Kingsley, (01)
There are certainly many situations where dependency on specific computer
operating environments and implementation languages cause some of the "tangled
mess". And striving to remove these dependencies from one's architecture and
implementation is certainly a big help. But my experience has been that these
dependencies often get blamed for interoperability problems that have nothing
to do with them, but are rather rooted in fundamental assumptions about what is
being represented and for what purposes it is being captures/represented. The
technical interoperability issues often mask these underlying representational
assumptions and frame of reference differences. (02)
Likewise, there is a tendency to point the finger at "poor design", often with
the implication that if "me" or "my team", or "the method I use" had designed
whatever system is at issue, things would be so much better. Certainly there
are examples of poor design out there. But, again, my experience is that more
often, what is really being complained about is that the system designer was
operating under different requirements assumptions and constraints than the
complainer would have made/chosen - assuming the complainer would have been
allowed to make such choices in the target system context. That's why the term
"legacy system" is more often than not used in a pejorative (and ironic) sense.
Some "legacy", huh? Hindsight is, of course, 20/20. (03)
I've fallen into this mode myself more often than I care to admit, especially
since I had a number of "system rescue" assignments and participated in many
"red teams" on late/over-budget/under-performing system development programs.
However, over time I have gained more respect for the constraints and scope
assumptions that most designers have to operate under. As John Sowa pointed
out not too long ago, much of modern mobile technology would not be possible if
it weren't for those much maligned legacy systems operating behind the scenes -
and bridging business model/world-model gaps/differences more so than
technical implementation differences. (04)
Hans (05)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Kingsley Idehen
Sent: Saturday, July 27, 2013 6:59 PM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum] Amazon vs. IBM: Big Blue meets match in battle for
the cloud (06)
On 7/27/13 5:49 PM, Hans Polzer wrote:
> Until you start including a description of the context(s) within which the
>representation of entities applies, you won't get very far. (07)
Yes. (08)
> Every legacy system has an assumed context and scope for what it does, and
>a specific perspective on that context, and specific frames of reference used
>to describe the relevant "world" from its perspective. (09)
Yes. (010)
> That's why we have a "tangled mess". (011)
Sorta. (012)
A system can separate orchestration oriented code from entities, relationships,
roles, and relations. The tangled mess is because orchestration oriented data
(code) and world view data (what ontologies are ultmately about) keep on
getting mangled. It's happened re: (013)
1. Object Oriented Languages
2. Object-Oriented Databases
3. Object-Relational Databases
4. SQL Relational Databases
5. 4GLs
6. 3GLs
7. etc.. (014)
>
> By the way, all systems are legacy systems - including the one you are
>designing this moment (from the perspective of the designer of the next
>system) - and no one will ever get the resources to redesign and reimplement
>all systems that do everything for everyone's purposes: live with it! (015)
Yes, no disagreement there. (016)
> Nothing operationally useful gets built without constrained scope, time,
>and resources, which limits the n-space of context ranges it can practically
>address.
