Comment 20 for bug 1119420

As I'm sure is the case with many other users and applications, this causes a regression / loss of previous functionality for Pidgin (despite some integration with the messaging app indicator, or whatever they call it): https://bugs.launchpad.net/ubuntu/+source/pidgin/+bug/1072637 .

I acknowledge that some people don't use a particular tray icon, or features from it, but that doesn't diminish the validity of the use case held by many others who do utilize them, and have diminished functionality (or complete inoperability in some cases) as a result of this.

That said, I fear Canonical is taking the approach that our loss is inconsequential, and abandoning these icons provides a greater benefit. The two "benefits" I can see them arguing for are:
1. Less code provides less of a maintenance burden for future changes and less opportunity for bugs.
2. Eliminating all those crufty old apps that don't want to fall in line with Canonical's new appindicator approach makes the system "cleaner" and more "unified."

While either of those is arguably true in its own right, I have trouble imagining that this particular code (a mere 42 lines pulled out, by my count) would impose an undue maintenance burden, particularly considering they're retaining a hard-coded whiltelist for wine and java and will continue to support the dconf whitelist for years in 12.04 LTS. And, since this already required a user to modify a dconf setting to expose this functionality, those users are more likely to be aware of the caveats for 'legacy' icons versus appindicator icons, and at the very least should prefer having the functionality at all, even if inconsistent, compared to none.