Test Information Space

Journal of Tech, Testing and Trends

Archive for July, 2007

Semantic Filing System Example

Posted by cadsmith on July 26, 2007

This entry has some observations on ways to approach the test infrastructure for those who have to deal with that as well as the verification efforts. It is basically a transponder ping from a moving target since there are a lot more considerations to make. For example, a number of industries are grouped together where each may have particular constraints for development paths.

Empirical logic uses knowledge or expectation in each step of activity. Many details may yet be unknown, so the approach to filling in the necessary gaps involves a rational process that can handle unassigned variables. A general series of tests may be expected, but results at any stage may divert the effort. This results in the need for logistical changes that ripple across the tester, subject under test & any dependent resources. For example, it may be expected that a particular machine will eventually need repair because most of the others of its type did at some point, but knowing precisely when this is necessary & an efficient means of assigning this value requires some experimentation. This may use up the subject to some degree & have its own costs associated with it. In cases where a finite number of testers are dealing with an ever-increasing number of subjects, then a way to organize the chaos becomes useful. Testers may change over time or have subsets of tests to complete so they are also a variable within the test population. The entire system adapts to a set of values within the context of the workplace culture so it automatically fulfills its function. An analogy might be the evolution of eyes which can easily identify threats or items that change unexpectedly so that the right information is understood within the process of discovery itself. Other devices such as motors with built-in sensors are examples of such a function. On the large scale, the semantic web might be an effort in this direction. A competitive element comes in from advantages which certain testers may have & be able to formalize in order to contribute to the system to achieve this level of quality across the board. Of course, design information also drops out of this at certain points, as well as ideas for improvements in production & management. This allows a form of synthetic discovery in service of quality assurance & organizational learning. Ongoing developments to any of these utilities, social software, & device interoperability may have applications. Devices have semantics, i.e. information about their internals for logic & user-specific. Translators can exist between native device forms, point-to-point, & external standards, e.g. XML or RDF, possibly also handling language translation of content. Proxies can represent user interests & be an adjunct to choice, or assume responsibilities for some decision-making if enabled, e.g. as a senator does for a constituency.

Adding some depth to the example, test results may be documented, &/or entered into a database. There may be additional media such as plans, configuration files, input data, & scripts. Metadata might include type of test, description of equipment & features under test, result summaries, metrics, & statistics. Healthcare might have a patient being monitored. A status site might show schedule & progress. There might be a tester network site that lists current activities & resources, personal details such as skills, specialties, vendor contacts, publications, blogs, test output traces related to them. Resources themselves might be constrained by availability schedules, maintenance, upgrade level, and so on, e.g. for facilities, benches, & test sets. There might be an equipment-related site that shows details such as manufacturing versions, h/w, s/w, f/w, links to component detailed specifications, test history, & repair history. Networked information includes stations, sites & servers which support these types of data. The effort & analysis may have some automation. It may be possible to access, search, navigate & edit data given authorization. Uniform ways of dealing with the data might include descriptive tags & annotations attached to any type of storage media entry which can be treated as text by generally available utilities. All of these can be distributed geographically & in different language domains. Visualization tools exist for design, detection, diagnostics, & decision-making. In the physical cases, these know about properties of materials & electronics, in physiological cases, properties of tissues & fluids. While much of this method is manual or automatic, some relationships allow it to progress to becoming more semantic, e.g. through the identification of surfaces. Test surfaces might include a set of tags & entities associated with tag, or a set of logical relationships such as frames, slots, RDF triples, etc. For example, a high-level surface might indicate roles that each person may fill & may have a graph for each indicating their level of trustworthiness for particular functions & resources.

Another simple surface, the lunch menu which may be accessed from the cell phone or web for remote ordering & possibly delivery:
o Trusted Menu, enhancement of contact list:
§ new Order(item, location) -> phone#, address, cost, directions,
Parameters may be functional feature: tastiest, lowest-cost, randomize subject &/or object, closest, …
§ Burger, McDonalds, …
§ Burger, Wendy’s, …
§ Chicken special, Panda House, …
§ Chicken pieces, KFC, …
§ Pizza, Family House, …
§ Calzone, Family House, …
§ Egg sandwich, Dunkin Donuts, …
§ Steak sub, D’Angelos, …
The cell can automatically exchange this information with a laptop or server. It can order in multiple languages if the other person is not a native speaker or one is visiting a foreign region. All of the expected social software functions such as suggestions for what else one might like, indicators of new items or locations, frequent customer promotions, wikis, social network sites, informational media, & productivity tools may be available from any given device. Where these devices are positioned within a 4D semantic capability map may be illustrated as follows:
a) Manual, Automatic, Semantic,
b) Content, Metadata, Tags, Surfaces, Proxies,
c) Search, Navigation, Discovery, Production, Management
d) Public, Encrypted, Airgapped

Specialized artifacts that fill in the above blanks become new products which allow economic growth & potentially improve the likelihood of success for that test group & its culture. Synthesized input data can be used with these to create simulation environments that may be of interest to researchers. Initially there is a metasemantic process that sorts these out & evolves into accepted practices & standards. Ideally these fulfill all of the requirements of that era, e.g. friendly to user, society, environment, & world. The folks in spacecraft may have additional perspectives.

Posted in Uncategorized | Tagged: , , | 1 Comment »