Re: AW: RAX Position Paper

Jean-Pierre,


Let me rephrase that ... when I said 'horrible', what I meant was 'a sophisticated serialization of the RDF graph which does not lend itself to being a source for transformation to other serializations, such as Turtle or JSON'.  At the time, these other serializations did not exist.  Arguably, if RDF/XML had been a cleaner representation of the underlying graph, they wouldn't have gained traction in the way they have.


Richard

________________________________
From: Evain, Jean-Pierre <evain@ebu.ch>
Sent: 20 July 2017 11:27:30
To: Richard Light; Lange-Bever, Christoph; Felix Sasaki; Dirschl, Christian
Cc: Jose Emilio Labra Gayo; Axel Polleres; public-rax@w3.org
Subject: RE: AW: RAX Position Paper

I have to disagree with saying that RDF/XML is horrible.

And I have written converters from xml to JSON-LD too. I am therefore not a RDF/XML integrist.

Still we can't take that sort of position or we might just lose an important part of the audience.
We must be agnostic to the RDF representation format or we may be throwing the baby with the bath water.

Jean-Pierre

________________________________
From: Richard Light [richard@light.demon.co.uk]
Sent: 20 July 2017 12:20
To: Evain, Jean-Pierre; Lange-Bever, Christoph; Felix Sasaki; Dirschl, Christian
Cc: Jose Emilio Labra Gayo; Axel Polleres; public-rax@w3.org
Subject: Re: AW: RAX Position Paper


Hi,


I have gone over the paper and tried to make the text clearer and more punchy.  Left a couple of comments for resolution.  One comment is about RDF/XML: shouldn't we state up front that there is a relationship between XML and RDF, and it's horrible?


As regards your point about mapping complex XML to RDF: there clearly needs to be a mechanism for declaring that mapping (it has to be a less-than-100% downstream conversion), and this could be one of our 'products'.


Best wishes,


Richard

________________________________
From: Evain, Jean-Pierre <evain@ebu.ch>
Sent: 20 July 2017 07:34:21
To: Lange-Bever, Christoph; Felix Sasaki; Dirschl, Christian
Cc: Jose Emilio Labra Gayo; Axel Polleres; public-rax@w3.org
Subject: RE: AW: RAX Position Paper

Another point:

about section 4.

To my opinion the key issue is that complex xml cannot be easily transformed into RDF:

- complex xml means that we can extend and complexify a model at will using e.g. intricated complex structures
- This cannot be easily transformed into RDF.
1.  it leads to creating new objects for which the need for unique identification is difficult to handle and blank nodes doesn't help either
2. it obliges  to derive differnet RDF model from the xml, application wise.

In fact this would lead to guidelines to either write RDF friendly XML or guidelines explaining the above.
________________________________________
From: Evain, Jean-Pierre
Sent: 20 July 2017 08:22
To: Lange-Bever, Christoph; Felix Sasaki; Dirschl, Christian
Cc: Jose Emilio Labra Gayo; Axel Polleres; public-rax@w3.org
Subject: RE: AW: RAX Position Paper

Section 3 - Motivation

First sentence:
I would rephrase "to address the Big Data" into "to address the challenges and needs of Big Data"

Big Data is not the most pressing issue but one of the most pressing issues.. :--)

One of the barriers is that a lot of data content providers have is at best XML. It is therefore required to provide tools to shift from the XML to RDF paradigm in order to benefit from semantic solutions.

Jean-Pierre

________________________________________
From: Lange-Bever, Christoph [Christoph.Lange-Bever@iais.fraunhofer.de]
Sent: 20 July 2017 03:12
To: Felix Sasaki; Dirschl, Christian
Cc: Jose Emilio Labra Gayo; Evain, Jean-Pierre; Axel Polleres; public-rax@w3.org
Subject: Re: AW: RAX Position Paper

Hi all,

Felix Sasaki <felix@sasakiatcf.com> on 19 July 2017 20:51:
> "About the contents...we should probably add some content because at this moment it is quite general...the problem that I can see is that if we start adding approaches or technologies, then the paper would  automatically grow in size...while if we try to keep it simple and just add a few, then it would be unfair for the technologies that we don't mention.."
>
> We still have space for about two paragaphs to takle this issue, but I have no good solution how to address this.

I passed over the whole paper, leaving some comments for you where I couldn't solve problems myself.

As we did already have some high-level conclusions about XSLT, the main topic on which we have done actual work, why not include them into the paper.

But to make this possible I had to shorten in other places.  Mainly by rephrasing and reformatting, not by removing ideas.  The paper didn't have the right format; now it has.  (One LNCS page is really small.)  Now it's still slightly too long, but I'm still in favour of at least having _one_ concrete contribution in the paper, other than just giving a high-level overview of the group's motivation and organisation. I (or you) might be able to come up with further ideas for shortening, or, if not, we could still remove the XSLT details.

Cheers,

Christoph

--
Dr. Christoph Lange
Fraunhofer IAIS: head of Enterprise Information Systems department
+49 2241 14-2428 (redirects to mobile); room B3-216
Universität Bonn: Smart Data Analytics, senior researcher
+49 228 73-4603; Römerstraße 164, Room A209
Further contacts (Skype etc.):  http://langec.wordpress.com/about
------------------------------------------------------------------------------

**************************************************
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed.
If you have received this email in error, please notify the system manager. This footnote also confirms that this email message has been swept by the mailgateway
**************************************************



------------------------------------------------------------------------------

**************************************************
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed.
If you have received this email in error, please notify the system manager. This footnote also confirms that this email message has been swept by the mailgateway
**************************************************

Received on Thursday, 20 July 2017 10:44:11 UTC