I'd like to add to the open discussion the question whether or not there should be a technical limit to the number of DT1-votes any DT1-member gets. For example: Is it really desirable that a DT1-member without any support from other DT1-members gets to include or exclude many other DT1-members? Or should there be a limit on the "voting power" within DT1 (possibly depending on the number of DT1-inclusions the user has)?
I know you've suggested before that DT2 inclusion should require at least two DT1 votes, and I tend to agree with this strategy. It would mitigate the TrustSelfScratching abuse that goes on and prevent many shady users from getting voted to DT2. I would suggest to add the same restriction for DT1 inclusion as well. Along with the merit-based votes, two DT1 inclusions would also be required to qualify for DT1. The system would still be decentralized, but a little more restrictive.