New section on "virtual things", clarifying terminology #696
Conversation
mlagally
requested review from
k-toumura,
egekorkan,
mmccool,
sebastiankb and
ashimura
| <h3>Virtual Things</h3> | ||
| <p> | ||
| A virtual thing is an abstraction of a physical entity that does not have a direct physical counterpart. | ||
| It can be used to describe abstract interfaces with a TD that are decoupled from individual devices. |
There was a problem hiding this comment.
Until this line I was OK but then after this line, it sounds too much like TM. It is also a bit wrong to mix the two.
There was a problem hiding this comment.
@egekorkan
Please be more specific about your concerns. Please review the virtual thing use case in the use case document.
There was a problem hiding this comment.
I agree, "abstract interfaces" sounds too much like what a TM is trying to do. It may be explained elsewhere, but that does not mean this statement isn't confusing. If you remove the word "abstract" it might be ok. For example, it could be a software-only service or simulation, which has an interface but is not a device. To me "abstract" means "without an instantiation" which describes a Thing Model, not a running simulation or service (which is instantiated. It is virtual, but not abstract).
There was a problem hiding this comment.
@egekorkan Please be more specific about your concerns. Please review the virtual thing use case in the use case document.
I had a look and the main issue seems to be the use cases document. Should I open an issue there?
| <p> | ||
| A virtual thing is an abstraction of a physical entity that does not have a direct physical counterpart. | ||
| It can be used to describe abstract interfaces with a TD that are decoupled from individual devices. | ||
| The purpose is to decouple a functional interface that defines the behavior |
There was a problem hiding this comment.
interface is not defined and can be interpreted differently by different people
| <p> | ||
| The virtual thing guarantees that | ||
| devices can be replaced without affecting the implementation of the consumer, | ||
| who can continue using the device as he did before. |
There was a problem hiding this comment.
A better wording around the gender is needed
There was a problem hiding this comment.
I don't understand your concern. Please propose text.
There was a problem hiding this comment.
We should avoid "he" since it projects a gender on the user for no good reason. Just replace with "they", which works fine here. I need to look at the context, but if "consumer" is the Consuming Thing, then "it" may also be appropriate.
There was a problem hiding this comment.
replacing he with a gender-neutral pronoun/subject is needed or the subject should be removed. proposal:
who can continue using the device as before.
There was a problem hiding this comment.
Actually, if you mean the person using the thing, using the word "user" would avoid confusion with Consumer. If you mean Consumer, capitalize and use "it".
| The interface remains stable and is independent from the actual implementation of things behind that interface. | ||
| </p> | ||
| <p> | ||
| The virtual thing guarantees that |
There was a problem hiding this comment.
Who gives this guarantee? It sounds a little too "serious" for a document/chapter that is this "loose".
There was a problem hiding this comment.
Please elaborate what you mean with "too serious"
There was a problem hiding this comment.
I think the issue here is that the spec is making a guarantee but is not in a position to enforce it. We could say that one use case is to provide a stable interface as a wrapper to another service that may be subject to change. It's an application of virtual things, not its sole purpose or something we can guarantee. I could also have a virtual thing that is a software service that is changing all the time. It may not be a good idea, but I might want to do it for various reasons.
There was a problem hiding this comment.
Well this section is non normative so we should not talk about any guarantees
| <a href="https://w3c.github.io/wot-usecases/#USE-CASES/Retail-virtual-thing">Virtual Thing</a> | ||
| <h3>Virtual Things</h3> | ||
| <p> | ||
| A virtual thing is an abstraction of a physical entity that does not have a direct physical counterpart. |
There was a problem hiding this comment.
could just say "a virtual thing is an abstraction that does not have a direct physical counterpart". I had to scratch my head over the "abstraction of a physical entity that does not have a direct physical counterpart". So does it have a physical referent, or not? I think you're trying to say it does not, so...
|
I think the main issue is that this PR contains the text of the use cases document for a use case that is submitted by Connexxus. The wording in the use cases document is confusing (mixing virtual things, TMs and composite Things/TD) into a single definition without realizing. We do not have to copy paste from use cases and instead should write it in a way that fits the other concepts we have in the WoT |
|
Arch call on 10.3.: |
fixes #682
Preview | Diff