|
On 9/20/13 7:06 PM, David Eddy wrote:
Kingsley -
But that's not the problem legacy systems need solved.
It is an major part of the problem. That said, legacy systems need
bridges to these contemporary tools.
Legacy systems do not contain public information.
Hmm.
Legacy systems may or may not be public. Likewise, as I've stated
repeatedly, nothing about these tools implies public use. You can
install DBpedia and other datasets in-house. You can even perform
local cross references with entities denoted using HTTP URIs that
depend on local area network scoped DNS servers.
One big here is is that "Web" has come to always mean "World Wide
Web" when in reality the subject matter theme at core of my
commentary and examples is one that all about making documents and
data resources web-like i.e., the content of documents (files) can
be web-like with specific emphasis on entity description
representation.
They contain corporate proprietary information. The
systems themselves are proprietary & the data in the
systems is proprietary.
I'm interested in the systems (e.g. the actual software),
NOT the data the systems produce & handle.
I have a small collection of 826 applications... how do I
find what's related to what?
You decouple the applications from the data. Then you map the data
to an entity relationship model graph (e.g., what RDF based Linked
Data delivers when you leverage HTTP URIs and RDF model theory re.,
baking machine- and human-compreshensible semantics into entity and
relation definitions) .
How do I do provenance research in these days of elevated
regulation & regulatory scrutiny?
For instance, you can make statements (i.e., express entity
relationships) about said research using
subject->predicate->object triple patterns using one of the
many notations that exist today.
This collection—50,000 programs, 38,000 control cards,
62,000 DB2 columns, 31,000 DB2 tables, 59 other kinds of
artifacts for a total of 1.7M artifacts—is behind the
corporate firewall & my guess is very little of it has
been webified.
"Web" != "World Wide Web" all the time. It can mean: making content
web-like. The World Wide Web didn't invent hypermedia, it simply
meshed it with the TCP/IP and DNS.
I don't know specifically but I'd assume at least some of
the data is accessible via internal & external web
mechanism, but the SYSTEMS have not be webified.
The systems are accessible in some form. The challenge is what tools
aid decoupling of application data from the application's
orchestration code.
A GUI browser is a terrible access mechanism since none of
the support staff has three arms.
Yes, in some cases. The command-line will be with use forever. In
many cases (these days) I am a zillion times more productive going
back to tools I used pre GUI explosion. BTW -- the World Wide Web is
also showing the power of intentionality via browser address bars,
utilities like curl, and URIs.
Having to remove hands from the keyboard to mouse around
immediately cuts productivity by 33%.
Yes, in a number of cases.
My guess would be that the suggestion of exposing this
collection—particularly the relationships between artifacts—to
"the web" would likely get me tossed off the roof.
Only if "Web" means "World Wide Web" :-)
I'm pretty sure I wouldn't be given the time to explain
that the objective is to use web mechanisms, BEHIND the
company firewall...
Remember... the value that I need here in tending &
enhancing legacy systems is to quickly discover the
relationships between artifacts...
First step is always decoupling the data from application code. Then
map the data (via views) to an entity relationship model oriented
graph comprised of domain entity types, relation types, and actual
assertions that describe how entities are associated (related).
Kingsley
_________________________________________________________________
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
--
Regards,
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
|
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)
|