IRC log of dig on 2012-07-20

Timestamps are in UTC.

00:45:20 [rszeno]
rszeno (~rszeno@79.114.93.230) has joined #dig
00:48:51 [presbrey]
presbrey (~presbrey@2001:4830:2446:b5:aede:48ff:fe00:6001) has joined #dig
05:56:29 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
06:41:40 [melvster]
melvster has quit (Read error: Operation timed out)
07:13:13 [bblfish]
bblfish has quit (Ping timeout: 246 seconds)
07:14:09 [bblfish]
bblfish (~bblfish@AAubervilliers-651-1-320-113.w83-200.abo.wanadoo.fr) has joined #dig
08:59:29 [deiu]
deiu (~andrei@unaffiliated/deiu) has joined #dig
09:17:28 [bblfish]
bblfish has quit (Ping timeout: 246 seconds)
09:18:06 [bblfish]
bblfish (~bblfish@AAubervilliers-651-1-320-113.w83-200.abo.wanadoo.fr) has joined #dig
09:25:24 [bblfish]
bblfish has quit (Ping timeout: 260 seconds)
09:30:41 [bblfish]
bblfish (~bblfish@AAubervilliers-651-1-320-113.w83-200.abo.wanadoo.fr) has joined #dig
09:39:10 [bblfish]
bblfish has quit (Ping timeout: 246 seconds)
09:59:16 [bblfish]
bblfish (~bblfish@ALagny-551-1-67-174.w86-218.abo.wanadoo.fr) has joined #dig
10:16:16 [cheater__]
cheater__ has quit (Ping timeout: 246 seconds)
10:29:15 [cheater__]
cheater__ (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
10:59:32 [melvster]
melvster (~melvin@p5797F256.dip.t-dialin.net) has joined #dig
11:08:47 [danbri]
danbri has quit (Ping timeout: 255 seconds)
11:13:22 [RalphS]
RalphS (Ralph@30-7-118.wireless.csail.mit.edu) has joined #dig
11:46:54 [fission6]
fission6 has quit (Ping timeout: 264 seconds)
12:18:17 [melvster]
melvster has quit (Remote host closed the connection)
12:28:13 [bblfish]
bblfish has quit (Remote host closed the connection)
12:28:27 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
12:31:29 [bblfish]
bblfish (~bblfish@ALagny-551-1-67-174.w86-218.abo.wanadoo.fr) has joined #dig
13:01:58 [danbri_]
danbri_ (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
13:04:56 [danbri]
danbri has quit (Ping timeout: 255 seconds)
13:18:37 [bblfish]
bblfish has quit (Ping timeout: 246 seconds)
13:24:10 [bblfish]
bblfish (~bblfish@AAubervilliers-651-1-320-113.w83-200.abo.wanadoo.fr) has joined #dig
13:36:51 [fission6]
fission6 (~fission6@cpe-74-66-1-166.nyc.res.rr.com) has joined #dig
14:01:33 [melvster]
melvster (~melvin@p5797F256.dip.t-dialin.net) has joined #dig
14:04:16 [melvster]
melvster has quit (Remote host closed the connection)
14:07:03 [melvster]
melvster (~melvin@p5797F256.dip.t-dialin.net) has joined #dig
15:20:52 [melvster]
melvster has quit (Read error: Connection reset by peer)
15:22:17 [melvster]
melvster (~melvin@p5797F256.dip.t-dialin.net) has joined #dig
15:32:08 [danbri_]
danbri_ has quit (Ping timeout: 255 seconds)
15:33:36 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
15:37:13 [cheater__]
cheater__ has quit (Ping timeout: 246 seconds)
15:45:06 [deiu]
deiu has quit (Quit: Leaving)
15:55:05 [cheater]
cheater (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
16:01:01 [cheater]
cheater has quit (Ping timeout: 246 seconds)
16:33:03 [bblfish]
betehess: not progressed much this week. Though we made a big progress with french student r_blin, who sometimes is on this channel
16:33:40 [bblfish]
He bought himself a new notebook 4 core 8 threads, with a new version of Linux and skype, and so I was able to help him debug rdflib.js
16:34:21 [bblfish]
His problem is that he did not really know how to work with JS - which is hard if you come from Java - since there is absoltely no type checking. All Type Checking has to be done by hand.
