papers from the Delft conference

This forum is for discussion of topics related to System Dynamics. You must be logged in to post but anyone who has registered on the System Dynamics web portal may post. Note that there may be a delay of a working day or two after registering before you can post anything.
Richard Dudley
Posts: 65
Joined: Sun Jan 11, 2009 1:17 pm
Location: Etna, New York, USA
Contact:

Re: papers from the Delft conference

Post by Richard Dudley » Tue Dec 23, 2014 11:36 pm

Len and others:

I must say that I am a bit disappointed at this new policy (that of not posting full papers from the conference). I find having access to the full papers helps greatly in understanding a complex problem that cannot be gotten across in a15 or 20 minute presentation. I certainly understand the problem of quality in the presented papers, especially when the society is trying to encourage newcomers to SD. Nevertheless I think that there are other, better, ways of accomplishing this task. (I will be happy to send my ideas... to whom?). I was always under the impression that ISDS papers were not meant to be fully completed, published, papers... but if they are meant to encourage discussion I would assume that we need to be to be able to read them.

I am sure, as Len stated, that a lot of thought went into this decision. But I am not convinced that it serves the society well to avoid web-posting of its own conference papers. Do we have some alternatives?

cheers, Richard

Jean-Jacques Lauble
Posts: 74
Joined: Fri Aug 23, 2013 3:49 pm

Re: papers from the Delft conference

Post by Jean-Jacques Lauble » Wed Dec 24, 2014 3:12 am

Hi Richard.

You can post your further ideas on this forum where you can include a supplementary file if you fear that the message might be too long. I think that this question is relevant to all the members of this forum and more generally to all SDS members and all people susceptible to be interested by the conference papers.

If you do not want to post at the forum, I am still interested.

My e-mail is jean-jacques.lauble@estvideo.fr

Best regards.
JJ.

Leonard Malczynski
Posts: 96
Joined: Fri Jan 16, 2009 11:12 am
Location: Albuquerque, NM USA

Re: papers from the Delft conference

Post by Leonard Malczynski » Wed Dec 24, 2014 11:46 am

Jean-Jaques and Richard,

Thanks for your comments and concerns.

The folks responsible for the decision are the Society Program Oversight Committee.
It has several responsibilities including thread chair selection, program chair selection, and overall program quality.
I am currently serving as the chairperson so email me directly at lamalcz@sandia.gov.
We had feedback from many Society members on this publication policy.

Hopefully I can address some of your questions and concerns.
The history that lead to the decision included criticism of paper quality, presentation quality, and model quality.

- Some Society members and/or attendees have voiced concern that conference quality influences their decision to attend.
- In addition there is a perception that placing incomplete or lower quality work on a public website does not support the mission of the Society.
- We also had an issue, raised by an unnamed journal, that a conference proceeding was indeed a publication and that work presented at our conference could not then be submitted to a journal, especially a journal other than the SDR. This is why conference submissions should be full papers and conference publications should be abbreviated.
- Some potential attendees cannot get their institution to fund their attendance without an acceptance of paper/poster/presentation.

There were other considerations too including conference financial viability.

The policy you have seen had the primary goal of improving the quality of submissions.
There are at least 3 other initiatives ongoing to address quality (models, papers, presentations) - details will be out soon.

As for models, I believe it is too much to ask reviewers and thread chairs to review models and there are logistic problems too.

Our hope is that interested persons will contact authors directly; not only to obtain models and papers, but to advise and collaborate.

Please don't hesitate to keep this discussion going especially if you have ideas to support the Society by improving the quality of our work.
Len

Richard Dudley
Posts: 65
Joined: Sun Jan 11, 2009 1:17 pm
Location: Etna, New York, USA
Contact:

Re: downloadable conference papers - a good idea

Post by Richard Dudley » Thu Mar 12, 2015 11:26 am

Sorry for the long delay, but here are some additional comments about the advisability of making SD conference papers available for download.

I think that papers that are to be presented at the SD conference should be available for download prior to the conference for the following reasons:
1) This will allow us access to more papers than would be physically possible to "see" at the conference.
2) Would provide access to "threads" that one doesn't normally attend.
3) It would allow us to get more detail of the thinking and background for the presented papers (more than can be squeezed into a 15 or 20 minute talk).
4) Will help us decide which papers we would like to attend at the conference.
5) Will give people who can't attend the conference a chance to participate and learn.
6) It will allow us to see, and examine, models used.
7) It will provide additional opportunity to have meaningful feedback from other society members.

Objections to having the papers posted seem to center around two issues: A)quality of the papers and models, and B)the difficulty of subsequently publishing material that some journals might then consider already published material.

(I note that these are clearly separate issues since if the quality is poor then the paper is likely not something that would subsequently be published without change.)

While the quality of papers could be improved by careful review, this is difficult for two reasons. Firstly the large quantity of papers to be reviewed in a short time and the nature of the review process limits the effectiveness of that process, and as Len pointed out, often the models can not be given more than a cursory review in the time available. Secondly the nature of the meeting is to encourage newcomers who may not, as yet, be good modelers. These are both issues that are not easily overcome.

There are two solutions that seem reasonable but would still allow access to the papers:
1) All papers are labeled in the upper right corner "Preliminary Paper, for Discussion only. Please feel free to provide comments to authors name at authors email address."

In any case, in reality, papers presented at the conference should actually be, for the most part, works in progress for discussion. Isn't this the idea of the conference? The alternative here is to get the paper accepted prior to the conference so it can be presented as "breaking news". If the paper prepared for presentation really is identical to one that has been, or will be, submitted to a journal then it shouldn't be a conference paper anyway, and hiding that fact is not the answer.

2) Perhaps more acceptable, but much more limiting, would be to limit access to the papers to members of the SD society. Login via the Web Portal, or some such arrangement.

I hope I can read your papers before I get to Boston! ;)

Jean-Jacques Lauble
Posts: 74
Joined: Fri Aug 23, 2013 3:49 pm

Re: papers from the Delft conference

Post by Jean-Jacques Lauble » Sat Mar 14, 2015 5:01 am

Hi Richard

Here is summarized what I would like the conference to be like.

1. There are exactly 6 weeks between the tentative program schedule and the start of the conference.
This is not much but it is possible to manage this short time at several conditions.

2. That schedule should be complete and not plenty of papers missing and only the abstract available.

3. A model that can be run with a player for example should be attached. If this condition is not respected the author should explain why and be credible.

4. Every single model included should satisfy at least two conditions if the software used allow it
(they generally all allow it): dimensional consistency and if needed a test that verifies that masses flow balance. Dimensional consistency makes the reading and understanding of a model much easier which is a necessity taken into account the large number of papers presented that must be eventually consulted.

5. The parallel sessions where generally the authors present their paper, should last longer than 30 minutes. I cannot say how long, but you cannot do anything serious in 30 minutes.

6. The objective of the session should be to answer questions from attendees who have previously fully studied the paper and used the model.

7. It should be possible for the attendees to send questions to the authors before the conference so that the authors can choose the more interesting and even post the main questions asked in the supporting text in the program schedule. Of course this is easier to do if the 6 weeks delay is longer.

8. To my opinion all the previous conditions can be respected only if the number of papers is severely reduced.

I do not expect all these conditions to be met at once. One can start by one and implement all the others progressively at later conferences to profit from the previous experiences.
This would make the conference attendee (client) oriented instead of being presently author oriented (producer). In any business the former works better than the other one.

Best regards.
JJ

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests