.

02:12:51 [bijan]

Then the rest is figureing out what predicates the built-in maps to, or just implementing it.

02:13:04 [bijan]

I'm unclear on type testing, failure modes, and the like.

02:13:40 [sbp]

if you can get a builtin to return a list, that'd be tremendous

02:13:44 [sbp]

that's a big goal

02:14:05 [bijan]

Ah, yeah. Lists in general.

02:14:16 [bijan]

I'm not sure if I'm going to condense lists.

02:14:19 [bijan]

Certainly not yet.

02:14:34 [bijan]

In which case, returning them is trivial.

02:14:42 [sbp]

e.g. ("http://example.org/#blargh" "#") string:split ?x .

02:14:59 [sbp]

?x : ("http://example.org/" "blargh")

02:15:08 [bijan]

Yeah, that will be read in as:

02:15:26 [bijan]

user4545 string:split ?x.

02:15:43 [bijan]

the builtin body will walk the list.

02:15:56 [bijan]

And then do a bunch of asserts.

02:16:37 [sbp]

returning mutiple triples would be good, too...

02:16:46 [bijan]

What do you mean?

02:17:28 [sbp]

("http://example.org/blargh" "#") string:split ?x . => ?x = _:Error . _:Error a cwm:Error . _:Error rdfs:label "no freaking # in the string, you dipstick!" .

02:17:32 [bijan]

Whta's that list look like.

02:17:35 [bijan]

In triples?

02:17:55 [bijan]

"http:..." daml:first...

02:18:01 [sbp]

[ a daml:List; daml:first "http://example.org/#blargh"; daml:rest [ daml:first "#"; daml:rest daml:nil ] ] .

02:21:59 [bitsko]

bijan: re Naggum and SGML, maybe "kicked it on the way down" is a better description ;)

02:22:14 [bijan]

Ok, that I believe.

02:22:23 [bijan]

He did come to loathe it, from what i've read/heard.

02:22:44 [AaronSw]

bitsko, how's this: http://rss.blogspace.com/bbcNews/world ?

02:22:52 [AaronSw]

I've added it to all rss.blogspace.com feeds

02:23:12 [AaronSw]

* AaronSw listens to Bicycle Race

02:23:31 [bijan]

Sean, I have a straw implementation of string:split.

02:23:51 [bijan]

built_in(S, 'http://www.w3.org/2000/10/swap/string#split', O, Context) :-

02:23:51 [bijan]

rdf(S, rdf:type, daml:list),

02:23:51 [bijan]

rdf(S, daml:first, String),

02:23:51 [bijan]

rdf(S, daml:rest, Second),

02:23:51 [bijan]

rdf(Second, daml:first, Delim),

02:23:52 [bijan]

%Could have a check for only two elements

02:23:54 [bijan]

split(String, Delim, O). %not a real predicate, but you get the idea

02:23:54 [sbp]

excellent

02:24:21 [sbp]

wow, that was quick

02:24:28 [bitsko]

AaronSw: do you happen to know the Syndic8 ids for these feeds?

02:24:28 [sbp]

does it return a list?

02:24:40 [bijan]

Depends on what split does

02:24:46 [AaronSw]

bitsko, there are a zillion, can't you use the grep function?

02:24:55 [bitsko]

heh, ok

02:25:00 [bijan]

What I'm *expecting* is that it will return a reference.

02:25:14 [bijan]

And that split will do all the proper assertions.

02:25:28 [bijan]

That's for a pure RDF approach.

02:25:38 [bitsko]

AaronSw: ah, can you add a version# in that URI?

02:25:46 [AaronSw]

Hm.

02:25:53 [AaronSw]

I don't really have versions

02:26:18 [bitsko]

then how will they know when bugs get fixed, hmm? ;)

02:26:39 [AaronSw]

hey, i'm centralized, boy

02:26:44 [AaronSw]

they're fixed when i say they're fixed

02:26:47 [bitsko]

you could just put a date in it, as in "this is the date I last thought I had everything fixed"

02:26:57 [AaronSw]

that's good

02:27:22 [AaronSw]

whoa, i screwed up that page...

02:28:18 [AaronSw]

tav's on the phone

02:28:34 [sbp]

say hi for me

02:28:43 [Morbus]

tell tav I miss his oppression.

02:28:48 [Morbus]

:)

02:28:58 [AaronSw]

* AaronSw says so

02:29:14 [Morbus]

* Morbus ear's start burning.

02:29:34 [sbp]

ear's?

02:29:37 [AaronSw]

tav: "don't mean to be rude but he tends to fill the channel with a lot noise"

02:29:54 [sbp]

yeah, like stray apostrophies

02:29:59 [Morbus]

yeah, we've had the discussion before.

02:39:30 [bijan]

Ohhh, sean, this will be *so* slow!

02:39:49 [bitsko]

AaronSw: if we end up using a purl.net URI for modules/admin, who's got privs to point that to the module?

02:40:07 [AaronSw]

I think I do, as does Rael

02:40:08 [bijan]

But I have an implementation.

02:40:28 [sbp]

slow: well, there's a trick

02:40:47 [bijan]

Well, it's slow because I'm asserting things left right and center.

02:41:05 [sbp]

[for builtin optimization] what you do is make a list of all the builtin namespaces. Then you match the namespaces first, and the URIs later

02:41:22 [bijan]

Huh?

02:41:56 [bijan]

No, what makes this slow is at least 3 rdf_asserts per list item.

02:42:07 [bijan]

"consing" with a vengence.

02:42:44 [bijan]

I'd be interested to know how quick *reading* or *walking* alist is, though.

02:42:49 [bijan]

That might be reasonably nippy.

02:43:09 [bitsko]

heh, I see Aaron's cheating with the version URI...

