Recommendation Document V0,5 for review

[matthe...@informationjunction.co.uk]

Dear Colleagues,

Thank you again for your various comments. They have been very helpful. Please find attached a revised version responding to those comments. Please check that my response is adequate and raise any remaining issued, or new ones that have been introduced.

Regards

Matthew West

[Steven Kraines]

Hi Matthew,

I just took a quick look at the paper.
It reads much more clearly, I think, but there
are still some typos and confusing phrasing.
Will you have the paper checked by an editor
before "going to press"?

I appreciated the example of the abnormal load - it
is probably relevant to the ASAM domain as well.
I think I understand what you are trying to convey
with this example, but I can't help wondering if
the kinds of people you are expecting to read this
paper will.

In particular, I understand that it is just the
"activities" (or the occurrents?) that express the
temporal changes (essentially movements) of the
vehicle itself that do not need to be duplicated, right?
For example, there should be an activity that is
not the vehicle corresponding to the statement
"the vehicle is loaded at the end of the day", correct?
In particular, the implication is that someone or
something different from the vehicle is the actor
of that activity. So this activity will be a different
object from the vehicle - is that right?

One other thing about the paper that bothers me a bit -
you write
"Considering the detailed content of these four, although each has some
valuable features, none encompasses all the features desirable for our
purposes with the Foundation Data Model."
But I cannot find anything about what specific features
each of the four alone do not encompass. In fact, the
overall paper (and the survey paper) seem to indicate that
the four ontologies pass all of the tests with flying
colors. Again, I do not know much about the potential
readership of the paper, but I wonder if that would not
appear at least somewhat suspicious.

I apologize for not raising this concern earlier -
I hope that it is not too late to address it (if in
fact it is a concern that does need to be addressed).

Best,

Steven

> --
> You received this message because you are subscribed to the Google
> Groups "UK NDT FDM" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to uk-ndt-fdm+...@googlegroups.com
> <mailto:uk-ndt-fdm+...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/uk-ndt-fdm/00bc01d67b8b%24bc94d770%2435be8650%24%40informationjunction.co.uk
> <https://groups.google.com/d/msgid/uk-ndt-fdm/00bc01d67b8b%24bc94d770%2435be8650%24%40informationjunction.co.uk?utm_medium=email&utm_source=footer>.

[matthe...@informationjunction.co.uk]

Dear Steven,

Hi Matthew,

I just took a quick look at the paper.
It reads much more clearly, I think, but there are still some typos and confusing phrasing.
Will you have the paper checked by an editor before "going to press"?

[MW] I'm certainly expecting it to go to an editor, and I haven't finished reviewing it myself yet.

I appreciated the example of the abnormal load - it is probably relevant to the ASAM domain as well.
I think I understand what you are trying to convey with this example, but I can't help wondering if the kinds of people you are expecting to read this paper will.

[MW] The primary audience for this paper is those who are going to endorse it "on behalf of the nation". If they are appropriate people to endorse it, whilst they do not have to be expert in this area (the idea is to have outside oversight) they should at least be able to grapple with the issues involved. Of the characteristics that we are making a recommendation on, stratification is the most abstract and difficult to get hold of (several reviewers commented on this). The example is added to help our endorsers get over that hump (rather than to have relevance to you). There is at some point a balance then between have we done a good enough job of explaining, and are the people involved appropriate to do the endorsing if they do not get it. I will be taking advice on this as part of the next stage for this document.

In particular, I understand that it is just the "activities" (or the occurrents?) that express the temporal changes (essentially movements) of the vehicle itself that do not need to be duplicated, right?

[MW] Not really. The key is that you are not insisting activities and physical objects are necessarily different things. It is not about abandoning one or the other.

For example, there should be an activity that is not the vehicle corresponding to the statement "the vehicle is loaded at the end of the day", correct?

[MW] That is OK. A state of the vehicle is probably a participant (spatial part) in that activity, but not all of it.

In particular, the implication is that someone or something different from the vehicle is the actor of that activity. So this activity will be a different object from the vehicle - is that right?

