Joining derp?

classic Classic list List threaded Threaded
27 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Joining derp?

metasj

Has this been considered?  It seems to apply to us in many ways.

http://news.yahoo.com/course-reddit-imgur-named-research-institute-derp-142950548.html


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Aaron Halfaker-3
Indeed.  Jonathan, Dario and I have been in contact with Tim Hwang and the other DERP organizers for months.  We're a big fan of the project and we're on the DERP mailing list.  Regretfully, there was some confusion around the time that DERP was going to "go live" that required us to back out (e.g. use of the WMF logo needed to be authorized).


On Wed, Sep 3, 2014 at 9:32 AM, Samuel Klein <[hidden email]> wrote:

Has this been considered?  It seems to apply to us in many ways.

http://news.yahoo.com/course-reddit-imgur-named-research-institute-derp-142950548.html


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l



_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Jodi Schneider-3
So does that mean that it's on the agenda for whichever group approves use of the WMF logo?


On Wed, Sep 3, 2014 at 2:36 PM, Aaron Halfaker <[hidden email]> wrote:
Indeed.  Jonathan, Dario and I have been in contact with Tim Hwang and the other DERP organizers for months.  We're a big fan of the project and we're on the DERP mailing list.  Regretfully, there was some confusion around the time that DERP was going to "go live" that required us to back out (e.g. use of the WMF logo needed to be authorized).


On Wed, Sep 3, 2014 at 9:32 AM, Samuel Klein <[hidden email]> wrote:

Has this been considered?  It seems to apply to us in many ways.

http://news.yahoo.com/course-reddit-imgur-named-research-institute-derp-142950548.html


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l



_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l



_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Ed Summers
In reply to this post by Aaron Halfaker-3
On Sep 3, 2014, at 8:36 AM, Aaron Halfaker <[hidden email]> wrote:
> Indeed.  Jonathan, Dario and I have been in contact with Tim Hwang and the other DERP organizers for months.  We're a big fan of the project and we're on the DERP mailing list.  Regretfully, there was some confusion around the time that DERP was going to "go live" that required us to back out (e.g. use of the WMF logo needed to be authorized).

Is the DERP mailing list public yet?

//Ed
_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Aaron Halfaker-3
When I started working with DERP, it was just a mailing list.  When DERP became something bigger (press release with logos == the WMF Comm Team's concerns), it needed wider consultation within the WMF.  For better or worse, it does not seem like that is a priority right now.  

For all intents and purposes, we have our own DERP right here.  We have a public mailing list.  We support volunteer & external researchers access to data (and we're actively working to make that easier[1]).   We hold regular outreach events (like the Research Hackathon @ Wikimania).  We're also organizing outreach events with other open ecosystem online communities.  For example, we have CSCW'15 workshop proposal submitted, and assuming it is accepted, we'll have participants from Imgur, Reddit and Zooniverse.


-Aaron


On Wed, Sep 3, 2014 at 3:05 PM, Ed Summers <[hidden email]> wrote:
On Sep 3, 2014, at 8:36 AM, Aaron Halfaker <[hidden email]> wrote:
> Indeed.  Jonathan, Dario and I have been in contact with Tim Hwang and the other DERP organizers for months.  We're a big fan of the project and we're on the DERP mailing list.  Regretfully, there was some confusion around the time that DERP was going to "go live" that required us to back out (e.g. use of the WMF logo needed to be authorized).

Is the DERP mailing list public yet?

//Ed
_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Jonathan Morgan
Sigh. Yes. We were early adopters of DERP. Then we had to pull out. But it might still happen someday. It's complicated :/


On Wed, Sep 3, 2014 at 7:42 AM, Aaron Halfaker <[hidden email]> wrote:
When I started working with DERP, it was just a mailing list.  When DERP became something bigger (press release with logos == the WMF Comm Team's concerns), it needed wider consultation within the WMF.  For better or worse, it does not seem like that is a priority right now.  

For all intents and purposes, we have our own DERP right here.  We have a public mailing list.  We support volunteer & external researchers access to data (and we're actively working to make that easier[1]).   We hold regular outreach events (like the Research Hackathon @ Wikimania).  We're also organizing outreach events with other open ecosystem online communities.  For example, we have CSCW'15 workshop proposal submitted, and assuming it is accepted, we'll have participants from Imgur, Reddit and Zooniverse.


-Aaron


On Wed, Sep 3, 2014 at 3:05 PM, Ed Summers <[hidden email]> wrote:
On Sep 3, 2014, at 8:36 AM, Aaron Halfaker <[hidden email]> wrote:
> Indeed.  Jonathan, Dario and I have been in contact with Tim Hwang and the other DERP organizers for months.  We're a big fan of the project and we're on the DERP mailing list.  Regretfully, there was some confusion around the time that DERP was going to "go live" that required us to back out (e.g. use of the WMF logo needed to be authorized).

Is the DERP mailing list public yet?

//Ed
_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l




--
Jonathan T. Morgan
Learning Strategist
Wikimedia Foundation


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Ed Summers
On Sep 3, 2014, at 11:00 AM, Jonathan Morgan <[hidden email]> wrote:
> Sigh. Yes. We were early adopters of DERP. Then we had to pull out. But it might still happen someday. It's complicated :/

If you can share why it's complicated I'd love to hear ; I suspect it's political, but oftentimes these politics have research implications.