02:43:18 [AaronSw]

:)

02:43:44 [bitsko]

AaronSw: the RSS I just pulled is empty, is that intentional

02:43:47 [bitsko]

?

02:43:49 [bijan]

Sean, I uploaded it. Check it out.

02:44:07 [AaronSw]

on the phone

02:44:18 [AaronSw]

what file?

02:44:35 [sbp]

* sbp slurps it off of the Web

02:44:45 [bijan]

it's not tested.

02:44:48 [deltab]

"off of"?

02:44:58 [bijan]

For it to work you need to declare the right namespaces.

02:45:09 [bijan]

And manually enter the triples.

02:45:16 [sbp]

yes, deltab, and Americanism. So what

02:45:19 [bitsko]

http://rss.blogspace.com/bbcNews/world

02:45:21 [sbp]

I say "color" too

02:45:25 [sbp]

a/and/an/

02:45:28 [sbp]

argh

02:48:47 [bijan]

heh. didnt' work.

02:49:23 [bijan]

Ahh buger. screwed up the prefix.

02:49:34 [sbp]

@stop. won't work

02:50:02 [bijan]

Really?

02:50:11 [bijan]

What did you enter?

02:50:16 [bijan]

ctrl-c will escape.

02:51:03 [sbp]

argh:-

02:51:03 [sbp]

?- compile_rules(user).

02:51:04 [sbp]

Yes

02:51:04 [sbp]

?- filter(user, test).

02:51:04 [sbp]

[]

02:51:04 [sbp]

No

02:51:05 [sbp]

?- apply_rules(user).

02:51:07 [sbp]

[]

02:51:09 [sbp]

No

02:51:11 [bijan]

That will not work.

02:51:13 [bijan]

What are you trying?

02:51:17 [sbp]

I entered...

02:51:17 [sbp]

?- n3_repl.

02:51:18 [sbp]

n3:- @prefix : <#> .

02:51:19 [sbp]

done read: [@, prefix, :, <, #, >, .]

02:51:21 [sbp]

n3:- @prefix string: <http://www.w3.org/2000/10/swap/string#> .

02:51:23 [sbp]

done read: [@, prefix, string:, <, http://www.w3.org/2000/10/swap/string#, >, .]

02:51:25 [sbp]

n3:- @prefix log: <http://www.w3.org/2000/10/swap/log#> .

02:51:27 [sbp]

done read: [@, prefix, log:, <, http://www.w3.org/2000/10/swap/log#, >, .]

02:51:29 [bijan]

oh oh oh

02:51:29 [sbp]

n3:- this log:forAll :x .

02:51:30 [bijan]

Stop.

02:51:31 [sbp]

done read: [this, log:forAll, :x, .]

02:51:33 [sbp]

S: this V: log:forAll O: default_:x C: user

02:51:35 [sbp]

n3:- { ("http://example.org/#blargh" "#") string:split :x } log:implies { :x :the :Result } .

02:51:36 [bijan]

Builtin's don't compile.

02:51:37 [bijan]

Yet.

02:51:37 [sbp]

n3:- @stop.

02:51:52 [bijan]

And list syntax doesn't work yet.

02:52:16 [bijan]

I pasted in to n3_repl the expanded list sytnax you gave me.

02:52:19 [bijan]

That's yoru list.

02:52:26 [bijan]

then you can use the built_in directly

02:52:42 [bijan]

via: built_in(anon_105, 'http://www.w3.org/2000/10/swap/string#split',O, user).

02:52:51 [bijan]

(Or whatever anon_ your list is)

02:53:01 [bijan]

Thought I'm still debugging the bulitin

02:53:36 [sbp]

O.K.

02:53:52 [sbp]

Hmm... I should have known about the list thing :-)

02:53:53 [bijan]

One minute.

02:53:58 [sbp]

* sbp is multitasking...

02:55:01 [bijan]

oops. lowercase List.

02:55:12 [bijan]

I put a new version.

02:55:32 [bijan]

Works.

02:55:46 [bijan]

I'm going to need a pretty printer RSN.

02:56:37 [bijan]

* bijan is pretty pleased.

02:56:59 [bitsko]

gotta run. tty'all later. Thanks Aaron!

02:57:00 [bitsko]

bitsko has quit ("Leaving")

02:57:09 [bijan]

Time to hack the compiler, sigh.

03:08:52 [AaronSw]

hey dudes and dudettes

03:09:11 [Morbus]

* Morbus grunts

03:09:13 [AaronSw]

tav's going to take an espian vote on whether we should be kicked off the cvs server

03:09:15 [Morbus]

hey AaronSw, you got soemtime

03:09:55 [AaronSw]

so if anyone feels a sudden urge to become an espian

03:09:58 [AaronSw]

i suggest you do so asap

03:11:29 [AaronSw]

Morbus, time for what?

03:11:43 [Morbus]

to look over this unfinished, but nearly done, draft and lemme know what you think

03:12:22 [AaronSw]

Hm. I sorta have to do some lobbying and other annoying political tasks right now.

03:12:42 [Morbus]

np

03:13:56 [sbp]

heh: "sorry, I can't: lobbying"

03:15:18 [AaronSw]

Heh, heh.

03:21:05 [Morbus]

i need some junk food.

03:21:11 [Morbus]

what do you guy's eat?

03:21:13 [Morbus]

at the computer?

03:21:26 [bijan]

Sean, did the built-in work for you?

03:21:42 [sbp]

* sbp downloads the latest version

03:21:50 [AaronSw]

Pizza seems popular, not exactly my taste.

03:24:58 [sbp]

O.K., I've entered the stuff... now what shall I do to make the rule work?

