So the choice is to risk all our stored solutions, recipes and anything else to be able to share between Beta and Developer, or have to put up with not being able to share?
That's a bad and worse choice.
If someone were to switch back from Dev to Beta, could they then load and re-save their own solutions so they could be shared with everyone else still on Beta, or are you just stuck with Developer once you go there?
I changed the status so that it would be visible again (a topic of another feedback).
The alternative is that we release an under-tested or even untested update straight to main, in which case you get no choice but to suffer the bugs. I'd imagine we would get far more complaints with that system than we do with the current one.
No B_2. This is first time as far I remember that dev->main saves not working. This is because save structure was changed. So dev can read both type and main only own ones.
Seth, jflat06 and I will talk about this bug in the chat tomorrow. We'll start 15 minutes earlier to discuss this.
This was never an issue when Beta and Developer could not work on the same puzzles. There was no need to share solutions between Beta (main) and Developer (beta).
I think Developer should work on copies of the "live" puzzles, but not share solutions with the live puzzles. That way the testing would be on the same type of puzzles that are being run live, but without affecting them.
We are getting back to where it is a dis-advantage to people who can't/won't/don't know about the Developer preview because those few who do run it are taking advantage of new features. This was the original argument back in 2008 when certain people were allowed to secretly use future releases of the client with distinctly unfair advantages, and the reason we ended up with the two version thing we have today, except that it got corrupted by letting the Developer version run live puzzles.
I was confused by "Beta and Developer" since to me those are the same: to use the "Developer preview" you set your options.txt to "beta".
I would not confuse this with the 'BETA' in the Foldit BETA icon since that is as different from this topic as my username: beta_helix.
(Just like when we talk about beta sheets we know that we're not talking about paper versions of the Foldit BETA game or something ;-)
We have 2 modes:
1) The default, regular Foldit that everyone uses; also referred to as "main" for its options.txt flag
2) Developer preview; also referred to as "beta" for its options.txt flag
Back to the topic at hand on 'main' vs 'beta'. Your solution of "copies of live puzzles" is a good one, except that the same problem we faced before arises. There will be fewer beta testers because all their time spent on playing the copies of these puzzle will be wasted if they can't use their solutions.
Brick, would you spend your time playing live puzzles in beta (assuming nothing crashed and it wasn't complicated to use) if you didn't get any credit for it whatsoever? I think you of all people would agree that getting points/ranks/rewards for your hard work in Foldit is necessary, right?
I know that we'd get a few people using beta out the goodness of their heart (or realizing that it's better to catch these bugs early) like we did "back in the old days" but as I pointed out before, we can't go back to the "old ways" because that's how we would end up with bad updates to main that would crash the whole game when we didn't have enough beta testers.
Hopefully we will be able to find a better system at the start of the developer chat today.
(Feel free to continue discussion here before then, especially if you can't make it to the chat)
I only have one system to run Fold.it. I cannot risk losing all such things as recipes and solutions with a bad Developer Preview. I've been there before, and it's why I won't even allow a download of "main" for a few days after release. If I had a second system to experiment with, I would.
If the Developer Preview installed in a separate file system and didn't affect the live client, then two clients could be run side by side on a sufficiently powerful machine. That could open up some very specific testing of improved functionality - start the same recipe in each client on two copies of the same puzzle.
Expanding the testing time schedule of Developer previews would allow more people to test. In the past, there has sometimes not been even 24 hours between the two releases.
Main now have same saves as beta. Closing.
