On 4 April 2016 at 21:14, Vladimir Sitnikov <[email protected]> wrote:
> sebb>I just don't think we should be expected to help with this beyond
> sebb>providing the information needed to download JMeter.
> Sebb, I think JMeter should provide users an ability to stick with a
> specific version.

That is already possible.

If a user downloads a copy of JMeter, it will be available until they delete it.
If they do accidentally delete it, then they can get it again quite easily.

> It is crucial for reproducible tests. Unexpected flips of JMeter
> versions would be bad.

It is not our problem if they start using a different version of JMeter.
We have no control over that, nor any responsibility.

> There are cases when "new version" breaks here and there as there are
> just way too many
> moving samplers.

That may be true but is not relevant to how JMeter is made available.

> I would not say JMeter breaks often, however I was personally bitten
> by those issues.

Ditto.

> Being able to ship multiple versions would simplify "try 3.0" kind of 
> scenarios.

We already provide all the published versions.
For all ASF projects. Since the ASF started.

> Bintray is a well-known distribution facility.
> It provides binary distribution for free for OSS projects, so I see absolutely
> no reason to ignore it.

I did not say ignore it, but it is out of scope for the JMeter project.

> Are there objections on not providing bintray binaries?

Yes, it is out of scope for the project.

> Apache Groovy uses bintray, so it should be doable.

Of course it is doable.
But it's not in scope for the JMeter project.

> There are three tasks:
> 0) Configure bintray repository owner somehow
> 1) Integrate bintray publish action into release process
> 2) Publish previous versions

Sorry, but these tasks are out of scope for the JMeter project.

> Vladimir

Reply via email to