03:26:09 [Morbus]

does this make any sense to anyone:

03:26:10 [Morbus]

"<Morbus> "we've got just enough time to chow down the last remnants of our "Salt and Vinegar" potato chips. We'll be sure to enunciate our words properly and with much weight.

03:26:10 [Morbus]

<Morbus> ""

03:26:12 [sbp]

compile_rules worked...

03:26:31 [bijan]

But it won't help :)

03:26:41 [BenSw]

BenSw has quit ("Client Exiting")

03:26:46 [AaronSw]

I get no sense from it

03:26:54 [Morbus]

yeah, didn't think so

03:27:03 [sbp]

Morbus: with a little bit of context...

03:27:09 [bijan]

I.e., the current compile won't detect builtins.

03:27:15 [Morbus]

i'm having difficulty saing "we got stinky breath and we're gonna blow it all over the people at the meeting"

03:27:24 [AaronSw]

Aha.

03:27:29 [BenSw]

BenSw (~yoda@12-249-96-16.client.attbi.com) has joined #swhack

03:27:31 [bijan]

But you can test the bi manually.

03:27:40 [sbp]

bijan: what shall I enter?

03:27:58 [bijan]

First, do you know the anon_FOO that names your list?

03:28:16 [sbp]

yep

03:28:18 [bijan]

rdf(X, rdf:type, daml:'List').

03:28:18 [sbp]

anon_11

03:28:20 [bijan]

Ok.

03:28:43 [bijan]

built_in(anon_11, 'http://www.w3.org/2000/10/swap/string#split', O, user).

03:28:58 [sbp]

oh crap, I didn't put the daml prefix in...

03:29:05 [bijan]

That's fine.

03:29:07 [bijan]

Don't worry about it.

03:29:12 [bijan]

It should still work.

03:29:14 [sbp]

?- built_in(anon_11, 'http://www.w3.org/2000/10/swap/string#split', O, user).

03:29:14 [sbp]

No

03:29:34 [bijan]

Hmm.

03:29:52 [bijan]

rdf(anon_11, X,Y).?

03:30:11 [sbp]

No

03:30:27 [sbp]

I had: n3implies(1, id(user, user11, user14), _G473):- (rdf(anon_11, damlfirst, literal(http://example.org/#blargh), user), rdf(anon_11, damlrest, anon_12, user), rdf(anon_11, http://www.w3.org/1999/02/22-rdf-syntax-ns#type, damlList, user), rdf(anon_11, http://www.w3.org/2000/10/swap/string#split, _G341, user), rdf(anon_12, damlfirst, literal(#), user), [...]

03:30:48 [bijan]

Right, that won't help you.

03:30:56 [bijan]

Ah, are you using list syntax?

03:31:18 [sbp]

yep [ a daml:List; daml:first [...]

03:31:32 [bijan]

That's the *only* thing you should put in user.

03:31:39 [bijan]

it looks like you did the rule.

03:31:59 [bijan]

Ok, try this.

03:32:11 [bijan]

the built_in, but with the last paran being user11

03:32:15 [bijan]

param

03:32:34 [bijan]

Basically, when you tucked it into that context, it isn't visible to the user context.

03:32:36 [sbp]

?- built_in(anon_11, 'http://www.w3.org/2000/10/swap/string#split', O, user11).

03:32:36 [sbp]

No

03:32:36 [sbp]

?-

03:32:41 [sbp]

ah...

03:32:49 [bijan]

rdf(X,Y,Z, user11).

03:32:50 [sbp]

so you have to put the rules in separately?

03:33:18 [sbp]

?- rdf(X,Y,Z, user11).

03:33:19 [sbp]

X = anon_11

03:33:19 [sbp]

Y = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#type'

03:33:19 [sbp]

Z = damlList ;

03:33:19 [sbp]

X = anon_11

03:33:19 [sbp]

Y = damlfirst

03:33:21 [sbp]

Z = literal('http://example.org/#blargh') ;

03:33:23 [sbp]

[...]

03:33:25 [bijan]

Ok.

03:34:06 [bijan]

Hmm.

03:34:13 [bijan]

That's odd.

03:34:34 [bijan]

What's the first few lines of the built_in source

03:34:40 [bijan]

(for string#split)

03:34:47 [bijan]

built_in(S, 'http://www.w3.org/2000/10/swap/string#split', O, Context) :-

03:34:47 [bijan]

rdf(S, rdf:type, daml:'List'),

03:34:47 [bijan]

rdf(S, daml:first, String),

03:34:48 [bijan]

rdf(S, daml:rest, Second),

03:34:51 [bijan]

Does it look like that?

03:35:17 [sbp]

Hmm... yep

03:35:25 [bijan]

Ok, try this.

03:35:27 [bijan]

rdf_clean.

03:35:31 [bijan]

n3_repl.

03:35:41 [bijan]

declare the daml prefix

03:35:57 [bijan]

enter the expanded form of the list

03:35:59 [bijan]

@stop.

03:36:09 [bijan]

built_in(...., user).

03:36:55 [sbp]

just the list?

03:37:08 [sbp]

or the whole implication?

03:37:16 [bijan]

Just the list.

03:37:32 [sbp]

done

03:37:39 [bijan]

Now @stop.

03:37:42 [bijan]

Then the built_in

03:37:46 [sbp]

?- built_in(anon_11, 'http://www.w3.org/2000/10/swap/string#split', O, user).

03:37:46 [sbp]

No

03:37:46 [sbp]

?-

03:37:51 [bijan]

Hmm.

03:37:56 [bijan]

Oopo!

03:37:59 [bijan]

not anon_ll.

03:38:10 [bijan]

you need to find what the list is named *now*.

03:38:31 [bijan]

rdf(X, Y, daml:'List').

03:38:38 [sbp]

?- built_in(anon_14, 'http://www.w3.org/2000/10/swap/string#split', O, user).

