Saku Ytti wrote: >> That is a fatal defect of SRVINFO (and URI), because, if a domain >> support 100 services with 2 transport protocols, there are 200 >> RRs, all of which must be replied for a single query, which is >> unacceptably inefficient, which is why it is necessary to use a >> service specific domain name for each service. > > Can and must are different things. Just because you can, does not mean > you should, must less that you must. Saying proposal is bad, because > you can do stupid things with it, is bizarre. That's why I didn't assume 100 transport protocols. But, it is stupid to forbid "https://example.com" and force using "https://www.example.com", which means, in a long run, a domain will support a lot of application protocols. Masataka Ohta I can have arbitrary > many A records for a name, does this mean A RR is bad as well? How is > SRV stopping you from creating stupidly large responses? > >