Quick Accent: Add currencies to VK_3 and VK_4 (most commonly used keys for currencies)

[ansalls]

Summary of the Pull Request

Added currencies to VK_3 and VK_4 (the most commonly used keys for currencies) and a couple of other minor currency updates - added £ to VK_L (base character) and VK_E (similar-presenting character), ¤ (generic currency symbol) to VK_O, and ₿ (bitcoin symbol) to VK_B. Few things were mapped to 3 and 4, so the lists aren't long even with the whole set of currency characters added.

Built and verified updates worked as expected.

PR Checklist

• Closes: Quick Accent Other Currency Keys #20339
• Communication: I've discussed this with core contributors in the noted issue.
• [NA] Tests: Quick Accent doesn't have tests.
• [NA] Localization: QA uses virtual keys which are locale agnostic.
• [NA] Dev docs: minor change
• [NA] New binaries: Added on the required places
  • [NA] JSON for signing for new binaries
  • [NA] WXS for installer for new binaries and localization folder
  • [NA] YML for CI pipeline for new test projects
  • [NA] YML for signed pipeline
• [NA] Documentation updated: If checked, please file a pull request on our docs repo and link it here: #xxx

Detailed Description of the Pull Request / Additional comments

Note: I'd investigated attempting to determine whether a virtual key represented a currency symbol to dynamically show the additional currency symbols only in that event. However, there wasn't a straightforward way to accomplish that and it would require significant changes to the application's structure to implement.

Validation Steps Performed

Built and manually verified that the additional characters appeared and were output following the update.

[ansalls]

@microsoft-github-policy-service agree

[Jay-o-Way]

Not a fan of alot of redundancy

[Jay-o-Way]

My opinion: the origin of the symbol - that is the letter L - outweighs a disputable visual resemblance. If it were we, I would not add it here.

[ansalls]

But does anyone know that's the origin symbol? I live in England and I would not have guessed that. This suggestion feels like sacrificing usability for semantics. Also, with this change, it's in both, which was the suggested fix in the later posts in the issue, so both views are addressed.

[Jay-o-Way]

Same. Just like there's no dollar sign under the "D".

[ansalls]

The pound sign is and was under P so this comment contradicts the existing code base.

[ansalls]

Not a fan of alot of redundancy

Not to be pedantic but this is quite literally not redundant. I think you mean repetitive but negating repetition requires a more reasoned retort given repetition serves an active purpose in a stable system. Alternative paths to useful behavior improve user experience.

On a side note, having utilization based priority of symbols will naturally order them in the most convenient way for each individual user, and I believe there's a request to opt out of specific locale characters, so repetition in the long term has very little negative impact on user experience, so the assumed (unstated) cons to this approach seem very minimal from my point of view.

[PesBandi]

Hi @ansalls, I stumbled upon this PR and I was wondering why you chose VK_3 and VK_4, so I decided to do some counting. I don't really understand where you got the idea of these being the most commonly used for currencies.
There are 214 keyboard layouts. I counted all three shift states: base, shift and ctrl+alt. I somewhat understand your choice of VK_4. Out of the 642 possible combinations, 102 are a currency symbol. If we assume that every layout has at most one shift state with a currency symbol, 47% of layouts have a currency on VK_4, however in reality this number is lower. This kind of makes sense.
I have a bigger problem with VK_3 though. The number is 9%. 9 or less percent of people would understand and expect (this is questionable) the behavior you are proposing. The other 91 (or more) would have 28 (!), seemingly random characters on keys where they only expect number-based symbols.

In my opinion this would make Quick Accent, already a relatively complex tool, unnecessarily more confusing and hard to use. Maybe my judgement is wrong, so message me with any counterarguments you might have – I am open to discussion. Best regards.

(Edit: turns out there are 217 layouts, I used an old list. I am not going to recount, as it would have minimal effect on the numbers)