If they scrape the updated comments again and ingest copyrighted text, you are poisoning the data.

I think you missed the part where you were strongly suggested “not” to use copyrighted text.

The point is not to get rid of the original text. It’s to “poison” the training data.

Nice write-up!

It’s been crazy to see the AI hype train amusement park ride lately.

Sure sure. Completely agree.

And employees may say no.

Maybe. I guess I hate the idea behind Jira more than Jira itself. I call it middle management driven Agile.

Also doesn’t help that Jira (and Confluence) were fucking slow for a long while. Nobody wants to use slow software as an integral part of the dev chain.

Anything simple. I’m using Linear at my current job, which is fine. I’ve used Trello in the past, also fine. Best experience so far was using the GitLab issue tracker, but it as not a product team so YMMV.

Jira

It’s not all bad! The Cleon part is pretty cool. You know… the part that is not in the books at all.

Thanks!

Could you expand a bit on how your setup works? I think I get the big picture, but not sure how tailscale fits into it.