///Ed
_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

metasj
On Wed, Sep 3, 2014 at 8:12 AM, Ed Summers <[hidden email]> wrote:
> On Sep 3, 2014, at 11:00 AM, Jonathan Morgan <[hidden email]> wrote:
>> Sigh. Yes. We were early adopters of DERP. Then we had to pull out. But it might still happen someday. It's complicated :/
>
> If you can share why it's complicated I'd love to hear ; I suspect it's political, but oftentimes these politics have research implications.

Ditto.

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Dario Taraborelli-3
Ed, SJ,

on top of what Aaron and Jonathan said (for example, the use of WMF’s logo) there were other concerns that we had to address.

The scope of DERP shifted from an informal communication broker between organizations and academic researchers to a body that – at least according to the original press release – would be responsible for: 

• setting guidelines for ethical research (for example, by prohibiting any kind of feature testing/experimentation)
• aggregating and sharing privacy sensitive data
• reviewing data requests

on behalf of the member organizations.

The Foundation has specific policies on what data can be collected from readers and contributors, how long this data is retained, how it’s released and licensed and under what conditions third parties can obtain access to private data hosted by the Foundation. We felt that we could not join an initiative that would commit the organization to make promises about data access to external parties, without having a conversation with our internal stakeholders first–not just WMF Legal or the WMF analytics team but also our volunteer community, chapters etc. 

The press release was later amended to remove language that would suggest specific commitments for member organizations but our concerns remain the same. 

Jonathan, Aaron and I are all interested in creating opportunities for cross-platform research on online collaboration and we have plans in the pipeline to make more data publicly available. We are big fans of DERP and are still involved in a personal capacity in the initiative, but we are not in a position to take Wikimedia with us at this time.

I hope that clarifies how the decision was made.

Dario

On Sep 3, 2014, at 1:03 PM, Samuel Klein <[hidden email]> wrote:

On Wed, Sep 3, 2014 at 8:12 AM, Ed Summers <[hidden email]> wrote:
On Sep 3, 2014, at 11:00 AM, Jonathan Morgan <[hidden email]> wrote:
Sigh. Yes. We were early adopters of DERP. Then we had to pull out. But it might still happen someday. It's complicated :/

If you can share why it's complicated I'd love to hear ; I suspect it's political, but oftentimes these politics have research implications.

Ditto.

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Kerry Raymond
I think many organisations would be reluctant to sign up to DERP on that basis. It seems somewhat ironic that a group apparently committed to ethical research seems not to have considered that existing data sets may have been collected under strict conditions of use and re-use as part of existing ethical processes.

It would seem that DERP needs to permit at least two levels of participation, one of which allows information to be provided about data set availability but retains separate control on access. Presumably it would easier for WMF to sign up for this lower level of engagement. 

Sent from my iPad

On 4 Sep 2014, at 6:46 am, Dario Taraborelli <[hidden email]> wrote:

Ed, SJ,

on top of what Aaron and Jonathan said (for example, the use of WMF’s logo) there were other concerns that we had to address.

The scope of DERP shifted from an informal communication broker between organizations and academic researchers to a body that – at least according to the original press release – would be responsible for: 

• setting guidelines for ethical research (for example, by prohibiting any kind of feature testing/experimentation)
• aggregating and sharing privacy sensitive data
• reviewing data requests

on behalf of the member organizations.

The Foundation has specific policies on what data can be collected from readers and contributors, how long this data is retained, how it’s released and licensed and under what conditions third parties can obtain access to private data hosted by the Foundation. We felt that we could not join an initiative that would commit the organization to make promises about data access to external parties, without having a conversation with our internal stakeholders first–not just WMF Legal or the WMF analytics team but also our volunteer community, chapters etc. 

The press release was later amended to remove language that would suggest specific commitments for member organizations but our concerns remain the same. 

Jonathan, Aaron and I are all interested in creating opportunities for cross-platform research on online collaboration and we have plans in the pipeline to make more data publicly available. We are big fans of DERP and are still involved in a personal capacity in the initiative, but we are not in a position to take Wikimedia with us at this time.

I hope that clarifies how the decision was made.

Dario

On Sep 3, 2014, at 1:03 PM, Samuel Klein <[hidden email]> wrote:

On Wed, Sep 3, 2014 at 8:12 AM, Ed Summers <[hidden email]> wrote:
On Sep 3, 2014, at 11:00 AM, Jonathan Morgan <[hidden email]> wrote:
Sigh. Yes. We were early adopters of DERP. Then we had to pull out. But it might still happen someday. It's complicated :/

If you can share why it's complicated I'd love to hear ; I suspect it's political, but oftentimes these politics have research implications.

Ditto.

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Joe Corneli-3
On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]> wrote:

> It would seem that DERP needs to permit at least two levels of
> participation, one of which allows information to be provided about data set
> availability but retains separate control on access.

<lol>    ... To boldly go where GIT has gone before.    </lol>

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

R.Stuart Geiger
Hi all, thanks for all the info. I'm a DERP fellow, which means I was planning on participating in this as a researcher (I'm doing some work on reddit, too) as well as serving as an advisory board. I apparently haven't been involved in the same threads/calls with the DERP organizers that Aaron, Jonathan, and Dario have been on, and I'm kind of shocked at what I'm hearing. I completely believe you guys, it just runs so opposite to what I've been told that I'm dreading the e-mail I think I'm going to have to write to the DERP folks.