>
> All attempts to define entities and relations in a context free/neutral
>manner as a solution to interoperability are doomed to have limited validity
>(i.e., to those systems/applications that share their implicit context/scope
>assumptions). (017)
No. (018)
Separate presentation, orchestration, and data models. The time-tested
Model-View-Controller (MVC) [1] and more recent Data, Context, and Interaction
(DCI) [2] patterns pretty much handle that. They key is to keep these patterns
devoid of any platform (operating system, programming language, database
management system) specificity, which is where it's always gone wrong in the
past, hence the mess to which you allude. (019)
> Only when we start being explicit about our context and scope assumptions
>within which we are describing entities and relationships will we be able to
>semi-reliably share meaning between arbitrary systems. So the first order of
>business should be an ontology for describing context and scope assumptions. (020)
You can't have one ontology to rule them all. You can have pieces that serve as
resuable components in different worlds. The World Wide Web demonstrates the
concept today via Web Resources (comprised of Webby
content) that take a variety of representational forms via content negotiation. (021)
Contextual Fluidity is a constant, and a good one at that, since perfection is
ultimately just a pursuit. Thus, the most important thing (in my eyes) is
turning structured data (resources) into reusable puzzle pieces via resolvable
URIs, negotiable representation formats, and logic. (022)
I am hopeful, and that's from someone who's always (over 20+ years) looked at
legacy systems as the starting point of any of these kinds of endeavors :-) (023)
Links: (024)
[1] http://en.wikipedia.org/wiki/Model–view–controller
[2] http://en.wikipedia.org/wiki/Data,_Context,_and_Interaction --- Data,
Context, and Interaction (025)
Kingsley
>
> Hans
>
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
>[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Kingsley Idehen
> Sent: Saturday, July 27, 2013 12:35 PM
> To: ontolog-forum@xxxxxxxxxxxxxxxx
> Subject: Re: [ontolog-forum] Amazon vs. IBM: Big Blue meets match in battle
>for the cloud
>
> On 7/27/13 10:06 AM, deddy wrote:
>> Michael -
>>
>>> your example becomes more and more specific and challenging :-) Yes,
>>> if all you have is the code, you are in trouble and SW technologies
>>> are not a magic bullet for solving it.
>>>
>> Welcome to the world of legacy systems.
> By "legacy systems" you mean a subjective tangled mess rife with contextual
>fluidity?
>
>> I hope against hope that somewhere in the SW stack of tools there just
>> might be something to help with understanding legacy systems.
> Yes, they help since "meaning" is critical to "understanding" anything.
> Thus, you can map out a tangled mess, rife with contextual fluidity, by
>decomposing the aforementioned mess into:
>
> 1. entities
> 2. entity relationships
> 3. entity relationship roles
> 4. entity relations .
>
> You can achieve the above with computer and human oriented languages.
>> It's a very SMALL hope.
> I have big hopes, the challenge lies in getting everyone to look at the task
>like they would a jigsaw puzzle game where every resource is a puzzle-piece,
>as exemplified by the World Wide Web.
>
>> What I do see is SW creating yet another tangled layer of undocumented,
>poorly understood systems.
> Of course not.
>
>>
>> I've been attending monthly MIT SW meetings for 4+ years. Once, by chance,
>I did catch TBL himself saying to the
>> audience that "Semantic Web" was a clever marketing label, but that in
>reality, "linked data" would be a more
>> appropriate description since there really isn't anything special about
>semantics in the SW.
> Hmm..
>
> I think TimBL was trying to unravel the obvious fact that the Web he
> envisioned was a read-write global graph comprised of:
>
> 1. entities -- things
> 2. entity relationships -- statements describing things
> 3. entity roles -- relationship roles e.g., subject, predicate, and object
> 4. entity relations -- sets of relationships scoped to common predicates
> 5. relation semantics -- exploitation of First-order logic as the
> foundation for relation semantics.
>
> In the context of the World Wide Web, HTTP URIs would serve as the
> denotation (naming) mechanism for the items above.
>
> The World Wide Web was always about a global entity relationship graph
> [1] where humans and machines would be able to comprehend entity
> relationship semantics [2]. Basically, the fidelity or entity
> relationship semantics of this global entity relationship graph would
> evolve (continuously) over time via crowd-sourcing.
>
>
>>
>> As you may have noticed, my passion is for a process—tool assisted, but
>requiring human knowledge at the beginnings
>> —to extract & make formal the MENSA_FL --> MEssage Notify Stop Action Flag
>--> "a collection of dunning flags"
>> process.
> We (certainly I) just need to find the right way to articulate that we
> (You and I) are on the same page. At the top of this response, what I
> outlined are steps that fall into your "beginnings" view point i.e.,
> that domain experts and systems analysts MUST be key participants in the
> process. That's totally different from the typical pattern where
> programmers (short on domain expertise and industry experience)
> generally make things up as they experiment and play with the latest and
> greatest programming language, where the real focus is parsing
> capabilities, language idioms, and data representation formats etc..
