Models Of Exposure

[FJ van Wingerde (UE, Vodafone)]

Hi all,

My name is FJ van Wingerde, I am a User Experience researcher inside
the Consumer Services group at Vodafone. I work in the offices at
Paddington, London, UK. I have designed and coded User Interfaces for
the last 15 years, focusing on mobile, and have been been online
socially on the Internet, starting in 1988 on Usenet.

Dan has reached out to me to take a look at OSW to see how the
technology fits the current landscape and can be used for both
research ideas floating around in Voda and pushing the state of social
networking beyond the current models OSW wants to connect.

Most of the needs we are seeing are in contrast to the Everyone In One
Group that Twitter and Facebook enable mostly. We are seeing needs in
the area of making families communicate as a unit better, or allowing
people to have multiple facets to their various groups of friends. As
OSW knits together the various systems people use for different
purposes, these needs become increasingly important.

I want to discuss some use-cases in this vein here and get your input
on them. To start off, I would like to share with you this
presentation from Google about how users socialize, and how badly the
current social media clients support their needs in this area:

http://www.slideshare.net/padday/the-real-life-social-network-v2

[Thomas Koch]

Hi,

see the explanation below about this excellent presentation. It was
enlightenmend for me and I would recommend it to everybody involved in
(distributed) social networking:
http://www.slideshare.net/padday/the-real-life-social-network-v2
(The presentation can be downloaded as PDF if you have a slideshare account.)

We should definitely discuss about the issues raised in the presentation. We
have the chance to not just copy twitter and facebook but to provide a better
user experience!

FJ van Wingerde (UE, Vodafone):

Thomas Koch, http://www.koch.ro

[Vamsi Sistla]

So, just to summarize the issues raised in this for everyone's benefit -

1. Users should be able to "create" different groups before they add "friends" or "follow" to their account - so that as and when you friend/follow someone, the User has an option to put them in each of the groups.
2. Groups should have different privacy and communication settings - ex - 1. Post this group only if its be explicitly identified, 2. Post to this group all common posts, etc, 
3. By default, each User account should have these groups to deal with "cold start" - "Family", "Close_Friends", "Colleagues", etc.
4. Every "post" by the User should have a target audience/group - everyone, family, close_friends, etc.
5. Each "post" by the User can be targeted to one or more groups. For example, post to Family and Close_Friends.

One of the issues with the current networks is lack of two way handshake between Users about privacy of "group" communication. For example, members of a User's "Family" group overlap with  members of "College_Friends" and if the members of one group respond to each other's posts that end up in front of members of other group - "world's collide" as George Costanza would say in Seinfeld :)
In the current networks, this is an issue with "like", "comment" and "share" posts.

If some works has already been done to address some of this, please let me know. If not, I will be happy to work with others to flush these out further,

Thank you,

@Vsistla

--
Vamsi Sistla

Twitter - @Vsistla

Quora - http://www.quora.com/vamsi-sistla

Blog - http://blog.sistla.com

Linkedin - http://www.linkedin.com/in/vsistla

[Simon Tennant (buddycloud)]

Facebook lets you group your friends.

I've yet to see any of my friends using this feature. I'm presuming it is becasue:

• maintaining a group is up-front work with little quick benefit
• posting to a particular circle of friends requires an extra step

I heard from a developer that OSW also has this feature and it's largely unused.

My view is that we are making this far more complicated than it needs to be. The future will look like many social networks provided by different groups.

• Your company server
• Your personal server
• A server run by your university
• A server run by your church or club
  
• A server run by...

You will notice that this fairly similar to email servers. You also have one email client that can log into multiple accounts. Expect future open social clients (at least buddycloud) to let you enter multiple accounts and interact in the same way you do with email:

• post from my work account to my work colleagues
• post from my personal account to my friends
• post from my university account to people on my course

This is a simple model that users already understand.

To me at least, my work social network has zero overlap with a personal social network. The accounts should be seperate and certainly in the case of work, maintained in a way that meets the corporate guidelines of whatever organisation.

S.

-- 
Simon Tennant
mobile: +49 17 8545 0880
office: +49 89 4209 55854      
office: +44 20 7043 6756 
xmpp: si...@buddycloud.com
build your own open and federated social network - http://open.buddycloud.com

[Antonio Tapiador del Dujo]

I agree with the multiple account/server approach. In fact, it is one of the
reasons I see for the federated social web. Many organizations would prefer to
host their own service and data.

However, I do think there are different types of contacts in all of those
escenarios. It is a recurring pattern. Examples are:

* In your company server: bosses, fellows
* In your personal server: family, friends, acquaintances
* School / University: teachers, fellows
* Club: members, followers

They are like classic roles but defined from the user point of view.

El Lunes, 28 de Marzo de 2011 21:53:46 Simon Tennant (buddycloud) escribió:
> Facebook lets you group your friends.
>
> I've yet to see any of my friends using this feature. I'm presuming it
> is becasue:
>

> * maintaining a group is up-front work with little quick benefit
> * posting to a particular circle of friends requires an extra step

>
> I heard from a developer that OSW also has this feature and it's largely
> unused.
>
> My view is that we are making this far more complicated than it needs to
> be. The future will look like many social networks provided by different
> groups.
>

> * Your company server
> * Your personal server
> * A server run by your university
> * A server run by your church or club
> * A server run by...

>
> You will notice that this fairly similar to email servers. You also have
> one email client that can log into multiple accounts. Expect future open
> social clients (at least buddycloud) to let you enter multiple accounts
> and interact in the same way you do with email:
>

> * post from my work account to my work colleagues
> * post from my personal account to my friends
> * post from my university account to people on my course

[FJ van Wingerde (UE, Vodafone)]

Well, one way to deal with these questions from a User Experience
point of view is creating a person that you think is representative of
your users, and writing up a scenario of how people would want to deal
with these questions. Between Dan and I we have come up with some of
these, augmented in our thinking by the questions our colleague Tracy
is finding in her research on group communications, and I would like
to share them with you as a thought experiment of how One Social Web
would fit.