This is the first time I've heard anything about DERP being much more than an informal communication broker between organizations and academic researchers. DERP was pitched to me as a big signaling mechanism to researchers, platforms, and the public that there are spaces outside of Facebook and Twitter to do research. Wikimedia obviously doesn't need DERP as much as some of the smaller platforms do, but I thought it would be great for Wikimedia's presence (yes, the logo) to be there, standing in solidarity with the lesser-researched platforms. As it was explained to me, all that was supposed to be involved in a platform joining DERP is 1) a public declaration that they are open to receiving requests from researchers via DERP and 2) a commitment to review and respond to proposals that were e-mailed from researchers to DERP. In one of the fellows calls, I actually think someone asked whether DERP would be like an Institutional Review Board that would independently approve/reject studies, and we all thought that it would be better for these to be done on a case-by-case basis between the researcher and the platform(s). 

Early on, I actually suggested adding some language about ethics. I suggested that as we started these projects, it would be great to develop an ongoing, informal set of best practices for doing computational social science in an academic/industry partnership -- particularly in the wake of the Facebook emotion contagion study. Something like a series of blog posts about the various ethical issues we encountered in the course of doing this kind of research across a bunch of different platforms, and ways that they were resolved. Perhaps that might synthesize into a mini workshop culminating in a whitepaper, but it wouldn't ever be binding. As I was told about it, DERP's direct role ends once the researcher has made successful contact with the platform, aside from very high-level community organizing things like discussions about best practices. Same thing with data standards -- it is a fool's errand to mandate those, but I was told that DERP might one day be a hub where people could talk about how to integrate data from different platforms.

I did see the language that "All research supported by DERP will be released openly and made publicly available," but I interpreted this as something even weaker than Green OA -- that even if you publish in a closed access journal, you have to write something up about the research. Kind of like what Aaron did with our ABS paper. [1] The idea was that you should't be able to do studies in the dark without anybody ever knowing about them. The fellows were told that this wouldn't apply to datasets at all. And given how many qualitative researchers are fellows and planning on doing interviews, the concern that we would have to release full interview transcripts was specifically brought up. Again, the idea was that DERP might later develop some optional, guiding best practices to make things easier, but any conditions of data access were supposed to be negotiated between the researcher and the platform. I remember asking if DERP was supposed to be some kind of central repository for storing data, and that was resoundingly rejected.

So if DERP has shifted beyond this, that would be a pretty serious matter of concern for me.

Best,
Stuart




On Wed, Sep 3, 2014 at 2:32 PM, Joe Corneli <[hidden email]> wrote:
On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]> wrote:

> It would seem that DERP needs to permit at least two levels of
> participation, one of which allows information to be provided about data set
> availability but retains separate control on access.

<lol>    ... To boldly go where GIT has gone before.    </lol>

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Jonathan Morgan
I don't think there's cause for you to be concerned, Stu. FWIW, we've talked to Tim since launch, and after we expressed our concerns he assured us that the model of DERP is still just facilitating connections in a non-exclusive way, rather than playing a role as a reviewing body or a data broker of any kind.

There were other reasons we decided to be a little more cautious about committing to this kind of initiative. As Toby Negrin pointed out recently: There is one major difference between the companies involved in DERP and ourselves -- they all use data collected from their users to make money and we explicitly do not. This is frankly a point of pride for many members of the foundation and certainly the community.

More pragmatically, the last week of organizing for the DERP launch just happened too fast for us (and happened during Wikimania, to boot!). Those of us in research-y roles hadn't had a chance to discuss all the evolving details as a team, and on the eve of the launch we didn't all feel we had a 100% clear idea of what commitments we would be making by joining. 

But we're still on the DERP mailing list, and (if the review gods are merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max Goodman at CSCW 2015. 

We like DERP! Don't stop DERPing!

- Jonathan




On Wed, Sep 3, 2014 at 3:31 PM, R.Stuart Geiger <[hidden email]> wrote:
Hi all, thanks for all the info. I'm a DERP fellow, which means I was planning on participating in this as a researcher (I'm doing some work on reddit, too) as well as serving as an advisory board. I apparently haven't been involved in the same threads/calls with the DERP organizers that Aaron, Jonathan, and Dario have been on, and I'm kind of shocked at what I'm hearing. I completely believe you guys, it just runs so opposite to what I've been told that I'm dreading the e-mail I think I'm going to have to write to the DERP folks.

This is the first time I've heard anything about DERP being much more than an informal communication broker between organizations and academic researchers. DERP was pitched to me as a big signaling mechanism to researchers, platforms, and the public that there are spaces outside of Facebook and Twitter to do research. Wikimedia obviously doesn't need DERP as much as some of the smaller platforms do, but I thought it would be great for Wikimedia's presence (yes, the logo) to be there, standing in solidarity with the lesser-researched platforms. As it was explained to me, all that was supposed to be involved in a platform joining DERP is 1) a public declaration that they are open to receiving requests from researchers via DERP and 2) a commitment to review and respond to proposals that were e-mailed from researchers to DERP. In one of the fellows calls, I actually think someone asked whether DERP would be like an Institutional Review Board that would independently approve/reject studies, and we all thought that it would be better for these to be done on a case-by-case basis between the researcher and the platform(s). 

