3 ms·

PR spam is a major problems for repo that run bounties. Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejecte

PR spam is a major problems for repo that run bounties. Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejected.

marginalx 5mo ago
Problem is the bots can create any number of github accounts and continue spamming. Though this would be a good simple defense to start with.
hiccuphippo 5mo ago
GitHub has not incentive for blocking AI. It's like asking an ad company to build an adblocker into their browser.
- rvnx 5mo ago
  It's called Brave
  - smaudet 5mo ago
    Which is not chrome and still has ads...(Ironically). The issue here is the core model is broken (misaligned incentives). That's not something you are going to fix with a github "downstream". A token system could help but it's easy to imagine ways that could be gamed, if not implemented well.
    - rvnx 5mo ago
      Ads are the main business model of browsers. If search ads are blocked on search engines, then there is no revenue for the browser. It's that simple (on top of that Brave has other revenues, but the majority is search ads). So it's a game of hoping that the majority won't change the default. This is the main reason Brave does not block search ads specifically by default, but still block the other ads. Blocking the other ads there are no consequences, since anyway this revenue is not shared back to the browser. This is why the business model of Brave is cynical. -> It's the same model as AdBlock and the "Acceptable Ads" (block all ads, except the acceptable ads, unless you disallow them)
      - expedition32 5mo ago
        Ads are the main business of the internet.
cdrnsf 5mo ago
GitHub and Microsoft are actively contributing to the problem, why would they admit fault?
microtonal 5mo ago
I feel like GitHub should have a system where you can give out tokens that are valid for e.g. 1 PR. If someone shows to engage in meaningful discussion and has a good idea to address an issue/feature, you initially give them one PR token. If the PR is of good quality, you can give them a few more, until they are contributors that can just create PRs as they like. A similar system would be nice for issues, though I'm not sure what it'd look like if issues are the springboard for contributing PRs. Not likely to ever happen (as others said), GitHub/MS want to sell CoPilot subscriptions/tokens and LLM-generated PRs are a part of that business model.
- ZeWaka 5mo ago
  My community does something vaguely similar, where you get credit for having bugfix PRs merged, and it's deducted when you get feature PRs merged.
- pbhjpbhj 5mo ago
  You could use a "OTP" to provide this: give out tokens ("OTP"), anyone with that token can submit, keep a record of the user (eg github username, email address) and token, run a bot to delete submissions that don't have a valid token, check the token+user pair, if there is a mismatch blacklist the user that token was given to whilst removing the PR.
  - sdsd 5mo ago
    If only there were a simple way to make tokens that weren't fungible and could be given to others /s
moraesc 5mo ago
[flagged]
godelski 5mo ago
I've gotten tons of spam on repos that were purely ml research code. Things I saw copy pasted over hundreds of repos. > Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejected. It's so bad I'd be okay with a lower bar where it's flagged if they're posting the same message over multiple repos... FFS they aren't even stopping this shit https://news.ycombinator.com/item?id=47964617 https://news.ycombinator.com/item?id=47964617
avs733 5mo ago
It seems like some better basic metrics should be made front and center with PRs in this day and age. Yes AI is the driving force behind the current crop of problems but there are other issues. Yes it’s accessible if you go look but the point is people don’t have time. the rate of comits/PRs total The rate of PRs to repos they don’t own The reject rate of PRs The number of ban An estimated “AI” or bot score or status flag There are a few better attempts at GitHub metrics calculators but I have not seen any that move beyond the paradigm of more vomits is default assumed good. It’s time to foreground quality not just quantity. The GitHub “4 kpis” are entirely action oriented.

PR spam is a major problems for repo that run bounties. Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejected.

marginalx 5mo ago

Problem is the bots can create any number of github accounts and continue spamming. Though this would be a good simple defense to start with.

hiccuphippo 5mo ago

GitHub has not incentive for blocking AI. It's like asking an ad company to build an adblocker into their browser.

rvnx 5mo ago
It's called Brave
- smaudet 5mo ago
  Which is not chrome and still has ads...(Ironically). The issue here is the core model is broken (misaligned incentives). That's not something you are going to fix with a github "downstream". A token system could help but it's easy to imagine ways that could be gamed, if not implemented well.
  - rvnx 5mo ago
    Ads are the main business model of browsers. If search ads are blocked on search engines, then there is no revenue for the browser. It's that simple (on top of that Brave has other revenues, but the majority is search ads). So it's a game of hoping that the majority won't change the default. This is the main reason Brave does not block search ads specifically by default, but still block the other ads. Blocking the other ads there are no consequences, since anyway this revenue is not shared back to the browser. This is why the business model of Brave is cynical. -> It's the same model as AdBlock and the "Acceptable Ads" (block all ads, except the acceptable ads, unless you disallow them)
    - expedition32 5mo ago
      Ads are the main business of the internet.

cdrnsf 5mo ago

GitHub and Microsoft are actively contributing to the problem, why would they admit fault?

microtonal 5mo ago

I feel like GitHub should have a system where you can give out tokens that are valid for e.g. 1 PR. If someone shows to engage in meaningful discussion and has a good idea to address an issue/feature, you initially give them one PR token. If the PR is of good quality, you can give them a few more, until they are contributors that can just create PRs as they like. A similar system would be nice for issues, though I'm not sure what it'd look like if issues are the springboard for contributing PRs. Not likely to ever happen (as others said), GitHub/MS want to sell CoPilot subscriptions/tokens and LLM-generated PRs are a part of that business model.

ZeWaka 5mo ago
My community does something vaguely similar, where you get credit for having bugfix PRs merged, and it's deducted when you get feature PRs merged.
pbhjpbhj 5mo ago
You could use a "OTP" to provide this: give out tokens ("OTP"), anyone with that token can submit, keep a record of the user (eg github username, email address) and token, run a bot to delete submissions that don't have a valid token, check the token+user pair, if there is a mismatch blacklist the user that token was given to whilst removing the PR.
- sdsd 5mo ago
  If only there were a simple way to make tokens that weren't fungible and could be given to others /s

moraesc 5mo ago

[flagged]

godelski 5mo ago

I've gotten tons of spam on repos that were purely ml research code. Things I saw copy pasted over hundreds of repos. > Maybe GitHub should temporarily block accounts from raising PRs if like 95%+ of them are getting rejected. It's so bad I'd be okay with a lower bar where it's flagged if they're posting the same message over multiple repos... FFS they aren't even stopping this shit https://news.ycombinator.com/item?id=47964617 https://news.ycombinator.com/item?id=47964617

avs733 5mo ago

It seems like some better basic metrics should be made front and center with PRs in this day and age. Yes AI is the driving force behind the current crop of problems but there are other issues. Yes it’s accessible if you go look but the point is people don’t have time. the rate of comits/PRs total The rate of PRs to repos they don’t own The reject rate of PRs The number of ban An estimated “AI” or bot score or status flag There are a few better attempts at GitHub metrics calculators but I have not seen any that move beyond the paradigm of more vomits is default assumed good. It’s time to foreground quality not just quantity. The GitHub “4 kpis” are entirely action oriented.