Another parallel effort?

[unhosted.N...@ordinaryamerican.net]

http://myprofile-project.org/
"MyProfile intends to provide a solution for managing the numerous
accounts and profiles that users have on the Internet. Its main
purpose is to provide a unified user account, or simply ‘user
profile’, which as opposed to current ‘silo’ profiles, would really be
under the user’s control, on a device controlled by the user."

[Melvin Carvalho]

One of my favourite projects anywhere right now. 

The author (Andrei Sambra) has been invited to spend some time working with Tim Berners-Lee at MIT, starting from November.  So expect it to get even better!
 

--

[Nick Jennings]

very cool

> --
>
>
>

[Michiel de Jong]

ha, nice to see they reused our CC-licensed website design. :)

iiuc, this would for me be 'yet another silo', and this extra online
identity would not replace any of the existing ones. it would not
allow me to Tweet without logging into twitter, to send email without
logging into gmail, or to tag myself in a photo without logging into
facebook.

so it doesn't really solve a problem i have, but i do think this is
important research into "what if we could somehow force everybody to
switch to this" (as an academic exercise).

> --
>
>
>

[Melvin Carvalho]

On 19 October 2012 13:49, Michiel de Jong <mic...@unhosted.org> wrote:

ha, nice to see they reused our CC-licensed website design. :)

iiuc, this would for me be 'yet another silo', and this extra online
identity would not replace any of the existing ones. it would not
allow me to Tweet without logging into twitter, to send email without
logging into gmail, or to tag myself in a photo without logging into
facebook.

so it doesn't really solve a problem i have, but i do think this is
important research into "what if we could somehow force everybody to
switch to this" (as an academic exercise).

The reason to like this project is that it's NOT a silo :)

Imagine the polyglot approach to identity of useraddress.net ... this is what my profile is doing ... breaking down the walls of identity and bringing all systems together.  Many other systems make this claim, but when you look at the details there's often one restriction or other that can eventually block you.

You very correctly point out that you cant tweet without logging in to twitter, but that's more about twitter than anything else.
 

On Fri, Oct 19, 2012 at 2:28 AM, Nick Jennings <ni...@silverbucket.net> wrote:
> very cool
>
> On Thu, Oct 18, 2012 at 9:29 PM, Melvin Carvalho
> <melvinc...@gmail.com> wrote:
>>
>>
>> On 18 October 2012 23:14, <unhosted.N...@ordinaryamerican.net>
>> wrote:
>>>
>>> http://myprofile-project.org/
>>> "MyProfile intends to provide a solution for managing the numerous
>>> accounts and profiles that users have on the Internet. Its main
>>> purpose is to provide a unified user account, or simply ‘user
>>> profile’, which as opposed to current ‘silo’ profiles, would really be
>>> under the user’s control, on a device controlled by the user."
>>
>>
>> One of my favourite projects anywhere right now.
>>
>> The author (Andrei Sambra) has been invited to spend some time working with
>> Tim Berners-Lee at MIT, starting from November.  So expect it to get even
>> better!
>>
>>>
>>>
>>> --
>>>
>>>
>>>
>>
>> --
>>
>>
>>
>
> --
>
>
>

--

[Fabio Barone]

I also agree with Melvin that this to me
actually looks like a
game changer!

We can further the idea of identity to encompass
organizations and projects in this way as well.

We then can do complex matching due to
RDF data, matching people, organizations, projects...

Something which has been tried very much in the latest
years but could not possibly happen because of the
"silo" problem.

2012/10/19 Melvin Carvalho <melvinc...@gmail.com>

--
 
 
 

[☮ elf Pavlik ☮]

Excerpts from Michiel de Jong's message of 2012-10-19 11:49:11 +0000:

> iiuc, this would for me be 'yet another silo', and this extra online
> identity would not replace any of the existing ones. it would not
> allow me to Tweet without logging into twitter, to send email without
> logging into gmail, or to tag myself in a photo without logging into
> facebook.

i find this statement rather confusing... if you want to post to your microblog, send and email or tag a photo, WebID which MyProfile uses can contribute to at some point making it all a very smooth experience.

but if you want to use twitter service, gmails service and facebook service you just need to depend on mercy of those very particular service providers...

once more, i really don't get your point here?

[Michiel de Jong]

right sorry, let me elaborate.

there are several problems with web2.0 as it is now. i would define
'the silo problem' as the sum of password silos, content silos, and
profile silos:

- password silos: there is no single sign-on (SSO) solution that is
both widely accepted and federated in the sense that anyone who
controls a DNS domain can be an identity provider. There are federated
SSO solutions like Mozilla Persona, OpenID and WebID, but they are not
widely accepted, and there are widely accepted ones like facebook
connect and (to a lesser extent) twitter connect, but the ideal of one
account that works across the majority of existing web2.0 services is
still a dream, at most. this means that if we want SSO (which
MyProfile states as a central goal iiuc), then we should choose one
(for instance WebID), and stop using the services that don't support
WebID.