[MW] The vehicle will not be the performer, its role will be passive.

One other thing about the paper that bothers me a bit - you write "Considering the detailed content of these four, although each has some valuable features, none encompasses all the features desirable for our purposes with the Foundation Data Model."
But I cannot find anything about what specific features each of the four alone do not encompass. In fact, the overall paper (and the survey paper) seem to indicate that the four ontologies pass all of the tests with flying colors. Again, I do not know much about the potential readership of the paper, but I wonder if that would not appear at least somewhat suspicious.

[MW] The survey paper does not go into the detail of the internals of any of the TLOs, but we have personal knowledge that that is the case (Chris or I were involved in all of them). I'll look at that bit again though and see if I can find a better way to put it.

I apologize for not raising this concern earlier - I hope that it is not too late to address it (if in fact it is a concern that does need to be addressed).

[MW] Not a problem. I was vaguely aware there was an issue here.
Regards
Matthew

Best,

Steven

On 2020/08/26 18:31, matthe...@informationjunction.co.uk wrote:
> Dear Colleagues,
>
> Thank you again for your various comments. They have been very helpful.
> Please find attached a revised version responding to those comments.
> Please check that my response is adequate and raise any remaining
> issued, or new ones that have been introduced.
>
> Regards
>
> Matthew West
>
> --
> You received this message because you are subscribed to the Google
> Groups "UK NDT FDM" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to uk-ndt-fdm+...@googlegroups.com
> <mailto:uk-ndt-fdm+...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/uk-ndt-fdm/00bc01d67b8b%24bc94d770%2
> 435be8650%24%40informationjunction.co.uk
> <https://groups.google.com/d/msgid/uk-ndt-fdm/00bc01d67b8b%24bc94d770%2435be8650%24%40informationjunction.co.uk?utm_medium=email&utm_source=footer>.

--
You received this message because you are subscribed to the Google Groups "UK NDT FDM" group.

To unsubscribe from this group and stop receiving emails from it, send an email to uk-ndt-fdm+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/4c803e12-b998-7494-ba36-841a312cbe60%40kraines.net.

[Steven Kraines]

Hi Matthew,

Thanks for your reply - basically this answers most of my concerns.

Regarding the example of the abnormal load, I guess that the point
you are trying to make is that a decision to stratify the 3D physical
object from the "story of its existence" (I like the comparison between
a person at one point in time and the person's life) ends up forcing
us to separate information that we would prefer to keep together.
So we want to know the "state" of the vehicle when it has an abnormal
load, but in the stratified version, we need to access a different
object in order to get that information.

For me, the simplest example is just of a vehicle moving along
some path. In the stratified version, to get the position of
the vehicle at any time, we need to access the activity object
that describes the movement of the vehicle - we cannot get the
position information directly from the vehicle object. This would
be like having a programming object that handles the time-independent
characteristics of the vehicle, and another programming object
(a vector) that contains the time dependent information.
In the non-stratified version, the dynamics of the
vehicle are captured by the vehicle object itself (as temporal
parts), so all of the information is kept together. The programming
counterpart would be to express the entire vehicle as a "vector"
(a collection of temporal parts) so that the time dependent
parameters are directly indexed to the vehicle object.

Is my understanding of what you want to say here correct?

Steven

[matthe...@informationjunction.co.uk]

Dear Steven,
Yes, that is close enough.
Regards
Matthew

To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/bf7a24ee-daf0-38c0-4199-a15fbe671aaa%40kraines.net.

[Steven Kraines]

Hi Matthew,

Thank you for the confirmation.
Would it make sense to include something along the lines of what
I have written in the paper itself?

Best,

Steven

[matthe...@informationjunction.co.uk]

Dear Steven,

Hi Matthew,

Thank you for the confirmation.
Would it make sense to include something along the lines of what I have written in the paper itself?

[MW] I don't know what you mean here. You did not attach a version with your comments in last time. Can you be specific please?

To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/83841f4a-1900-c564-4eb4-f1e512b8c69d%40kraines.net.

[Steven Kraines]

Hi Matthew,

Sorry - what I meant is
"do you think we should add X to the paper?"
where X is something along the lines of what I have written
in my previous email:

>> Regarding the example of the abnormal load, I guess that the point
you are trying to make is that a decision to stratify the 3D physical
object from the "story of its existence" (I like the comparison between
a person at one point in time and the person's life) ends up forcing us
to separate information that we would prefer to keep together.
>> So we want to know the "state" of the vehicle when it has an abnormal
load, but in the stratified version, we need to access a different
object in order to get that information.
>>
>> For me, the simplest example is just of a vehicle moving along some
path. In the stratified version, to get the position of the vehicle at
any time, we need to access the activity object that describes the
movement of the vehicle - we cannot get the position information
directly from the vehicle object. This would be like having a
programming object that handles the time-independent characteristics of
the vehicle, and another programming object (a vector) that contains the
time dependent information.
>> In the non-stratified version, the dynamics of the vehicle are
captured by the vehicle object itself (as temporal parts), so all of the
information is kept together. The programming counterpart would be to
express the entire vehicle as a "vector"
>> (a collection of temporal parts) so that the time dependent
parameters are directly indexed to the vehicle object.

Best,

Steven

[matthe...@informationjunction.co.uk]

Dear Steven,
So you presumably mean replace the abnormal load example with just a vehicle journey example.
I agree that is a simpler example, but another commenter specifically asked for the abnormal load example to be included, I think just because it is non-trivial.

To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/5549f637-7703-13e7-2360-3fff4cb16230%40kraines.net.

[Steven Kraines]

Hi Matthew,

No, I actually mean "add" some material.

Possibly something like:

"The point is that a decision to stratify the 3D physical object from
its activities through time ends up forcing us to separate information
that we would prefer to keep together. In this case, we want to know
when the vehicle has an abnormal load, but in the stratified version,
we need to access a different object (the activity) in order to get that
information. In the non-stratified version, the temporal changes of the

vehicle are captured by the vehicle object itself (as temporal parts),
so all of the information is kept together."

Best,

Steven

[matthe...@informationjunction.co.uk]

Dear Steven,
Ah. OK. I'll take a look at it.

Commenting on the Recommendation document is now closed for this phase.

Regards
Matthew West

To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/0536e51d-a8cc-9de0-cf72-db0d6ca2d4d9%40kraines.net.

[matthe...@informationjunction.co.uk]

Dear Steven,
I've added this:
" The point is that a decision to stratify the physical object from its activities forces us to separate information that can be kept together. In this case, we want to know when the vehicle is an abnormal load, but in the stratified version, we also need a different object (the activity) for some of the information. In the non-stratified version, the temporal part of the vehicle is also the activity, so all of the information is kept together."

See notes on why below.

Hi Matthew,

No, I actually mean "add" some material.

Possibly something like:

"The point is that a decision to stratify the 3D physical object

[MW] The object does not have to be 3D. You could stratify with 4D objects as well (you would not be able to have extensionality - you are thinking BFO).

from its activities through time ends up forcing us to separate information that we would prefer to keep together. In this case, we want to know when the vehicle has

[MW] The vehicle *is* an abnormal load.

an abnormal load, but in the stratified version, we need to access a different object (the activity) in order to get that information.

[MW] The same temporal part of the vehicle is the activity.

In the non-stratified version, the temporal changes of the vehicle are captured by the vehicle object itself (as temporal parts),

[MW] See above, the temporal part of the vehicle is the activity, that is an abnormal load.

so all of the information is kept together."

Regards
Matthew West

To view this discussion on the web, visit https://groups.google.com/d/msgid/uk-ndt-fdm/0536e51d-a8cc-9de0-cf72-db0d6ca2d4d9%40kraines.net.

[Steven Kraines]

Thanks, Matthew.

I do think that this addition makes the example (and the
reason for including it) easier to understand.

Best,

Steven