@glS Sorry for not getting back to you on this, but I have a different idea - instead of (practically impossible to undo) synonymising two tags, I can simply merge the tags so that everything that was previously operator-sum now becomes kraus-representation (I'm hesitant to make synonyms without some meta discussion beforehand, or if it's necessary, while this is simple to either synonymise or undo)

@glS In practical terms of what questions we have, quantum-operation has questions about specific unitary operations, so while these are technically Kraus operators, describing them as such is overcomplicating it, to my mind

So I think, for the latter, the solution would be to go through all the quantum-operation questions and properly re-tag, then synonymise or something

(... Welp, that's my job, isn't it?)

Ahh, no, of course, if everything is currently tagged as operator-sum, it's easy to undo a synonymisation of them

@Mithrandir24601 sounds good. I didn't know synonymising was difficult to undo

@glS In this case it's not, but generally, yeah, the only way I know of would be to go through every single one of the posts and see what it was originally tagged as

@Mithrandir24601 I'm not thinking as the quantum-operation matter as related to kraus-representation. My questions is simply the relation between quantum-operation and quantum-channel. The difference would be that quantum-channel refers to CPTP maps, while quantum-operation to more general CP maps

we can also have a meta discussion of course, just there are not many people on meta so I'm not sure we'd get many more opinions anyway

It just means I'm going to have to re-tag a few questions

@Mithrandir24601 that's true, but I don't have a big problem with tagging questions about gates and unitary operations as quantum-operation. My beef is that it's hard to enforce, and sometimes tell, the difference between questions asking specifically about CPTP maps and those asking about more general maps. We might as well just have a tag for general quantum-operations, which refers more to the formalism and techniques associated

let me get a couple examples

@glS Yeah, that's fair

I guess I just want a parent-child tagging system :/

yea that might be useful. Thought as a counterargument, I never found very useful tags that are too generic

like, when is mathematics or quantum-state ever going to be useful?

One of my examples is this: quantumcomputing.stackexchange.com/q/11445/23

@glS I have no idea

not to speak of quantum-information

It is a quantum operation, but I'd rather it be labelled as 'unitary-operation' or something

@Mithrandir24601 right, that should be quantum-gate and not quantum-operation at all

(although labelling it as both would be fair)

or unitary-operation or whatever

@glS yup

I mean actually my examples are most of the questions tagged as quantum-operation: quantumcomputing.stackexchange.com/questions/tagged/…. Most of them could be argued to be about quantum-channel, but it's hard to tell

@glS Yeah, for sure...

so, replacing channel with operation then?

ok there's plenty of mistagged questions. But to be fair the excerpt itself is confusing

we should make it clear that it is about the quantum maps formalism, not just about the general idea of "an operation"

@glS ... Copied from Wikipedia...

the latter would be a useless tag anyway, essentially anything is about some "operation" in a sufficiently loosed interpretation of the word

Anyone reading this who's making tag descriptions: please don't copy from Wiki

@Mithrandir24601 usually a good advise

anyway, let's fix it now and then we'll slowly deal with the mistagged questions in time

better fixed now than later

how do you feel about For questions related to completely positive (CP) maps. Use both for quantum channels (trace-preserving CP maps) and more general CP maps. as new excerpt?

maybe For questions about completely positive (CP) linear maps between quantum states. Can also be used for trace-preserving CP maps (quantum channels). is a bit better

@glS For the mis-tagged questions, does quantum-gate sound better than (what would be a new tag) unitary-operation? (or something similar)

@Mithrandir24601 I would actually go with synonyms on that one

I mean I cannot see a situation in which we would want to distinguish between them

because a 'for questions about unitary operations, please use [this tag]', would be useful

but many situations in which a user might want to use one rather than the other

@Mithrandir24601 you mean added to the quantum-operation tag excerpt? Yes that's a good idea

@glS Yeah, exactly

done

if you can merge quantum-channel in it we're set

@glS I want to get through re-tagging the questions first, as this is easier pre-synomynising

@Mithrandir24601 helped you a bit =)

on quantum-operation at least

@Mithrandir24601 so we are doing the synonyms? I thought we preferred merging

@glS Yeah, I noticed, thanks :) I've also found things that need actual editing, so it'll take me a few minutes

in the process I also noticed again the usefulness of quantum-information

@glS So, merging just replaces all current instances of one tag with another, but doesn't change its use in the future. If we want to change how something is used permanently, it needs to be synomynised

@Mithrandir24601 well ok, but if the tag disappears new users won't use it, no?

I seem to remember there is a rep cap for creating new tags

@glS This is true. I'll try merging then

@Mithrandir24601 I think I've untagged most of them (though I've scanned very quickly through them)

in the process, also noticed we have unitarity. Shouldn't this be the same as quantum-gate or whatever?

@glS Not so sure - it could be about gates, circuits, whether or not something is unitary...

@glS Good enough for me, thanks :)

And done (turns out they actually had a fairly large overlap)

@Mithrandir24601 that was my thought yes

great!

can I ask your thoughts about quantum-information?

I mean it makes a lot of sense on physics.SE, it's the main tag I use to follow questions there, but on quantumcomputing it seems to generic

maybe we can ask on meta about it

@glS Yeah, this is one for meta, I think