03:38:38 [sbp]

O = anon_17 ;

03:38:38 [sbp]

No

03:38:38 [sbp]

?-

03:38:45 [bijan]

Yes!

03:38:48 [bijan]

it worked.

03:38:51 [sbp]

* sbp was one step ahead there

03:38:55 [bijan]

rdf(X,Y,Z).

03:38:56 [sbp]

how do I find the result?

03:39:03 [bijan]

And you'll see the results.

03:39:08 [bijan]

And the non-results :)

03:39:17 [bijan]

but anon_17 will be the head pointer.

03:39:21 [sbp]

ooh:-

03:39:22 [sbp]

X = anon_17

03:39:22 [sbp]

Y = 'http://www.daml.org/2001/03/daml+oil#first'

03:39:22 [sbp]

Z = 'http://example.org/' ;

03:39:22 [sbp]

X = anon_17

03:39:22 [sbp]

Y = 'http://www.daml.org/2001/03/daml+oil#rest'

03:39:23 [sbp]

Z = anon_18 ;

03:39:25 [sbp]

X = anon_18

03:39:27 [sbp]

Y = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#type'

03:39:29 [sbp]

Z = 'http://www.daml.org/2001/03/daml+oil#List' ;

03:39:31 [sbp]

X = anon_18

03:39:33 [sbp]

Y = 'http://www.daml.org/2001/03/daml+oil#first'

03:39:35 [sbp]

Z = blargh ;

03:39:37 [sbp]

[...]

03:39:39 [sbp]

great stuff!

03:39:40 [bijan]

There you go.

03:39:59 [bijan]

Note that that's all done in raw rdf.

03:40:04 [bijan]

no prolog list code invoked.

03:40:08 [sbp]

why is example.org quoted, but blargh isn't?

03:40:17 [sbp]

I guess Prolog needs to quote the /?

03:40:21 [bijan]

Yes.

03:40:32 [bijan]

Ah, that's prolly a bug, too.

03:40:39 [bijan]

they shoudl be literals

03:40:46 [sbp]

yeah

03:44:59 [bijan]

Ok, got a fix.

03:46:25 [bijan]

literalize_list([],[]).

03:46:25 [bijan]

literalize_list([H|T], [literal(H)|New_t]) :-

03:46:26 [bijan]

literalize_list(T, New_t).

03:46:45 [bijan]

If I move to using prolog strings for literals, some of that will disappear.

03:46:49 [bijan]

Or be a tad easier.

03:46:59 [sbp]

ooh. Prolog is the balls

03:47:17 [sbp]

* sbp really should pick up a manual...

03:47:21 [bijan]

Yep.

03:47:45 [bijan]

Or get ORA to hire me to write "Learning Prolog (with web applications)

03:47:46 [AaronSw]

jeremiah, want to become a "community member"?

03:48:06 [sbp]

* sbp mails Derrick

03:48:18 [AaronSw]

heh

03:48:28 [bijan]

Actually, CWMClone would be a great learning project.

03:48:29 [AaronSw]

Derrick just does Mac stuff, dude/

03:48:34 [sbp]

Pff

03:48:39 [sbp]

[at Aaron]

03:48:44 [sbp]

and pff

03:48:48 [sbp]

[at bijan] :-)

03:48:51 [bijan]

I was thinking of trying to write up a "literate" version focused on teaching prolog and cwm.

03:48:57 [bijan]

and n3 etc.

03:49:14 [sbp]

that would make for an incredible article series

03:49:20 [sbp]

"Roll your own CWM"

03:49:21 [bijan]

...that no one would pay for ;)

03:49:25 [MorbusIff]

MorbusIff (~Morbus@s126.terminal3.totalnetnh.net) has joined #swhack

03:49:28 [sbp]

heh, true

03:49:40 [sbp]

Morbus! have a peanut

03:49:40 [bijan]

Not even teh w3c.

03:49:49 [sbp]

well, I'd pay for it

03:49:54 [bijan]

But not very much :)

03:49:55 [sbp]

but I'm odd like that

03:50:01 [sbp]

I might manage a dollar

03:50:02 [bijan]

You're means limited.

03:50:07 [sbp]

indeed

03:50:13 [MorbusIff]

* MorbusIff munches.

03:50:15 [MorbusIff]

MMmMm!

03:57:09 [Morbus]

Morbus has quit (No route to host)

04:10:29 [MorbusIff]

MorbusIff is now known as Morbus

04:22:18 [Morbus]

Morbus has quit ("http://www.disobey.com/")

04:40:44 [bijan]

bijan has quit ("Leaving")

04:48:42 [AaronSw]

jeremiah, are you around?

05:49:33 [redmonk]

redmonk (sid@ip68-2-179-170.ph.ph.cox.net) has joined #swhack

05:50:24 [redmonk]

redmonk is now known as rm

05:52:49 [rm]

rm is now known as rm_lrkr

05:57:00 [AaronSw]

* AaronSw downloads "We Can Work it Out"

06:30:04 [rm_lrkr]

rm_lrkr has quit ("i think you all need to grow up. nighty night folks.")