Early on, I actually suggested adding some language about ethics. I suggested that as we started these projects, it would be great to develop an ongoing, informal set of best practices for doing computational social science in an academic/industry partnership -- particularly in the wake of the Facebook emotion contagion study. Something like a series of blog posts about the various ethical issues we encountered in the course of doing this kind of research across a bunch of different platforms, and ways that they were resolved. Perhaps that might synthesize into a mini workshop culminating in a whitepaper, but it wouldn't ever be binding. As I was told about it, DERP's direct role ends once the researcher has made successful contact with the platform, aside from very high-level community organizing things like discussions about best practices. Same thing with data standards -- it is a fool's errand to mandate those, but I was told that DERP might one day be a hub where people could talk about how to integrate data from different platforms.

I did see the language that "All research supported by DERP will be released openly and made publicly available," but I interpreted this as something even weaker than Green OA -- that even if you publish in a closed access journal, you have to write something up about the research. Kind of like what Aaron did with our ABS paper. [1] The idea was that you should't be able to do studies in the dark without anybody ever knowing about them. The fellows were told that this wouldn't apply to datasets at all. And given how many qualitative researchers are fellows and planning on doing interviews, the concern that we would have to release full interview transcripts was specifically brought up. Again, the idea was that DERP might later develop some optional, guiding best practices to make things easier, but any conditions of data access were supposed to be negotiated between the researcher and the platform. I remember asking if DERP was supposed to be some kind of central repository for storing data, and that was resoundingly rejected.

So if DERP has shifted beyond this, that would be a pretty serious matter of concern for me.

Best,
Stuart




On Wed, Sep 3, 2014 at 2:32 PM, Joe Corneli <[hidden email]> wrote:
On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]> wrote:

> It would seem that DERP needs to permit at least two levels of
> participation, one of which allows information to be provided about data set
> availability but retains separate control on access.

<lol>    ... To boldly go where GIT has gone before.    </lol>

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l




--
Jonathan T. Morgan
Learning Strategist
Wikimedia Foundation


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Stuart A. Yeates
What I think we really need is better standardisation of description
of datasets, so that they can shared in machine-readable ways. Then we
can have as many different groups working with different sets of
datasets as we like and still search, find and publish globally.

cheers
stuart

On Thu, Sep 4, 2014 at 11:29 AM, Jonathan Morgan <[hidden email]> wrote:

