[tor-relays] Unwarranted discrimination of relays with dynamic IP

[Rana]

@grarpamp

>Please see and contribute to the following...
>https://trac.torproject.org/projects/tor/wiki/doc/HardwarePerformanceCompendium

The Pi info there is indeed totally out of date. I opened an account on the wiki. However, after 10 (!)  tries to pass the totally unnecessary captcha which blocked my access AFTER I logged in, I have given up on trying to upload my data there. 

>If the source code and network technically permits any given node, it is valid for discussion.

Not only the network and code permit Pi-based relays from residential premises with ANY kind of Internet connection bandwidth, the texts on Tor page encourage people to run relays without telling them that their relays may be unwanted or useless if their connection is not fast enough. I have no firm data on this but my gut feeling is that the use of small residential relays can be optimized and made useful; and if it can't as some knowledgeable people on this forum seem to opine, then this info should be openly available for all and not just for the initiated or for people like me who spend the time to dig into the discussion on this forum for 3 weeks in order to find this out.  

>I've often suggested that all node selection and testing / ranking / node trust pki metrics / geoip / etc all be left as subscription style services and/or configurable parametrics for clients to >choose from or configure themselves. With some default "Tor Project" set shipped as fine for most users, in which Tor Project acts as one such supplier of such params.
>That leave only malacting nodes and 'net useful' nodes up to dirauths themselves. With 'useful' being no excuse to not make efforts to scale networks to the next level.

I could not agree more.