He's not dead, he's resting

Today In PMS FUD: PMS Shot JFK

Patrick Lauer is complaining that one of his patches to PMS was rejected. What he is not telling you is why it was rejected, so I shall explain here:

The version of bash to be required by ebuilds was decided by the Council, as was the specification defining it. The PMS project quite rightly doesn’t have the authority to override a Council decision. Patrick was told this, and was told that the process for getting EAPI definitions retroactively changed was to go to the gentoo-dev list and from there to the Council. Such changes have been made in the past, and they’re open to being made again in the future.

Unfortunately it looks like he’s not interested in getting his concern fixed, and is instead just trying to cause trouble.


5 responses to “Today In PMS FUD: PMS Shot JFK

  1. ant September 22, 2009 at 3:02 pm

    He may be a huge troll, and lying through omission, and fluent in LUE/4chan slang, but he’s got a point. Why bother following Gentoo’s rules correctly? Gentoo obviously doesn’t.

    Even though you’re in the right, I think you’re fighting a losing battle here. PMS needs to take a direction more like HTML 5 than XHTML 2; try telling a million idiots that <!-- -- --> is invalid and you’ll get nowhere.

    • Ciaran McCreesh September 22, 2009 at 3:31 pm

      The solution to that is to make package managers enforce the spec. If web browsers started refusing to accept invalid input, people would have to fix their code. Of course, the browser vendors don’t have the balls to do that, and unfortunately it looks like Gentoo doesn’t either.

      • florian September 25, 2009 at 12:55 am

        browser vendors don’t do it, because it would be a pain.

        general input/output rule: be tolerant in what you accept, strict in what you give out.


        (ein jeder kehr vor seiner eigenen tuer, da hat er dreck genug dafuer :) )

        • Ciaran McCreesh September 25, 2009 at 1:08 pm

          The more I see that the more I become convinced it’s a really bad rule. It leads to people writing bad code, which means things end up only working on one implementation, and then future versions of that implementation have to include horrible hacks so that things they accidentally accepted the first time around still work.

          The C++ standard guys have it right: never accept any non-standard input without issuing a warning.

        • Steven Oliver September 25, 2009 at 5:47 pm

          That sounds ass backwards to me. It’s that kind of thinking that leads to SQL injection.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s