W3C

- DRAFT -

RDF WG FtF Breakout on JSON

14 Apr 2011

See also: IRC log

Attendees

Present
cycri, StevenPemberton, mischat, Guus, tomlurge, NickH, tomayac
Regrets
Chair
Guus
Scribe
mischat_

Contents


<mischat> hello all

<mischat> 26632

<mischat> zakim code

<cygri> tomayac, can you hear us?

<Steven> tomayac, can you hear us?

<mischat> FWIW http://steveharris.tumblr.com/post/4590579712/construct-json

<mischat> what steve and I talked about last night

Hello all

cygri: Summarising yesterday, linked-data-api and talk about the cut down BBC json

<tomayac> is there a link to the bbc json story? thanks.

Guus: asked if there is a document describing the work

<mischat> http://www.bbc.co.uk/programmes/b00sbqnb.json

<cygri> Linked Data API: http://code.google.com/p/linked-data-api/

<mischat> the link from yesterday ^^ bbc "cut down json"

NickH: stated that BBC are hoping to write up the json work

cygri: asked if the BBC work is a generic rdf -> json ?

<NickH> RDF/XML: https://gist.github.com/919228

NickH: Stated that the work was designed to handle generic rdf

<tomayac> memorizing the NYT json again: http://data.nytimes.com//60694995023816375851.json

<NickH> JSON: https://gist.github.com/919230

<NickH> not sure if this is an ideal example

tomayac: Do say if you would like go talk about the nytimes json

<NickH> not, it isn't

<NickH> Will get some proper examples on to the mailing list

<NickH> with some code snippets

<tomayac> mischat_: simply mentally counter-posed it to bbc's json

So we have three 3 use-cases :

1st use case, which json-ld was presented as a potential starting point

The use case is that you want to give developers the choice to parse the data as json or with the rdf stack

<NickH> tomayac: is there a spec for the format of the nytimes json?

<tomayac> NickH: not one that i'm aware of. it's easy enough to derive it by "reverse-engineering" (blown-up word) the sample code

<NickH> sure

cygri: To be fair to manu et al and json-ld Has been presented to cover more than just the 1st use case

<Guus_> http://code.google.com/p/linked-data-api/wiki/API_Deployment_Example

<tomayac> _about is the subject

<tomayac> on the same level as predicates

tomayac: Talked to Evan sandhaus from The nytimes... some points about the json

It has a comment in it, which they had to do for licensing purposes, if you pass a callback function it will strip out the comment

All of the slash escaping was

<scribe> Done so that the json could be put in <script/> .

<tomayac> mischat_: small correction: if a string contains </script> ("blabla </script> blalala") this opens the doors for XSS attacks, so they forward all forward slashes

Thanks tomayac

<tomayac> escape, not forward

Nick asked about rdfa profiles and how

They relate to the json work

<NickH> mapping that is used to generate both the RDF and JSON:

<NickH> https://gist.github.com/919232

<tomayac> sparql results json seems adequate for results that are well-represented as a table (many rows, similar structure each time), but not so well for just one, deeply nested result

Sure tomayac I agree, was just playing devils advocate

<tomayac> got it, mischat_

We need to capture the inputs on the wiki, we have 3 inputs it seems for the use-case 3

These are 1 linked data api, 2 BBC json, and 3 nytimes json

<tomayac> they threw away a lot...

<tomayac> (not necessarily bad)

<tomayac> no dot notation

<tomayac> need to obj['cu:rie']

<tomayac> cant do obj.cur:ie

Guus_: Is talking about the recipes and how useful a note would be to start the conversation

<tomayac> manu and nathan are for sure in

<tomayac> (so am i, but need the email for concrete todos [hard to follow on the phone only])

Of course

<tomayac> (plus missed all the discussions from y'day)

Again I don't do work which other people would like to do

<tomayac> (so documented stuff helps a lot)

Will get some text up over the weekend

Anyways ... We are grabbing lunch now

<tomayac> thanks, mischat_! highly appreciated

<tomayac> bon app├ętit!

<mischat> np

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2011/04/14 12:18:04 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: mischat_
Inferring Scribes: mischat_

WARNING: No "Topic:" lines found.

Default Present: tomayac, mischat
Present: cycri StevenPemberton mischat Guus tomlurge NickH tomayac
Got date from IRC log name: 14 Apr 2011
Guessing minutes URL: http://www.w3.org/2011/04/14-rdf-json-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


[End of scribe.perl diagnostic output]