I haven't tried many of the services on this list, but I would recommend Feedly. It can be a bit slow, and the fact that it requires a browser extension is a bit off-putting, but those things are being improved I believe.
They seem to be willing to listen to their users (in particular all of the ex-Google Reader users) and are improving the service very quickly, trying to cater for everyone's seemingly different needs.
The Android app is great as well, and is a much more visual reading experience than Google Reader was (but it doesn't have to be- it seems to be very highly configurable).
The fact that it's free is a bit worrying, but in a survey they sent out last month they asked if user's would be willing to pay a small fee for it- so this might be where it heads in the future (I'd hope so, anyway).
I suppose the main thing you'd need is a data source mapping the stations and lines, then build it in whatever technology from that. The information page doesn't mention where the data came from, so I imagine it was manually created. I don't think TfL would be too willing to provide that data either, judging by the comments regarding licensing etc. in this thread.
This could actually have a very helpful application- generating maps with customised colour schemes. Anyone with a form of colour blindness will know that some of the lines on the standard map are basically indistinguishable from each-other, and while TfL does provide a black-and-white map, it's designed for ALL forms of colour blindness, so they've had to come up with a load of different patterns, making it a bit messy (http://www.tfl.gov.uk/gettingaround/14091.aspx).
With this CSS map a user could customise the colour/pattern of just a couple of problem lines (I could already in Firebug, but it would be great if there was a control panel to fiddle these things and then provide print/save links). Just an idea!
This is pretty fun to play around with, if a bit difficult to navigate- I kept flying all over the place when trying to look at a particular episode in the initial view. Amazing work anyway, and as a (heavy) Lostpedia contributor (around the season 3 days) it's nice to see the data being used in interesting ways.
Edit: Oops, SparkleShare is mentioned on the page. Looks like this will work better with larger files. He does mention that SparkleShare is a GUI and not 'just a folder', although in my experience it is 'just a folder', like Dropbox.
Interesting proposals, but like Andy Rutledge's NYT re-design (http://andyrutledge.com/news-redux.php) limited by the practicalities of implementing them. I don't think the Wikimedia foundation has the resources to do any kind of major re-design, at least not quickly.
I got the impression that by doing "it" he meant searching Google for it, not actually doing it. I think most people realise that Google works by tracking what you search for, so it makes sense that if you want something to remain unknown, you shouldn't search Google for it.
I found it horribly slow from about 18 hours ago (3pm GMT) onwards. I think the last time I tried it was about 9pm GMT, and it was slow them. It took about 30-60 seconds to load a page, maybe more.
It's even smaller than I thought it would be, and so much more portable than a normal UK plug.