Re: e-triples

On Fri, Jul 24, 2009 at 9:11 AM, Martin Hepp
(UniBW)<martin.hepp@ebusiness-unibw.org> wrote:
> Dear Danny,
>
> Danny Ayers wrote:
>>
>> So Kingsley, you know this stuff. OpenLink provides good products. Oh yes.
>>
>> This is so in your scope, dude...ready?
>>
>> plug&play out of the box e-commerce solution.
>>
>> Personally I'd grumble a bit with Martin's modelling, but it is
>> usable. The Good Relations vocab is good enough.
>>
>
> Thanks - if you could serialize the grumbling into change requests, that
> would help ;-)
>
> Seriously, any suggestions for improvement are welcome!
>
> Note that GoodRelations tries to strike a balance between
>
> a)  a clean, reusable conceptual model and
> b) ease of creating annotations / populating knowledge bases from existing
> sources.
>
> In case of conflict, a) is more important for GoodRelations than b), because
> most transformations will take place in scripts anyway, and for simplifying,
> we have several shortcuts like gr:includes now.
>
> The key goal is that the data can be reused in as many contexts as possible,
> so GoodRelations maintains conceptual distinctions that e.g. commerce
> microformats don't (for example, product makes and models vs. actual
> products). That may be a bit of the burden initially, but will pay out in
> the long run.
>>
>> Front end a la Amazon, back end triples. Linked triples.
>>
>>
>
> Exactly. On a massive scale.
>>
>> Make it so - I reckon it could be a huge earner.
>>
>>
>
> I think so ;-) By the way: also for society, because the future wealth of
> developed economies will depend on our ability to coordinate the exchange of
> highly specific goods and services at much lower costs than today. Note that
> there is economic evidence that 50% of the US GDP goes into maintaining the
> infrastructure and institutions for exchange and trade.

Enjoyed this talk, looking at the contributing factors to transaction
costs, it seems to me that linked e-triples offer a unique way to
bring down that value from the 50%.  Looking at each item you mention:

1. Low Transaction Frequency

This ties in with what has been talked about on this thread.  Put the
data out there and let the web do the REST.  So the first step is
marking up your data using good relations.  We then need mechanisms to
route interest in an good or service back to the producer, without
necessarily incurring a high overhead.  How do we do it (HTTPS POST?
SPARUL?  Web Forms?)  I think we need a proof of concept here.

I guess this also means building order and payment gateways on a
generic, and per business, basis.  Important also to obey regulatory
conditions and compliance.  This could be a lucrative business area
for anyone that is prepared to specialise in it, or perhaps some banks
are well suited to take this on.

2. Specifity

For this we need aggregators of the market, perhaps ad servers too.
Search will be a key application, so perhaps GR can help instruct
robots that want to crawl goods together and provide offerings in the
marketplace.  Yahoo is a good start, but we need many more hubs of
semantic goods, exchanges and marketplaces with innovative designs.
Semantic tagging will help with specific search, as will geo location
of the business.

3. Uncertainty / Opportunism

This was the area I found the most interesting, and also the area
where e-triples offer a unique advantage.  The concept of linking an
Agent to an ID (eg URL / email / #phone) is gaining traction these
days.  The great benefit of this is that it can be an entry point into
linked data and then proceed into trust and reputation, in order to
reduce the costs related to uncertainty / opportunism.  Ebay ratings
is a micro version of this, however imagine a global version, where
you can see reviews and ratings of a particular offering, and where a
business can see that a user is securely authenticated, and has a
reputation for making good on their transactions.

It may be interesting to mock up a system where a business can go from
marking up a product to receiving an order (perhaps by email at
first), and looking at how all the pieces fit together, then expanding
to other cases.  I havent seen the GR oscommerce plugin, so maybe that
does some of this already.  It may be interesting to work out how we
could do this technically.

>
> See
> "The World Wide Web and the Wealth of Nations: Does IT Matter?"
> http://tr.im/inaugurallecture
>>
>> Cheers,
>> Danny.
>>
>>
>
> --
> --------------------------------------------------------------
> martin hepp
> e-business & web science research group
> universitaet der bundeswehr muenchen
>
> e-mail:  mhepp@computer.org
> phone:   +49-(0)89-6004-4217
> fax:     +49-(0)89-6004-4620
> www:     http://www.unibw.de/ebusiness/ (group)
>        http://www.heppnetz.de/ (personal)
> skype:   mfhepp twitter: mfhepp
>
> Check out GoodRelations for E-Commerce on the Web of Linked Data!
> =================================================================
>
> Webcast:
> http://www.heppnetz.de/projects/goodrelations/webcast/
>
> Recipe for Yahoo SearcMonkey:
> http://tr.im/rAbN
>
> Talk at the Semantic Technology Conference 2009: "Semantic Web-based
> E-Commerce: The GoodRelations Ontology"
> http://tinyurl.com/semtech-hepp
>
> Overview article on Semantic Universe:
> http://tinyurl.com/goodrelations-universe
>
> Project page:
> http://purl.org/goodrelations/
>
> Resources for developers:
> http://www.ebusiness-unibw.org/wiki/GoodRelations
>
> Tutorial materials:
> CEC'09 2009 Tutorial: The Web of Data for E-Commerce: A Hands-on
> Introduction to the GoodRelations Ontology, RDFa, and Yahoo! SearchMonkey
> http://tr.im/grcec09
>
>

Received on Friday, 24 July 2009 11:39:57 UTC