On 2/1/13 5:20 PM, Hans Polzer wrote:
> Railing at technology or vendor-driven incompatibilities won't really
> accomplish much. (01)
That isn't my goal. I am highlighting issues. (02)
> Even if all those issues were to be addressed/resolved, we
> will still be left with silos, albeit silos that might be a bit more
> translucent and porous. (03)
No, there will be no silos left, bar those where opportunity costs
haven't reached the point of palpability. (04)
Note, we've seen this movie before. At the advent of the Web there was
resistance to outbound links. Today, publishing a an HTML silo on the
Web is a misnomer. (05)
> Nothing gets done without silos, this forum
> included. (06)
See my comment above. (07)
> You can substitute the words "focus" or "project" or "domain" or
> "enterprise" or "ontology" (or any one of many similar words) for "silo". (08)
In this context, my focus is structured data representation, access,
integration, and dissemination. (09)
> They all imply some limited scope, with some local context and associated
> frames of reference and perspectives on reality (and often alternate
> realities as well). (010)
Until the law of diminishing returns kicks in. (011)
> Busting silos won't work. (012)
It always works. Exhibit #1 the World Wide Web. (013)
> You can only make them more
> aware of other silos and more willing to work/interact with those other
> silos. (014)
I disagree. (015)
> No one wants to expose every bit of information about themselves to
> everyone else. (016)
Linked Open Data is all about Linked Open Standards based Structured
Data. It isn't about discarding privacy, far from it. At this juncture,
the hot bed of Linked Data activity all about privacy, access control
lists and policies etc.. These are the critical components that will
accentuate the nascent Read-Write dimension of the Web [1][2][3]. (017)
> And the degree of information that they actually are willing
> to share depends on the relationship they feel they have with those who will
> have access to that information. (018)
Amen! (019)
That's what I am referring to above re. data access policies. Basically,
policies that leverage social relationship semantics [4]. (020)
> In other words, silos of varying width and
> height, and varying degrees of opacity/permeability - where "silo" is
> associated with the nature of the relationship; who is inside or outside the
> "silo".
>
> I actually prefer the "lava lamp" metaphor to that of the "silo" metaphor.
> All such conceptual entities are more like dynamic blobs in "n-space" that
> overlap with each other (like supply chains and countries vice
> corporations), merge, and separate (like new specialties/domains or
> corporate spin-offs/divestitures). (021)
I call them Data Spaces :-) (022)
>
> The key issue is how to represent the scope/permeability of a "silo" in an
> "open" and "network-accessible" way so that those "cryptic" database keys
> are less "cryptic" and more understandable outside the "silo", at least to
> those who are viewed as meriting access to the silo's info/services based on
> their relationship to the silo. (023)
I would say: leverage the power of those hyperlink based super keys.
They are the conduits to entity relationship graphs that enable the
application of logic, at Web-scale to the following: (024)
1. Spam
2. Verifiable Identity
3. Data Access Policies. (025)
> Not everything will ever be understandable
> or accessible to everyone on the network. Let's stop fighting to make it so
> and instead start grappling with how to represent scope/frame of
> reference/relationship differences among silos in an open,
> network-discoverable, and relationship-driven way. By the way, WebID,
> S/MIME, and PKI are all very fragmentary attempts in this direction. (026)
Not really, see the demo link collection below :-) (027)
Links: (028)
[1]
http://kingsley.idehen.net/DAV/home/kidehen/linked-data-transformation-wikipedia-page-picasso-2.png (029)
-- try looking up the resource at that URL (Address) .
[2] http://bit.ly/U2uKLI -- Open Identity & Web Access Control demo
(showcases the use of a variety of identifiers and authentication &
authorization protocols) .
[3] http://bit.ly/UDlwc6 -- Yet another demo showcasing the above .
[4] http://bit.ly/UuWZSI -- various posts about working with social
relationship semantics and data access policies . (030)
Kingsley
>
> Hans
>
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Kingsley Idehen
> Sent: Friday, February 01, 2013 4:05 PM
> To: ontolog-forum@xxxxxxxxxxxxxxxx
> Subject: Re: [ontolog-forum] Natural Language based SPARQL Generator
>
> On 2/1/13 10:16 AM, Michael Brunnbauer wrote:
>> Hello Kingsley,
>>
>> On Fri, Feb 01, 2013 at 09:43:31AM -0500, Kingsley Idehen wrote:
>>> Google, Bing, and others are silos. I am in the business of
>>> silo-busting via Web architecture and open standards.
>>> I am most interested in a
>>> global distributed database (offering equal billing to extensional
>>> and intensional functionality) where hyperlinks are super keys that
>>> resolve to entity relationship graphs endowed with machine and human
>>> comprehensible entity relationship semantics.
>> A database that can only be explored via cryptic primary keys is of
>> limited use - even if the entities are connected via those keys.
> An today you are using DBMS engines with what kinds of keys? Can you
> reference a record in Oracle from DB2?
>
> When I speak of super keys I am referring to dbms, operating system, and
> application agnostic foreign keys. These keys resolve to descriptors that
> describe of their referents. The descriptors are viewable via your browser.
> You can use said keys to share data with anyone that has access to a
> browser.
>
> Examples:
>
> 1. http://dbpedia.org/resource/Linked_data -- a hyperlink based super key 2.
>
> http://dbpedia.org/resource/describe/?uri=http://dbpedia.org/resource/Linked
> _data
> -- an URL that resolves to a descriptor document endowed with faceted
> navigation capability.
>
> Look at the footer or #2 it exposes the same data in a myriad of formats.
> You can even use the CSV URL to directly import data into Google
> Spreadsheet, Excel, or Open Office.
>
>> This is the reason
>> Google has been so successful.
> You are missing the point. Google being successful doesn't mean others are
> unsuccessful. Google is currently focused on an aspect of the Web i.e., its
> Information Space dimension. I and others in the Linked Data realm are
> interested in the Web's Data Space dimension.
>
>
>> The reason that Google is still very successful is that they are
>> doing such a good job that a buch of geeks with distributed hash
>> tables is not able to beat them. One of the problems those geeks are
>> facing is fighting spam in a distributed environment.
> An Spam is one of the problems that Linked Data addresses with aplomb, once
> the noise settles this will be another pain point that aids Linked Data
> appreciation via the fusion of WebID, S/MIME, and PKI [1].
>
>> Maybe WebID has a future
>> there...
> Yes, as per my comment above.
>
>
> Links:
>
> 1. https://plus.google.com/s/spam%20idehen%20webid%20linked%20data -- Spam,
> WebID, and Linked Data (example of where Google is very useful :-) )
>> Regards,
>>
>> Michael Brunnbauer
>>
> (031)
-- (032)
Regards, (033)
Kingsley Idehen
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen (034)
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (01)
|
