Talk:LSL Protocol/Restrained Love Relay/Comparison

Before we list 50 different relays and all their derivatives: on what basis should we decide to list a relay in this page?

I am not too sure about listing my "Damn Fast Relay"... it is more something I made for fun "Let's script a relay in 5 minutes and less than 50 lines of LSL." than anything else... although its lightweight approach and prioritization of performance before any other criterion are interesting.

All the other listed relays (except Dahlia's) meet the criterion of being very popular. My own inquiries show they actually are the top 4 and cover more than 80% of the "market". Other relays have less than 3% of market share... and most often than not are derivatives of either Maike's or my implementation.

I added Dahlia's because it is one of the most feature complete relays and also another brand new implementation from scratch of RLVR and ORG standards. This can be useful to test the robustness of the specifications. For a short description, Dahlia's is very similar to mine, feature-wise and UI-wise, but it uses the multi-prim approach (like TkPBA).

What other implementations should be mentioned? Relevant criteria on the positive side could be:

• popularity
• interesting new concept
• particularly good coverage of ORG
• new implementation derived on no other
• or if it is based on some other implementation, it has to have significant differences (for instance: OpenCollar's w.r.t. my MultiRelay: OpenCollar's focuses on collar integration and stability, whereas my HUD MultiRelay focuses on features).

On the negative side:

• old implementations that are not supported by their authors anymore

Other ideas?

--Satomi Ahn 09:01, 21 October 2010 (UTC)

The initial list was "relays Winter has in her inventory at the moment". =p I was hesitant to include OpenCollar's but I felt it was different enough, while being very widely used, to warrant an entry. My feeling is that a relay that meets any of the criteria would be good to include. In that light, the Damn Fast relay would fall under the "new implementation" category. The main thing to learn from it, I think, is that optimizing for scans can be valuable. (And also, that a minimal relay can be very very small.)

On a related note, I think it'd be useful to have a survey of relay use. I'll likely start that as another page.

Winter Seale 10:37, 21 October 2010 (UTC)