> I don't think there's cause for you to be concerned, Stu. FWIW, we've talked
> to Tim since launch, and after we expressed our concerns he assured us that
> the model of DERP is still just facilitating connections in a non-exclusive
> way, rather than playing a role as a reviewing body or a data broker of any
> kind.
>
> There were other reasons we decided to be a little more cautious about
> committing to this kind of initiative. As Toby Negrin pointed out recently:
> There is one major difference between the companies involved in DERP and
> ourselves -- they all use data collected from their users to make money and
> we explicitly do not. This is frankly a point of pride for many members of
> the foundation and certainly the community.
>
> More pragmatically, the last week of organizing for the DERP launch just
> happened too fast for us (and happened during Wikimania, to boot!). Those of
> us in research-y roles hadn't had a chance to discuss all the evolving
> details as a team, and on the eve of the launch we didn't all feel we had a
> 100% clear idea of what commitments we would be making by joining.
>
> But we're still on the DERP mailing list, and (if the review gods are
> merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max
> Goodman at CSCW 2015.
>
> We like DERP! Don't stop DERPing!
>
> - Jonathan
>
>
>
>
> On Wed, Sep 3, 2014 at 3:31 PM, R.Stuart Geiger <[hidden email]> wrote:
>>
>> Hi all, thanks for all the info. I'm a DERP fellow, which means I was
>> planning on participating in this as a researcher (I'm doing some work on
>> reddit, too) as well as serving as an advisory board. I apparently haven't
>> been involved in the same threads/calls with the DERP organizers that Aaron,
>> Jonathan, and Dario have been on, and I'm kind of shocked at what I'm
>> hearing. I completely believe you guys, it just runs so opposite to what
>> I've been told that I'm dreading the e-mail I think I'm going to have to
>> write to the DERP folks.
>>
>> This is the first time I've heard anything about DERP being much more than
>> an informal communication broker between organizations and academic
>> researchers. DERP was pitched to me as a big signaling mechanism to
>> researchers, platforms, and the public that there are spaces outside of
>> Facebook and Twitter to do research. Wikimedia obviously doesn't need DERP
>> as much as some of the smaller platforms do, but I thought it would be great
>> for Wikimedia's presence (yes, the logo) to be there, standing in solidarity
>> with the lesser-researched platforms. As it was explained to me, all that
>> was supposed to be involved in a platform joining DERP is 1) a public
>> declaration that they are open to receiving requests from researchers via
>> DERP and 2) a commitment to review and respond to proposals that were
>> e-mailed from researchers to DERP. In one of the fellows calls, I actually
>> think someone asked whether DERP would be like an Institutional Review Board
>> that would independently approve/reject studies, and we all thought that it
>> would be better for these to be done on a case-by-case basis between the
>> researcher and the platform(s).
>>
>> Early on, I actually suggested adding some language about ethics. I
>> suggested that as we started these projects, it would be great to develop an
>> ongoing, informal set of best practices for doing computational social
>> science in an academic/industry partnership -- particularly in the wake of
>> the Facebook emotion contagion study. Something like a series of blog posts
>> about the various ethical issues we encountered in the course of doing this
>> kind of research across a bunch of different platforms, and ways that they
>> were resolved. Perhaps that might synthesize into a mini workshop
>> culminating in a whitepaper, but it wouldn't ever be binding. As I was told
>> about it, DERP's direct role ends once the researcher has made successful
>> contact with the platform, aside from very high-level community organizing
>> things like discussions about best practices. Same thing with data standards
>> -- it is a fool's errand to mandate those, but I was told that DERP might
>> one day be a hub where people could talk about how to integrate data from
>> different platforms.
>>
>> I did see the language that "All research supported by DERP will be
>> released openly and made publicly available," but I interpreted this as
>> something even weaker than Green OA -- that even if you publish in a closed
>> access journal, you have to write something up about the research. Kind of
>> like what Aaron did with our ABS paper. [1] The idea was that you should't
>> be able to do studies in the dark without anybody ever knowing about them.
>> The fellows were told that this wouldn't apply to datasets at all. And given
>> how many qualitative researchers are fellows and planning on doing
>> interviews, the concern that we would have to release full interview
>> transcripts was specifically brought up. Again, the idea was that DERP might
>> later develop some optional, guiding best practices to make things easier,
>> but any conditions of data access were supposed to be negotiated between the
>> researcher and the platform. I remember asking if DERP was supposed to be
>> some kind of central repository for storing data, and that was resoundingly
>> rejected.
>>
>> So if DERP has shifted beyond this, that would be a pretty serious matter
>> of concern for me.
>>
>> Best,
>> Stuart
>>
>>
>> [1] http://www-users.cs.umn.edu/~halfak/publications/The_Rise_and_Decline/
>>
>>
>> On Wed, Sep 3, 2014 at 2:32 PM, Joe Corneli <[hidden email]>
>> wrote:
>>>
>>> On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]>
>>> wrote:
>>>
>>> > It would seem that DERP needs to permit at least two levels of
>>> > participation, one of which allows information to be provided about
>>> > data set
>>> > availability but retains separate control on access.
>>>
>>> <lol>    ... To boldly go where GIT has gone before.    </lol>
>>>
>>> _______________________________________________
>>> Wiki-research-l mailing list
>>> [hidden email]
>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>
>>
>>
>> _______________________________________________
>> Wiki-research-l mailing list
>> [hidden email]
>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>
>
>
>
> --
> Jonathan T. Morgan
> Learning Strategist
> Wikimedia Foundation
> User:Jmorgan (WMF)
> [hidden email]
>
>
> _______________________________________________
> Wiki-research-l mailing list
> [hidden email]
> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Kerry Raymond
Hmm, a meta data standard. there must be a few dozen on the shelf to pick from ... I recall writing a few myself once :-)