06:45:24 [AaronSw]

* AaronSw is away: sleep

06:46:51 [sbp]

Gotta run

08:31:03 [deltab]

deltab has quit (carter.openprojects.net irc.openprojects.net)

08:33:58 [deltab]

deltab (deltab@mewtwo.espnow.com) has joined #swhack

09:20:07 [tansaku]

tansaku has quit (Read error: 110 (Connection timed out))

13:51:14 [Morbus]

Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack

14:16:28 [redmonk]

redmonk (sid@ip68-2-179-170.ph.ph.cox.net) has joined #swhack

14:16:59 [redmonk]

redmonk is now known as rm

14:25:30 [rm]

rm has quit ("Snak 4.8.3 - http://www.snak.com")

15:33:18 [Rooth4ckus]

Rooth4ckus (Rooth4ckus@211.225.231.14) has joined #swhack

15:39:19 [tansaku23]

tansaku23 (~sam@n145-141.tokyu-net.catv.ne.jp) has joined #swhack

16:33:03 [tansaku23]

tansaku23 has quit (Read error: 110 (Connection timed out))

16:49:16 [tansaku23]

tansaku23 (~sam@h131-241.tokyu-net.catv.ne.jp) has joined #swhack

17:22:12 [hazmat]

hazmat (~ender@adsl-66-123-57-58.dsl.lsan03.pacbell.net) has joined #swhack

17:42:35 [hazmat]

.time est

17:42:35 [xena]

Feb. 27, 2002 12:44 pm US/Eastern

17:42:47 [AaronSw]

howdy, hazmat

17:42:54 [AaronSw]

tav told me you wanted to become an espian

17:42:56 [hazmat]

good morning AaronSw

17:43:11 [hazmat]

its been a while, but yes. i want to change the world ;)

17:43:18 [AaronSw]

Heh, heh

17:44:07 [hazmat]

i'm still not clear on what it means to be an espian..

17:44:45 [AaronSw]

yeah, i wasn't either...

17:45:22 [AaronSw]

and it got me into trouble

17:47:28 [hazmat]

ic

17:48:08 [hazmat]

a leap of faith then?

17:48:32 [AaronSw]

since it's not really written down, tav seems to take it as a license to tell you what to do :-(

17:48:56 [AaronSw]

when i disagree with things he says, he just responds "why didn't you speak up when you first became an espian?"

17:50:38 [hazmat]

ok, so what does tav want to do?

17:52:48 [AaronSw]

well, he wants the plex to be espian controlled

17:56:12 [Morbus]

whee, i'm loved.

17:56:38 [Morbus]

.email root@mobile.totalnetnh.net

17:56:38 [xena]

email successfully sent.

17:56:46 [AaronSw]

heh!

17:57:00 [Morbus]

just testing a new server

17:57:18 [AaronSw]

whee, i'm gonna get paid!

17:57:25 [Morbus]

? for what?

17:57:31 [AaronSw]

i can't say

17:57:37 [sbp]

neat

17:57:50 [Morbus]

.email majordomo-owner@mobile.totalnetnh.net

17:57:50 [xena]

email successfully sent.

17:57:52 [sbp]

Gotta run

18:25:16 [denshi]

denshi (~chatzilla@adsl-216-62-223-193.dsl.rcsntx.swbell.net) has joined #swhack

18:53:15 [sbp]

* sbp wonders who Rooth4ckus is

18:53:19 [sbp]

and denshi

18:53:26 [sbp]

lots of new people today!

18:54:18 [denshi]

sbp: denshi is probably a bot.

18:54:36 [sbp]

ah. probably

18:54:45 [monokrom]

spooky

18:54:52 [sbp]

can it do any cool tricks, I wonder?

18:55:06 [denshi]

I took the blue pill.

18:55:46 [sbp]

Er... Mmmkay

18:56:01 [sbp]

* sbp is unimpressed with that trick

18:56:42 [denshi]

is there a chatlog for this channel?

18:57:05 [sbp]

yep

18:57:21 [sbp]

http://blogspace.com/swhack/chatlogs/

18:57:30 [sbp]

to say something off-log, prefix it with "# "

18:57:39 [Morbus]

sbp!

18:57:45 [Morbus]

man, lots of new people coming lately.

18:57:58 [sbp]

yeah, I noticed

18:58:14 [sbp]

although one of them is (and I quote) "probably a bot"

18:58:37 [monokrom]

haha

18:58:42 [Morbus]

mmhmm

18:59:35 [sbp]

Gotta run

19:00:00 [Morbus]

oh, i see how it is.

20:03:24 [sbp]

* sbp returns

20:10:33 [AaronSw]

sbp, what do you think of the idea of W3C Phone BOFs.

20:10:45 [AaronSw]

s/./?/

20:11:19 [sbp]

quite a good idea

20:11:27 [sbp]

where was that raised?

20:11:34 [AaronSw]

sandro and I are chatting about it

20:12:02 [sbp]

it is a good idea

20:13:53 [Morbus]

Florps Pie?

20:29:16 [Rooth4ckus]

Rooth4ckus has quit ()

20:29:31 [Morbus]

la la la.

20:31:58 [bijan]

bijan (bparsia@login1.isis.unc.edu) has joined #swhack

20:33:19 [bijan]

So, I'm way off :)

20:34:11 [sbp]

yes

20:34:35 [sbp]

* sbp comes up with a normalization script to canonicalize the NTriples output from CWM

20:36:02 [sbp]

Hmm... it only seems to have applied a couple of the rules

20:37:14 [bijan]

oh my.

20:37:17 [bijan]

i jsut got a bunch.

20:37:18 [sbp]

the funny thing is, there are no rules with "subClassOf" as predicate in the ante.

20:37:25 [sbp]

a bunch of what?

20:37:30 [sbp]

flowers? chips? :-)

20:37:38 [bijan]

Stuff in a filter.

20:39:22 [bijan]

Now I have 308!

20:40:01 [bijan]

Something about my forward chainer isn't quite working.

20:44:11 [bijan]

Ok, first application gets 15. (in a filter)

20:44:20 [bijan]

Second applicatoin ups it to 79

20:44:40 [bijan]

Third, 194

20:44:54 [bijan]

Fourth, 308.

20:44:59 [bijan]

That's it.

20:45:41 [sbp]

eek

20:45:52 [sbp]

well, the third is the closest :-)

20:46:41 [sbp]

* sbp continues work on his normalization script

20:46:44 [bijan]

Werid that apply rules isn't working automatically.

