Re: RFC: tent.io (protocol for social networking)

On 24 September 2012 17:35, Michiel de Jong <michiel@unhosted.org> wrote:

> On Mon, Sep 24, 2012 at 5:01 PM, Daniel Harris <daniel@kendra.org.uk>
> wrote:
> > I feel our model has to cater for multiple namespaces for contacts. Can
> it? Can we create an meta namespaces-for-contacts wrapper?
>
> yes, this is exactly what i mean when i say one namespace. i mean one
> 'meta wrapper'. useraddress.net aims to be a search tool that can
> support that meta wrapper. it supports webfinger/OStatus as well as
> facebook, twitter, diaspora, friendica, and i want it to support
> xmpp/buddycloud too.
>

Your interop possibilities are only as strong as your identifiers.

Consider a telephone network in berlin.  Everyone has a phone and everyone
in berlin can call everyone else.  Everyone is happy.  A few people in
Hamburg get a telephones too, they are happy.  Now someone from Hamburg
wants to call someone in Berlin and says, "what's your area code?".  In
Berlin they say, "we dont need an area code, our system works just fine.
Come to Berlin and join us".  The Hamburg person says, "I want to stay in
Hamburg, cant we just all have an area code".   After much discussion every
german district gets an area code and you can get german phone calls.   Now
what happens when someone from the UK wants to call someone from Germany.
You need the whole reworking of identifiers again.  Identifiers are the key
to interop.  It's the same on the social web and FSW.  Everyone thinks
their system works fine, but silos will *never* scale beyond themselves,
interop is impossible.  You only will find this out once you try and "call"
another system and it fails.

Case in point.  I think useraddress.net is an awesome initiative.  But it's
got the HTTP URIs wrong for both me and tim, which gives incorrect
semantics, incorrect comparison and incorrect interop.  And yes I have
raised the issue on github.  How long will it take to fix, I dont know ...
perhaps never.  It's small details like this that prevent intercomm.
Historically getting these little details right, for whatever reason, in
the FSW takes months and years, rather than hours and days.  Any plan to
speed things up would be welcome!


>
> > And do it quickly and quietly before someone else creates an
> incompatible meta namespaces-for-contacts wrapper? And maybe that won't
> matter actually as a wrapper should be able to wrap another wrapper, right?
>
> correct, once we start wrapping, the more people work on wrappers the
> better. we'll wrap it all into one big ball where more and more things
> get included with each wrap. :) i mean, there should not be one
> wrapper competing with any other wrapper, all wrappers should include
> each other wrapper and vice versa. then we get one network, and we all
> win.
>
> As we already said on the phone, i personally will be working on this,
> and will report my results and contributions to this mailing list.
>
> I don't need any money myself, but i'll be more than happy to
> coordinate my efforts with what other people are working on so we get
> more momentum on this topic, with multiple full-time and/or part-time
> people working specifically on tackling the fedsocweb problem in a
> sort of a 'team work' way if you like. Maybe even splitting it up like
> "you do this, you do that, we reconvene and discuss the progress in
> six weeks". Indeed SWAT0 was a good start for establishing such a
> coordination, but we may need some new Plan now.
>
> That way i'm sure we can make some good progress over the coming months.
>
>