I think you are on the right track, now take it to the end. All of this was settled 30 years ago --- in most disciplines. If "perfection" is not worth it from the point of view of the BUYER, then the acceptance criteria states what likelihood of latent error is acceptable and that becomes the specification for Quality. The therapeutic effect on both buyer and producer is well documented, in healthcare, and other venues as well as in manufacturing. I recommend that Quality of the Zero Defects kind be given serious consideration regarding ontologies even if some software people find it daunting. Jack On Dec 16, 2010, at 10:38 AM, Ali Hashemi wrote: Wouldn't it rather be the case that for certain applications, a binary interpretation of Quality is a necessary requirement, whereas for many others, that tail end of getting the Quality to 100% is not worth the time, resource or monetary effort to achieve? So instead of achieving Quality = T or F, we have a gradation, where one is in a process of continually improving the quality of the product with respect to some evaluation?
To make it more concrete - if I'm designing a social networking tool to connect people, expertise and problems, the amount of resources required to make it completely bug / error free might alter my implementation of quality. I can probably tolerate some false positives between problems and people. Moreover, given quite real budgetary and timeline concerns, I will almost certainly satisfice to achieve the constraints of the project - until which time more resources are allocated to improve the system. It is less clear how a binary understanding of quality would help here (except perhaps with reference to a baseline functionality).
Best, Ali On Thu, Dec 16, 2010 at 12:22 PM, Jack Ring <jring7@xxxxxxxxx> wrote:
On Dec 16, 2010, at 12:43 AM, Matthew West wrote:
Dear Jack,
Well actually I had decided that your twisting of my counter-example was so transparently desperate that it did not even warrant a reply, and Chris was doing an entirely adequate demolition job. But if you insist...
Quality as binary only works when there is a clear binary specification that you can say something meets or does not meet. Software is rarely that simple. If we assume a piece of software has no bugs that prevent it from functioning at all, it is usually the case that it is still less than perfect in the way that it performs and supports the business process that the business wishes to use it for. This rarely prevents the software from being used, but typically increases the effort in terms of work-arounds that the business has to employ, i.e. it increases the cost of ownership. Given several software packages that nominally support the same business function, they will vary in the degree to which these work arounds are necessary, and thus vary in the cost of their use.
Business people will consider that these systems vary in quality, and if you tried to tell them that quality was binary and because they all met some nominal specification they had the same quality they would think you were a quality geek and off your rocker.
On Dec 16, 2010, at 2010 12:43:41 AM MST, Matthew West wrote:
MW: What is called for is being able to distinguish a binary situation from
a non-binary situation. The original quality work of the distinguished
gentlemen you mention was around the manufacture of mechanical parts for
which a specification could be created which they either met or did not. It
was simple. If it did not meet the spec you did not think of using it, it
was scrap. Not everything is as simple, and certainly not software.
Matthew, Thank you for this. It is an eloquent description of why software, after 40 years of Software Engineering, is still the lousiest artifact ever devised by mankind.
A simple lack of belief in and dedication to Zero Defects(tm). If nurses dropped infants at the rate software people commit errors society would put them all in jail. Fortunately, a growing number of software engineering efforts now heed the quality/six sigma gurus (instead of the sick stigma view of quality).
For whatever reason you find it necessary to disparage the messenger. So be it. I will simply make sure that those who might consider pursuing the development of formal ontologies are aware of your view of quality.
Caveat Emptor, indeed. Onward, Jack Ring ps. regarding your use of "perfect" is that a binary?
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
-- www.reseed.cawww.pinkarmy.org(•`'·.¸(`'·.¸(•)¸.·'´)¸.·'´•) .,.,
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
|
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (01)
|