comment on "radiation ENVO:01001023"

[Bayesianworld]

In a discussion between the people working on the Radiation Biology Ontology and the (emerging) Radiation Therapy Ontology, we were interested in your class "radiation" which is located in the process taxonomy.

Our concern is that typically in radiation-oriented biology and medicine, the term "radiation" refers to continuants. We prefer the use of "irradiation" or perhaps "radiation process" or something that differentiates it from the more physics-oriented concept of "radiation".

The problem arises since this term is already defined within OBO community, it makes it difficult to map out an ontology that is unambiguous in the radiation science communities.

Thank you for your consideration of this,

Mark Phillips
[email protected]

[pbuttigieg]

Thanks @Bayesianworld - could you explain or point to sources describing how radiation is a continuant? At time instant t, any waves/particles participating in a radiation process are radiation ?

For reference: The RBO:radiation class http://purl.obolibrary.org/obo/RBO_00015000

[cmungall]

I think like a lot of concepts (e.g. energy), convincing philosophical arguments can be made for material entities, qualities, disposition, processes.

In fact I see that RBO uses "energy" as the genus in the textual definition. There is a whole paper about the upper level classification of energy here (the authors conclude it is a quality, whereas it's a ME in RBO).

But is there an actual use case where this makes a difference? In what way is the current state of affairs ambiguous?

Re-rooting something at this stage could cause a lot of disruption. Additionally, I would like ENVO to stay consistent with OEO, which models energy as a process.

[pbuttigieg]

Relevant classes

ENVO:energy =def. "A disposition which is realized during the execution of work, the emission of heat, or the possession of mass."

PATO:energy =def. "A physical quality inhering in a bearer by virtue of the bearer's capacity to do work."

I can't find OEO.

Aristotlean concepts aren't energy friendly, so yes, many ways (including potentiality) could work and be equally flawed.

I think the disposition that is realised by a process is a good model. In that model, radiation is a process that can realise various energy forms that radiate (sound, electromagnetism)

ENVO:radiation =def. "A process during which energy is emitted or transmitted in the form of waves or particles through space or a material medium."

with subclasses for acoustic, particulate, electromagnetic and other radiation processes.

[pbuttigieg]

The ENVO definition of energy needs some sharpening. The performance of work is one way to detect energy, but not a form of its realisation in the disposition sense.

[cmungall]

I can't find OEO.

See #1053

[pbuttigieg]

OEO viewer: https://openenergyplatform.org/viewer/oeo/

Additionally, I would like ENVO to stay consistent with OEO, which models energy as a process.

It's a quality. I prefer disposition, as realisation is needed by some process.

[Edit: the OEO definition relies on the notions of "capacity" and "manifests", which are dispositional, but also more about the detection and mathematical abstraction of energy rather than its metaphysics. It also ignores that (likely or effectively) massless photons and possibly gluons in the standard model are not material in the usual Aristotlean sense, but have energy by virtue of momentum: "Energy is a quality of material entities which manifests as a capacity to perform work (such as causing motion or the interaction of molecules)"]

[Bayesianworld]

From a physics point of view, radiation is a particular form of energy, which is definitely a continuant (mass and energy are, at some level, the same thing). From our point of view, radiation is the form of energy inherent in photons. So, if irradiation is the process of an object (or space) being subject to photons impinging on it, then radiation is the name given to the collection of photons, and photons are a form of energy.
This is getting a bit convoluted, but in my defense :) it is how we think of these things in our field. We are sensitive to the issues of ontological stability, so I will start a discussion with RBO and the radiation therapy group to see if we can make peace with things.

[pbuttigieg]

Wikipedia

Irradiation is the process by which an object is exposed to radiation.

In physics, radiation is the emission or transmission of energy in the form of waves or particles through space or a material medium.

Both processes. The former about the exposure, the latter about the movement / transition

Merriam-Webster concurs, but offers your sense of radiation ("a radiation") in 2 and 3

irradiation
noun
1 : exposure to radiation (such as X-rays or alpha particles)
2: the application of radiation (such as X-rays or gamma rays) for therapeutic purposes or for sterilization (as of food)

radiation
1a: the action or process of radiating
1b: the process of emitting radiant energy in the form of waves or particles
1c(1): the combined processes of emission, transmission, and absorption of radiant energy
1c(2): the transfer of heat by radiation
compare conduction, convection
2a: something that is radiated
2b: energy radiated in the form of waves or particles
3: radial arrangement

ENVO would express 2 or 3 as a portion or collection of things (smearing the matter/energy and wave/particle issue, as Aristotlean metaphysics can't deal with it) being radiated (i.e. participating in a process called radiation).

