Closed PingWIN7 closed 6 years ago
Description does not explains it.: This was created as we had a bit of a concern about the meeting id won't be easily readable. The issue here is that in case someone enters the room ID without using capslock (or similar) they won't be able to enter as they used lower case letters.
As well as make sure all checks are case insensitive, we could also generate the codes using unambiguous characters (i.e. ones that are hard to misread). https://github.com/swanson/friendly_code does it from the set of "3 6 7 C D F G H J K M N P R T W X" which seems like a reasonable and gives you 17 possible values at each position.
It's also nice, from a usability point of view, to break the chars up for easier reading. Groups of 3 are pretty common and we used 3 groups of 3 (e.g. "3FG-CDR-T6X") that'll give 118587876497 possible values which should be plenty :)
If the user does type in the "-" we should remove those (they're only there for a UI), and don't require those parts.
Should I be concerned about duplicates? The odds for this to happen is around 1:101,559,956,668,416 in case if it's length is 9 with the current implementation but with an unambiguous characters it'll be 1:24,137,569 for 6 character and 1:118,587,876,497 for 9 characters. (If I counted it correctly)
Yea, I think 6 or 9 unambiguous characters should be fine. 6 makes it a bit easier to type as well.
Added an unambiguous meetingIDGenerator (Implemented it by copying an existing generator and replaced the characters). For the main screen it should not be case sensitive anymore.
Might be related to the enter room method.