20:46:59 [bijan]

Do we have any idea what the correct inference is?

20:47:31 [bijan]

OH OHOHOH!

20:47:38 [bijan]

I *know* what the discrepancy is.

20:47:41 [bijan]

Or I have a good bet.

20:47:45 [denshi]

sbp, could you break down the function of that script you're writing?

20:47:54 [bijan]

Lists!

20:48:24 [bijan]

Maybe nt :)

20:48:57 [bijan]

Yes, I have a number of daml:Lists type triples.

20:49:08 [bijan]

Oops. no.

20:49:34 [bijan]

Well, note that that's a possibility.

20:49:41 [bijan]

If I stick in more triples for some things.

20:50:21 [bijan]

yay! I see a list.

20:52:03 [sbp]

denshi: do you want me to just paste it in?

20:52:24 [denshi]

just clue me in as to what this group is about and the project you are working on...

20:52:32 [sbp]

ah, right

20:52:56 [sbp]

Welcome to #swhack :-) Basically, we chat about anything here, but at the moment, we're working on stuff to do with the Semantic Web (http://www.w3.org/2001/sw/)

20:53:29 [sbp]

the script that I'm writing at the moment normalizes NTriples - a subset of Notation3 - since CWM can't output it properly

20:53:32 [sbp]

.google NTriples

20:53:33 [xena]

NTriples: http://www.w3.org/2001/sw/RDFCore/ntriples

20:53:36 [sbp]

.google Notation3

20:53:37 [xena]

Notation3: http://www.w3.org/DesignIssues/Notation3.html

20:53:39 [sbp]

.google CWM

20:53:40 [xena]

CWM: http://www.fme.nl

20:53:47 [sbp]

the last one is incorrect...

20:53:51 [sbp]

.google "Closed World Machine"

20:53:52 [xena]

"Closed World Machine": http://infomesh.net/2001/cwm

20:54:01 [sbp]

Bijan here is working on a clone of CWM

20:54:06 [sbp]

(written in Prolog)

20:54:25 [sbp]

and doing a rather brilliant job of it, I might add

20:54:32 [bijan]

* bijan blushes.

20:54:44 [bijan]

hardly! but it does seem to be shaping up.

20:56:46 [sbp]

ooh, my normalizer worked. Time to send it to www-archive...

20:57:01 [bijan]

Sean,i just uploaded the current version, if you want to try rules.n3

20:57:26 [bijan]

I don't think anony handling is going to make a difference.

20:57:31 [bijan]

I don't note any [] in your rules.

20:58:05 [sbp]

* sbp gets the latest version...

20:58:26 [bijan]

You want my 0_95, too.

20:58:32 [bijan]

As I had to strip out the """

20:58:49 [sbp]

O.K.

20:58:57 [bijan]

it's in cwmclone/

20:59:01 [bijan]

Obviously named :)

20:59:16 [sbp]

denshi: http://lists.w3.org/Archives/Public/www-archive/2002Feb/0043 and http://lists.w3.org/Archives/Public/www-archive/2002Feb/att-0043/01-norm

21:00:05 [sbp]

O.K. bijan, I've got the files, and CWMClone is running...

21:00:18 [AaronSw]

sbp is going for the most-posts-to-www-archive award

21:00:32 [bijan]

You want to consult_n3('path/to/rules.n3', user).

21:00:35 [sbp]

heh, indeed :-)

21:00:59 [denshi]

I'm trying to read all of this very fast...

21:01:03 [sbp]

bijan: it did it

21:01:09 [sbp]

denshi: :-)

21:01:21 [bijan]

If you compile_rules(user). you should get 17 rules.

21:01:38 [denshi]

i'm just missing the purpose of N-Triples

21:01:50 [sbp]

er... can't count them, but it kinda works

21:01:59 [bijan]

Look at the last rule.

21:01:59 [sbp]

NTriples are useful because they're very easy to parse

21:02:02 [bijan]

There shoudl be a number there.

21:02:04 [bijan]

In the header.

21:02:04 [sbp]

whereas N3 isn't

21:02:17 [sbp]

ah, you're right. The last one is 17

21:02:22 [bijan]

NTriples doesn't good chunks of n3.

21:02:33 [sbp]

I wrote an NTriple parser in about 10-or-so lines :-)

21:02:34 [bijan]

Now, if I filter(user, test). until I can't any more...

21:02:43 [bijan]

doesn't do good chunks...

21:02:48 [bijan]

but it's much easier, indeed.

21:03:10 [bijan]

Line orientation helps, too.

21:03:15 [sbp]

[[[

21:03:16 [sbp]

?- filter(user, test).

21:03:16 [sbp]

[4, 4, 4, 4, 4, 4, 4, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 8, 9, 9, 15, 15, 15, 15, 15, 15, 15, 15, 15,

21:03:17 [sbp]

[...]

21:03:18 [sbp]

]]]

21:03:42 [bijan]

findall(X, rdf(X,_,_, test), Xs), length(Xs, N).

21:03:47 [bijan]

Will give you a triple count for test.

21:03:48 [denshi]

back it up a bit... N3 & N-Triple fit where? In RDF description/discovery? Where in an RDF feed system is it used?

21:04:02 [sbp]

[[[

21:04:03 [sbp]

X = _G428

21:04:04 [sbp]

Xs = ['http://www.daml.org/2001/03/daml+oil#Class', 'http://www.daml.org/2001/03/daml+oil#Datatype', 'http://www.daml.org/2001/03/daml+oil#Restriction">http://www.daml.org/2001/03/daml+oil#Restriction', 'http://www.daml.org/2001/03/daml+oil#Restriction', 'http://www.w3.org/2000/01/rdf-schema#ConstraintProperty', 'http://www.w3.org/2000/01/rdf-schema#containerMembershipProperty']

21:04:16 [sbp]

N = 6 ;

21:04:17 [sbp]

No

21:04:17 [sbp]

]]]

21:04:17 [bijan]

Right, you can keep filtering until it, er, falters.