- content silos: content we want to read, and content we want to
publish, stays inside whichever silo we post it into. This is a
problem, because that means if you choose MyProfile as your platform,
and your friend chooses Twitter, then you won't see each other's
tweets. So even if it's possible for one person to switch from one
silo to another, there is currently no federation of content delivery
across silos. this is of course a much bigger problem than federation
of login/identity.

- profile silos: there is no cross-silo user search. this means that
even where content federation is possible, addressing people in other
silos is cumbersome because the addressbooks usually only contain
contacts that are 'local' to the silo. for instance: facebook exposes
smtp and xmpp interfaces for messages and chat respectively, but when
you search for a person in facebook, only facebook profiles will show
up in the search results. for federated social networks like
StatusNet, Diaspora and MyProfile, the same is true. It seem MyProfile
seems at first glance to have slightly better characteristics than
StatusNet and Diaspora in this sense: it seems you can do free text
search for anybody who has a WebID. Still, it will not let you find
people with a Buddycloud ID or a Friendica ID. Now you can say
'MyProfile will be perfect as soon as all the non-believers will stop
using these other platforms', but i guess it's clear that that's not a
very realistic position.

so imho projects like MyProfile, StatusNet, Diaspora, Friendica,
Buddycloud, app.net, tent, secushare, and 40 others, solve none of
these three problems /unless/ we all somehow pick one and switch to
it. There has been some work in trying to agree on an 'esperanto'
language that all software providers can agree on, the most notable
candidates were OStatus and xmpp. Apparently MyProfile supports
neither OStatus nor xmpp, and neither do many other software
providers. So that's why i said that to me, my friends on buddycloud
are as much a silo as my friends on MyProfile and my friends on
Diaspora. They are 'better' silos because they're internally
decentralized, but they still are platforms into which content gets
locked in.

But all hope is not lost, there is another approach to this which is
what i call the 'polyglot approach': we're currently working on a
server that can speak many protocols, thus being able to send content
into all these silos, as well as extracting it out of them. It
consists of two parts: user search across all silos, and a little
'polyglot server' which a user runs on their DNS domain and allows
them to interact with people in multiple silos.

MyProfile has some nice properties that some of the other silos don't
have, and if there is enough traffic on there to make it worthwhile
then we'll definitely add its protocols to the polyglot server and
make sure user search works for the foaf documents MyProfile exposes.


Sorry if it sounds pessimistic to call MyProfile a 'silo', but unless
we globally decide to all switch to it, for me it would really be 'yet
another place to check'. so this is why i personally gave up on the
'esperanto approach' altogether, and i no longer believe there will be
1 project that will be the solution. i now embrace the fact that there
will always be 30 or 40 projects like MyProfile, and most of them will
not even try to be compatible with any of the others. That's why i'm
now working on establishing 'polyglot' bridges between all such
projects.

i hope this explains it a little bit! :)

cheers,
Michiel

[Jan-Christoph Borchardt]

Yep, I completely agree.

Any new federated network doesn’t solve a problem if it doesn’t
federate with the other existing solutions. It just creates yet
another thing where you have to create an account, add friends and
cross-post. We don’t need to develop another »network«, we rather need
to find a way to connect the existing ones.

> --
>
>
>

[Andrei Vlad S]

Hey! 

I'm the person responsible for MyProfile, and I would like to clarify a couple of issues, if you'll allow me to. :)

This project started as an idea I had for my PhD, so it moves a bit slowly due to this fact. The main goal of the project is to allow users to host their own content on a machine they control, both in terms of access control as well as _physical_ control (think FreedomBox). But it's not only content you are used to put on social web applications, such as photos, videos, links, etc. It's also about our communication, may that be instant messages, liking stuff on other websites, actually pretty much all interaction we have online. Having said that, overall it's more of an identity platform than yet another decentralized social network. I've tried to explain it a bit in the online manifesto, though it may require some updating.

There are two important aspects of MyProfile. First, it uses the semantic web (linked data) to provide interoperability. This is a huge gamechanger, as it will allow applications to basically "speak" the same language. The LDP group at W3C (http://www.w3.org/2012/ldp/wiki/Main_Page) will help a lot with that, and it is one of the reasons I'm holding back on implementing certain features until a final specification draft is available. The second important aspect is that MyProfile uses the WebID protocol (http://webid.info) for authentication and identity. This protocol cures the headache of having to use different forms of authentication, yet remaining very simple and most of all, secure!

In the end, I have to agree with Michiel on the fact that since people prefer app X over Y, and X and Y don't speak the same language, then we still have a problem. The semantic web hopes to resolve this problem, so I hope that in the near future we will see more applications supporting W3C standards and being able to understand each other.

Feel free to contact me with any questions you may have. :)

Andrei

[☮ elf Pavlik ☮]

Excerpts from Michiel de Jong's message of 2012-10-21 17:03:51 +0000:

> right sorry, let me elaborate.
>
> there are several problems with web2.0 as it is now. i would define
> 'the silo problem' as the sum of password silos, content silos, and
> profile silos:
>
> - password silos: there is no single sign-on (SSO) solution that is
> both widely accepted and federated in the sense that anyone who
> controls a DNS domain can be an identity provider. There are federated
> SSO solutions like Mozilla Persona, OpenID and WebID, but they are not
> widely accepted, and there are widely accepted ones like facebook
> connect and (to a lesser extent) twitter connect, but the ideal of one
> account that works across the majority of existing web2.0 services is
> still a dream, at most. this means that if we want SSO (which
> MyProfile states as a central goal iiuc), then we should choose one
> (for instance WebID), and stop using the services that don't support
> WebID.

Myself I would like to support 3 OPEN authentication systems your mentioned, Mozilla Persona, OpenID Connect and WebID. Leaving to people choice over which one they PREFER to use and at the same time simply not support Facebook, Twitter and other PROVIDER SPECIFIC ones. In terms of aligning notion of identity among those 3 open standards, WebID can also support Webfinger identifiers (but currently not required) which would bring it closer with OpenID Connect, when it comes to Mozilla Persona since it uses email (like?) identifiers could also play nicely with Webfinger...

Still I see topics of AUTHENTICATION and IDENTITY having certain distinct challenges!

>
> - content silos: content we want to read, and content we want to
> publish, stays inside whichever silo we post it into. This is a
> problem, because that means if you choose MyProfile as your platform,
> and your friend chooses Twitter, then you won't see each other's
> tweets. So even if it's possible for one person to switch from one
> silo to another, there is currently no federation of content delivery
> across silos. this is of course a much bigger problem than federation
> of login/identity.

Here I see no problem comparing MyProfile, Diaspora and Buddycloud since one can host any/all of them on domain one controls. But Twitter, Facebook etc. don't fit here since they don't leave such FREEDOM!

Myself when it comes to content I like to look for approach where I publish everything to my Personal Data Space of some sort, and then this content gets syndicated, embedded in other contexts. In such approach Linked Data seams very handy so without getting familiar with implementation details of MyProfile I guess it can offer some solid foundation here....

>
> - profile silos: there is no cross-silo user search. this means that
> even where content federation is possible, addressing people in other
> silos is cumbersome because the addressbooks usually only contain
> contacts that are 'local' to the silo. for instance: facebook exposes
> smtp and xmpp interfaces for messages and chat respectively, but when
> you search for a person in facebook, only facebook profiles will show
> up in the search results. for federated social networks like
> StatusNet, Diaspora and MyProfile, the same is true. It seem MyProfile
> seems at first glance to have slightly better characteristics than
> StatusNet and Diaspora in this sense: it seems you can do free text
> search for anybody who has a WebID. Still, it will not let you find
> people with a Buddycloud ID or a Friendica ID. Now you can say
> 'MyProfile will be perfect as soon as all the non-believers will stop
> using these other platforms', but i guess it's clear that that's not a
> very realistic position.

Again in my opinion you use term *silos* in inconsistent way. I find issues with Facebook quite different then with federated 'self-hosted' systems. One IMO huge difference, we can update our 'self-hosted' systems to newer versions with improved cross-system interoperability. While SERVICE offered by Facebook doesn't include such freedom!

>
> so imho projects like MyProfile, StatusNet, Diaspora, Friendica,
> Buddycloud, app.net, tent, secushare, and 40 others, solve none of
> these three problems /unless/ we all somehow pick one and switch to
> it. There has been some work in trying to agree on an 'esperanto'
> language that all software providers can agree on, the most notable
> candidates were OStatus and xmpp. Apparently MyProfile supports
> neither OStatus nor xmpp, and neither do many other software
> providers. So that's why i said that to me, my friends on buddycloud
> are as much a silo as my friends on MyProfile and my friends on
> Diaspora. They are 'better' silos because they're internally
> decentralized, but they still are platforms into which content gets
> locked in.
>
> But all hope is not lost, there is another approach to this which is
> what i call the 'polyglot approach': we're currently working on a
> server that can speak many protocols, thus being able to send content
> into all these silos, as well as extracting it out of them. It
> consists of two parts: user search across all silos, and a little
> 'polyglot server' which a user runs on their DNS domain and allows
> them to interact with people in multiple silos.

