Dear Thomas, An anecdote that supports at least your view of IT HR was a time when I was hiring data modellers/ontologists. The HR guy suggested I set some metrics to judge the volume/quality of work that was being performed, helpfully suggesting entity types per day as something suitable. I wrote back saying that such a metric was problematic because a larger data model might easily be worse than a smaller one for the same scope, and that one with fewer entity types might be better and take longer to produce, but if he could think of metrics that took this into account I would be very interested to hear about them. I didn’t hear from him, but I did get the data modellers I wanted. Regards Matthew West http://www.matthew-west.org.uk +44 750 338 5279 From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Thomas Johnston
Sent: 17 March 2015 15:45
To: [ontolog-forum]
Subject: Re: [ontolog-forum] SME (subject matter experts) and Ontology developement - principle? - Ethics? ".... I would emphasize that really good SMEs who can articulate the implicit assumptions in their expertise are rare. The few who have
those abilities are in high demand and can command high fees for their
work. Since your employment at Shell gave you access to experts in the industry, you had a luxury that is rarely available in most projects.."
I have worked for, consulted for, or contracted with twenty-four enterprises over the course of a forty-five year career in commercial IT, a third of them Fortune 500 companies (and a third of those Fortune 100). I can't speak for how much value companies put on the (extremely rare) articulate SMEs you refer to, since they are on the business side of the house. But I can comment on how much value companies put on BAs who can extract ontological commitments from relatively inarticulate SMEs (which are most of them), who can clarify the SME's initially fuzzy requirements statements, and who can design databases that are ontologically well-founded. That value is -- nothing. Every company I have worked for has regarded BA work and data modeling work as commodity tasks, as empty cubicles (or contractor "pits") which can be filled by anyone who can impress an interviewer (who usually doesn't know how to interview). The determining factor is always money, because the hire decisions are made by low- to mid-level management for whom ten dollars an hour less on a one-year contract is quite a big deal. This is not sour grapes. When my rates were forced down by the fact that hiring managers had no understanding of the added value of my skills as BA and data modeler, or when I lost out to novices who could just about get the definitions of the first three normal forms right (about all the expertise a hiring manager could recognize), well, there were some sour grapes back then. But I just want to dispel the myths that management, in any commercial enterprise, can differentiate experts from the rest. They can't. There are occasionally non-management personnel in commercial IT who command remuneration that busts the upper limits that HR sets for the job title they have. Some of them are pretty good, but none of them are that good. Most of them have a personal style that a senior manager -- VP or even C-level executive -- has been impressed by. Most of them are distinguished from the rest of us by nothing more than that. I don't think that much can be done about this. Hiring managers in IT departments are where they are because of the Peter Principle, and so the notion that they would or could ever recognize the value of a BA or data modeler who had a real facility with uncovering implicit ontologies, is nothing but fantasy. Again, I have attempted not to exaggerate. I have found things to be really this bad. I would be interested in hearing from anyone else with real world commercial IT experience, especially anyone whose experience is different than mine. Dear Matthew,
JFS
> For the future, I believe that we can and should develop*automated* or
> at least *semi-automated* tools that can help extract the underlying
> assumptions from anyone -- even a SME.
MW
> I admire the ambition. Generally, you can only automate things where
> you have reliable and repeatable procedures. We don't. At best we have
> methodologies that need training in, and experience of execution before a
> new member of the fraternity can be welcomed. Indeed at present if I had to
> choose one of the SME or KE to develop an ontology for a domain, I'd go for
> the KE. It will generally be easier for him to become an expert in some
> domain, than for the SME to become a KE.
I agree. But there are huge numbers of tasks that could benefit from
ontologies, but only large corporations or governments can afford to
train KEs to that level of expertise. Even large businesses don't
use ontologies as well or as often as they could because they require
a very clear "business case" before they would consider that expense.
Hashtags and folksonomies are a "poor man's" replacement for ontologies.
The next step up are things like Schema.org, which are slightly more
than systematic terminologies. There are millions of people on that
continuum from hashtags to folksonomies to Schema.org to the highly
structured ontologies used at Shell.
And there are many people (VivoMind included) that are designing novel
technologies for addressing the continuum. For examples, look at the
methods for automatically extracting ontologies from documents:
http://www.jfsowa.com/talks/goal7.pdf
This is still a research area, but work is proceeding at many
institutions. I expect the new tools to revolutionize the methods
of software design and development. The examples in goal7.pdf are
promising steps. And such tools can also be used to enable today's
KEs to address new areas of expertise more rapidly and thoroughly. |
_________________________________________________________________
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)
|