[exia]

Can you please add a checkbox for users that will enable a "Beta Mode" for a Bot?

What is beta mode?
Beta mode would be a version of the bot that is compiled from the head of the author's SVN repo instead of from the revision that they have pushed into the queue.

Why do I want this?
I want to develop my bot and make it premium. I don't have the time to test the bot for hours or have the desire to do so. There are plenty of leeches in the community who would test it willingly and would be able to collect valuable error reports on new features. There should be a way I can rapidly push out updates so I can speed up development time and ultimately release my bot to be premium.

The large user base can find bugs and exceptions much more easily than I can and having to wait 48hrs+ every time I push an update makes it difficult to utilize the user base for testing.

How would it work?
The user checks a box when they start the bot and gets the head of the repo instead of the approved revision. They would also get a warning when they check the box that "Beta version has experimental features and has not been verified by RM staff and could contain bugs or malicious code"

In fact this could be integrated into the main bot development cycle, by simply requiring a certain threshold of hours/session in beta before a bot can be approved as stable and ready for general use.

There is great possibility here because everybody wins:
Win for the Author: because we don't have to spend hours testing a bot.
Win for RM: The easier it is for authors to develop, there will be more quality bots and a higher quality of product, and the increased possibility of premium bots being made and meeting the required quality mark.
Win for the general user base: Even if they receive some buggy bots they also get a lot of free quality content mixed in, with the addition of faster updates and bleeding edge features.

@Arbiter @Cloud @SlashnHax

Thanks @Stunt for the amazing idea.

 

[party]

While I feel the likelihood of this happening is very, very low, I support it. Arbi is too security conscious to ever let this through.

It's really frustrating having to wait (up to) 2 days for a push, not that I'm blaming Slash for that; there's a lot to push and he has better things to do with his time. There should be some kind of faster track for those of us trustworthy enough.

 

[exia]

While I feel the likelihood of this happening is very, very low, I support it. Arbi is too security conscious to ever let this through.

It's really frustrating having to wait (up to) 2 days for a push, not that I'm blaming Slash for that; there's a lot to push and he has better things to do with his time.

Maybe there could be a trusted author rank that would be allowed to use this. There are a few of us who commit consistently good code, and the bot would still need to be approved when it becomes the current version.

 

[snufalufugus]

Great idea, but I doubt it would ever get implemented. All it takes is one angry author or one stupid author who gets their account hijacked, and RM would have a very angry customer base.

 

[botterking]

I would love for this to be implemented I would be more then happy to test out bots before they are pushed, as I have about 10 dummy username accounts just for this purpose

 

[party]

Great idea, but I doubt it would ever get implemented. All it takes is one angry author or one stupid author who gets their account hijacked, and RM would have a very angry customer base.

The scope of this is actually very small, you're overestimating the number of author's this would be intended for.

 

[exia]

Great idea, but I doubt it would ever get implemented. All it takes is one angry author or one stupid author who gets their account hijacked, and RM would have a very angry customer base.

I think it's also a use at your own risk feature, which is why there would be a big warning when a user selects the feature.

 

[proxi]

I'd actually love to see some variation of this idea be implemented.

I also like the 'Trusted' authors idea.
 
Or maybe each author and/or bot can have a 'Tester' group in it, and the bot author can assign users that are allowed to grab the most recent commit from the SVN (with a warning of course).

Idk, I think it'd make the Bot Author -> End User relationship a little better <3

 

[exia]

I'd actually love to see some variation of this idea be implemented.

I also like the 'Trusted' authors idea.
 
Or maybe each author and/or bot can have a 'Tester' group in it, and the bot author can assign users that are allowed to grab the most recent commit from the SVN (with a warning of course).

Idk, I think it'd make the Bot Author -> End User relationship a little better <3

I'd like to see it open to all users so I can get the most people throwing test cases and runtime at my bot.

 

[proxi]

I'd like to see it open to all users so I can get the most people throwing test cases and runtime at my bot.

Well yea, I'd like to see that as well. I was just meaning to give an alternative yet similar solution if Cloud and/or Arbiter didn't like the idea.

 

[snufalufugus]

The scope of this is actually very small, you're overestimating the number of author's this would be intended for.

I don't think I am; I'm assuming that only people with staff/tester rank would have access to this. Regardless of who the feature is available to, all it takes is one angry or stupid person to cause a big problem.

I think it's also a use at your own risk feature, which is why there would be a big warning when a user selects the feature.

Users never pay attention to warnings. People try to start fishing bots with no equipment and complain on the forums saying the bot is broken.

Maybe authors could distribute beta access on a user-specific basis, similar to private bots now? This would block the feature out from the kinds of people I just mentioned and ensure that there is an understanding between the author and user, but it would also take away pretty significantly from the potential benefits of the system. It would be extra work for the author (still a lot less than testing yourself, though) and significantly less testers than if it were open to anyone.

 

[awesome123man]

Support
Even though this has no chance of getting passed as party said earlier. So i understand if this doesnt make it through, but it is an amazing idea

 

[cloud]

Not going to happen for a variety of reasons including security, the fact that not every revision is compiled, etc

 

[yuuki asuna]

Not going to happen for a variety of reasons including security, the fact that not every revision is compiled, etc

So what? Just have it say WARNING and its not your fault.. If you cared that much about your userbase you'd have the Payment FAQ done before multiple people had problems.. This should be added imo.

 

[exia]

Not going to happen for a variety of reasons including security, the fact that not every revision is compiled, etc

One one extra revision needs to be compiled. If it fails for any reason, then no beta will be available.

I don't see the security concern at all. You should already have automated detection in place to detection things like the Reflection packages being used. It would be a use at your own risk feature, if some accounts get stolen that is the user's problem and if they complain just direct them to the terms of service and make sure it has a clause about Beta usage. ¯\_(ツ)_/¯

We either need this, or we need a streamlined version of the current bot approval method, right now it simply takes too long.

 

[jhinn]

One one extra revision needs to be compiled. If it fails for any reason, then no beta will be available.

I don't see the security concern at all. You should already have automated detection in place to detection things like the Reflection packages being used. It would be a use at your own risk feature, if some accounts get stolen that is the user's problem and if they complain just direct them to the terms of service and make sure it has a clause about Beta usage. ¯\_(ツ)_/¯

We either need this, or we need a streamlined version of the current bot approval method, right now it simply takes too long.

I don't think RM wants a reputation where once a user uses a 'beta' bot, you risk your account details + possibly more. Even if it's their own risk, I still think Cloud and/or Arbiter don't want to deal with that stuff.

Don't get me wrong, I think it's a good idea, but I understand why it won't be implemented.

 

[exia]

I don't think RM wants a reputation where once a user uses a 'beta' bot, you risk your account details + possibly more. Even if it's their own risk, I still think Cloud and/or Arbiter don't want to deal with that stuff.

Don't get me wrong, I think it's a good idea, but I understand why it won't be implemented.

Yeah, makes me sad.

On a side note @Cloud what about priority/expedited approval for premium scripts?

 

[aidden]

Yeah, makes me sad.

On a side note @Cloud what about priority/expedited approval for premium scripts?

Or just no need to approve bots for extremely trusted authors...

 

[wet rag]

Or just no need to approve bots for extremely trusted authors...

Especially for staff members at least....

 

[quantum]

one word : Gh0$t