You may remember me showing you diagram at: https://buddycloud.org which includes (Facebook and Twitter under 'Legacy Gateway') For FSW summit in 2011 they even prepared paper: Bridging Federated Protocol Stacks [1]

I would also like to submit right away a pull request to your polyglot server with this adjustment in it's continuous integration environment :)
echo '127.0.0.1 useraddress.net' >> /etc/hosts

>
> MyProfile has some nice properties that some of the other silos don't
> have, and if there is enough traffic on there to make it worthwhile
> then we'll definitely add its protocols to the polyglot server and
> make sure user search works for the foaf documents MyProfile exposes.
>
>
> Sorry if it sounds pessimistic to call MyProfile a 'silo', but unless
> we globally decide to all switch to it, for me it would really be 'yet
> another place to check'. so this is why i personally gave up on the
> 'esperanto approach' altogether, and i no longer believe there will be
> 1 project that will be the solution. i now embrace the fact that there
> will always be 30 or 40 projects like MyProfile, and most of them will
> not even try to be compatible with any of the others. That's why i'm
> now working on establishing 'polyglot' bridges between all such
> projects.

Polyglot approach sounds great, as long as you can design it without introducing dependency on some central point like useraddress.net!
And once more I wouldn't toss services like Facebook/Twitter/etc. in the same bucket with systems like Buddycloud/MyProfile/Diaspora etc. which one can host on domain under one's own control!!!

>
> i hope this explains it a little bit! :)
>
> cheers,
> Michiel

Thanks for taking your time to share more on those topic,
Cheers!

[1] http://d-cent.org/fsw2011/wp-content/uploads/fsw2011-Bridging-Federated-Protocol-Stacks.pdf

[unhosted.N...@ordinaryamerican.net]

On Sun, Oct 21, 2012 at 1:09 PM, Andrei Vlad S - andrei...@gmail.com wrote:
<snip>

> There are two important aspects of MyProfile. First, it uses the semantic
> web (linked data) to provide interoperability. This is a huge game changer,
> as it will allow applications to basically "speak" the same language.

The key word here is "allow". Other communities centered around other
data formats, for example JSON, consider the W3C offerings
unnecessarily verbose. Personally, I think they don't, yet,
understand the complete problem. Verbosity is a cost generated by a
need they don't recognize.

> The LDP group at W3C (http://www.w3.org/2012/ldp/wiki/Main_Page) will help a lot with that, and it
> is one of the reasons I'm holding back on implementing certain features until a final specification
> draft is available. The second important aspect is that MyProfile uses the WebID protocol
> (http://webid.info) for authentication and identity.

Thank you for cooperating with the efforts of others. Until all the
Federated net approaches recognize the need for cooperation, they
can't offer a comprehensive solution; their programmer-hours are
consumed by duplicate effort.

<snip>

[Melvin Carvalho]

+1000
 

<snip>

--

[Paul Frazee]

I have a very specific take on this.

I think silos are created in the Web by the security model of document sandboxes.

It's not possible to share a connective session between two applications without involving remote hosts. All traffic has to go out to a server to reach other tabs, and it only happens if the two hosts have channels to each other. As a result, the functional connectivity of web applications is determined by the providing services.

Put another way, I can't tell the the facebook interface to download all of my posts and hand them over to my google docs interface. I might tell the google docs server to use a protocol like OAuth to get permission for a remote session, so the docs server could download the posts from the facebook server-- but that scenario is only happening if both parties implement it and agree to it, which is cumbersome. So as of right now, even if the facebook interface can easily grab a list of my posts, they're staying in the interface.

And it's actually a little worse than that. Even with a perfectly connective world of hosts, where every service has a channel to the other service you need, the arrangement is difficult to trust. Exactly what am I granting permission to by connecting services? How could I know? It's only vaguely apparent in the permissions interfaces, and the decisions are rarely case-by-case. "Will read all your mail and post to your feed (and communicate with its third-party host)." Great, I really want to roll the dice on that.

Again, this is all because there are no permeable barriers between browser tabs for information to pass through, nor are there the tools to easily configure and safely mediate that barrier if it existed. And since we rely on domain-based security, there aren't any good tools for mediating remote connections either.

So how do we solve it? I think we need to update the browser environment to support more fine-grained connectivity and security policies, which is the purpose of LinkAP. Once you get a sandbox that can inter-communicate with other sandboxes, you're off to the races-- and that's what web workers give you. After that, it's a matter of setting up a strong protocol for routing messages, configuring endpoints, and enforcing security.

There are probably a lot of ways you can do it, but the choice I made was to use HTTP so that local applications can behave as invisible proxies to remotes. That allows programs to be gatekeepers to their services, usually with some permissioning tools from the environment. HTTP also allows links/forms/ajax to work as before on the in-browser apps. There are a lot more details (some still developing) but I'll leave that for another time.

So while protocols are a key part of interconnectivity, I don't think they're the most pressing need right now. I think it's more important to lower the barrier between client-side applications, so they can become more modular, single-purpose and cooperative.

prf