16:35:16 [bblfish]
Also he learned about debugging JS, stepping through the code, reading rdflib - only way to get to know it , as there is no documentation at present - and understanding the difference beteween a graph and a store.
16:35:34 [bblfish]
so he should be a lot more productive now.
16:35:51 [bblfish]
that he understands all those concepts
16:36:10 [bblfish]
All these interventions do slow me down.
16:36:13 [bblfish]
...
16:37:05 [bblfish]
( also it's a huge hurdle to get people to learn to work in distribute teams )
16:37:47 [bblfish]
eg: having a skype coding session - a la pair programming made a huge difference today - was not possible before .
16:56:19 [cheater]
cheater (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
17:00:52 [cheater]
cheater has quit (Ping timeout: 246 seconds)
17:12:59 [timbl]
timbl has quit (Quit: timbl)
17:13:47 [timbl]
timbl (~timbl@host86-144-247-27.range86-144.btcentralplus.com) has joined #dig
17:13:49 [timbl]
timbl has quit (Client Quit)
18:25:12 [danbri]
danbri has quit (Remote host closed the connection)
19:26:45 [cheater]
cheater (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
20:22:25 [RalphS]
RalphS has quit ()
20:38:34 [cheater]
cheater has quit (Ping timeout: 246 seconds)
20:50:31 [bblfish]
that Web browser implementors are not willing to risk breaking any such deployments, however convoluted that makes the resulting technology. http://lists.w3.org/Archives/Public/public-webappsec/2012Jul/0068.html
20:50:53 [bblfish]
(forgot quotes)
20:51:12 [bblfish]
apparently that explains why CORS cannot GET pulbic resources
20:51:45 [dsheets]
bblfish: hixie also believes that using URIs for identifiers that may not resolve over the network is a Bad Idea because it is Confusing
20:52:32 [bblfish]
really :-)
20:52:40 [dsheets]
one moment
20:54:00 [dsheets]
http://lists.w3.org/Archives/Public/public-webapps/2011JulSep/1584.html
20:54:41 [cheater]
cheater (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
20:55:01 [dsheets]
This thread was cited as a philosophically-sound refutation of my proposal to lift WebGLSL extension names into URI: http://www.khronos.org/webgl/public-mailing-list/archives/1205/msg00138.html
20:55:57 [dsheets]
fwiw, imho if you have security-sensitive information protected by IP, you are doing it wrong and your information must not be that sensitive
20:55:58 [bblfish]
ah ok, he does not understand name spaces.
20:56:14 [bblfish]
dsheets: very good point
20:56:16 [dsheets]
bblfish: who? hixie or glenn?
20:56:41 [bblfish]
hixie, it looks like
20:57:00 [dsheets]
yes, exactly, and that ignorance is spreading among the html5-head
20:57:00 [dsheets]
s
20:57:27 [bblfish]
well in fact in html5 they are using a uri. They just have a hidden namespace
20:57:41 [bblfish]
it's http://ieff.org/mimes/text/html
20:57:47 [bblfish]
( or something like that )
20:58:20 [dsheets]
yup but you can't show that to poor authors, they might get confused
20:58:37 [bblfish]
but ok. In any case they are stymining debate by saying they can't change anything because browser vendors don't want it
20:59:08 [bblfish]
the close one gets to showing that this would work the more vociferous they get
20:59:11 [dsheets]
bblfish: browser vendors are not the constituency that matters… merely the gatekeepers
20:59:35 [bblfish]
who are the getekeepers?
