Re: Syntactical interoperability: a strategic consideration?

Hi.
See Stamper's Ladder in "A SEMIOTIC THEORY OF INFORMATION AND
INFORMATION SYSTEMS" by Ronald Stamper (1993) 
https://research.utwente.nl/files/5383733/101.pdf
I don't know whether this is an agreed definition of semantic/syntactic 
interoperability, but it is what I use to teach data governance and 
information systems. It is helpful here as the paper proposes that 
syntactic interoperability is needed in order to establish semantic 
interoperability - which is Ruben's argument.

Regards, Harsh

On 16/01/2025 15:55, Theissen-Lipp, Johannes wrote:
> Dear all,
> 
> Thank you, Ruben, for pointing out the layers and importance of existing pain points. I too support the continued development of standards. And I agree with your statement: We also need to address the practical / broader challenges that exist at a meta level. With the image from your SW identity crisis blog post in mind, I agree that the remaining 20% task of simply putting solutions into practice often requires another +80% effort instead. This means that the community should drive both foundational standards and practical solutions for broad user bases in parallel.
> 
> By the way: Is there an agreed-upon definition of (semantic/syntactic) interoperability? I have never interpreted it as a boolean "yes/no" property, but as some sort of distance metric between two artifacts.
> 
> Best,
> Johannes
> 
> -----Original Message-----
> From: Ruben Verborgh (UGent-imec) <Ruben.Verborgh@UGent.be>
> Sent: Donnerstag, 16. Januar 2025 15:26
> To: semantic-web@w3.org
> Subject: Syntactical interoperability: a strategic consideration?
> 
> Dear all,
> 
> The Semantic Web community aims to build systems with semantic interoperability.
> 
> As many of us have probably experienced at some point, convincing others that RDF is a great fit for a particular use case, can sometimes be tricky.
> However, we always have the promise of long-term interoperability as a key argument.
> 
> Recently, I became aware that the upcoming RDF 1.2 standards might update the existing definition of text/turtle and others.
> For those interested, the technical discussion is happening at https://github.com/w3c/rdf-star-wg/issues/141
> 
> However, there is potentially a larger strategic consideration that concerns all of us.
> If we arrive at a situation where two systems that both indicate to accept text/turtle, cannot talk to each other, we have a massive reputation problem.
> How can we claim semantic interoperability, if we do not have explicit syntactical interoperability?
> 
> I fear this adds brittleness to RDF-based systems, to the extent that it might cause people to conclude “RDF doesn’t work”.
> And I’d be inclined to actually agreeing with them, rather than attempting a conversation about RDF 1.1 versus 1.2 details, while selling them on RDF was the hard thing in the first place. It endangers the credibility of our entire technology stack, in my opinion.
> 
> I’m not looking to add pressure on the WG or the technical side of the argument.
> Rather, I’m interested in your opinions on whether the strategic and reputational risk is as big as I perceive it, and hopefully to find people who can defuse my arguments and worries.
> 
> Thanks in advance,
> 
> Ruben
> --
> Ruben Verborgh (they/them)
> Professor of Decentralized Web Technology IDLab, Ghent University – imec https://ruben.verborgh.org/ – @RubenVerborgh
> 

-- 
---
Harshvardhan J. Pandit, Ph.D
Assistant Professor
ADAPT Centre, Dublin City University
https://harshp.com/

Received on Thursday, 16 January 2025 18:33:39 UTC