The problem is that it is almost always just one lf them. Let’s say that v0.20 is called “Fuck Spez” and v0.21 is called “YouKnowWhatFuckMuskToo”.
Most people are going to refer to them by either the number or the name, almost never are both used. The biggest problem with names is that they are rarely sortable (google did it with android, for a bit but not anymore), so in the future it is hard to know which is which without resorting to looking at a list of releases.
For example, in the future when we are on v0.30 someone might say “ah, but this has been an issue since “Fuck Spez”.” And then most likely you have to look it up to know what they are talking about. If we coulld force everyone to alwaya write “version “Fuck Spez” (v0.20)” then it would be great, but that never happens.
I personally prefer just semantic versioning for this reason.
What if the names were picked from whatever news was prominent during the release week. During last summer we could have had something like blackout or APIcalypse.
People can search the name or SemVer and get the other name. It’s not hard, not time-consuming, and it’s way more fun. I genuinely miss when Android had dessert-themed names because it made the updates that much more exciting
I think that’s the case and why I said it’s fine in my OC, but the more I think about it I’m slightly concerned it might cause some fragmentation “Does anyone know how to force a user sync in Lightfoot?” and “How can I sync users (forced) in v13.8.12?”
Yeah it muddles with searching for answers, especially when the names are common things. Apple can get away with naming macOS on places in California, but it’s going to be tough searching “having trouble posting comments on Thor” or whatever. The naming scheme can work, but it has to be very unique
Eh, if someone on the team is feeling creative I don’t mind fun version numbers but semantic versioning is quite searchable and reduces confusion.
One does not exclude the other. You can have a fancy name and a semantic versioning.
The problem is that it is almost always just one lf them. Let’s say that v0.20 is called “Fuck Spez” and v0.21 is called “YouKnowWhatFuckMuskToo”.
Most people are going to refer to them by either the number or the name, almost never are both used. The biggest problem with names is that they are rarely sortable (google did it with android, for a bit but not anymore), so in the future it is hard to know which is which without resorting to looking at a list of releases.
For example, in the future when we are on v0.30 someone might say “ah, but this has been an issue since “Fuck Spez”.” And then most likely you have to look it up to know what they are talking about. If we coulld force everyone to alwaya write “version “Fuck Spez” (v0.20)” then it would be great, but that never happens.
I personally prefer just semantic versioning for this reason.
What if the names were picked from whatever news was prominent during the release week. During last summer we could have had something like blackout or APIcalypse.
People can search the name or SemVer and get the other name. It’s not hard, not time-consuming, and it’s way more fun. I genuinely miss when Android had dessert-themed names because it made the updates that much more exciting
I think that’s the case and why I said it’s fine in my OC, but the more I think about it I’m slightly concerned it might cause some fragmentation “Does anyone know how to force a user sync in Lightfoot?” and “How can I sync users (forced) in v13.8.12?”
Yeah it muddles with searching for answers, especially when the names are common things. Apple can get away with naming macOS on places in California, but it’s going to be tough searching “having trouble posting comments on Thor” or whatever. The naming scheme can work, but it has to be very unique
I have exactly this problem when trying to fix issues on my MacBook :(
Hm… Good point.