Re: Normative vs Informative

On 28 Nov 2011, at 20:06, Peter Williams wrote:

> 
>  THere is nothing complex about webid. Its an HTML page with 3 strings, that a CGI script reads given an URL in a cert. It then does a rather trivial string match producing yes/no.

Exactly! WebId is not complicated. I started working on it because I was looking for the simplest possible thing to get people to understand the value of the semantic web. Looking for that we found accidentally the simplest possible way to brings security together with linked data in a useful way that is people centric. So don't be surprised that it is simple. 

>  Its only the semantic web and its wider goals that is making this hard. Ive seen some progress here, however, with folks stripping out from the spec biases which promote some or other agenda. 

Come on. The biggest issues we have been having here is not the semantic web ones, but the cryptographics ones, the weird  binary formats such as ASN.1 that are pre web, 1970ies technology, the hidden features hidden around in RFCs everywhere, etc. etc...

But anyway, "simple" is viewpoint relative. What is simple for you - X509, TLS, crypto wars - is complex for others. Where semantic web is simple for others it is complex for you. Each has something to bring to the table, but for that people have to be open to learn across their field of expertise, to be open to new ideas. Those who do that will discovers whole new spheres of possibilities that were just not visible before, as I have discovering crypto, and you are discovering linked data.

>  
> The spec is slowly saying (effectively) what I said above, and without sounding like a philosophy lecture.
>  
> Webid is a test for me: is the semantic web (in its "foaf project" form) CULTURALLY ready for mainstream identity management?

It would be nice if you said this without sounding like you were yourself lecturing. I don't know if you noticed, but identity management is a pretty big shambles. So it's not like there is anyone who is in a position to lecture here.

> Is it mature yet (or still full of research angst, and the usual claim/counter-claim style of argument that facilitates ongoing research)? Are things sufficiently mainstream for consumers to impose no learning significant curve? Given a hint or two that removes gotchas, can an engineering student who has taken a couple of programming classes make something work, in the usual open systems fashion? Do we have to wait 2 years for a Microsoft or Apple or Oracle to do X to the platforms?

And how will you tell if it is researchy or not? Don't wait on others: build it yourself. You have the opportunity to do something big. If you don't do it now you'll look like the person who could have bought Microsoft stock in 1979, but did not because they wanted to wait to make sure everything was secure.  There is nothing to be gained when the risk is gone.

>  
> A seperate measure is whether industrial-grade vendors are onboard, and ready to apply professional data center engineering, including hardware crypto. But, until the first test is met, who cares. hardware folks only react once the software solutions have reached their limit, including economic limits.

"Industrial", "Professional grade", ... all of that is empty marketing speak. People who create things don't measure themselves on those yard sticks.

Please, let's stop this type of empty talk, and start building cool apps.

Anyway, thanks for helping improve the spec. You detailed criticism of the text was to the point and helpful.

	Henry


>  
>  
>  
>  
> > From: henry.story@bblfish.net
> > Date: Mon, 28 Nov 2011 17:35:09 +0100
> > CC: kidehen@openlinksw.com
> > To: ddooss@wp.pl; public-xg-webid@w3.org
> > Subject: Re: Normative vs Informative
> > 
> > Dominik is referring to the discussion in Berlin documented on the wiki
> > 
> > http://www.w3.org/2005/Incubator/webid/wiki/Berlin_meetup_2011
> > 
> > especially the first photo of a diagram drawn up by Dan Brickley showing the complexity chart of WebID. It is potentially explosive if we don't limit ourselves. We know we can grow to accept the full potential. We need to be a bit more than just one format - because then we can't tell the semantic story so well, and we want to prepare people for the fact that we are inclusive. But at the same time we can't start with everything either, because we would be like fools drowning under our own self imposed workload. So we decided there to keep things simple enough. There was pretty good support of that there, and I think that support remains here.
> > 
> > Henry
> > 
> > On 28 Nov 2011, at 13:48, Dominik Tomaszuk wrote:
> > 
> > > On 28.11.2011 00:00, Kingsley Idehen wrote:
> > >> On 11/27/11 6:46 AM, Dominik Tomaszuk wrote:
> > >>> On 27.11.2011 03:14, Henry Story wrote:
> > >>>> Great idea. Let's start with a wiki.
> > >>>> Open one page: Formats, and have it point to an example page in every
> > >>>> other format
> > >>>> you can think of - so that each format can get the attention it
> > >>>> deserves.
> > >>>> 
> > >>>> Here I wrote up the initial page:
> > >>>> 
> > >>>> http://www.w3.org/2005/Incubator/webid/wiki/Formats
> > >>> OK, but first I think we should discuss in mail list. For example I
> > >>> disagree that json-ld is important now to WebID. I'm very big fun of
> > >>> JSON-LD, but now it's community draft and we have RDF/JSON draft [1]
> > >>> from RDF WG.
> > >> 
> > >> The issue isn't about JSON-LD or RDF/JSON etc.. WebID shouldn't be about
> > >> Syntax. Leave syntax to the implementation details bucket. The concepts
> > >> behind WebID has zilch to do with syntax.
> > > I don't know what's the point... I've just create wiki pages with examples how WebID profile can look like.
> > > We talk about this many times in Berlin, in private mails, teleconf. and IMO there is consensus: WebID must be simple and core of WebID should support minimum of standards, ontologies, URI schemes, formats, protocols etc.
> > > Of course we have a lot of possible serialization, but only two of them should be supported in spec. There is only questions:
> > > 1) Which of then should be supported via normative section?
> > > 2) Which of then should be supported via innormative section?
> > > 3) Which of then shouldn't have reference in spec?
> > > I disagree that WebID shouldn't show any examples of syntax, because it should be simple. Simple isn't mean that you link and say: "read and analyze:
> > > 1) RDF Concepts and Abstract Syntax spec
> > > 2) RDF Semantics spec
> > > 3) RDF Schema spec
> > > 4) FOAF spec
> > > 5) Cert ont
> > > 6) RDF/XML spec
> > > 7) Turtle spec
> > > 8) etc
> > > and then you know how the syntax should be written".
> > > "Simple" for me means to see some examples that I can copy/paste change some small things and it should works. Do not scary people Semantic Web standards.
> > > I do not want add to spec (in normative section) all of serialization possibles! This wiki is only starting point of discussion. Additionally, this examples may be useful for implementations. That's all.
> > > 
> > > Best regards,
> > > Dominik 'domel' Tomaszuk
> > 
> > Social Web Architect
> > http://bblfish.net/
> > 
> > 

Social Web Architect
http://bblfish.net/