These scenarios are based on the idea that a lot of these interactions
will be based on using mobile devices, because that is where we really
see social computing going, and on the idea that One Social Web
clients could be more valuable by not just connecting social networks
but supporting the ways people interact with groups as discussed in
the slideset. While the easy answer to these scenarios is to make a
new social network that does it all, the response that has a better
chance of uptake is making clients that can work over multiple social
networks and identities to give users the tools they need to manage
all this variety.

It could be that these scenarios are best supported by having
different servers, or very good group management in the clients, or
using other measures. These scenarios are just meant as an anchor to
the discussion to ground us together.

Let me know if you can't open this.

https://docs.google.com/document/d/1i26v07dUUYSBMqBUt3yD96dnq57dZPUKS5vwWu2Nmso/edit?hl=en&authkey=CILA8dUP

[Daniel E. Renfer]

While I agree that it is possible to create different accounts on
different servers to represent different aspects of ones social life,
that can't be the only way to acheive this goal. We'll always have the
people that have a list a mile long of all of their different usernames
that they can hide behind and take care that none of them can ever be
traced back to their real identity, but for the rest of us tend to make
our online identities represent us, as real people.

I have a handful of email addresses: One from my work, an old one from
school, a couple gmail accounts, maybe an old Yahoo account, one or two
accounts for every domain I own, but then I have my one, primary
identity. I've chosen that identity to represent myself whenever
possible. When a site asks for a email address for me that it'll use as
a public identifier for me, I use that one. If you Google that email
address, you'll find me and links to the other services I use.

I hate having to create new accounts, and I know I'm not the only one.
This is the very reason behind technologies such as OpenId and WebId.
The goal is to allow people to create as many or as few identifiers as
they want, and to use those identifiers with as many services as they want.

Now, even if that number is 1, we still need the ability to control what
activities we perform are visible to whom. I don't care if people know
who I am, but might not want everybody to know what I say.

I, personally, try to live pretty much as an open book when it comes to
"friend locking" posts, but I know there are plenty of others (I asked
my wife about her opinions on this) that consider the ability to choose
which of their friends can see a post.

tl;dr - I think it's better to have the ability to friend lock a post
and create accounts as the situations require than to be required to
create different accounts in order to separate social contexts

Daniel E. Renfer (look for du...@kronkltd.net and you'll find me)

[Jorge Araluce]

Hello 

I am an entrepreneur working on an project to give the Persona the ability to take control of their activities, contacts and presence across the multiple social systems they use .

Talking from the use-case scenario , make sense to have a user-centric approach instead of the current service centric; what I mean is instead of having to go to Facebook to see what their facebook friends are doing, and then to twitter etc.. , have an approach where the user define a context or dimension that makes sense for him and include in it the "pieces" that relate to it ( a few friends from here and there , maybe an Rss feed o whatever other services) . That would allow to have a more fine grained control of what is relevant to the user and also avoid the "social shyness " of broadcast an updated to everybody, as one could taylor the message to what is relevant to that specific social dimension.

I am sure there must be a few ways to achieve this, one being a kind of mashup a la Tweetdeck but my feeling is that if is done at the base at a protocol level, that will enrich the possibilities and will keep the control on the user side vs an arbitrary decision from the service side ( like the twitter client warning from a few week ago ..) Can you tell me if i may be mixing things here?  

2011/3/31 Daniel E. Renfer <duck1...@gmail.com>

--
Jorge Araluce
www.UtilidadesdeRed.com
http://twitter.com/jaraluce
http://es.linkedin.com/in/jaraluce
skype; jaraluce75

[imaginator]

On Thursday, March 31, 2011 4:04:19 AM UTC+2, Daniel E. Renfer wrote:

While I agree that it is possible to create different accounts on
different servers to represent different aspects of ones social life,
that can't be the only way to acheive this goal. We'll always have the
people that have a list a mile long of all of their different usernames
that they can hide behind and take care that none of them can ever be
traced back to their real identity, but for the rest of us tend to make
our online identities represent us, as real people.

I am quite happy to have my work identity separate from my other identity. My company is welcome to run a channel server. And if they pay me, I'll use it from 9-5pm.

I am in difficult region of the world, quite happy to "hide behind another identity"

At the end of the day there are many organisations that offer social services. Expecting them to all come from one provider is just not realistic. I'm happy to work from the point of distrusting most of them rather than bundling my lot in with one.  

But, expecting them to federate with one-another is key.

I hate having to create new accounts, and I know I'm not the only one.
This is the very reason behind technologies such as OpenId and WebId.
The goal is to allow people to create as many or as few identifiers as
they want, and to use those identifiers with as many services as they want.

Now, even if that number is 1, we still need the ability to control what
activities we perform are visible to whom. I don't care if people know
who I am, but might not want everybody to know what I say.

Technology will never be able to control your message going to 100 trusted people and none of them leaking it. That's a social problem that no amount of PGP message signing contortions will ever solve. 

I, personally, try to live pretty much as an open book when it comes to
"friend locking" posts, but I know there are plenty of others (I asked
my wife about her opinions on this) that consider the ability to choose
which of their friends can see a post. 

tl;dr - I think it's better to have the ability to friend lock a post
and create accounts as the situations require than to be required to
create different accounts in order to separate social contexts

That would be nice... but try persuading your company's IT staff that they should be using an identity outside of their firewall, their control and the logging and compliance requirements imposed on them from outside (SOX for one) and you will be fighting a steep uphill battle.

S.