The definition of "quality" is pretty simple "meets requirements" the issue is
how to define requirements. ISO 22745-30 does a pretty good job. (01)
Data and Information are two different concepts, their characteristics are
different. Timeliness and relevance are characteristics of information, the
data is what it is. Finally there is no such thing as accuracy only assertions
of accuracy. A lot of this ground is covered in ISO 8000 (02)
Peter
Cell: +1 610 462 5923 (03)
On Jan 22, 2011, at 9:18 AM, "Brian K Lucas" <lucasb@xxxxxxxx> wrote: (04)
> Greetings all,
>
> I like this discussion, and have a few thoughts of my own (as you come to
> know me more, you'll discover that to be the norm ;=)
>
> A) @Jack : I agree that "adequate, accurate and timely" is a worthy goal.
> I'd like to hear more about your definition of "adequate" and "accurate".
>
> B) @Jack : Regarding quality: In my experience, "fitness for purpose" is
> trinary, when taken from a single consumer's point of view: not fit, fit,
> overly fit. In the real world, however, very few offerings are consumed by
> a single entity. Therefore, I believe that, when taken in the context of
> the offering, and including all consumers, fitness for purpose/quality is
> indeed scalar, if the measurement is being taken from the perspective of the
> producer. If you complete a histogram/pie chart of these three values for
> all consumers, you get three counts of valuable metrics, which should not be
> combined: % fit for purpose, % not fit for purpose, and %overly fit for
> purpose. The requirements not met for the "not fit" group represent
> offering "defects"; the requirements met for the "overly fit" group
> represent waste of effort for that consumer group. If requirements have
> been implemented that NOBODY in the "fit" group needs, then that is wasted
> producer effort - unless, of course, they serve a future consumer. My
> conclusion? There is value in measuring "quality" across the existing and
> intended user base, and improving the offering to move more consumers into
> the "fit for purpose" count, without removing anyone who is already there.
> And, because consumer requirements usually change over time, continuous
> improvement of "quality" is desirable. Then, add in that even for a single
> consumer, they may present KANO-like ranking of requirements (must have,
> should have, could have, etc.), and binary gets even a little more fuzzy for
> me.
>
> C) Producer cost is also key here. If the value exchange received by the
> consumer does not support the cost to meet their requirements, then a "not
> fit" offering may still be of value to the consumer, as they may augment it
> with other offerings. An example of this is Microsoft Word. It does NOT
> meet all of my requirements - and yet I find it "fit for purpose" because I
> can work around the "defects".
>
> D) @Nicola : I agree with the case for varying levels of "quality"; however,
> I also think that case studies are notoriously hard to quantify. In my
> experience, it is usually an opinion that one work method over another
> produced "better" results, because most people don't try it both ways and
> actually measure fitness for purpose afterwards.
>
> E) I have been working on an ontology of organizations and human value
> exchange. I have tried traditional ERD-style modeling, object-oriented
> class modeling, and now OWL ontology modeling. Each has strengths as a
> modeling method; each has "defects", or "fitness of purpose" for my work.
> It may be an understanding defect on my part, but one of the primary issues
> that I have been facing is polymorphism of modeled objects. As soon as you
> "declare" something to be of a "type" of some modeling class/entity/etc, you
> constrain it from taking on characteristics of other classes that it may
> also play a role in. An example is a human - most people would model
> "human" as a class; but in my domain, a "human" is also the offering of an
> educational process (with an improved knowledge metric). My solution so far
> has been to only use inferred classes to discover a thing's class by its
> relationships. So, from my perspective, I am finding ontologies "not fit
> for purpose" in their current description language and implementation. I am
> interested in thoughts that the other summit participants may have to help
> me remove this obstacle - even if it is more training for me on building
> ontologies. :=)
>
> Brian K. Lucas
> Sponsor, Worldwide Institute for Organization Ontologics
> Lucasb-at-wio2-dot-org
>
> -----Original Message-----
> From: ontology-summit-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontology-summit-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Tim Wilson
> Sent: Saturday, January 22, 2011 8:11 AM
> To: ontology-summit@xxxxxxxxxxxxxxxx
> Subject: Re: [ontology-summit] Invitation to a brainstorming call for the
> 2011 Ontology Summit
>
> Gentlemen,
>
> If I may jump in here. This discussion makes me think of the 85/15 rule
> where finding and fixing 85% of all software bugs is relatively easy, the
> last 15% is much more difficult in terms of time and effort. There comes a
> point where developers have to say that the ontology is 'good enough'. Jack
> is arguing that this does not constitute 'high quality'
> and therefore the comment on quality being binary. Some person or persons
> must make a decision that the product is good enough (until the next serious
> bug is uncovered). You may think that there are no more blue balls in the
> bin, but yet one is found. Quality instantly goes from "1" to "0" until the
> issue is analyzed and a choice is made to either ignore it or fix it.
