Kubernetes moves to end ‘permanent beta’ for some APIs

The Kubernetes project has decided the time has come to stop existing in a state of permanent beta. The decision, included in the Changelog for version 1.19 of the container-wrangling code and explained in a blog post, reflects the fact that Kubernetes offers plenty of REST APIs and they can evolve … or not. The project’s new …

[Pascal Monett]

That's a good thing

I totally agree with this position. Way too many projects end up gathering dust on the shelves after the initial feature set is more or less working.

This will at least level the field - either someone is actively taking care of the project, or the project dies the death that is waiting for it.

[Mike 137]

"to help slow upgraders"

So "upgrading" is now an obligation that some folks "shirk"? I thought IT was tools for running businesses, but it appears that's only part of its function. The other (and possibly greater) part seems to be to keep vendors' income streams flowing.

We have a mechanical workshop where some of the tools are almost 50 years old. They're still in use because:

[a] the makers got them right in the first place

[b] they still work

If both those criteria were applied to IT, most users would be much happier and much safer.

[martinusher]

Re: "to help slow upgraders"

Its difficult to explain to a modern programmer that you can't keep upgrading production code all the time. In my area when a customer orders a product part of the SKU is the firmware build that they want with that product. The reason for this is that there's often a test and certification cycle associated with the system associated wth that product, its just a small cog in a very big machine, so its thought to be a lot more desirable to incorporate a product with known bugs than introduce code that potentially has unknown bugs. This then extends into software maintainance where key customers' products firmware are maintain on a code archive branch with only critical bug fixes migrated to that branch (and then only after its been green lighted by the customer).

A short way of describing this is that we don't want to end up in Windows 10 type Hell -- there's too much riding on the integrity of the machines to risk a shotgun approach to software updates.

(Then there is the well known engineering maxim -- "If it works, don't mess with it".)

[fidodogbreath]

The Kubernetes project has decided the time has come to stop existing in a state of permanent beta.

So they're moving away from Agile, then?

[Bitsminer]

Fear and Loathing

a significant subset of Kubernetes end-users fail to upgrade within the previous nine-month support period,

In other words, users are afraid an upgrade will break things.

Kubernetes is now legacy software, then.