>
> To conclude, we just need to align our own entity relationship semantics
> as we discuss these matters, en route to common understanding :-)
>
> Links:
>
> [1] http://bit.ly/10Y9FL1 -- Proof that Relationship Semantics & Linked
> Data were part of original World Wide Web design and proposal
> [2] http://bit.ly/16EVFVG -- Illustrating the loose-coupling of
> Identifiers (e.g., URIs), Structured Data, and Logic exemplified by Web
> Architecture
> [3] http://slidesha.re/18CtxGK -- Blogic Presentation by Pat Hayes.
>
>
> Kingsley
>
>
>>
>>
>> Does "dunning" have a German meaning? It means the process to send someone
>a series of (increasingly firm) bills to
>> collect a debt. R.G. Dun was an early (1840s) credit rating business here
>in the States & eventually merged to become
>> Dun & Bradstreet, which survives to this day.
>>
>> ______________________
>> David Eddy
>> Babson Park, MA
>> 781-455-0949
>>
>>
>>> -------Original Message-------
>>> From: Michael Brunnbauer <brunni@xxxxxxxxxxxx>
>>> To: [ontolog-forum] <ontolog-forum@xxxxxxxxxxxxxxxx>
>>> Subject: Re: [ontolog-forum] Amazon vs. IBM: Big Blue meets match in
>battle for the cloud
>>> Sent: 2013-07-27 08:50
>>>
>>>
>>> Hello David,
>>>
>>> your example becomes more and more specific and challenging :-) Yes, if
>all
>>> you have is the code, you are in trouble and SW technologies are not a
>magic
>>> bullet for solving it.
>>>
>>> Regards,
>>>
>>> Michael Brunnbauer
>>>
>>> On Fri, Jul 26, 2013 at 10:44:17AM -0400, David Eddy wrote:
>>> > Michael -
>>> >
>>> > On Jul 26, 2013, at 9:13 AM, Michael Brunnbauer wrote:
>>> >
>>> > > distributed in the heads of two experts for one or the other system.
>>> >
>>> > Let us assume the experts are not readily available...
>>> >
>>> > - I'm too green to formulate a coherent question
>>> >
>>> > - experts do not like to be pestered by clueless newbie questions
>>> >
>>> > - experts are simply too busy
>>> >
>>> >
>>> > The "knowledge" that has trickled down to me is as most technical
>documentation severely stripped of useful
>>> > context & content.
>>> >
>>> >
>>> >
>>> >
>>> > > once you have discovered that M0760 and MENSA-FL are the same
>>> >
>>> > That's the hard part... how are M0760 & MENSA-FL discovered to be the
>same?
>>> >
>>> > Remember, we're looking at a data structure with 1700 data elements &
>analysts/programmers are pawing over
>> this stuff on a regular basis.
>>> >
>>> > In theory there should be documentation... but situations like this
>typically come down to: "the code is the
>> documentation."
>>> >
>>> > - David
>>> >
>>>
>>> >
>>> > _________________________________________________________________
>>> > 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
>>> >
>>>
>>> --
>>> ++ Michael Brunnbauer
>>> ++ netEstate GmbH
>>> ++ Geisenhausener Straße 11a
>>> ++ 81379 München
>>> ++ Tel +49 89 32 19 77 80
>>> ++ Fax +49 89 32 19 77 89
>>> ++ E-Mail brunni@xxxxxxxxxxxx
>>> ++ http://www.netestate.de/
>>> ++
>>> ++ Sitz: München, HRB Nr.142452 (Handelsregister B München)
>>> ++ USt-IdNr. DE221033342
>>> ++ Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer
>>> ++ Prokurist: Dipl. Kfm. (Univ.) Markus Hendel
>>>
>>> _________________________________________________________________
>>> 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
>>>
>>
>> _________________________________________________________________
>> 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
>>
> (026)
-- (027)
Regards, (028)
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 (029)
_________________________________________________________________
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 (030)
|