21:04:24 [sbp]

denshi: N3 and NTriples are useful alternate serializtions of RDF

21:04:33 [bijan]

4 times.

21:04:37 [sbp]

N3 is useful because it has extra logic constructs, e.g. formulae

21:04:45 [bijan]

N3, actually, is a logic langauge which is a proper superset of RDF.

21:04:50 [bijan]

And NTriples for that matter.

21:04:56 [sbp]

NTriples are useful because they don't suck like XML RDF

21:05:20 [bijan]

NTriples standardizes the common "triple view" of an RDF graph.

21:05:40 [bijan]

Which means you can actually use it for interchange.

21:05:46 [sbp]

yep, making it a lot easier to work out what triples you get from the serialiation

21:06:06 [sbp]

plus, NTriples (potentially) allows you to express things that XML RDF can't, as bijan said (a proper superset)

21:06:21 [bijan]

Well, both ntriples and n3 are proper supersets.

21:06:22 [sbp]

e.g. XML RDF doesn't let you end predicates with non XML-name chars

21:06:27 [bijan]

Ntriples is a *lot* closer.

21:06:44 [bijan]

NTriples is prolly RDF.

21:06:56 [bijan]

RDF/XML will turn out to be a rather poor serialization of it.

21:07:28 [bijan]

y'know, sean, it's *really* wacky that there should be multi-filterings.

21:07:53 [sbp]

what's going on?

21:07:58 [bijan]

I'm not sure.

21:08:11 [bijan]

What's *supposed* to be going on with a filter is that first you test to see what rules apply.

21:08:17 [bijan]

That's the list of numbers.

21:08:35 [bijan]

Then you run through that list applying the corresponding rule.

21:08:48 [bijan]

Not all the tests will yield correct applications.

21:08:58 [bijan]

As applying the rule might block a conclusion.

21:09:20 [bijan]

But why, then, are the rules working the second time around?

21:09:34 [bijan]

Theoretically, all the possible applications of the rules should be in the target context.

21:09:56 [bijan]

The target context *isn't* available to supply new premises.

21:10:47 [bijan]

Hmm. The no may be a clue.

21:10:53 [bijan]

it must be aborting early.

21:11:18 [bijan]

That would explain why apply_rules needs to be kickstarted.

21:13:36 [bijan]

Hmm. Weird, it stoped on rule 8, but then picked up on the next filter.

21:17:44 [bijan]

Seems like one extra positive gets in.

21:20:18 [bijan]

Ok, fixed that.

21:20:32 [AaronSw]

.google sandro hawke

21:20:33 [xena]

sandro hawke: http://www.w3.org/People/Sandro

21:20:37 [bijan]

I still end up with 95 in the target context.

21:21:41 [sbp]

.google Sandro SWAN site:www.w3.org

21:21:41 [xena]

Sandro SWAN site:www.w3.org: http://www.w3.org/2000/12/swan/swan-1.0.html

21:22:22 [AaronSw]

I'm writing up the Phone BOF proposal and needed a link

21:24:21 [bijan]

Ok, I still get 308 in the filter.

21:24:42 [bijan]

consult_n3('path/to/hacked/0_95.n3',user).

21:25:09 [bijan]

And then filter by hand until it stops.

21:25:45 [sbp]

it worked, but I don't know how many output triples I got...

21:26:14 [bijan]

findall(X, rdf(X,_,_, test), Xs), length(Xs, N).

21:26:31 [sbp]

N = 6

21:26:35 [sbp]

(again)

21:29:52 [bijan]

What?

21:30:03 [sbp]

N = 6

21:30:10 [bijan]

For which?

21:30:24 [sbp]

for findall(X, rdf(X,_,_, test), Xs), length(Xs, N).

21:30:29 [sbp]

and then "; No"

21:30:53 [bijan]

What did you do?

21:31:15 [bijan]

Doyou have both files consulted into user?

21:31:39 [sbp]

not sure...

21:31:46 [sbp]

shall I run it again from the start?

21:31:55 [sbp]

and note that I don't have the latest version

21:31:55 [bijan]

How many times did you filter?

21:32:00 [bijan]

Ah.

21:32:58 [denshi]

denshi has quit (Remote closed the connection)

21:33:10 [bijan]

That means you need to filter repeatedly until it stops.

21:34:27 [bijan]

But I uploaded the new one.

21:35:04 [sbp]

* sbp gets it

21:35:14 [sbp]

I have so many damn versions of the thing now :-()

21:35:17 [sbp]

:-)

21:37:35 [AaronSw]

Hm, I have trouble balancing proper link text with Fitt's Law.

21:37:49 [AaronSw]

Please review: http://logicerror.com/phoneBOFs

21:39:59 [AaronSw]

(sbp, that means you ;)

21:40:28 [sbp]

yep, that seems good as a proposal

21:40:35 [sbp]

can't you use singular "they", though?

21:40:56 [AaronSw]

I suppose, felt a little awkward.

21:41:04 [sbp]

perhaps you should give some examples of why BOFs would be needed

21:41:18 [sbp]

i.e. with respect to current cross WG problems, not hypothetical

21:41:28 [AaronSw]

Specific problems, hmm.

