<span style="font-family: Arial, Helvetica, sans-serif; font-size: 10pt">This is quick summary since I'm tired, and probably starting (command economy) work tomorrow, to Andrea whom I have invited to join the AE list, and whom I am CC'ing for now. <br />
<br />
One key point is that AE is <em>deliberately non-prescriptive</em> since it is designed to support multiple top layers (people can use this data how they like, and may wish to improve on the default cheap and cheerful upper layer). So it defines a meaning for the data, rather than who it should be used. Compare for example, now HTML assigns a meaning to the link sturucture of <a href="..."> which was defined a long time before Google (or others) chose to interpret such links through the pagerank algorithm. Specifically AE aims to define two sorts of data on top of the F2F network:<br />
1) How people feel about particular things that happened (c.f. AE10)<br />
2) People's (multi-dimensional) exchange rates, which includes their sympathy, since this is a particular case (e.g. preparedness to swap my time for your time, c.f. slide 6, AE9)<br />
<br />
One likely use of sympathy is to direct energy to those whom people care for the most, other things being equal. I'm not quite sure what you mean by "AE assumes that there are gains and losses". AE is a system of self-expression, and as such it assumes that people will wish to report +/- things about the experiences they have. Is this what you're saying? <br />
<br />
You wonder how "people more altruistic are rewarded". The system tracks evaluations of what others say about you, so if you have positive evaluations due to past altruism, strangers can detect this and may be more favourably inclined to help you. I am presuming that most people will choose to be generous to those who have been generous to others (particularly to them and those that they have sympathy for), though this is not forced, and other dynamics will likely present in other circumstances.<br />
<br />
There is no checking per se of the self-reported data (by design, since it's decentralised system). It is expected that people will make mistakes or deliberately attempt fraud. Hence the need for circumspection in how to interpret this data (a tension explored in http://www.altruists.org/ideas/economics/altruistic/measuring_value/self-evaluation/credibility/ ) - which should probably be calibrated. See also AE12, slide 14 for a loose end which I never tied up. <br />
<br />
I'm pleased that you see such commonality in our approaches. The key elements of AE are:<br />
1) Decentralised, self-evaluation of interactions<br />
2) Self-assigned, unlimited # currencies available<br />
3) The exchange rate structure, including<br />
a) sympathy<br />
b) resource processing<br />
<br />
I'm sorry not to have looked further at your work, and will do so as soon as I have moved computers, now that I am safely back in Bangladesh.<br />
<br />
Robin<br />
<br />
P.S. My CCC talk on Plutocracy went well, and is at http://www.youtube.com/watch?v=51qTb2O6eYM<br /></span>