[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Social-discuss] Interoperability?

From: Nathan
Subject: Re: [Social-discuss] Interoperability?
Date: Fri, 28 May 2010 10:07:17 +0100
User-agent: Thunderbird (Windows/20100228)

I'll second both Mischa and Dan's comments,

We already have more than enough protocols to do everything we need; caveat some of them may need used together in some tasty ways, but the whole world and their dog is working on that right now.

So far the highest value I've seen come out of this particular group are the ideas and use-cases - this work is valuable to all (and places social-discuss at the center of what's going on).

thus (imho):
1 - support & make sure the many protocols that are out there work
2 - focus on use cases & requirements + being a central 'log' of all the efforts that are going on.



Mischa Tuffield wrote:
Hello All,
Following on from my rant last night, and danbri's email this morning (/me bows to the awesomeness 
that is danbri), I should clarify my use of the term protocol. In my ideal world, the 
"protocol" that I mentioned would be built on top of existing protocols and standards, as 
in the ones mentioned in the email below, (RSS/ATOM, OpenID, OAuth, FOAF, XMPP, FOAF+SSL, RDF, and 
so on), I was NOT implying that this group should try and think up their own "protocol" - 
and if it can access my bad.

From my POV if this group could identity a number of interactions which would/should be supported by a distributed social network and a standards way of communicating things like : "a friend request", "tagging a photo", and so on and as stated in the email below perhaps build libraries to support this, then people can choose to implement their part of this future distributed social network as they wish (i.e. could use whatever language, whatever data format the developers wished).
Would be awesome if we could spec out the use cases and requirements for this future we all want to 
live in, spec'ing out the utility use cases and the privacy use cases and then coming to a happy 
medium at the intersection of the two. I say "utility" and "privacy" use cases 
'cause I want my future social network tool to do stuff, and well I want it to privacy aware. FWIW, 
I would love to contribute to the writing of such UCRs.

Post development of such use cases, and the identification of existing 
standards/protocols best fitted to make them work, even if (jumping the guns 
here) it turns out that and RDF based approach for making nodes talk to each 
other (friend requests etc) makes the most sense, libraries from GNUSocial (or 
someone else for that matter) would mean that you could choose to implement 
your node on the distributed social network any way you wish (i.e. sans RDF for 


On 28 May 2010, at 08:22, Dan Brickley wrote:

Excuse the vanity, but I'm resending this, to 
see if it is now closer to the project's direction than when it was written.

  "What I would value most from GNU is an effort to make sure the supporting 
software libraries for standards-based social Web interop are solid, tested and up to 
date, and that they are integrated throughout the GNU collection of software packages and 
the wider software scene. Why doesn't Mailman do oauth or openid? Why even today do my 
attempts to use my l 
Wordpress's openid provision to log into my locally installed MediaWiki often fail? GNU's 
reputation is in worldclass free software; I'd suggest sticking close to that and 
focussing on asking ourselves what we can do to take this massive network of free 
software installations, and integrate them to improve people's social experience of the 
Web.  "

"We already have 'social networks' scattered across the entire Internet/Web, and this is as it should be. The challenge isn't to move them all to one giant replacement service or network, but to patch them together, the way the Internet itself was patched together from its constituent ancestors. Take IRC for example: the popular Freenode IRC network is apparently powered by GNU software including its ircd, ... ... now thousands of people happily use IRC daily to socialise, share and communicate. Thousands of others use GNU Mailman to do similar in email. Let's not get distracted by the impossible dream of cloning the facebook experience in a 'free' way, when we have real vibrant online communities already, thanks to GNU software. I look to the GNU Social initiative not to add just another software package to the mix, but to take a lead - by workshops, evangelism, free beers, code reviews, whatever
it takes - in getting more integration and standards support across the existing 
suite of GNU social software. Why can't we better integrate the IRC community on 
Freenode with the network of mailman installations out there? Work on XMPP support 
in ircd to modernise the underlying standards, or integrate IRC's notion of user 
identity (nickserv) with that of the Web? "

I wrote that back in March, and at the time it seemed that GNU Social was 
heading strongly towards making yet another PHP-based Web thingy for people to 
hang out in, rather than attempting the 'bigger picture' of global integration 
that I was hoping for.

At the time I tried arguing that GNU's strength isn't that; there are already 
thousands upon thousands of lines of 'social' code out there in daily use. 
What's needed now is integration, coherence and usability. I believe this can 
be built by making careful use of existing protocols and standards (RSS/Atom, 
Activity Streams, OpenID/Oauth, FOAF, XMPP, OpenSocial, RDFa/Microformats), 
OStatus, GPG ...

It sounds like this project is now moving more in that direction, although talk of 
"a protocol" worries me. When we're talking on this scale (the entire Web) it's 
probably worth stepping back and figuring out what we're trying to write a protocol to 
do, before leaping in and doing it. I hope the first drafts to come out of the inner 
circle attempt to set some kind of scope first, rather than jumping straight in with 
protocol design for an undocumented problem.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]