Syndication or import/export?
1 view
Skip to first unread message
Ben Werdmuller
Apr 30, 2008, 12:39:06 PM4/30/08
to Open Data Definition
I was explaining ODD to someone at a large tech company this
afternoon, and their initial impression was that import / export would
be a harder sell, whereas the syndication capabilities were far more
attractive. Simply put, ODD can act as RSS for the things that RSS
doesn't capture.
Given that, I was wondering about the difference on two levels:
As a user, which would you value higher: the ability to import and
export your data, or the ability to syndicate non-bloggy data, RSS-
style, so that other people can subscribe to it?
If you're a software developer, do you feel the same way?
afternoon, and their initial impression was that import / export would
be a harder sell, whereas the syndication capabilities were far more
attractive. Simply put, ODD can act as RSS for the things that RSS
doesn't capture.
Given that, I was wondering about the difference on two levels:
As a user, which would you value higher: the ability to import and
export your data, or the ability to syndicate non-bloggy data, RSS-
style, so that other people can subscribe to it?
If you're a software developer, do you feel the same way?
Chris Messina
Apr 30, 2008, 1:08:50 PM4/30/08
to open-data-...@googlegroups.com
Not to be overly twerpish, but there's no value in exporting or
syndication, from a normal-human perspective if there are no
applications or consumers of said data.
syndication, from a normal-human perspective if there are no
applications or consumers of said data.
For example, Basecamp offers a complete XML export of your projects.
Guess how many apps use this data?afaik, none! So, just because the
data is available, doesn't mean anyone will benefit from or use it.
The Data Portability group is drunk on hype without seeming to provide
rationales, justification or real world examples of how having
portable data (subscribable/exportable) really makes services better
for regular folks. Sure, I support the cause in spirit, but I've
barely begun to see new classes of applications emerge built on
portable data, even between two vendors.
Anyway, beyond ranting, my question is about priorities and drivers --
syndication or export is irrelevant -- it's what you can *do* that you
couldn't [easily] do before. What vision do you have for where ODD
goes, presuming you get RSS-like adoption?
Chris
Sent from a typo-prone iPhone.
Ben Werdmuller
Apr 30, 2008, 2:32:42 PM4/30/08
to Open Data Definition
Hi Chris,
> Not to be overly twerpish, but there's no value in exporting or
> syndication, from a normal-human perspective if there are no
> applications or consumers of said data.
That's not twerpish at all; it's common sense.
> Anyway, beyond ranting, my question is about priorities and drivers --
> syndication or export is irrelevant -- it's what you can *do* that you
> couldn't [easily] do before. What vision do you have for where ODD
> goes, presuming you get RSS-like adoption?
The first thing you need to know to understand our perspective is that
we're not (just) a service provider - our software runs as a
platform. The social networking platform we offer has been downloaded
hundreds of thousands of times, and is installed as the basis of an
unknown number of social networks (but it numbers in the thousands at
a minimum). Each of these has slightly different features and types
of data, and one of the things we've been asked about most is moving
data from one to another. Often these are running at schools or on
intranets within certain groups, and therefore you can see how moving
from one to another might be a desirable feature (for example, if you
change schools or workgroups).
This summer we're releasing a rewritten version of our engine, and we
wanted to build in that kind of functionality - alongside the other
kinds of portable functionality that people are developing. I agree
that there aren't many apps out there yet, but I think they will
develop as distributed installs of web applications become more
prevalent (not just in the form of server-based software like ours,
but also OpenSocial apps and so on). It seemed to us that other
people might want to solve the same problem, so we decided to make the
format generic and open the conversation. Elgg 1.0 already has ODD
support, and any decisions made here will be reflected in the
release. We'll also separately release a generic set of ODD
libraries.
I've covered import / export from our perspective. A simple
syndication use case is a Friendfeed-style dashboard without the need
for implementing dozens of APIs. But take an install base of
thousands of small social networks, add ODD syndication and OpenID,
and you have a larger cloud social network. If you're on Service A
and I'm on Service B, if both services support ODD, I can log into
Service A using my Service B credentials and make you a friend.
During the logon, both services alert each other to the fact that they
support ODD. I can then subscribe to your updates and content on your
home site and see them over on my dashboard on my home site.