Sent from my iPad

> On 4 Sep 2014, at 9:53 am, "Stuart A. Yeates" <[hidden email]> wrote:
>
> What I think we really need is better standardisation of description
> of datasets, so that they can shared in machine-readable ways. Then we
> can have as many different groups working with different sets of
> datasets as we like and still search, find and publish globally.
>
> cheers
> stuart
>
>> On Thu, Sep 4, 2014 at 11:29 AM, Jonathan Morgan <[hidden email]> wrote:
>> I don't think there's cause for you to be concerned, Stu. FWIW, we've talked
>> to Tim since launch, and after we expressed our concerns he assured us that
>> the model of DERP is still just facilitating connections in a non-exclusive
>> way, rather than playing a role as a reviewing body or a data broker of any
>> kind.
>>
>> There were other reasons we decided to be a little more cautious about
>> committing to this kind of initiative. As Toby Negrin pointed out recently:
>> There is one major difference between the companies involved in DERP and
>> ourselves -- they all use data collected from their users to make money and
>> we explicitly do not. This is frankly a point of pride for many members of
>> the foundation and certainly the community.
>>
>> More pragmatically, the last week of organizing for the DERP launch just
>> happened too fast for us (and happened during Wikimania, to boot!). Those of
>> us in research-y roles hadn't had a chance to discuss all the evolving
>> details as a team, and on the eve of the launch we didn't all feel we had a
>> 100% clear idea of what commitments we would be making by joining.
>>
>> But we're still on the DERP mailing list, and (if the review gods are
>> merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max
>> Goodman at CSCW 2015.
>>
>> We like DERP! Don't stop DERPing!
>>
>> - Jonathan
>>
>>
>>
>>
>>> On Wed, Sep 3, 2014 at 3:31 PM, R.Stuart Geiger <[hidden email]> wrote:
>>>
>>> Hi all, thanks for all the info. I'm a DERP fellow, which means I was
>>> planning on participating in this as a researcher (I'm doing some work on
>>> reddit, too) as well as serving as an advisory board. I apparently haven't
>>> been involved in the same threads/calls with the DERP organizers that Aaron,
>>> Jonathan, and Dario have been on, and I'm kind of shocked at what I'm
>>> hearing. I completely believe you guys, it just runs so opposite to what
>>> I've been told that I'm dreading the e-mail I think I'm going to have to
>>> write to the DERP folks.
>>>
>>> This is the first time I've heard anything about DERP being much more than
>>> an informal communication broker between organizations and academic
>>> researchers. DERP was pitched to me as a big signaling mechanism to
>>> researchers, platforms, and the public that there are spaces outside of
>>> Facebook and Twitter to do research. Wikimedia obviously doesn't need DERP
>>> as much as some of the smaller platforms do, but I thought it would be great
>>> for Wikimedia's presence (yes, the logo) to be there, standing in solidarity
>>> with the lesser-researched platforms. As it was explained to me, all that
>>> was supposed to be involved in a platform joining DERP is 1) a public
>>> declaration that they are open to receiving requests from researchers via
>>> DERP and 2) a commitment to review and respond to proposals that were
>>> e-mailed from researchers to DERP. In one of the fellows calls, I actually
>>> think someone asked whether DERP would be like an Institutional Review Board
>>> that would independently approve/reject studies, and we all thought that it
>>> would be better for these to be done on a case-by-case basis between the
>>> researcher and the platform(s).
>>>
>>> Early on, I actually suggested adding some language about ethics. I
>>> suggested that as we started these projects, it would be great to develop an
>>> ongoing, informal set of best practices for doing computational social
>>> science in an academic/industry partnership -- particularly in the wake of
>>> the Facebook emotion contagion study. Something like a series of blog posts
>>> about the various ethical issues we encountered in the course of doing this
>>> kind of research across a bunch of different platforms, and ways that they
>>> were resolved. Perhaps that might synthesize into a mini workshop
>>> culminating in a whitepaper, but it wouldn't ever be binding. As I was told
>>> about it, DERP's direct role ends once the researcher has made successful
>>> contact with the platform, aside from very high-level community organizing
>>> things like discussions about best practices. Same thing with data standards
>>> -- it is a fool's errand to mandate those, but I was told that DERP might
>>> one day be a hub where people could talk about how to integrate data from
>>> different platforms.
>>>
>>> I did see the language that "All research supported by DERP will be
>>> released openly and made publicly available," but I interpreted this as
>>> something even weaker than Green OA -- that even if you publish in a closed
>>> access journal, you have to write something up about the research. Kind of
>>> like what Aaron did with our ABS paper. [1] The idea was that you should't
>>> be able to do studies in the dark without anybody ever knowing about them.
>>> The fellows were told that this wouldn't apply to datasets at all. And given
>>> how many qualitative researchers are fellows and planning on doing
>>> interviews, the concern that we would have to release full interview
>>> transcripts was specifically brought up. Again, the idea was that DERP might
>>> later develop some optional, guiding best practices to make things easier,
>>> but any conditions of data access were supposed to be negotiated between the
>>> researcher and the platform. I remember asking if DERP was supposed to be
>>> some kind of central repository for storing data, and that was resoundingly
>>> rejected.
>>>
>>> So if DERP has shifted beyond this, that would be a pretty serious matter
>>> of concern for me.
>>>
>>> Best,
>>> Stuart
>>>
>>>
>>> [1] http://www-users.cs.umn.edu/~halfak/publications/The_Rise_and_Decline/
>>>
>>>
>>> On Wed, Sep 3, 2014 at 2:32 PM, Joe Corneli <[hidden email]>
>>> wrote:
>>>>
>>>> On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]>
>>>> wrote:
>>>>
>>>>> It would seem that DERP needs to permit at least two levels of
>>>>> participation, one of which allows information to be provided about
>>>>> data set
>>>>> availability but retains separate control on access.
>>>>
>>>> <lol>    ... To boldly go where GIT has gone before.    </lol>
>>>>
>>>> _______________________________________________
>>>> Wiki-research-l mailing list
>>>> [hidden email]
>>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>>
>>>
>>>
>>> _______________________________________________
>>> Wiki-research-l mailing list
>>> [hidden email]
>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>
>>
>>
>> --
>> Jonathan T. Morgan
>> Learning Strategist
>> Wikimedia Foundation
>> User:Jmorgan (WMF)
>> [hidden email]
>>
>>
>> _______________________________________________
>> Wiki-research-l mailing list
>> [hidden email]
>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>
> _______________________________________________
> Wiki-research-l mailing list
> [hidden email]
> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Brian Butler
Equally important are "standards" (wrong term really) for dataset descriptions so they can be shared in HUMAN readable/comprehensible ways.
 
This is often the bigger problem, resulting in publishing and sharing efforts that don't really work.



