What about world space canvases?
“Now, when the Quick Menu or Main Menus are shown, we find all user-created screen space canvases and automatically hide them (if not already).”
I worry that vket ex menu is being affected in 2022 open beta (yes I’ve made a report elsewhere) as well as other tricks we’ve done like our map alternate world’s “rift wristband” which has artist information and other settings.
We actually use the quick menu buttons to call up the menu, too.
My question for the community is have you tried some of the “extended canvas based menus” in worlds in open beta and have you see other instances of it break?
I can say for sure it no longer works in 2022 from vket 3 and beyond all the way to summer 2023.
Part of the problem here is that a lot of people that do that now will upload it on a second account. So if the account gets banned, it doesn’t matter. They will create a new account and reupload them.
VRChat could contextualize the account and check if there are any other VRChat accounts from that same IP address. But when to use that and how is contentious because of privacy, nevermind that people can probably still use VPNs.
I am also frustrated by this, but I didn’t have the words to describe how. Thanks for clarifying that.
They don’t even need to do that. It is completely reasonable to have, for example: "Udon" 2.x.x, or to have "Udon 2" 1.x.x.
From my understanding, minor in semvers works the way it does because, sure, you can release a feature on minor that is compatible with older content, but new features always have a chance to introduce bugs. Like how I can open up content made in blender 2.9 in 3.6, or vice versa. But the functionality is not guaranteed to work.
I’m sure we all know that people are going to see the 3 dots and assume it’s semver, but it’s going to cause problems down the line if VCC or whatever introduces functionality based on those versions.
for the screenspace canvas changes. Does this change affect all platforms? Disabling makes sense on desktop or mobile where the menus and quick menus are in a fixed location locked to the screen, but on pcvr and quest the menus aren’t locked to the screen (unless you set it up for quick menus).
As a primarily PCVR user, on the one hand I’m not sure if I’d like to be taken out of an experience that uses screenspace effects if I want to change something through the quick menu. But on the other hand, having a quick way to instantly stop an overbearing effect just by opening the menu sounds like a nice method.
I’m sorry that you find it very frustrating, our hope is that clarifying the version scheme going forward will reduce frustrations that we haven’t been following SemVer properly.
Yes, you could use ~3.4.x to ensure that your package is not used with a newer version which might have breaking changes.
We discussed the approach of having SDK 3 4.0.0, and thought that beginners would have an especially hard time understanding if that means it’s SDK3 or SDK4.
Did SDK 3 already go through 1.x.x, 2.x.x, and 3.x.x? Then you had this problem already, no? Ideally, it would just be VRC SDK 4.0.0?
You can avoid numbers altogether and just focus on the major version being a proper major version. It seems you dug a hole for yourselves by calling it “SDK 3”. Or really, if it’s just the VRC SDK, version 3.x.x, then it already is VRC SDK.
I don’t see the point in calling it SDK 3 4.0.0, if 2.x.x was SDK 2. It would just be VRC SDK, 4.0.0, paraphrased as “SDK 4”, no?
You’re implementing your own semver which contradicts the actual standard, because of a problem you guys created, from what I am gathering. There is no reason everyone else should be forced to use the same broken standard instead of the actual semver standard that the majority of software engineers are used to.
Edit: Nevermind, I guess I missed this part:
It’s still really silly. Just bump to SDK 4.0.0 and tell people that there are no breaking changes between 3 and 4. It’s literally that simple. You now have created a problematic paradigm that is going to confuse more people than it helps.
The only reason I saw was that SDK 3 4.0.0 would be confusing. Yes, it would. That’s why it should be SDK 4.0.0, just like how SDK 3 is SDK 3.0.0, or should be.
And in the feedback post, it even mentions how 2022 is a breaking change and requires 4.0.0. I don’t see why you can’t just use normal semver. Starting SDK 3 out on 3.0.0 makes sense, since it is the SDK… version 3.0.0.
I just simply don’t understand why you are choosing to make things infinitely more complicated. If SDK 3 was branding, it should have started at SDK 3 1.0.0. And in this case, you could have done SDK 4 1.0.0 if you really desired to.
Edit: Also, that feedback post has 7 upvotes, for an opinion… IDK anymore man.
SDK 4 would have been fine. I don’t appreciate people that pre-fix perceived issues, but then ignore real issues or feature requests that have been longstanding for years. You’re letting the made-up marketing terms rule over how you do versioning for software. Just do it right and elaborate if needed, people will understand.
This is great till more devs enter the development fold for VCC packages and might get more confused over not properly following semver when literally every other package creator will be, along with other package managers also following this standard. Alongside that, this is in a way a breaking change to the VCC dependency manager itself since packages have been assuming that everything IS semver, and now that isn’t true, and since we can’t even yank packages we can’t even fix this for previous releases
I will admit I was a bit passive aggressive. But it was for a good reason.
Asking us to stop with the long-winded discussions that are difficult to read: Sure
Telling us that they vaguely will moderate “off topic” posts, and then directly violating that rule: Bad
It was super vague, I asked for clarification, and it is still relatively vague. VRChat has a massive PR problem that they are not willing to solve. I just wish they’d check their posts with other team members, or not reactively post when representing VRChat.
If things are unclear, I am going to ask for clarification. If people are going to take me being literal, direct, and asking legit questions as rude or passive aggressive, idk what to tell them.
I was mostly meaning on a larger scale - you don’t have any public surveys or anything to get mass opinions. The communication is isolated and anecdotal, with only small pipelines like in person or places like here. The result is bias and inaccuracy due to sample size.
If you started doing surveys on what general opinions were on things, such as what things users consider most important (performance vs features, avatar features vs world features, UX features vs creator features, nuance/granular features vs large features, what gives people fatigue or nausea, votes on whether people are satisfied with a feature or think it needs improvement) etc, i have little doubt you’d get different results than expected. Most people want UX improvement before more things to make cool things with (even though the latter can generate more hype) because they have to deal with it constantly. Such data would be extremely invaluable due to the infancy of social VR.…you could even sell the anonymous data for other companies to inform their own development
Surveys should be available via twitter and discord announcements, or better yet, ingame with a menu popup or survey tab.
Recently they added some languages that will be supported in the future. If you have an interest in their statuses, check the dedicated Discord server “VRChat Localization” (The invite link is here: Suggesting Localization Changes )
I was wondering if there are any plan to revise both PC Vr and Standalone Vr avatar rating or it is going to be untouched after the imposter system comes out.
With the quest poly count being so hard to reach for regular player, it creates the opposite effect to what the performance ranking is trying to do. People see it as “all or nothing”. “If the avatar is so hard to reach good/medium, why should I bother?” Which ends up into a scenario with users completely giving up on the idea of optimizing their avatar on the quest platform.
Anyone playing the game on quest or on PC will notice how little people care about avatar optimization, its either by laziness or because of difficulty to reach certain numbers.
For the people in the second category, they have difficulty to reach, for example, the extremely low poly count (10k), and they end up just giving up on it.
I don’t appreciate people that pre-fix perceived issues, but then ignore real issues or feature requests that have been longstanding for years. You’re letting the made-up marketing terms rule over how you do versioning for software. Just do it right and elaborate if needed, people will understand.