This kind of thing might not be useful straight away for the larger
names, but it makes sense for networks of smaller sites.
Ben
> Not to be overly twerpish, but there's no value in exporting or
> syndication, from a normal-human perspective if there are no
> applications or consumers of said data.
> Anyway, beyond ranting, my question is about priorities and drivers --
> syndication or export is irrelevant -- it's what you can *do* that you
> couldn't [easily] do before. What vision do you have for where ODD
> goes, presuming you get RSS-like adoption?
we're not (just) a service provider - our software runs as a
platform. The social networking platform we offer has been downloaded
hundreds of thousands of times, and is installed as the basis of an
unknown number of social networks (but it numbers in the thousands at
a minimum). Each of these has slightly different features and types
of data, and one of the things we've been asked about most is moving
data from one to another. Often these are running at schools or on
intranets within certain groups, and therefore you can see how moving
from one to another might be a desirable feature (for example, if you
change schools or workgroups).
This summer we're releasing a rewritten version of our engine, and we
wanted to build in that kind of functionality - alongside the other
kinds of portable functionality that people are developing. I agree
that there aren't many apps out there yet, but I think they will
develop as distributed installs of web applications become more
prevalent (not just in the form of server-based software like ours,
but also OpenSocial apps and so on). It seemed to us that other
people might want to solve the same problem, so we decided to make the
format generic and open the conversation. Elgg 1.0 already has ODD
support, and any decisions made here will be reflected in the
release. We'll also separately release a generic set of ODD
libraries.
I've covered import / export from our perspective. A simple
syndication use case is a Friendfeed-style dashboard without the need
for implementing dozens of APIs. But take an install base of
thousands of small social networks, add ODD syndication and OpenID,
and you have a larger cloud social network. If you're on Service A
and I'm on Service B, if both services support ODD, I can log into
Service A using my Service B credentials and make you a friend.
During the logon, both services alert each other to the fact that they
support ODD. I can then subscribe to your updates and content on your
home site and see them over on my dashboard on my home site.
This kind of thing might not be useful straight away for the larger
names, but it makes sense for networks of smaller sites.
Ben
Danny Ayers
Apr 30, 2008, 2:44:36 PM4/30/08
to open-data-...@googlegroups.com
I've clogged the list enough recently, so responded via blog:
http://blogs.talis.com/n2/archives/59--
http://dannyayers.com
~
http://blogs.talis.com/nodalities/this_weeks_semantic_web/
http://blogs.talis.com/n2/archives/59
http://dannyayers.com
~
http://blogs.talis.com/nodalities/this_weeks_semantic_web/
Chris Messina
Apr 30, 2008, 3:16:47 PM4/30/08
to open-data-...@googlegroups.com
On Wed, Apr 30, 2008 at 11:32 AM, Ben Werdmuller <ben...@gmail.com> wrote:
> Anyway, beyond ranting, my question is about priorities and drivers --The first thing you need to know to understand our perspective is that
> syndication or export is irrelevant -- it's what you can *do* that you
> couldn't [easily] do before. What vision do you have for where ODD
> goes, presuming you get RSS-like adoption?
we're not (just) a service provider - our software runs as a
platform. The social networking platform we offer has been downloaded
hundreds of thousands of times, and is installed as the basis of an
unknown number of social networks (but it numbers in the thousands at
a minimum).
Yep, I'm familiar with Elgg. ;)
Each of these has slightly different features and types
of data, and one of the things we've been asked about most is moving
data from one to another. Often these are running at schools or on
intranets within certain groups, and therefore you can see how moving
from one to another might be a desirable feature (for example, if you
change schools or workgroups).
This is not specific enough for my brain. I understand the concept of portability and moving data around and of interop (opening Rich Text documents in Word or TextEdit, for example). However, I'm trying to get to a real *actual* use case where this has happened and where this portable data has enabled something new to happen. I'm not saying that it hasn't happened or that there's not value, I'm just trying to get concrete here.
This summer we're releasing a rewritten version of our engine, and we
wanted to build in that kind of functionality - alongside the other
kinds of portable functionality that people are developing.
