Lunie: Releases of Voyager should identify which testnet(s) they work on

[nylira]

UI Version: develop

Description: I think that version 0.9.x of Voyager functioning with gaia-8000 doesn't make much sense. Wouldn't it be better if we just publicly labeled this release as Voyager 8000?

[NodeGuy]

Yes, strong agreement.

[jbibla]

i suppose we can match the testnets and increment changes like gaia-8001.0, gaia-8001.1, gaia-8001.3?

implement semver after 1.0.0?

[faboweb]

Does this have any implications for releases on GitHub or naming or the about section?

[NodeGuy]

SemVer is more important for libraries than for applications. I don't think we need it at all.

[jbibla]

how will we increment version numbers after mainnet launch?

[faboweb]

how will we increment version numbers after mainnet launch?

we would switch back to semver I suppose

[NodeGuy]

package.json requires SemVer.

How do we want to handle this?

[faboweb]

It says version is optional if you don't publish the package. Does something complain when we do not use semver?

[NodeGuy]

Running a script with npm fails. I was unable even to commit the change because Husky failed to run the precommit script.

[NodeGuy]

How about we keep our current version scheme but also display the default_network value from config.toml in the About box and in the release binaries' file names?

E.g.,

Cosmos_Voyager-v0.10.6-gaia-8001-macOS.zip

[faboweb]

Sounds reasonable. Just brainstorming: Can we use funky semver by putting the network number in there? Like 0.8001.1. After 1.x we could change that again.

[NodeGuy]

It would need to go in the major version to comply with SemVer (8001.10.6) because our testnets are not backwards compatible and we couldn't lower the version number to 1.x.x later because it's strictly increasing.

[faboweb]

It maybe doesn't comply with semver, but it also wouldn't crash any scripts I guess.

[NodeGuy]

We won't know but we'll always be in risk of it, which is what happens when we don't comply with expected standards. I don't see that risk being worth the benefit, IMHO.

[faboweb]

A decision was made to do it, there is no technical problem doing so as 0.x.y is compatible to a semver version schema. After 1.x we implement true semver. So let's do it.

[jbibla]

i would propose we stick with semver for testnets and after mainnet launch. let's just change the title of the release for the testnet releases...

Cosmos Voyager Alpha v0.10.6 (UNSAFE)
...
Cosmos Voyager Alpha v.11.0 (Gaia-9000)(UNSAFE)
Cosmos Voyager Alpha Gaia-9000 v.11.0 (UNSAFE)

ps. when do we move from Alpha to Beta?

[faboweb]

👍 Bowing to the majority on this one then.

[NodeGuy]

Don't bow yet! I forgot that we don't need to comply with the normal SemVer rules before 1.0.0. We agreed in the meeting to go with Fabo's proposal.