For the sake of argument, if bots are allowed but are tied to accounts (that can function normally within the API) what is to prevent bots from just creating a normal account, if it would bypass restrictions (being a full privileged user).
If a normal account is an obvious bot, it would be banned. It also wouldn't have a "bot page" with a "request this bot for your tribe" button and other similar functionalities.
How do you differentiate an user from a bot? Turing test?
I like the reddit aproach since the user needs to comply to several rules, like spam rules/automod rules, etc. But what you are suggesting bots inside accounts, means that a second time the snapzus programers need to write code to deal with spam, automod rules, etc.
Most bots are obviously bots, we are talking about xkcd transcribers and the like.
To you. The problem comes for snapzu to become sentient and differentiate them.
I would think that internally they would be considered users linked to an owner account.
Then they use the same interface as users. You are sugesting making special accounts for bots. Then you are suggesting that a Real Account has bots accounts. All of this is more code, more complexity, more procesing time.
Easily solved by bots using the already Users Interface, and leveling up to prove they are worthy.
For the sake of argument, if bots are allowed but are tied to accounts (that can function normally within the API) what is to prevent bots from just creating a normal account, if it would bypass restrictions (being a full privileged user).
If a normal account is an obvious bot, it would be banned. It also wouldn't have a "bot page" with a "request this bot for your tribe" button and other similar functionalities.
How do you differentiate an user from a bot? Turing test?
I like the reddit aproach since the user needs to comply to several rules, like spam rules/automod rules, etc. But what you are suggesting bots inside accounts, means that a second time the snapzus programers need to write code to deal with spam, automod rules, etc.
Most bots are obviously bots, we are talking about xkcd transcribers and the like.
I would think that internally they would be considered users linked to an owner account.
To you. The problem comes for snapzu to become sentient and differentiate them.
Then they use the same interface as users. You are sugesting making special accounts for bots. Then you are suggesting that a Real Account has bots accounts. All of this is more code, more complexity, more procesing time.
Easily solved by bots using the already Users Interface, and leveling up to prove they are worthy.
To admins and moderators.
So the spam filter would be Humans? That may be too much on them. An algorithm may reduce work.