I share Julian's sentiment. While we're not opposed to using AI tools, submitting multiple Jiras and PRs all at once instantly creates resistance. The normal workflow is to first create a Jira ticket for discussion, then write the code—following this process is much more efficient. Although this kind of behavior is discouraged by the community, the motivation to contribute is something we want to encourage. As Alessandro said, perhaps we should guide him first before making any decisions.
Best regards, Zhen ---- 回复的原邮件 ---- | 发件人 | Alessandro Solimando<[email protected]> | | 日期 | 2026年09月23日 03:38 | | 收件人 | [email protected] | | 抄送至 | | | 主题 | Re: Frank Chen's AI slop | It's probably fair to try to warn them at least once, if they do it again, we can consider this as spamming and block them. Best regards, Alessandro On Tue, Sep 22, 2026, 19:05 Julian Hyde <[email protected]> wrote: > Frank Chen has logged several Jira issues and PRs, all of them clearly > authored by a bot. > > I don't have time to review these. I doubt anyone does. But someone > will probably shrug and say "looks ok", and some of his PRs will get > merged. > > I took a look at just one: https://github.com/apache/calcite/pull/5283 > justifies a change to RexShuttle and SqlShuttle that could have > significant performance implications with one extremely narrow > benchmark. > > What should we do about this? Should we ban Frank? He is logging > issues and PRs faster than I can close them. > > Julian > > https://issues.apache.org/jira/browse/CALCITE-7805 > https://issues.apache.org/jira/browse/CALCITE-7806 > https://issues.apache.org/jira/browse/CALCITE-7807 > https://issues.apache.org/jira/browse/CALCITE-7808 > https://issues.apache.org/jira/browse/CALCITE-7809 > https://issues.apache.org/jira/browse/CALCITE-7810 >
