Beta to Live transition

[Adrian Hudson]

Hi Steve,

In addition to David's questions ( https://cumulus.hosiene.co.uk/viewtopic.php?f=27&t=12975 ) and your replies regarding licensing/donationware v paid etc:

Beta testing is difficult in view that most (all?) people don't have two weather stations (or virtual VP). What I do is save the whole Cumulus1 folder, do my testing, restore the Cumulus1 folder and let Cumulus1 download all the data again from the Davis console. I guess some people don't even have this luxury.

What I am asking is your thoughts around using MS permanently - i.e. migrating to it and staying that way.

Obviously you can't advise people to do that - it being beta software with the (seemingly minimal - because it seems so relatively bug free - or it will be in a few iterations) risks that using such software incurs.

But, what you could do is give some idea of your intention and the likelihood that MX will become a live product and whether you commit to continuity of availability from beta to production.

If you can give that commitment - i.e. that there will be an operational beta version one day and a live version the next - then, we testers could make the individual decision to migrate to MX now and stop keeping Cumulus1 up to date. This would relieve a the burden of keeping two versions current - with all the risks of data loss that doing that incurs.

We realize that this software is currently Donationware (and may or may not stay that way) and thus is supported voluntarily by you but the data stored within the application is important - at least to us. Knowing that we can - probably - rely on the new version's availability would go a long way to helping us decide how to do our tests and how we are likely to be able to proceed out of beta.

Adrian

[steve]

I've been using the engine for my live system for over four months now. I am reasonably confident that the engine works pretty much as well as Cumulus 1 does, for Davis stations, on UK systems, at least. In theory, because MX uses the same data format as Cumulus 1, I could go back to using Cumulus 1 if necessary, with my current data. I haven't actually tried this recently, but I have no desire to go back to Cumulus 1 anyway, ever. Obviously not everyone has this option at the moment as the rest of the code is less well tested, and isn't yet working at all with some stations. Everyone can make their own decision based on their experience so far.

There are obviously still issues to be resolved regarding the user interface and its interaction with the engine. Hopefully these can all be resolved.

My intention is that the engine part at least will become the replacement for Cumulus. Possibly with a different user interface if the current one eventually proves untenable for whatever reason.

[Adrian Hudson]

Thanks Steve.

There is an art to wording posts on forums and I don't possess that art. I hope I didn't irritate you

Adrian

[steve]

Not at all.

[duke]

Possibly with a different user interface if the current one eventually proves untenable for whatever reason.

I sincerely hope not. Especially when users get used to it and learn how straight forward it is to customise to their own 'tastes'.