Rename globals to avoid duplicate BCD keys from a case-insensitive perspective

[Elchi3]

What type of issue is this?

Other

What information was incorrect, unhelpful, or incomplete?

As discussed in London, it is not ideal that BCD exposes these:

api.crypto <-> api.Crypto
api.performance <-> api.Performance
api.scheduler <-> api.Scheduler

So, we think it would be better to change the global properties to something like api.crypto_property etc.

cc @queengooborg @fiji-flo @claas

What browsers does this problem apply to, if applicable?

No response

What did you expect to see?

No response

Did you test this? If so, how?

No response

Can you link to any release notes, bugs, pull requests, or MDN pages related to this?

No response

Do you have anything more you want to share?

No response

MDN URL

No response

MDN metadata

No response

[skyclouds2001]

discussion has occurred in https://github.com/orgs/mdn/discussions/360 and content update has been done in mdn/content#33019

---

if working on, the first step is to split current api/_global/**.json into two entries

its main entry to move to the Window.json, the worker_support entry to move to the WorkerGlobalScope.json

then the documentation should update too

[ddbeck]

This came up on today's BCD call. We discussed it briefly, but realized that no one on the call could clearly remember why we should rename these properties.

I had a look through the linked discussions. It was clearly an issue for mdn/content, but that seems to have been resolved. The only mentions of BCD in those discussions seems to be making BCD consistent with MDN's URLs. I'm not sure that's a particularly compelling reason to do this work.

That said, if the case sensitivity is a problem for consumers, that would make a difference and it would be nice to hear about it.

[skyclouds2001]

I think the reason should be, there is no real "global" in web api. Current the feature we say as "global" in BCD (and previously in content) are only available in window scope and worker scope, via Window and WorkerGlobalScope, through WindowAndWorkerGlobalScope mixin. Those features are not available in other scope (like PaintWorkletGlobalScope, SharedWorkletGlobalScope, AudioWorkletGlobalScope and more in spec definition but hasn't got implement yet). These features are actually properties or methods both on Window interface and WorkerGlobalScope interface, they are not real global at all.

Also notice though the linked discussions are only for content, the cases are the same for BCD. And case sensitivity is truly an issue, but it is not the major concern of this proposal.

[fiji-flo]

I think the issue was that our BCD API is case-sensitive

https://bcd.developer.mozilla.org/bcd/api/v0/current/api.Crypto.json

vs

https://bcd.developer.mozilla.org/bcd/api/v0/current/api.crypto.json

Since it's powered by static files it can only be built in an case-sensitive file system. However, I'm strongly in favor of keeping the current version and not let OS shortcomings leak into design.

[skyclouds2001]

My proposal is changing api.crypto to be api.Window.crypto, and api.crypto.worker_support to be api.WorkerGlobalScope.crypto, to make global properties and methods to be subfeatures for Window interface and WorkerGlobalScope interface.

And this is what https://github.com/orgs/mdn/discussions/360 and mdn/content#33019 suggests. Honestly it doesn't match the proposal of this issue.

[skyclouds2001]

maybe for https://github.com/orgs/mdn/discussions/360 and mdn/content#33019, it is better to create a new issue, rather than linking to the current #18739 issue