search

tdwg / bdq

Biodiversity Data Quality (BDQ) Interest Group
https://github.com/tdwg/bdq
43 stars 7 forks source link

TG2 - Link to Specification Source Code #165

Open ianengelbrecht opened 6 years ago

ianengelbrecht commented 6 years ago

Hi all, thanks for the opportunity to participate. In line with @mjcollin's initiative to develop implementation libraries (#150), would it be of any value to include pseudocode in the specification table for each test rather than a link to a specific implementation? As I understand it we're hoping to develop in a number of languages so it might be good to have a common pseudocode reference to use.

mjcollin commented 6 years ago

I think that was discussed at one point (I've only recently taken over our participation in this effort) and the feeling was that it'd be more direct to write a primary implementation in one language with any additional languages using that as a reference. I'd let @ArthurChapman or @chicoreus weigh in though.

chicoreus commented 6 years ago

@ianengelbrecht, yes it would be of great value to add a pseudocode specification to each test. The TG1 framework includes an element for specification that could contain this, we haven't yet actually framed such a specification for any of the tests, and pseudocode would be a very good way to do this.

chicoreus commented 6 years ago

To be more explicit: "Specification" has a formal meaning in the TG1 framework. A validation method, measurement method, and enhancement method (and problem method) all compose a specification with a (dimension,criterion,enhancement) in context. Where what we have specified so far in the TG2 test descriptions is at the level of dimension/criterion/enhancement in context. Adding a specification to an X in context moves from the description of general concepts to the level of data quality solutions. See the TG1 Framework Cheat Sheet in the wiki.

ArthurChapman commented 6 years ago

@chicoreus @Tasilee I am wondering if using pseudocode (with links to actual implementations) would be suitable for a TDWG Best Current Practices Document to document the tests? Just a thought

Tasilee commented 6 years ago

As previously raised, I think it may be expedient to use @chicoreus code as the base due it being largely done. In theory, pseudocode would be nice and this was the original idea.