On Sep 4, 2014, at 12:05 AM, Kerry Raymond wrote:

> Hmm, a meta data standard. there must be a few dozen on the shelf to pick from ... I recall writing a few myself once :-)
>
>
>
> Sent from my iPad
>
>> On 4 Sep 2014, at 9:53 am, "Stuart A. Yeates" <[hidden email]> wrote:
>>
>> What I think we really need is better standardisation of description
>> of datasets, so that they can shared in machine-readable ways. Then we
>> can have as many different groups working with different sets of
>> datasets as we like and still search, find and publish globally.
>>
>> cheers
>> stuart
>>
>>> On Thu, Sep 4, 2014 at 11:29 AM, Jonathan Morgan <[hidden email]> wrote:
>>> I don't think there's cause for you to be concerned, Stu. FWIW, we've talked
>>> to Tim since launch, and after we expressed our concerns he assured us that
>>> the model of DERP is still just facilitating connections in a non-exclusive
>>> way, rather than playing a role as a reviewing body or a data broker of any
>>> kind.
>>>
>>> There were other reasons we decided to be a little more cautious about
>>> committing to this kind of initiative. As Toby Negrin pointed out recently:
>>> There is one major difference between the companies involved in DERP and
>>> ourselves -- they all use data collected from their users to make money and
>>> we explicitly do not. This is frankly a point of pride for many members of
>>> the foundation and certainly the community.
>>>
>>> More pragmatically, the last week of organizing for the DERP launch just
>>> happened too fast for us (and happened during Wikimania, to boot!). Those of
>>> us in research-y roles hadn't had a chance to discuss all the evolving
>>> details as a team, and on the eve of the launch we didn't all feel we had a
>>> 100% clear idea of what commitments we would be making by joining.
>>>
>>> But we're still on the DERP mailing list, and (if the review gods are
>>> merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max
>>> Goodman at CSCW 2015.
>>>
>>> We like DERP! Don't stop DERPing!
>>>
>>> - Jonathan
>>>
>>>
>>>
>>>
>>>> On Wed, Sep 3, 2014 at 3:31 PM, R.Stuart Geiger <[hidden email]> wrote:
>>>>
>>>> Hi all, thanks for all the info. I'm a DERP fellow, which means I was
>>>> planning on participating in this as a researcher (I'm doing some work on
>>>> reddit, too) as well as serving as an advisory board. I apparently haven't
>>>> been involved in the same threads/calls with the DERP organizers that Aaron,
>>>> Jonathan, and Dario have been on, and I'm kind of shocked at what I'm
>>>> hearing. I completely believe you guys, it just runs so opposite to what
>>>> I've been told that I'm dreading the e-mail I think I'm going to have to
>>>> write to the DERP folks.
>>>>
>>>> This is the first time I've heard anything about DERP being much more than
>>>> an informal communication broker between organizations and academic
>>>> researchers. DERP was pitched to me as a big signaling mechanism to
>>>> researchers, platforms, and the public that there are spaces outside of
>>>> Facebook and Twitter to do research. Wikimedia obviously doesn't need DERP
>>>> as much as some of the smaller platforms do, but I thought it would be great
>>>> for Wikimedia's presence (yes, the logo) to be there, standing in solidarity
>>>> with the lesser-researched platforms. As it was explained to me, all that
>>>> was supposed to be involved in a platform joining DERP is 1) a public
>>>> declaration that they are open to receiving requests from researchers via
>>>> DERP and 2) a commitment to review and respond to proposals that were
>>>> e-mailed from researchers to DERP. In one of the fellows calls, I actually
>>>> think someone asked whether DERP would be like an Institutional Review Board
>>>> that would independently approve/reject studies, and we all thought that it
>>>> would be better for these to be done on a case-by-case basis between the
>>>> researcher and the platform(s).
>>>>
>>>> Early on, I actually suggested adding some language about ethics. I
>>>> suggested that as we started these projects, it would be great to develop an
>>>> ongoing, informal set of best practices for doing computational social
>>>> science in an academic/industry partnership -- particularly in the wake of
>>>> the Facebook emotion contagion study. Something like a series of blog posts
>>>> about the various ethical issues we encountered in the course of doing this
>>>> kind of research across a bunch of different platforms, and ways that they
>>>> were resolved. Perhaps that might synthesize into a mini workshop
>>>> culminating in a whitepaper, but it wouldn't ever be binding. As I was told
>>>> about it, DERP's direct role ends once the researcher has made successful
>>>> contact with the platform, aside from very high-level community organizing
>>>> things like discussions about best practices. Same thing with data standards
>>>> -- it is a fool's errand to mandate those, but I was told that DERP might
>>>> one day be a hub where people could talk about how to integrate data from
>>>> different platforms.
>>>>
>>>> I did see the language that "All research supported by DERP will be
>>>> released openly and made publicly available," but I interpreted this as
>>>> something even weaker than Green OA -- that even if you publish in a closed
>>>> access journal, you have to write something up about the research. Kind of
>>>> like what Aaron did with our ABS paper. [1] The idea was that you should't
>>>> be able to do studies in the dark without anybody ever knowing about them.
>>>> The fellows were told that this wouldn't apply to datasets at all. And given
>>>> how many qualitative researchers are fellows and planning on doing
>>>> interviews, the concern that we would have to release full interview
>>>> transcripts was specifically brought up. Again, the idea was that DERP might
>>>> later develop some optional, guiding best practices to make things easier,
>>>> but any conditions of data access were supposed to be negotiated between the
>>>> researcher and the platform. I remember asking if DERP was supposed to be
>>>> some kind of central repository for storing data, and that was resoundingly
>>>> rejected.
>>>>
>>>> So if DERP has shifted beyond this, that would be a pretty serious matter
>>>> of concern for me.
>>>>
>>>> Best,
>>>> Stuart
>>>>
>>>>
>>>> [1] http://www-users.cs.umn.edu/~halfak/publications/The_Rise_and_Decline/
>>>>
>>>>
>>>> On Wed, Sep 3, 2014 at 2:32 PM, Joe Corneli <[hidden email]>
>>>> wrote:
>>>>>
>>>>> On Wed, Sep 3, 2014 at 10:14 PM, Kerry Raymond <[hidden email]>
>>>>> wrote:
>>>>>
>>>>>> It would seem that DERP needs to permit at least two levels of
>>>>>> participation, one of which allows information to be provided about
>>>>>> data set
>>>>>> availability but retains separate control on access.
>>>>>
>>>>> <lol>    ... To boldly go where GIT has gone before.    </lol>
>>>>>
>>>>> _______________________________________________
>>>>> Wiki-research-l mailing list
>>>>> [hidden email]
>>>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Wiki-research-l mailing list
>>>> [hidden email]
>>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>>
>>>
>>>
>>> --
>>> Jonathan T. Morgan
>>> Learning Strategist
>>> Wikimedia Foundation
>>> User:Jmorgan (WMF)
>>> [hidden email]
>>>
>>>
>>> _______________________________________________
>>> Wiki-research-l mailing list
>>> [hidden email]
>>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>>
>> _______________________________________________
>> Wiki-research-l mailing list
>> [hidden email]
>> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
>
> _______________________________________________
> Wiki-research-l mailing list
> [hidden email]
> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l


