NTR: plastics

[GoogleCodeExporter]

Suggested definition: 
any of numerous organic synthetic or processed materials that are mostly 
thermoplastic or thermosetting polymers of high molecular weight and that can 
be made into objects, films, or filaments

Definition source(s):
merriam-webster

Suggested parent class:
anthropogenic environmental material(ENVO:0010001)
Anthropogenic material in or on which organisms may live.

Suggested synonyms:
thermoplastic


children/narrower concepts suggested in FAO's AGROVOC: 
http://aims.fao.org/skosmos/agrovoc/en/page/c_5998

These could be expanded to include other plastic types
(list expanded from 
http://eurovoc.europa.eu/drupal/?q=request&uri=http://eurovoc.europa.eu/1837):
polyester
polyethylene
polypropylene
polyurethane
polyurethane foam
polyvinyl chloride
polyethylene terephthalate

Original issue reported on code.google.com by duhai...@umich.edu on 18 Mar 2015 at 5:30

Attachments:

• [Screenshot 2015-03-18 12.58.10.png](https://storage.googleapis.com/google-code-attachments/envo/issue-134/comment-0/Screenshot 2015-03-18 12.58.10.png)

[GoogleCodeExporter]

I'd be happy to discuss the best way to fit in the above term, I already have 
some suggested amendments. For instance, "anthropogenic litter" could be added 
as a child of "anthropogenic environmental material" and "plastic" a child of 
"anthropogenic litter", which would leave room for expansion of the litter 
class to include glass, metal, etc. The proposal for these classes is preceding 
a large effort to sequence the microbiome of aquatic debris. I am currently 
developing recommendations for metadata to collect with plastics samples.

Original comment by duhai...@umich.edu on 18 Mar 2015 at 5:38

• Added labels: ****
• Removed labels: ****

[pbuttigieg]

"plastic" was added to revision 238 of envo-edit.owl. Will be released soon!
+1 for the AGROVOC link!

Which of those sub-classes do you need?
A note on "litter" - this is more a role than a material in itself. There's increased motivation to create environmental roles (e.g. "pollutant", "litter", "nutrient", etc) to keep orthogonal and prevent the hierarchies getting crossed.

@cmungall: Is there an ontology of BFO:role classes? Can we simply add them to ENVO with ENVO IDs? Right now, I see no specific need to differentiate "environmental role" from "role".

[cmungall]

'environmental role' doesn't do a lot of work. Before adding roles though we should plan ahead a bit. Are we sure everything we want to add is strictly a bfo role? Might some things be dispositions, qualities? Might some things be better modeled using processes? E.g. output-of some pollution process. Are some things in other ontologies? E.g. CHEBI has 'nutrient role'. However, it's somewhat hardcoded to humans. How do we relate to this? Some roles should relative - a nutrient to one organism may be toxic to another. One reason to keep 'environmental role' is to give us a bin we can fudge things a bit without making too strong an ontological commitment with a view to refactoring later.

[pbuttigieg]

Good points - at this stage (especially to address user requests) I think
fudge is an appropriate material. I am leaning towards a role commitment,
but the realisable entity story can get complicated once context is
developed. I think we can import quite a lot from CHEBI (e.g.
http://purl.obolibrary.org/obo/CHEBI_78298)

On 31 March 2015 at 00:22, Chris Mungall notifications@github.com wrote:

'environmental role' doesn't do a lot of work. Before adding roles though
we should plan ahead a bit. Are we sure everything we want to add is
strictly a bfo role? Might some things be dispositions, qualities? Might
some things be better modeled using processes? E.g. output-of some
pollution process. Are some things in other ontologies? E.g. CHEBI has
'nutrient role'. However, it's somewhat hardcoded to humans. How do we
relate to this? Some roles should relative - a nutrient to one organism may
be toxic to another. One reason to keep 'environmental role' is to give us
a bin we can fudge things a bit without making too strong an ontological
commitment with a view to refactoring later.

—
Reply to this email directly or view it on GitHub
#134 (comment)
.