21:41:46 [sbp]

if there aren't any, then you have to question the point of BOFs :-)

21:42:01 [sbp]

I'm sure that there are plenty. They seem to be ten a penny in the WAI

21:42:04 [AaronSw]

oh, there are tons (see the "Why?" section) but i sorta wanted to keep it general

21:42:30 [sbp]

I think that some scenarios would really help the proposal. Just a couple, at the end (as an appendix, perhaps)

21:42:42 [sbp]

as DanC says: ground it in use cases!

21:42:45 [AaronSw]

ok.

21:42:48 [Morbus]

Morbus has quit ("http://www.disobey.com/")

21:42:52 [AaronSw]

you don't think that goes in "Why?"

21:43:10 [sbp]

Hmm... it would fit either end, I guess

21:43:18 [AaronSw]

gotta run, i'll fix it when i get back

21:43:24 [sbp]

O.K. c'ya

21:43:55 [AaronSw]

feel free to spool requests here

21:49:11 [bijan]

sean, did it work?

21:53:26 [sbp]

oops, sorry _ ihaven't had the chance to test it yet, and Fraiser's going to be on quite soon... I'll have to attend to it when I get back. Sorry

21:53:34 [bijan]

S'ok.

22:19:06 [Morbus]

Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack

22:19:38 [Morbus]

bwahah, this rocks.

22:19:40 [Morbus]

http://saladwithsteve.com/osx/2002_02_01_archive.html#10192247

22:27:54 [MorbusIff]

MorbusIff (~morbus@morbus.totalnetnh.net) has joined #swhack

22:28:16 [Morbus]

Morbus has quit (Read error: 104 (Connection reset by peer))

22:29:07 [bijan]

bijan has left #swhack

22:33:53 [AaronSw]

Cool, that was fun

22:35:07 [AaronSw]

Heh, after driving all that way i keep expecting the chair to lurch forward

22:38:47 [MorbusIff]

heh, heh.

22:38:53 [MorbusIff]

hey, AaronSw, do you read Forward: OS X?

22:39:09 [MorbusIff]

one of the guys was bitching about os x rss readers, and four people mentioned amphetadesk to him (he hadn't known of it before).

22:39:11 [MorbusIff]

that's kinda cool.

22:39:18 [AaronSw]

you mean salad with steve?

22:39:20 [MorbusIff]

yeah.

22:39:25 [AaronSw]

i don't read it, but I know all the writers

22:39:32 [MorbusIff]

you know "george" something or other?

22:39:35 [AaronSw]

they all hang out on infoanarchy

22:39:44 [AaronSw]

and stuff like that

22:39:48 [MorbusIff]

george something or other does?

22:39:54 [AaronSw]

Hm, dunno about him

22:40:13 [AaronSw]

george scriban... sounds familiar

22:40:24 [AaronSw]

oh, the blogaritaville guy

22:40:27 [MorbusIff]

yeah, that guy

22:44:08 [AaronSw]

sbp, need some non-rdf topics of discussion

22:47:35 [MorbusIff]

documentation!

22:48:04 [AaronSw]

heh

22:49:31 [AaronSw]

.google swad

22:49:32 [xena]

swad: http://swadautotransporters.com

22:49:37 [AaronSw]

.google swad w3

22:49:37 [xena]

swad w3: http://www.w3.org/2000/01/sw

22:56:10 [MorbusIff]

MorbusIff has quit (Read error: 104 (Connection reset by peer))

23:15:50 [Morbus]

Morbus (~Morbus@s116.terminal3.totalnetnh.net) has joined #swhack

23:16:04 [sbp]

Hi Morbus

23:18:25 [Morbus]

hey there sbpster!

23:18:31 [sbp]

Morbster!

23:19:05 [Morbus]

how are ya?

23:19:15 [sbp]

yes, thanks

23:19:17 [Morbus]

i hope to do something with amphta tonight.

23:19:19 [sbp]

I mean: fine :-)

23:19:21 [Morbus]

after i get all my other junk done

23:19:36 [sbp]

cool. what are you going to do to it? Nothing too perverted, I hope :-)

23:19:51 [Morbus]

nope, i'm gonna focus on the myChannels.opml file.

23:20:07 [Morbus]

adding unique id's, allowing autoimports of radio files, etc.

23:20:21 [Morbus]

and switching the template over to text::template.

23:20:26 [Morbus]

i'm hoping it'll go smoothly

23:23:22 [Morbus]

Morbus has quit ("http://www.disobey.com/")

23:42:23 [Morbus]

Morbus (~Morbus@63.173.138.115) has joined #swhack

23:43:49 [Morbus]

* Morbus grumbles

23:48:24 [Morbus]

.google barry ween

23:48:24 [xena]

barry ween: http://www.barryween.com

23:50:25 [AaronSw]

.dns cvs.plexnow.com

23:50:25 [xena]

cvs.plexnow.com - 209.216.14.26

23:51:32 [sbp]

lol!

23:51:36 [sbp]

this is hillarious:-

23:52:33 [AaronSw]

we're waiting

23:52:35 [AaronSw]

* AaronSw does the sonic the hedgehog foot tap

23:54:20 [sbp]

"2. Human readable information is good. Humans can sue it" - TimBL on www-tag

23:54:22 [sbp]

- mid:062601c1bfea$19ae2460$0301a8c0@w3.org

23:54:35 [AaronSw]

I was about to conclude that sbp had been abducted by Dr. Robotnic

23:54:43 [sbp]

heh, heh

23:54:45 [AaronSw]

Heh, that is funny.

23:54:55 [Morbus]

AaronSw: there's no rpobelms with blogify, right? i can still mention it to people?

23:55:08 [AaronSw]

yeah, you can

23:55:13 [Morbus]

k, thanks

23:55:18 [Morbus]

trying to get surlyrobot.com

23:55:22 [AaronSw]

but it's slow and buggy and i'm not going to fix their problems. so ha!

23:55:27 [AaronSw]

:-)