>
> Tim Wilson
>
> On 12/15/2010 3:22 AM, Matthew West wrote:
>> Dear Jack,
>>
>>> MW,
>>> Standing on the shoulders of Deming, Crosby, Juran, etc. I would
>>> first ask
>> the
>>> owner a) Is the fifth one guaranteed irrelevant
>> MW: I am assuming it is relevant.
>>
>>> and b) what is your level of
>>> confidence there are not 6 errors?
>>> Jack
>> MW: Indeed, but then by the same token how can you be certain anything
>> is defect free, even if no defects are apparent?
>>
>> MW: I think it is more useful to think of quality as the degree to
>> which requirements are met. Then when you fix some bugs you have
>> improved the quality, though you may not have met all the requirements.
>>
>> Regards
>>
>> Matthew West
>> Information Junction
>> Tel: +44 560 302 3685
>> Mobile: +44 750 3385279
>> matthew.west@xxxxxxxxxxxxxxxxxxxxxxxxx
>> http://www.informationjunction.co.uk/
>> http://www.matthew-west.org.uk/
>>
>> This email originates from Information Junction Ltd. Registered in
>> England and Wales No. 6632177.
>> Registered office: 2 Brookside, Meadow Way, Letchworth Garden City,
>> Hertfordshire, SG6 3JE.
>>
>>
>>
>>
>>> On Dec 14, 2010, at 3:45 PM, Matthew West wrote:
>>>
>>>> Dear Jack,
>>>>
>>>>> Regarding Nicola's quite relevant concern (below) it may be useful
>>>>> to
>> note
>>>>> that
>>>>> a) quality is binary, not a scalar (Crosby, Deming, Juran, etc.)
>> Quality
>>>>> signifies conformance to requirements, Yes or No, therefore 'high
>>>> quality' is
>>>>> meaningless.
>>>> MW: So presumably you would argue that if an ontology has 5 defects,
>>>> and
>> 4
>>>> of them are fixed, there is not improvement in quality as a result....
>>>>
>>>> Regards
>>>>
>>>> Matthew West
>>>> Information Junction
>>>> Tel: +44 560 302 3685
>>>> Mobile: +44 750 3385279
>>>> matthew.west@xxxxxxxxxxxxxxxxxxxxxxxxx
>>>> http://www.informationjunction.co.uk/
>>>> http://www.matthew-west.org.uk/
>>>>
>>>> This email originates from Information Junction Ltd. Registered in
>> England
>>>> and Wales No. 6632177.
>>>> Registered office: 2 Brookside, Meadow Way, Letchworth Garden City,
>>>> Hertfordshire, SG6 3JE.
>>>>
>>>>
>>>>
>>>>
>>>>> b) note carefully that from the usage viewpoint the requirements
>>>>> amount
>> to
>>>>> 'fit for purpose' (Checkland) or 'satisficing' (Simon).
>>>>> c) both proof of correctness and exhaustive test are futile,
>>>>> therefore
>> not
>>>>> included.
>>>>> d) the goal becomes warranty that the ontology of interest is
>>>>> devoid of internal faults and external incompatibilities wherein
>>>>> warranty means
>> zero
>>>>> false positives and false negatives.
>>>>> e) an appropriate theme may be "Making the case for adequate,
>>>>> accurate
>> and
>>>>> timely ontologies" which embraces both the result and the
>>>>> development activity.
>>>>> f) whether any ontology is viable or not depends on both the
>>>>> ontology
>> and
>>>> the
>>>>> intended usage.
>>>>> g) this means that any cadre of ontology developers must include
>> members
>>>> who
>>>>> are dedicated to independent and objective assessment of the
>>>>> viability
>> of
>>>> any
>>>>> ontology or patch thereof or ordered set of patches.
>>>>> h) fortunately, technologies, tools and methods exist (or are
>>>>> imminent)
>>>> for
>>>>> viability assessment of algorithms of all classes and types with
>> respect
>>>> to
>>>>> intended usage. This includes ontologies. Even the spaghetti code
>>>>> in
>> most
>>>> OWL-
>>>>> based examples can be assessed, even simplified, and potentially
>>>>> made
>> more
>>>>> "lean" without inducing 'brittle.'
>>>>> i) this is one reason why I suggested to Steve Ray that one corner
>>>>> of
>> the
>>>>> Summit allow open-mind dialogue regarding new technologies.
