This essay originally appeared on secure scuttlebutt at %3k6qAo85Q/1hjMW6xc3S0MNt+PsBCM00S354HeXOUco=.sha256
I just spent like an hour fiddling with a (very short) script in order to get the RSS it generates to work in somebody’s preferred RSS reader — one that, I guess, was using a standard XML parser (and therefore couldn’t handle ampersands in URLs) and that could only support RFC882 dates (for some reason).
I have no idea if it worked. I can’t run their RSS readers, and all the readers I have can read this feed just fine. And when I search for documentation on the RSS format, the examples I see of well-formed RSS often contain the things these clients apparently reject.
Basically: why is RSS so broken?
The answer, I think, is that it was created with a webtech mentality.
On the one hand, RSS uses XML — which, on top of being absurdly verbose and fiddly to read and write (and difficult to write an efficient or reliable parser for, and subject to strange failure modes), is not at all suited to an application whose primary purpose is to transmit URLs. This is because an extremely common character in URLs, the ampersand, must be escaped (and therefore expanded into three or four characters). Meanwhile, standard HTML escape sequences aren’t technically part of XML so XML parsers won’t necessarily handle them! So, for the titles I automatically process out of the HTML of the linked documents, I would probably need to find HTML escape…
