<div dir="ltr">Thanks for the summary of responses, Ivan. <div><br></div><div>Just thought I'd jump-in on the last noted point about AWS S3: it is tricky indeed, but our aim with OSF is to make it much more accessible. In OSF, you simply need to create your AWS S3 account, connect it to your OSF project, and then you can interact with it drag-and-drop through OSF. AWS S3 will be the storage, but OSF can provide the interface. Happy to demo it for you if that would help. Just let me know.<div>
<br></div><div>Andrew</div></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Mar 27, 2014 at 5:04 PM, Ivan Gonzalez <span dir="ltr"><<a href="mailto:igonzalez@mailaps.org" target="_blank">igonzalez@mailaps.org</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto" style="word-wrap:break-word">Hi,<div><br></div><div>Thanks to all for your comments! We had some delays with our paper, but finally is posted in the arXiv. We plan to submit it to a real journal next week. Regarding the data, we haven't taken a final decision yet. As for now, we are adding our data to a GitHub repo [1]. We will add this repo to OSF or move all to Figshare (which might be required by our publisher.)</div>
<div><br></div><div>I'm making a brief recap of what I found out for future reference:</div><div><br></div><div>- in general, I find the whole process pretty difficult for an average scientist. I guess it's natural as all the tools are relatively new.</div>
<div><br></div><div>- the approach I liked most is OSF (<a href="https://osf.io" target="_blank">https://osf.io</a>), which looks based (?) in GitHub, making easier for GitHub users to jump in. A killer point is that you can add other GitHub repositories or AWS S3 buckets, so to putting your project there is just adding your code and data in the form that they already are. It displays, among many other filetypes, iPython notebooks, which is a good way for people to interact with your data.</div>
<div><br></div><div>- Figshare (<a href="http://figshare.com" target="_blank">figshare.com</a>) is pretty similar to OSF and looks pretty nice, but I didn't find a way to add many files easily. With OSF, I can just put them in a repo with one command in the command line and then add the repo as a whole to OSF. </div>
<div><br></div><div>- Dryad (<a href="http://datadryad.org" target="_blank">datadryad.org</a>) also looks nice, but it seems focussed to material that has been already published, or at least in its final form.</div><div>
<br>
</div><div>I'll keep you posted on how all this ends up.</div><div><br></div><div>Best,</div><div><br></div><div>Ivan</div><div><br></div><div>[1] Andrew recommended using AWS S3. I think it would be more suited too, but man, you need a couple of PhDs in Computer Science to figure out how it works. I tried, but gave up after half an hour... </div>
<div><br></div><div><br></div><div><div style="direction:ltr"><div>El 10/03/2014, a las 07:43, Neil Chue Hong (SSI) <<a href="mailto:N.ChueHong@software.ac.uk" target="_blank">N.ChueHong@software.ac.uk</a>> escribió:</div>
<br><blockquote type="cite"><div><div class="h5"><div dir="ltr">Hi Ivan,<br><div class="gmail_extra"><div class="gmail_quote"><br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div style="word-wrap:break-word"><div>- I've heard about Figshare, which looks good, but don't know exactly how it works, in particular whether people can interact easily with the data. Any experience with that? </div>
</div></blockquote><div><br></div><div>FigShare have been continually developing their tooling since their launch to support different types of file formats. For CSV files, they provide a couple of simple viewers: table view and variable view. These automatically identify things like column headers, but are obviously not as functional or flexible as a specialist viewer.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div style="word-wrap:break-word"><div>- What about a combination of Figshare and a Github repo? What about Fidgit [1]? </div>
</div></blockquote><div><br></div><div>My understanding is that Fidgit is going the other way round, i.e. making an explicit link between source code in a GitHub repo and source code deposited (and therefore assigned a DOI) in FigShare. </div>
<div><br></div><div>What it seems you're looking to do is the "traditional" method, where the code resides in GitHub, and the data in FigShare, and the tools you've created can easily access the data. </div>
<div><br></div><div>I believe that the current FigShare API [1] doesn't support direct read of datasets (it's a repository, rather than a filesystem) but I haven't tried using it, so you may want to get an opinion from someone with experience.</div>
<div><br></div><div>I guess the main question is how you expect people to "manage" the datasets you're providing. If the expectation is that they will download certain datasets and the tools to their own machine, then storing the data in GitHub or FigShare both seem like reasonable options. If they have to download many datasets along with the tools, then you might find it easier to have everything in one repository (presumably GitHub). A final alternative, if you wanted to make it easy for people to explore the data without downloading is to store the data on a cloud storage service and make the tools available via a cloud instance as well. This would be in addition to storing the "source" code and data in a repository.</div>
<div><br></div><div>best regards</div><div>neil</div><div><br></div><div>[1] <a href="http://api.figshare.com/docs/index.html" target="_blank">http://api.figshare.com/docs/index.html</a></div><div> </div></div><div><br></div>
-- <br><div dir="ltr">
Neil Chue Hong<br>Director, Software Sustainability Institute<br>EPCC, University of Edinburgh, JCMB, Edinburgh, EH9 3JZ, UK<br>Tel: <a href="tel:%2B44%20%280%29131%20650%205957" value="+441316505957" target="_blank">+44 (0)131 650 5957</a><br>
<a href="http://www.software.ac.uk/" target="_blank">http://www.software.ac.uk/</a><br>
<br>LinkedIn: <a href="http://uk.linkedin.com/in/neilchuehong" target="_blank">http://uk.linkedin.com/in/neilchuehong</a><br>Twitter: <a href="http://twitter.com/npch" target="_blank">http://twitter.com/npch</a><br><div>
ORCID: <a href="http://orcid.org/0000-0002-8876-7606" target="_blank">http://orcid.org/0000-0002-8876-7606</a></div>
</div>
</div></div></div></div><div class="">
_______________________________________________<br>Mozillascience mailing list<br><a href="mailto:Mozillascience@mozilla.org" target="_blank">Mozillascience@mozilla.org</a><br><a href="https://mail.mozilla.org/listinfo/mozillascience" target="_blank">https://mail.mozilla.org/listinfo/mozillascience</a><br>
</div></blockquote></div><br></div></div><br>_______________________________________________<br>
Mozillascience mailing list<br>
<a href="mailto:Mozillascience@mozilla.org">Mozillascience@mozilla.org</a><br>
<a href="https://mail.mozilla.org/listinfo/mozillascience" target="_blank">https://mail.mozilla.org/listinfo/mozillascience</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br><div dir="ltr"><div>Center for Open Science (<a href="http://centerforopenscience.org" target="_blank">http://centerforopenscience.org</a>)</div><div>Partnerships, Collaborations, and Funding</div>
Twitter (asallans)<div>Skype (andrew.sallans)</div></div>
</div>