Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There's two ideal ways to deal with this:

1. Accept that the new standard is that ip addresses should be supported, and write a new rfc document with that change. 2. Encourage people to not support this.

Simply asking all mailservers to support this non-standard feature (or de facto standard) is arguable the worst outcome, because now every mail server implementer is expected to implement the standard + a bunch of institutional knowledge.

Luckily with mail there's a lot of incentive to set things up correctly, because getting parts of it wrong likely results in a higher number of emails going to spam.



There’s also the option to tell them that “you shouldn’t do it this way, but I’m going to tolerate it ... for now.”

I made up the tongue-in-cheek HTTP 397 for this case:

https://pastebin.com/TPj9RwuZ


The problem with this is if a few popular servers are lenient, people are going to start relying on this lenient behavior.

More people relying on this means more servers will have to start supporting this bug and you end up with the 'worst case' again.

Being strict if you can tolerate it is usually the better option.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: