Zawinsky principle of social software, "... is single", "... is in a relationship", which (new?) verb?
Tantek Çelik
software (but maybe not by that name)
* "Social software" is about making it easy for people to do other
things that make them happy: meeting, communicating, and hooking up.
http://www.jwz.org/doc/groupware.html
In that regard, there seems to be quite a big hole in ActivityStreams
(or I've missed something obvious).
I searched the archives of this list, and wiki.activitystrea.ms and
found nothing that corresponds to the items that you see on services
like Facebook that say/show things like:
♥ Sally just became single.
♥ John is in a relationship.
etc.
For the relationship case, if a direct object is specified, e.g.
♥ John is in a relationship with Dana.
Then this can be represented by "Post"ing a "Status" with an embedded
XFN link surrounding "Dana".
However, my understanding of the justification given for introducing
new AS verbs is to provide an opportunity for AS consuming
applications to provide a different user interface / look+feel for
different activities, and potentially who/what they are done with/to,
thus this use case is likely deserving of a different verb than "post"
(or even "update").
Is there already an ActivityStreams answer for this?
None of the current verbs in the schema seem specific enough:
http://activitystrea.ms/head/activity-schema.html#anchor4
Nor does there appear to be any examples of how you do this on the wiki or list.
In addition, this has revealed two use cases that XFN doesn't handle,
1) updating relationship status to explicitly "single", and 2)
updating relationship status to "in a relationship" without a known
(or perhaps an undisclosed) direct-object. However, I'm in favor of
basing any microformat solution for this on an ActivityStreams
solution.
So, bottom line:
What is the (potentially new) AS verb/object for stating "... is
single", "... is in a relationship" ?
Tantek
--
http://tantek.com/
I made an HTML5 tutorial video/book! http://tantek.com/html5
Monica Keller
have only modeled one type of XFN relationship: friending
We definitely want to be able to model this with pure activity stream
concepts.
To me there are 2 major approaches
The first generic approach is to leverage the verb: Update which is
fairly new and define an Update Activity to have a field name and
value thus allowing us to model general profile changes like "Monica"
changed her last name to "Wilkinson", "Monica set spouse to Matt",
"Sally set her relationship status to Single"
Or we can model this specifically for relationships under the argument
that this are fairly significant
So we would then need some new verbs "date", "marry", "break up",
"reconcile"... ok im being funny but yes that would follow the current
model we have with friend and follow.
To me this latter approach is better suited for human consumption and
the first one is better suited for data synchronization.
We may need both
Christian Crumlish
So we would then need some new verbs "date", "marry", "break up",
"reconcile"... ok im being funny but yes that would follow the current
model we have with friend and follow.
--
Christian Crumlish
MY NEW BOOK: Designing Social Interfaces. http://designingsocialinterfaces.com
Get It. Read It. Love It. Review it. on Amazon: http://www.amazon.com/dp/0596154925/
Nathan
> On Fri, Jul 23, 2010 at 12:58 PM, Monica Keller <monica...@gmail.com>wrote:
>
>> So we would then need some new verbs "date", "marry", "break up",
>> "reconcile"... ok im being funny but yes that would follow the current
>> model we have with friend and follow.
>>
>
> hmm, isn't there an ontology problem in trying to map out the contours of
> human relationships, especially at a meta-level like this? is there some way
> to abstract a grammar for it so that different endpoints can share their
> possible relationship states and work within them?
wouldn't you need to use URIs for that rather than just string literal
verbs?
Christian Crumlish
wouldn't you need to use URIs for that rather than just string literal verbs?
Tantek Çelik
On Fri, Jul 23, 2010 at 1:06 PM, Christian Crumlish <xi...@pobox.com> wrote:
> On Fri, Jul 23, 2010 at 12:58 PM, Monica Keller <monica...@gmail.com>
> wrote:
>>
>> So we would then need some new verbs "date", "marry", "break up",
>> "reconcile"... ok im being funny but yes that would follow the current
>> model we have with friend and follow.
>
> hmm, isn't there an ontology problem in trying to map out the contours of
> human relationships, especially at a meta-level like this?
In short no.
These kinds of questions were brought up with the launch of XFN seven
years ago, but they were all theoretical. XFN is broadly implemented
web-wide today.
http://microformats.org/wiki/hcard-xfn-supporting-friends-lists
> is there some way
> to abstract a grammar for it so that different endpoints can share their
> possible relationship states and work within them?
Again I'd say no. "abstract" is the wrong approach.
Right approach: sample *actual* uses on sites like Facebook, and model
them or at least the 80% of such real world use cases. Anything more
can be left to academic research and paper writing based on the
results of our implementations.
Let's solve the simple cases simply. Nothing more, nothing less.
Hi Nathan,
On Fri, Jul 23, 2010 at 1:10 PM, Nathan <nat...@webr3.org> wrote:
>
> wouldn't you need to use URIs for that rather than just string literal
> verbs?
In AS, all verbs are URIs. See (RTM)
http://activitystrea.ms/head/activity-schema.html#anchor4
Tantek
--
Kevin Marks
--
You received this message because you are subscribed to the Google Groups "Activity Streams" group.
To post to this group, send email to activity...@googlegroups.com.
To unsubscribe from this group, send email to activity-strea...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/activity-streams?hl=en.
Christian Crumlish
Let's solve the simple cases simply. Nothing more, nothing less.
These kinds of questions were brought up with the launch of XFN seven
years ago, but they were all theoretical. XFN is broadly implemented
web-wide today.
http://microformats.org/wiki/hcard-xfn-supporting-friends-lists
Nathan
> On Fri, Jul 23, 2010 at 1:10 PM, Nathan <nat...@webr3.org> wrote:
>> wouldn't you need to use URIs for that rather than just string literal
>> verbs?
>
> In AS, all verbs are URIs. See (RTM)
>
> http://activitystrea.ms/head/activity-schema.html#anchor4
apologies, I was thinking of the 'Turning the schema into a keyword
registry rather than using URIs' thread and had assumed that this would
be the case moving forwards.
Best,
Nathan
Monica Keller
verbs and object types so we can use labels.
The nice thing is that this can be expressed easily with json-schema.
See the end of http://activitystrea.ms/json-schema/activity.json
"links": [
{"rel": "describedby", "href": "http://activitystrea.ms/json-
schema/verbs/{verb}.json"}
]
So if the verb says "make-friend" the default schema will be at
http://activitystrea.ms/json-schema/verbs/make-friend.json
Publishers who wish to extend can just provide their own "describedby"
property
fyi - I am working on the validator using Kris Zyp's dojo libs
http://bugs.dojotoolkit.org/browser/dojox/trunk/json?rev=17531
@Tantek yes we should model what is in the wild today. A lot of the
profile info changes have not been very granular so far
Tatu Saloranta
Not sure why this would be necessary -- URIs have no more explicit
semantic information than keywords. They just provide namespacing.
But there was a lengthy thread on this earlier for anyone who wants to
revisit arguments for and against ditching lengthy URIs in favor of
more concise key works (tokens, whatever).
-+ Tatu +-
Tantek Çelik
On Fri, Jul 23, 2010 at 1:28 PM, Christian Crumlish <xi...@pobox.com> wrote:
> On Fri, Jul 23, 2010 at 1:21 PM, Tantek Çelik <tan...@cs.stanford.edu>
> wrote:
>>
>> Let's solve the simple cases simply. Nothing more, nothing less.
>>
>
> I'll buy that (and the proposed "pave the cowpaths" approach).
Ok, let's do this.
https://wiki.activitystrea.ms/heart
> re:
>>
>> These kinds of questions were brought up with the launch of XFN seven
>>
>> years ago, but they were all theoretical. XFN is broadly implemented
>>
>> web-wide today.
>>
>> http://microformats.org/wiki/hcard-xfn-supporting-friends-lists
>
> I'm relatively familiar with XFN but wasn't sure how broad (or deep)
> adoption has been. Seems like there are some cute artifacts in XFN (like
> "muse") - don't they go beyond supporting the 80% cases?
They supported the 80% cases at the time (blog rolls and mentions of
people on blogs).
However this demonstrates that even *with*
design-by-empirical-research as a reality check, it is possible to
overdesign a schema etc.
Just another reminder to keep things ridiculously minimal simple and
then iterate.
When in doubt cut it out.
The counter-example is of course rel="me", which we decided to not put
into XFN 1.0, only to find out from folks that they wanted it, put it
in XFN 1.1, and now it's perhaps the most useful XFN value, or second
most used next to rel="contact". Though even there, little harm was
done by excluding it at first, and only adding it later when it was
more obvious that it was needed.
Hi Monica,
On Fri, Jul 23, 2010 at 1:38 PM, Monica Keller <monica...@gmail.com> wrote:
>
> @Tantek yes we should model what is in the wild today. A lot of the
> profile info changes have not been very granular so far
Right, and I think this one is particularly useful/important for
Activity Streams per the Zawinsky social software principle.
I've started a research page on the wiki:
https://wiki.activitystrea.ms/heart
Does anyone have screenshots of "is single" or "is in a relationship"
status posts?
@factoryjoe.com - do you have screenshots of this in your Flickr
compendium somewhere?
Please add more examples to the wiki, of what you've personally *seen*
on Facebook (not just "want"), and *especially* other sites too.
Does anyone know of any other sites that show relationship changes in
any significantly different way like Facebook does? MySpace? Windows
Live? Anybody?
Thanks,