We can add a clarifying comment to our class explaining this. The primary label can be changed to "radiation process" with a related synonym "radiation".

From a physics point of view, radiation is a particular form of energy, which is definitely a continuant (mass and energy are, at some level, the same thing).

ENVO doesn't take any other point of view. But I would never commit to a definitive categorisation of something like energy in an Aristotlean upper level and OWLish logic. Recall the E=mc2 is just the simplification for rest masses, excluding the momentum (and thus temporally extended, processual) term. Any continuant can be treated as a process in that frame, inertial effects in pion clouds, but we have to make a mesoscopic choice.

[Bayesianworld]

Discussion with RBO today highlighted some of these points that pbuttigieg posted. Several thoughts along these lines:

How bad is it to have the same term for two different classes, i.e. radiation as a process and radiation as a material entity?
Similarly, "energy" as disposition (ENVO) and "energy" as a material energy. The IRI resolves the definitions of these different labels.

[cmungall]

How bad is it to have the same term for two different classes, i.e. radiation as a process and radiation as a material entity?

If it's in two ontologies that are not intended to be combined in any way? no problem. If I need to import both your ontologies? Recipe for major confusion and headaches.

But even if you give them different labels (and have some preferred-label-for-community-X scheme, which is common), bifurcating the same concept into two classes is an anti-pattern that has led to over-ontologization, confusion, ragged asymmetric lattices etc in multiple other ontologies. I'd urge everyone to keep it simple and not worry too much about the upper-ontology class

[Bayesianworld]

The issue isn't that the same concept is being bifurcated into two classes; it is that there are two concepts with the same name. This language issue is outside of our control. How do we deal with it? In our case, sometimes we are extending names to make things clearer, but that gets a bit unwieldy at times.

[pbuttigieg]

How bad is it to have the same term for two different classes, i.e. radiation as a process and radiation as a material entity?

it's valid, but can be very irritating when engineering / editing and software like Protégé decide that my ENVO glacier snout has an UBERON collection of hair on snout as a part, just because the expression editor got overexcited with its autocompletion. Don't get me started about hair ice.

It's better to have unique and verbose lables to avoid this, using synonyms to show variants.

As I said above, we can change our primary label to include the "process" suffix. This will help us differentiate between radiations in the sense of spatial arrangements. However, when that time comes, we may have the same label conflict.

Similarly, "energy" as disposition (ENVO) and "energy" as a material energy. The IRI resolves the definitions of these different labels.

I'm assuming we're still talking about radiation and not energy as a class.

As above, it's technically valid, but will inevitably cause errors, axiomatisation issues, slow down workflows etc unless the terms are kept out of ontologies that duplicate their labels.

I still find RBO's radiation branch inconsistent (representing lingo rather than metaphysics - photons have no mass, but are material entities? no energy superclass, that would force the statement that energy is a material entity that - as a continuant - preserves its identity over time despite being converted to different forms?), but that's your domain.

[pbuttigieg]

But even if you give them different labels (and have some preferred-label-for-community-X scheme, which is common), bifurcating the same concept into two classes is an anti-pattern that has led to over-ontologization, confusion, ragged asymmetric lattices etc in multiple other ontologies. I'd urge everyone to keep it simple and not worry too much about the upper-ontology class

Well, it doesn't look like there'll be resolution here (not uncommon across OBO), so a label fix is at least something.

[PaulNSchofield]

It would be really helpful if you could change the envo label "radiation" by adding the suffix "process" which would make it unambiguous. The issue of the definition of Energy, and who owns that class, is a bit more complicated. We think that this may be better described in an upper level ontology because of its importance and complexity.

[cmungall]

Can we reset this thread? Can those in favor of bifurcating the concept outline the complete proposal: - propose the definitions for the two bifurcated root energy terms - document the use cases for having two different branches, make it crystal clear when curators should use terms from one branch rather than the other - describe how the two branches will be populated and synchronized (e.g. will there be both "ionizing radiation continuant" and "ionizing radiation process"? if so, what axioms link them? what is the direction of existential dependence? how will ragged lattices be avoided?)

…

On Thu, Jan 2, 2025 at 10:09 AM PaulNSchofield ***@***.***> wrote: It would be really helpful if you could change the envo label "radiation" by adding the suffix "process" which would make it unambiguous. The issue of the definition of Energy, and who owns that class, is a bit more complicated. We think that this may be better described in an upper level ontology because of its importance and complexity. — Reply to this email directly, view it on GitHub <#1490 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAMMOM7IXHBQCTCXNTJLMD2IV6GDAVCNFSM6AAAAABTXI4XC2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNRYGE3TANBQGE> . You are receiving this because you commented.Message ID: ***@***.***>