>>>>>
>>>>> Onward,
>>>>> Jack Ring
>>>>>
>>>>>
>>>>> On Dec 14, 2010, at 5:00 AM, Nicola Guarino wrote:
>>>>>
>>>>>> Dear colleagues,
>>>>>>
>>>>>> I also agree very much with John and Matthew concerning the
>>>> importance
>>>>> of high quality ontologies, and on their observation that the quest
>>>>> for
>>>> high
>>>>> quality data models in software engineering definitely reflects a
>>>> sensitivity
>>>>> to important ontological aspects much higher than what we find in
>> people
>>>> just
>>>>> focusing on ontology languages.
>>>>>> In the light of this, I suggest to specify a bit more the overall
>>>> theme
>>>>> of our Summit, which in my opinion could be "Making the case for
>>>> ontological
>>>>> analysis" instead of "Making the case for ontology". An alternative
>> could
>>>> be
>>>>> "Making the case for high-quality ontologies".
>>>>>> The reason for this proposal should be self-evident, I believe.
>>>> Deciding
>>>>> how much effort to put in developing a particular ontology is a
>>>>> crucial choice, and it is very important to distinguish the cases
>>>>> where a
>> proper
>>>>> ontological analysis pays off, and is indeed a crucial aspect of
>> success,
>>>> from
>>>>> those where a "lightweight" approach is sufficient.
>>>>>> Just brainstorming...
>>>>>>
>>>>>> Talk to you soon,
>>>>>>
>>>>>> Nicola
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 9 Dec 2010, at 16:03, John F. Sowa wrote:
>>>>>>
>>>>>>> Dear Matthew and Peter,
>>>>>>>
>>>>>>> MW:
>>>>>>>> ... my forthcoming book "Developing High Quality Data Models".
>>>> Substitute
>>>>>>>> ontology for data model and the same argument applies. The
>>>>>>>> benefits
>>>> come
>>>>>>>> from improving and automating decision making through
>> fit-for-purpose
>>>>>>>> information to support those decisions.
>>>>>>> I very strongly agree. Software engineers have been doing
>>>>>>> ontology (avant la lettre, as they say) for a very long time.
>>>>>>> And much of
>> that
>>>>>>> work has been very good -- sometimes much better than what people
>>>>>>> are doing with so-called ontology languages.
>>>>>>>
>>>>>>
>>>>>> _________________________________________________________________
>>>>>> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
>>>>>> Subscribe/Config:
>>>> http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
>>>>>> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
>>>>>> Community Files:
>>>>>> http://ontolog.cim3.net/file/work/OntologySummit2011/
>>>>>> Community Wiki:
>>>> http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
>>>>>> Community Portal: http://ontolog.cim3.net/wiki/
>>>>>
>>>>> _________________________________________________________________
>>>>> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
>>>>> Subscribe/Config:
>>>> http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
>>>>> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
>>>>> Community Files:
>>>>> http://ontolog.cim3.net/file/work/OntologySummit2011/
>>>>> Community Wiki:
>> http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
>>>>> Community Portal: http://ontolog.cim3.net/wiki/
>>>>
>>>> _________________________________________________________________
>>>> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
>>>> Subscribe/Config:
>> http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
>>>> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
>>>> Community Files:
>>>> http://ontolog.cim3.net/file/work/OntologySummit2011/
>>>> Community Wiki:
>> http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
>>>> Community Portal: http://ontolog.cim3.net/wiki/
>>>
>>> _________________________________________________________________
>>> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
>>> Subscribe/Config:
>> http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
>>> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
>>> Community Files:
>>> http://ontolog.cim3.net/file/work/OntologySummit2011/
>>> Community Wiki:
>>> http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
>>> Community Portal: http://ontolog.cim3.net/wiki/
>>
>> _________________________________________________________________
>> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
>> Subscribe/Config:
>> http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
>> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
>> Community Files: http://ontolog.cim3.net/file/work/OntologySummit2011/
>> Community Wiki:
>> http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
>> Community Portal: http://ontolog.cim3.net/wiki/
>>
>
> --
> Timothy C. Wilson
> Graduate Student in Knowledge Management Kent State University Expected
> Completion: August 2011
>
>
> _________________________________________________________________
> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
> Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
> Community Files: http://ontolog.cim3.net/file/work/OntologySummit2011/
> Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
> Community Portal: http://ontolog.cim3.net/wiki/
>
>
>
> _________________________________________________________________
> Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
> Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
> Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
> Community Files: http://ontolog.cim3.net/file/work/OntologySummit2011/
> Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
> Community Portal: http://ontolog.cim3.net/wiki/ (05)
_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2011/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011
Community Portal: http://ontolog.cim3.net/wiki/ (06)
|