FOSS4G'13

The working site for the conference committee of FOSS4G 2013

Abstract/Paper Submissions Systems

Posted by Barry Rowlingson on September 11, 2012

We're discussing possible systems for submission of papers and refereeing for the Academic Track. We should be able to get something on the OSGeo servers if we want to run it ourselves. However, do we use something that will handle AT papers and Main Track abstract submissions? I don't like the idea of running two services where one will do.

So, as AT member I ask:
I'm pretty sure that either OCS or Indico can be coerced into handling both an anonymously refereed process for the AT and a less formal review process for MT submissions, but we need to get a handle on the workflow.

Comments

Barend Köbben on September 11, 2012:

Thanks for that info Barry. Completely agree on not using two systems. That also seems much better for keeping content of past years archived and searchable.
Clearly you are the better man for setting things up and fiddling with the software, but if you can bear with some stupid questions asked from the sideline, I'd like to be in{volved|formed} too in setting things up, for the sake of future Foss4g ATs.

Steven Feldman on September 11, 2012:

Rollo is the lead on the programme for the main tracks and the paper selection process so he should respond on how he envisages running it.

Barry Rowlingson on September 11, 2012:

Cape Town in 2008 used an OCS instance hanging off the OSGeo servers:

http://conference.osgeo.org/index.php/foss4g/2008

this was for submission to both tracks, assembling the programme, and even
had a registration system for a while until they moved to a system run by
the venue team (I think).

It also has announcements and info and sponsors logos and can be styled
with bits of CSS. What worries me now is having that and the Wordpress site
duplicating things.

I'll submit a ticket to the OSGeo admin team to get access to the server so
we can create a 2013 OCS instance for experimentation at first.

Rollo Home on September 11, 2012:

I'm looking at:
- the programme format for the MT
- the theme and topic areas for the main tracks (the AT will not be interested in these 'constraint's I suspect - although I don't want anyone to feel constrained, more 'guided' by them).
- the outline timeline for calls etc.. (but needs to fit the schedule as outlined at the F2F meeting)

Submission process:
- I'm happy with the concept that is proposed above for the actual submission handling. I will need to look at the detail of the systems. Is Jo on board (I've added her to this conversation)?

Review process:
- selection is based on the AGI process, with a 'community' component (to be agreed, and the methodology to be reviewed)

Jo Cook on September 11, 2012:

Happy to use an existing system, please just make sure that you give us enough time to link from/integrate with the existing website. I'm pretty snowed under for the next two weeks but after that should find it easier. I can give other people access to the web server if required but I will need to figure out how to do that (it's an amazon instance with pem files for sshing)

Jo

Barend Köbben on September 12, 2012:

The EasyChair website is back online. I already knew it as an end-user (both author and reviewer)  and liked it in that respect. Checked some more and what I like is its relative simpleness: hosted (no server hassle), concentrates on submission & review, no possibilities for  website with schedules, tracks, proceedings online etcetera, such as OCS and Indico have.

My idea is that it would be a good option if for some reason we do the AT track submission separate from the overall submission/web system.  In other words: if I myself would be on my own in having to manage a system, this is the simpler/safer one... But if others are ready to take care of the more involved managment of Indico, I am fine with that.

Question: Should/could  Barry and Rollo (+Jo?) make a decision on the appropriateness of Indico for the overall system? And if so, when could that decision be made?

Rollo Home on September 13, 2012:

I'm not sure that the decision of whether or not the MT and AT are merged should be based on the software. Is it not more of a case of what we want to achieve with the CfP's? In my mind the AT has a very specific task/time-table that differs from the MT which would lead me to think a separate CfP process is best. I understood that was what the AT committee were thinking? I believe that we can be very flexible with the MT process/schedule - the hardest element being how we handle the 'public vote'. 

Barend Köbben on September 13, 2012:

Agreed. To clarify: my point was not about merging MT and AT, but about using the same software for both. My proposal is that MT+ Barrry (included because he would look at the possiblities of having this hosted somewhere useful) look if they could/would use the Indico software for the website and the MT submission system. If they do decide to use it, AT just goes along with that, because it fits our needs. If not, we will "go our own way".
Does that make sense...?

Barry Rowlingson on September 13, 2012:

Exactly that. Except that since OSGeo already has a running OCS instance I
suggest we investigate that before Indico. It would mean no need to setup a
server and sysadmin it (OSGeo SA group admin the servers).

Implications include:

* Submitters may have to specify a track for their submissions, or we have
to filter them manually
* Submissions will be full papers for AT and abstracts only for MT [yes?]
[so the s/w needs to handle this, shouldn't be a problem]
* Integration between the Wordpress site and the OCS site. OCS can give
you a comprehensive conference front page, but we seem settled on
Wordpress, so its probably just a matter of hyperlinks from one to the
other...

Barry Rowlingson on September 15, 2012:

Having played with Indico for a bit I've hit on a problem which might apply to OCS too - anonymity. Reviewers in Indico get sent to the details page for each submission, and this lists the authors and affiliations - I see no way to turn this off.

The workflow for Indico also appears to be two-stepped - first you submit an abstract, then that has to be accepted before you can upload the paper, and then the paper is refereed before it gets accepted into the conference.

This got me thinking that what we are trying to do in the AT is not a conference but a journal, and that we should use OJS instead of OCS. This lists double-blind reviewing as a feature - which I think means reviewers don't know who wrote the paper and authors don't know who reviewed it.

What did the last AT people use? And was it double-blind reviewing? And is that a strict requirement for us anyway?

Barry Rowlingson on September 17, 2012:

Just been experimenting with OCS on my desktop while waiting for the OSGeo sysadmins to give us access. Install was straightforward.

it can do blind anonymous reviewing, so that's good. The admins can put in a checklist for authors, so you can include '[ ] I have removed my name from the title page' as one of those.

Can we use it for the main track as well? I think yes, but with some limitations...

OCS divides the world up into conference series (eg FOSS4G) then conferences (eg 2013 Nottingham) then tracks (Academic Track, Main Track). But the tracks have to have a common submission policy and review process.

The submission policy controls what materials are submitted with a submission: abstract only; paper only; abstract + paper; abstract followed by paper after abstract acceptance (2 review rounds). On OCS forms an abstract is a text box, a paper is an uploaded document.

For the academic track I assume we want paper or abstract + paper. Main track may just want long text abstracts? Rollo?

One way round this would be to have two separate OCS "conferences", or to have all submissions by uploaded document.

If anyone wants a login for the 2013 Space Otter Conference let me know and I'll get you the URL.