Research talk:Breaking into new Data-Spaces/Proposal

Splitting challenges and strategies in goals

Kristen asks:

could 'identify common challenges' and -identify novel strategies be broken into two different goals? I feel like they are unique enough that this would improve clarity and really help show that there are different approaches as well as replicable approaches to using OOC datasets

My sense is that challenges and strategies for circumventing such challenges are two sizes of the same coin. It would be awkward to discuss strategies for solving problems without discussing the problems that are solved. --EpochFail (talk) 14:57, 13 October 2015 (UTC)

I follow your logic, on keeping challenges and solutions together. If we do that I would recommend providing an example so that the relationship is clear and a framework for discussing relationships between challenges and solutions is well defined (or at least how we would like workshop participants to discuss how they identify challenges and develop solutions). Does that make sense?

I think it does. I hope that this will come out clearly in the long-form discussions for each major section: methods, technology and metadata. --EpochFail (talk) 18:00, 13 October 2015 (UTC)

Use of the term "break into" when discussing working with a new dataset/data space

I chose to use this term because it could be part of a catchy title. "to enter or open a (place, vehicle, or container) forcibly" It seems like an apt term for the experience of picking up a dataset and figuring out how to use it. Lots of reverse engineering as well as (metaphorical) hammers and duct tape are usually involved. But I might be missing something. Kristen proposes that we just call it "seeking to use a new dataset". Thoughts? --EpochFail (talk) 15:01, 13 October 2015 (UTC)

I think that the phrase could work if it is used throughout the proposal... I only noticed it appearing once before.

Fair point. That was my intention. I agree that it should be used throughout the proposal if we are going to use it at all. We should also probably have define the circumstances we are referring to when using the term "breaking into" a dataset. --EpochFail (talk) 17:59, 13 October 2015 (UTC)

Is it safe to assume the title should be "break into" and not "beak"? --(thiry)

Ha! Yes. Fixed. --EpochFail (talk) 22:29, 13 October 2015 (UTC)

Just a doubt: as we always speak of OOC in the proposal, shouldn't the title say "open online community science"? If it reads too long, no problem, just a doubt. Btw, I like the "Breaking into" title. Sdivad (talk) 16:57, 15 October 2015 (UTC)

Liz has no clue how to use the talk page

First off, I have to admit that I have never used a talk page before. So i have no clue really what i am doing.

Based on the call we still need to address the number of participants and the means of recruiting participants. I don't think has to be long but we do need to address it. I would recommend putting it with the ' Participant survey -- Datasets and studies to replicate' section. I would also suggest changing the heading to 'Participants Recruiting'

  It is our goal to accommodate up to X participants, participants will be recruited via announcements made via XX mail lists as well as personal connections. The participants will be selected based on XXX. 

Something to also considered, I also noticed that we are well over the 3000 word suggested size if we include all of our bios in the count.

For the schedule - for the submission I would not include who is doing which section. Also I won't put a survey on the agenda, we may want it a little more informal as I want the participants to focus on their goals of getting into data, while it is our goal to understand their process by observation.

Reporting out - My vote is having participants report out, no surveys or interviews.

Awesome! Thank you. (BTW, your talk page use is spot on except you should use ~~~~ at the end of your comment to make a signature like this: EpochFail (talk) 22:37, 13 October 2015 (UTC)

The concern about length is very real. I think that we'll need to do some trimming, but I think we can do that pretty well without dropping much meaning. I'll take that on. Also, +1 for reporting out and no surveys.

For recruiting, you're right that we should have a plan. I don't have one. I was just going to post on twitter and reach out to personal connections. Does anyone use mailing lists anymore outside of Wikimedia? --EpochFail (talk) 22:37, 13 October 2015 (UTC)

Yes, academia is still built on mailing lists, AFAIK. Jmorgan (WMF) (talk) 15:04, 14 October 2015 (UTC)

Saving space by not using bios

The CFP says that the proposal needs to describe the background of the organizers, but we can do that a lot more efficiently than we are now. I suggest that we do not include our bios in the proposal. Instead, lets include this single sentence, or something like it, to describe our backgrounds. That frees up, like, 800 words. Our names and affiliations will of course be listed in the masthead of the proposal. Jmorgan (WMF) (talk) 15:20, 14 October 2015 (UTC)

Updates to Metadata & indexes section

I just overhauled the Metadata & indexes section, primarily for length. Here's what I did:

1. changed the title to 'metadata & taxonomies' because 'indexes' has many different technical definitions, not all of which are easy to understand in this context.
2. moved a bunch of content down to the bitbucket section, and attempted to integrate the most important points from that content back into the body of the proposal.

Let me know if you think this was too drastic, or if I've misrepresented something. Jmorgan (WMF) (talk) 21:03, 14 October 2015 (UTC)