_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Ed Summers
In reply to this post by Jonathan Morgan
On Sep 3, 2014, at 7:29 PM, Jonathan Morgan <[hidden email]> wrote:
There were other reasons we decided to be a little more cautious about committing to this kind of initiative. As Toby Negrin pointed out recently: There is one major difference between the companies involved in DERP and ourselves -- they all use data collected from their users to make money and we explicitly do not. This is frankly a point of pride for many members of the foundation and certainly the community.

Wikimedia is a nonprofit, but that doesn't mean it can't bring in money based on data collected from its users. I think we all know that this is exactly what it does. As a non-profit WMF is just prevented from making a profit, right?

More pragmatically, the last week of organizing for the DERP launch just happened too fast for us (and happened during Wikimania, to boot!). Those of us in research-y roles hadn't had a chance to discuss all the evolving details as a team, and on the eve of the launch we didn't all feel we had a 100% clear idea of what commitments we would be making by joining. 

Ok, this is much more plausible. I'm new to the idea of DERP, but based on what Stuart just wrote it does sound like a useful effort to be a part of.

But we're still on the DERP mailing list, and (if the review gods are merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max Goodman at CSCW 2015. 

Do you know how to get on the DERP mailing list?

//Ed

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Aaron Halfaker-3
Do you know how to get on the DERP mailing list?

It looks like [hidden email] is the right place to ask.  




On Thu, Sep 4, 2014 at 8:52 AM, Ed Summers <[hidden email]> wrote:
On Sep 3, 2014, at 7:29 PM, Jonathan Morgan <[hidden email]> wrote:
There were other reasons we decided to be a little more cautious about committing to this kind of initiative. As Toby Negrin pointed out recently: There is one major difference between the companies involved in DERP and ourselves -- they all use data collected from their users to make money and we explicitly do not. This is frankly a point of pride for many members of the foundation and certainly the community.

Wikimedia is a nonprofit, but that doesn't mean it can't bring in money based on data collected from its users. I think we all know that this is exactly what it does. As a non-profit WMF is just prevented from making a profit, right?

More pragmatically, the last week of organizing for the DERP launch just happened too fast for us (and happened during Wikimania, to boot!). Those of us in research-y roles hadn't had a chance to discuss all the evolving details as a team, and on the eve of the launch we didn't all feel we had a 100% clear idea of what commitments we would be making by joining. 

Ok, this is much more plausible. I'm new to the idea of DERP, but based on what Stuart just wrote it does sound like a useful effort to be a part of.

But we're still on the DERP mailing list, and (if the review gods are merciful) we plan to co-organize a CSCW workshop with Tim Hwang and Max Goodman at CSCW 2015. 

Do you know how to get on the DERP mailing list?

//Ed

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l



_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

aaron shaw
In reply to this post by Jonathan Morgan

On Thu, Sep 4, 2014 at 8:52 AM, Ed Summers <[hidden email]> wrote:
Wikimedia is a nonprofit, but that doesn't mean it can't bring in money based on data collected from its users. I think we all know that this is exactly what it does.

Sorry Ed, I don't think we all know that. In fact, I'm unaware of any way in which Wikimedia makes money based on data collected from its users. To my knowledge, the Foundation is supported almost entirely through private donations[1].

If there's something else you have in mind here, it would be helpful to know what you mean. Otherwise, I think it's quite unhelpful to suggest that the Foundation is somehow making money on the backs of users.

all the best,
Aaron



_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Reply | Threaded
Open this post in threaded view
|

Re: Joining derp?

Ed Summers

On Sep 4, 2014, at 11:20 AM, aaron shaw <[hidden email]> wrote:
Sorry Ed, I don't think we all know that. In fact, I'm unaware of any way in which Wikimedia makes money based on data collected from its users. To my knowledge, the Foundation is supported almost entirely through private donations[1].

Ok, try this on for size:

An edit to a Wikipedia article is data collected from its users. WMF receives millions of dollars of donations a year because of this data, and its accessibility. 

//Ed

_______________________________________________
Wiki-research-l mailing list
[hidden email]
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
12