Yes, this is the model of the DiSo Project (http://diso-project.org). We're starting with WordPress, then will hit Moveable Type and Drupal -- in order to create a distributed social network (hence DiSo). ;)
It seemed to us that other
people might want to solve the same problem, so we decided to make the
format generic and open the conversation. Elgg 1.0 already has ODD
support, and any decisions made here will be reflected in the
release. We'll also separately release a generic set of ODD
libraries.
Yes, I agree that this is the appropriate approach, but I'm unclear about your particular format, though it's great that you've put your work out there and have solicited feedback. Ultimately there will be a format and a human-readable expression of this data and I will support it and implement it in the DiSo project; I'm unclear if ODD is the format that I need though.
Can you point me to an actual ODD document in the wild? Would love to take a look at what one of these feeds might look like and how, for example, it differs from the OpenSocial activity stream model:
--
Chris Messina
Citizen-Participant &
Open Source Advocate-at-Large
Work: http://citizenagency.com
Blog: http://factoryjoe.com/blog
Cell: 412.225.1051
IM: factoryjoe
This email is: [ ] bloggable [X] ask first [ ] private
Ben Werdmuller
May 1, 2008, 4:59:54 AM5/1/08
to Open Data Definition
On Apr 30, 8:16 pm, "Chris Messina" <chris.mess...@gmail.com> wrote:
we've been asked for. Say you've got an elementary and a junior high
school, each running their own networks, and a student jumps from one
to the other. Or two districts running networks and a student moves
house. Right now, their accounts would need to be recreated, and
you'd have a hard time bringing their work with them. Educational
networks are a big use case for import / export, and the educational
community tends to come up with 80 page spec documents that throw in
the kitchen sink - formats that people struggle for months to output,
and can rarely import.
> Yes, this is the model of the DiSo Project (http://diso-project.org). We're
> starting with WordPress, then will hit Moveable Type and Drupal -- in order
> to create a distributed social network (hence DiSo). ;)
I actually wasn't familiar with DiSo, but have just read up on it and
applied to join the group.
> Can you point me to an actual ODD document in the wild? Would love to take a
> look at what one of these feeds might look like and how, for example, it
> differs from the OpenSocial activity stream model:
I think that would help greatly. We haven't release our new core, so
it's hard to point to examples in the wild, but we are building out
OpenSocial support, and it's probably a smart move to provide examples
using that kind of data. We'll take a look.
Ben
> On Wed, Apr 30, 2008 at 11:32 AM, Ben Werdmuller <benw...@gmail.com> wrote:
>
> This is not specific enough for my brain. I understand the concept of
> portability and moving data around and of interop (opening Rich Text
> documents in Word or TextEdit, for example). However, I'm trying to get to a
> real *actual* use case where this has happened and where this portable data
> has enabled something new to happen. I'm not saying that it hasn't happened
> or that there's not value, I'm just trying to get concrete here.
The schools model is a concrete example - or at least a use case that
>
> This is not specific enough for my brain. I understand the concept of
> portability and moving data around and of interop (opening Rich Text
> documents in Word or TextEdit, for example). However, I'm trying to get to a
> real *actual* use case where this has happened and where this portable data
> has enabled something new to happen. I'm not saying that it hasn't happened
> or that there's not value, I'm just trying to get concrete here.
we've been asked for. Say you've got an elementary and a junior high
school, each running their own networks, and a student jumps from one
to the other. Or two districts running networks and a student moves
house. Right now, their accounts would need to be recreated, and
you'd have a hard time bringing their work with them. Educational
networks are a big use case for import / export, and the educational
community tends to come up with 80 page spec documents that throw in
the kitchen sink - formats that people struggle for months to output,
and can rarely import.
> Yes, this is the model of the DiSo Project (http://diso-project.org). We're
> starting with WordPress, then will hit Moveable Type and Drupal -- in order
> to create a distributed social network (hence DiSo). ;)
applied to join the group.
> Can you point me to an actual ODD document in the wild? Would love to take a
> look at what one of these feeds might look like and how, for example, it
> differs from the OpenSocial activity stream model:
it's hard to point to examples in the wild, but we are building out
OpenSocial support, and it's probably a smart move to provide examples
using that kind of data. We'll take a look.
Ben
Reply all
Reply to author
Forward
0 new messages