20:59:44 [dsheets]
browser vendors
20:59:55 [bblfish]
ah
20:59:56 [dsheets]
but their users are the most important constituency
20:59:56 [bblfish]
:-)
21:00:10 [cheater]
cheater has quit (Ping timeout: 260 seconds)
21:00:20 [bblfish]
ah I see how to parse your sentence
21:00:23 [dsheets]
and the browser vendors operate under the mistaken belief that they act in the interest of their user base
21:00:50 [bblfish]
well I myself doubt that he is fully representing the browser vendors in this case
21:01:09 [bblfish]
if he were, then the simplification that would result may be something they could agree on
21:01:23 [bblfish]
s/if he were//
21:01:25 [dsheets]
in this case, the browser vendors are claiming that the best course of action for all users is to protect those few who operate sensitive web resources with weak security policies
21:01:48 [bblfish]
well I am not sure. Who in that discussion is a browser vendor?
21:01:52 [bblfish]
or representing one?
21:02:30 [dsheets]
Anne van Kesteren works for opera
21:02:36 [bblfish]
they could just be lazy employees that don't want to do the heavy work of thinking this through, and want to get their standard published
21:02:39 [bblfish]
ah
21:02:58 [dsheets]
hixie is obv. a standardista that works with vendors closely to codify their brokenness
21:03:13 [dsheets]
viz html5
21:03:59 [dsheets]
if you allowed unauthenticated GET from page agents, the web would be more decentralized and browser vendors (and the major central repositories some of them manage) would be less powerful
21:04:22 [bblfish]
ah, I am not sure that that is their thinking
21:04:32 [bblfish]
I think they are just frightened chickens
21:04:46 [dsheets]
frightened of?
21:04:52 [bblfish]
thinking
21:04:54 [bblfish]
or working
21:05:05 [bblfish]
because it is a slightly tricky issue
21:05:14 [bblfish]
if they get it wrong they could get a lot of bad heat
21:05:30 [bblfish]
( security flaw on the headline news! )
21:05:56 [dsheets]
sure… they are currently working to consolidate and centralize their control, though
21:06:05 [bblfish]
so they are probably not taking the risks that their employees expect them to
21:06:12 [dsheets]
the recent major push is to expand js API surface area and codify broken HTML
21:06:26 [bblfish]
But it makes no difference. With a CORS proxy you get the data anyway
21:06:31 [dsheets]
why would they innovate when they are busy maturing and digging moats?
21:06:49 [cheater]
cheater (~cheater@g230223075.adsl.alicedsl.de) has joined #dig
21:07:03 [dsheets]
you get the data but cause an intermediary to use 2x bandwidth
21:07:07 [bblfish]
well, that's why I think this probably makes no difference to the vendors. They are probably just want to go to final call
21:07:55 [bblfish]
true dsheets, but I don't think these people think at the level of trying to consolidate
21:08:14 [dsheets]
so they serve those interests blindly?
21:08:26 [bblfish]
no, they are just frightenened chickens
21:08:29 [dsheets]
or lobotomize to stop the dissonance?
21:08:39 [bblfish]
and they don't have a big picture interest to go for better quality
21:08:47 [bblfish]
so they go for what is easier
21:09:05 [dsheets]
ok so they are lazy and tired and thinking 1yr ahead instead of 50yrs
21:09:14 [bblfish]
( anyway, just a hypothesis: trying to explain their behavior while keeping it rational - ala Donal Davidson )
21:09:25 [bblfish]
instead of 5-10
21:09:48 [dsheets]
that seems like a reasonable hypothesis to me… how do you leverage that to win over the community and draw out supporters?
21:10:16 [dsheets]
offer the huge upside, the minor/silly downside, and the futility due to proxies?
21:10:39 [dsheets]
those speaking with some authority (hixie) seem to have shut down those avenues
21:10:51 [bblfish]
well, perhaps the W3C has policies on short term thinking such as that one
21:11:15 [bblfish]
perhaps there is a security group who is there to say: oops! don't think so short term. it costs us money
21:12:11 [dsheets]
it is crazy… when i was at w3c in summer 2006, timbl concluded that tabulator needed to be a ff extension to work around this nonsense… and here we are 6yrs later...
21:12:12 [bblfish]
One would need to do some research on those questions I brought up, with careful answers.
