Re: Exercise: LOD questions (R)†was ( Do we need another list(s)? )

This is a series of questions that I received from a member in our Semantic
Web Austin group. I can't seem to answer all of them thoroughly. Hope we can
get more answers on this and we can start creating material for the FAQ I am
proposing.

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

These are some things I'd like to see in an FAQ. Here are a few questions
from the
point of view of a business person who is vaguely aware of LOD but not
clear on its use:

- My company has recently released an API for access to structured
(database) data about 55 million companies and 35 million people. Do
you think I should release this in an LOD format? How would my
customers benefit.

- Can you give a use case for mixing LOD with privately supplied data
(from my companies own data sources or from user-generated content) to
produce a useful application?

- What commercial applications are there that use LOD?

- What are some of the major limitations of today's software that
would be improved upon by using LOD?

- How does LOD fit into the bigger Semantic Web picture? Write a
seperate sentence or two for each of the following terms that states
how that topic relates to LOD:
RDF
OWL
Resource
URI
Ontologies
Agent
Service discovery
Triples

Example: "RDF is a standard format that can be used to publish LOD
datasets."

- Okay, I'd like to use LOD for a pilot of a commercial project. I'm
going to include 1 million triples. What production-environment
resources will I need to set up. What will my architecture include?
Will there just be a giant RDF file or a big set of them? Will they
just be front-ended by a web server? Will a database be needed?

- Can I build a proprietary closed source application that
incorporates LOD? How would I combine free and fee-based data? I know
how to do it with an API. How would I do it with linked data?

Juan Sequeda, Ph.D Student
Dept. of Computer Sciences
The University of Texas at Austin
www.juansequeda.com
www.semanticwebaustin.org


On Fri, Dec 5, 2008 at 10:32 PM, Juan Sequeda <juanfederico@gmail.com>wrote:

> Giovanni
>
> Great answers! I really hope other people will start commenting on this
> questions, giving answers, or making more answers.
>
> You are right, maybe education is not the right word. However I do think we
> need to do outreach. With respect to the brainstorming that you suggest; I
> think this is what we are doing now. I truly believe that this community
> should brainstorm more about how we should do the outreach :)
>
> I propose to create a FAQ about Linked Data (hopefully on the official
> linked data web site). But to do so, we need the frequently asked questions!
> Hopefully we can start putting this together.
>
> Juan Sequeda, Ph.D Student
> Dept. of Computer Sciences
> The University of Texas at Austin
> www.juansequeda.com
> www.semanticwebaustin.org
>
>
> On Fri, Dec 5, 2008 at 6:01 PM, Giovanni Tummarello <
> giovanni.tummarello@deri.org> wrote:
>
>> i agree on all your comments and believe me by talking to actual web
>> 2.0 people you're way ahead.
>> i'll try to answer some of your questions
>>
>> > I then asked if they new the value of Linked Data. The answer I got was
>> > "well, i would think that my site would be easier to find right? i mean,
>> i
>> > would link stuff on my site better"
>>
>> lets see the key point here:
>>
>> * There is a site
>> * There are human visitors as HUMANS Bring money/business not machines
>> * There is a perception that metadata can help to find things better.
>>
>> >
>> > Question 2: Would my site be easier to find then using Linked Data?
>>
>> Answer: no, matter of fact you open your data to being used without
>> getting any visitors.
>>
>> >
>> > Question 3: So are microformats in my pages doing Linked Data?
>>
>> they are not doing "linked data" but in practice the do answer the
>> questions above or practically go well in that direction. see next one
>>
>> >
>> > Question 4: By what method are these things linked?
>> >
>>
>>
>> 2 pages have the same vcards = you can link them. They have a me link
>> = they are linked they have 2 events on the same date, same city =
>> they should be grouped they might be interesting to show together to
>> the user.
>>
>> They are in practice linked by simple, practical use cases which
>> involve finding/related pages (real sites which want to get traffic)
>> for users (real people who want to get pages)
>>
>>
>> > After explaining somebody what linked data was, and giving them the
>> existing
>> > links about it, question 5 came up:
>> > Question 5: "I see value in the data and the data being linked together
>> but
>> > i don't see practically how i would use it"
>>
>> big technical barrier in using it with the Lod model.
>>
>> on the other hand querying Freebase is infinitely simpler solving :
>>
>> * the access problem . a single language accesses all the datasets
>> they have integrated, no hopping around, very fast
>> * the data omogeneity and quality problem, they care about the dataset
>> and import only clean stuff
>> * identifiers omogeneity, big efforts are made to smush things together
>> * Ontology issues: both a clear taxonomy is defined AND all the
>> sources that are integrated are harmonized to it.
>> * the multiple points of failure problem
>>
>> So since i believe querying large datasets of structured, matched data
>> is in fact very useful once one gets a slightely bit creative i think
>> they'll have success. Could i buy some of their shares i would do it
>> :).
>>
>> I dont think its a coincidence that some of the smartest people who
>> worked on semantic web now work for them.  (but of course there is
>> much more than a good idea for a successful business so they might go
>> bust anyway obviously)
>>
>> >
>> > A final quote "people like me don't a) know about this and b) don't
>> > understand how to use it once they do? I would say some additional
>> education
>> > is necessary to make this understood... i would also say that in a
>> broader
>> > sense the semantic web message has gotten lost under a mass of acryonyms
>> and
>> > theory"
>>
>> for a more articulate attept at an explanation of what happened i
>> agree a lot with this post
>>  http://inamidst.com/whits/2008/technobunkum by Sean Palmer
>>
>> I dont think "more education" is needed Juan, one really should teach
>> something if .. the answer is known else its called brainstorming or
>> handwaving (according to weather you're in  good faith or not)
>>
>> note that this is all but a bashing on the power of handling loosely
>> structured data and RDF. I think on the other hand RDFa will triumph
>> and so people will be probably making their own little vucabolaries..
>> but starting from the web 2.0 approach and practical "how do i bring
>> visitors, how to do simple site to site integration" use cases.
>>
>> Giovanni
>>
>
>