[O] Unwanted percent-encoding capturing data from Firefox

[Bill Jacobson]

For some months now, I've been successfully capturing Firefox data as 
explained here: http://orgmode.org/worg/org-contrib/org-protocol.html
But the results below demonstrate a problem. The 2nd and 3rd represent 
what was captured into Emacs, the quoted line in each case being what 
was highlighted in Firefox.
Emacs version is: GNU Emacs 24.0.50.1 (x86_64-unknown-linux-gnu, GTK+ 
Version 2.22.0) of 2011-05-03

Manually copied:
[[http://www.tbray.org/ongoing/When/201x/2011/05/31/Browsers][ongoing by 
Tim Bray · Me and My Browsers]]
"I like that Chrome’s fast, and I really like that it’s robust"

Org-mode version 7.4:
[[http://www.tbray.org/ongoing/When/201x/2011/05/31/Browsers][ongoing by 
Tim Bray · Me and My Browsers]]
"I like that Chrome’s fast, and I re­-ally like that it’s ro­-bust"

Org-mode version 7.4 (release_7.5.409.g4f3a3) == latest
[[http%3A%2F%2Fwww.tbray.org%2Fongoing%2FWhen%2F201x%2F2011%2F05%2F31%2FBrowsers][ongoing 
by Tim Bray %C2%B7 Me and My Browsers]]
"I like that Chromeâ%80%99s fast%2C and I re%C2%ADally like that 
itâ%80%99s ro%C2%ADbust"

I don't yet have the Elisp chops to see where this should be mended.

(The bogus hyphens are only a minor concern but they're a puzzle. 
Apparently some Javascript is sprinkling the text with soft hyphens 
which are removed when copying via the browser but not when capture 
copies using window.getSelection())

Thanks, Bill