21:12:44 [dsheets]
so the long term thinking has been there, the apps have been there, the idea has been there and it hasn't caught on or been pushed
21:13:08 [bblfish]
true
21:13:31 [bblfish]
the apps need to get virally good
21:13:38 [bblfish]
then of course they will be forced to move
21:13:41 [dsheets]
chicken/egg
21:13:50 [bblfish]
not really. The CORS proxy helps
21:14:18 [bblfish]
in fact we can do all this through a CORS proxy on our Freedom box
21:14:22 [dsheets]
cors proxy for image textures is expensive
21:14:40 [bblfish]
ah, how do image textures come in?
21:14:41 [dsheets]
does that exist?
21:14:55 [dsheets]
bblfish: can't read image data out of cross-domain images
21:15:09 [dsheets]
bblfish: because it might be your bank capcha or sth
21:15:41 [bblfish]
ah but all we are talking about is publicly available resources
21:15:59 [bblfish]
for the moment in this CORS discussion
21:16:21 [bblfish]
bank captcha would be private right? (mhh, you are right, could be public)
21:16:59 [bblfish]
for a good social web through a freedom box we probably need to go 2 ways
21:17:09 [bblfish]
1. have the FB do all the work
21:17:24 [bblfish]
2. build thick clients on the desktop that do - no need for browsers there anymore
21:17:39 [bblfish]
( sadly Apple by not supporting Java has made 2 more expensive )
21:19:42 [bblfish]
but well building browsers across platforms has always been expensive
21:20:58 [bblfish]
Still I think in the end that is what is going to happen: specialised linked data browsers such as an rdfTunes
21:22:23 [bblfish]
there one could be a lot more flexible with security policies.
21:22:32 [bblfish]
( which are tricky for sure )
21:23:13 [bblfish]
dsheets: where are you working?
21:23:25 [dsheets]
san francisco
21:24:07 [bblfish]
ah cool https://www.youtube.com/watch?v=LuDN2bCIyus
21:26:07 [bblfish]
how the driver does that?
21:26:15 [bblfish]
neither do I :-)
21:26:32 [dsheets]
that appears to be newtonian physics and nice automotive design and operation
21:26:46 [dsheets]
i don't understand how the venue was acquired and liability was considered
21:27:29 [dsheets]
"expensively"?
21:27:54 [bblfish]
very likely very expensive
21:28:01 [bblfish]
( the car has the number 43 on it )
21:30:02 [dsheets]
43?
21:32:22 [bblfish]
22 million people viewed this
21:32:26 [bblfish]
that's how he finances it
21:32:29 [bblfish]
a blockbuster
21:33:11 [dsheets]
viewed for free… the adverts/sponsorship must compensate
21:33:16 [dsheets]
the driver needs compensation
21:33:47 [dsheets]
the economic impact to the city is multiple millions on dollars and the liability is multiple millions of dollars… doesn't seem worth it
21:36:28 [dsheets]
ironically, turing complete and wide-surface JS can run cross-domain but data can't...???
21:37:05 [dsheets]
What about those intranets that have sensitive JS?
21:37:46 [dsheets]
<script src="http://sekkrit/sensitive/script.js"></script> ?
21:37:57 [bblfish]
I think it's a series, he must get his money back on it.
21:38:07 [bblfish]
SF probably pays for the advertising it gives SF
21:38:19 [bblfish]
the car manufacturer pays for the cars
21:38:29 [bblfish]
Red Bull pays a lot
21:40:22 [bblfish]
Not sure how we can shift the discussion there
21:41:49 [dsheets]
dunno… with their current attitude the non-secure security policy of network topology will be enshrined forever at the expense of public technical capability
21:42:02 [dsheets]
have a good pint
22:28:37 [melvster]
melvster has quit (Ping timeout: 240 seconds)
23:05:25 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
23:11:08 [amy]
amy has left #dig
23:15:21 [danbri]
danbri has quit (Remote host closed the connection)
23:35:14 [betehess]
betehess has quit (Quit: Leaving)
23:42:03 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
23:42:21 [danbri]
danbri has quit (Remote host closed the connection)
23:55:27 [danbri]
danbri (~danbri@cpc6-aztw25-2-0-cust83.aztw.cable.virginmedia.com) has joined #dig
23:59:35 [danbri]
danbri has quit (Remote host closed the connection)