There’s a softer non-technical side to this topic, which I think @zettawatt explores really well in the OP. The business / marketing / sales perspective of MaidSafe having something to sell is kinda cool. Opening the company up to building relationships via sales of NRS names is a pretty strong means of introduction. Maybe too strong and it becomes a burden? There’s definitely positives on the non-technical perspective.
But also there are negatives. Already MaidSafe must do a ‘controlled launch’ so they can secure the 15% safecoin needed. That’s not a big deal since it’s been there from very early onand is part of the expectations. But the more stuff they control (or even appear to control) the stronger the perception of it being ‘their’ network or an illegitimate network, and I am not in favor of that.
This will probably be an unpopular opinion and I’m adding it here not because I support it all that strongly but because it’s part of the mix and I feel is worth including.
Why not use something like vanity addresses instead of NRS? Mutable Data addresses are chosen by the uploader so they can simply be chosen to start with the desired characters. The trailing characters can be random or chosen specifically, whatever.
The root motive for this idea is that NRS is a net-negative to include from the start. Initially only xorurls should be available. This may be a little unwieldy but early users will pioneer. Has bitcoin suffered specifically because of the address format and there being no human-meaningful way to send funds? I think no. Has ethereum or namecoin specifically benefitted because of their NRS equivalents? I think no. Happy to be shown otherwise. I think NRS does the opposite of solve a problem, it creates one.
Excluding NRS solves the problem of squatting - there’s nothing to squat, no decision about burn or hold-for-later-sale or any of that.
Excluding NRS also solves the problem that punycode exists to resolve, eg visually similar characters and commonly confused characters, homoglyphs, wikipedia rabbit hole will show how deep this goes; as well as unicode normalization, banned characters such as zero-width-space, just so many issues once we depart from the ascii charset.
Getting off topic, I would rather see improved ergonomics of xorurls by competitive innovation rather than an NRS style system. I have no idea what that looks like but I’m happy to leave out NRS at the start of the network. I think it’s better for users in the long run.
Also slightly off topic and previously discussed in this topic, it’s tempting to treat NRS as an identity verification sort of system, and it is not that. It’s just for convenience. So users potentially accidentally treating it as convenience and verified identitication is to me a poor mix. It’s like accidentally mixing DNS and SSL (which do admittedly have close relations to each other) and NRS is perhaps perpetuating a potentially dubious mixing of concepts. Just potentially, not definitely, maybe a good debate for another topic, convenient identities and/or verified identities.
Overall I don’t really support NRS, I would prefer to see competition and innovation make xorurls more ergonomic (which is what NRS is aiming for but without the competition). I also feel that having NRS links scattered as links through immutable data webpages is potentially a disaster. IMO all links would be better as xorurls.
And lastly, @JimCollinson has once again expressed opinions I agree with. Really impressive insights.