Is That Your Final Answer?
Tim Bray: the increase in the total fetch count is for the moment unexplained
Theory: Perhaps some are subscribed to tbray.org which is returning a 302 a.k.a. a temporary redirect. This means that consumers will continue to fetch the original address, and then follow the redirect. Each time.
More likely, it is due to consumers not properly interpreting a 301 status code. Radio UserLand used to have this problem, but that particular tool has long since been fixed.
Apparently, others haven’t. Despite having coded a permanent redirect over three months ago, I still received 69,142 requests last week for feeds in one of several RSS formats.
I would think that three months is more than sufficient for those with a clue to catch on, so accordingly, this is the last entry that such readers will see.
And speaking of problems with readers, Google Reader assigned the following attribute value for the link to this article:
href="" I have seen this happen with Reader many times, mostly on this blog. Any clue what’s going on there? They seem to get it right about 95% of the time.
Posted by Scott Johnson at
Interestingly enough, it seems that Bloglines still is using [link] and not [link].atom, despite the redirects. Current subscribers on Bloglines totals 2,130 for that feed.
OK, just what is the correct Atom feed URL? Bloglines is using index.atom, this article points to Final_Answer.atom, and the post at [link] points to 2215.atom.
Or are those the feeds for the comments? Firefox wants to open the .atom files externally, instead of internally, and I don’t feel like running the files down just to see what is going on. That’s an annoyance when trying to get some feeds subscribed under Bloglines, since it’s easier to just open the XML file with Firefox, and then tell Bloglines to subscribe from that (tool bar button).
As far as hell freezing, Blogger is supporting RSS now.
(Now, to be either damned and/or blessed by Sam Ruby and Dave Winer - in the same post no less!)
Posted by Bryan Price atI was baffled for a moment by those 302s, then realized those are the people who come to tbray.org as opposed to www.tbray.org; they get 302’d over to the www. version and then get the 301. Sigh.
Posted by Tim Bray at
Summary of user agents of fetches to one or more of my various RSS feed URIs.
First glance, this is not an instance of people who consciously decided that HTTP Status codes can’t be trusted, but rather something that Jon Udell has referred to in the past as space junk.
Posted by Sam Ruby atRegarding the Google Reader bug, we have just pushed a fix (Sam switched to using relative URLs for links, until now we weren’t resolving those against the feed URL, like the spec says).
Posted by Mihai Parparita at
The permanent/temporary distinction turns out to be irrelevant, because (buggy) hotel/public wireless access points that require a web sign-on sometimes issue permanent redirects. NetNewsWire hit this bug sometime last year.
Posted by Robert Sayre at