Fire Banner

Thursday, July 10, 2014

SiF 2014 – New city, new conference sessions, same old story?

Views from a post-doc

The 8th international conference on Structures in Fire was held in the massive and ever changing city of Shanghai from 11th to 13th June 2014. A large delegation from Edinburgh attended the conference, many for their first taste of this conference series, and many of the delegation had the opportunity to present their research, with 7 oral papers and one poster paper being presented.  The conference attracted over 250 delegates from all around the world (apart, yet again, from Africa).

Bund - Shanghai (Copyright D.Rush)
This year 282 two-page extended abstracts were submitted and reviewed by the scientific committee, with around 150 papers being selected for the 80+ oral and 60+ poster presentations.  All of the research that was presented, whether through a poster or oral presentation, also had an 8-page research paper included within the conference proceedings, which rightly gives all the papers the same level of rigour for a conference striving to present the most current research being performed in the area of structures in fire, and the organisers should be commended for this.

The high number of accepted papers meant that for the first time in the history of this conference series, parallel sessions were introduced during the middle part of each day.  Whilst members of the SiF community may be strongly for or strongly against having parallel sessions, I personally can see the positives and negatives. More papers can be accepted, if their quality is good enough, and it means that sessions can be more specialised. However, parallel sessions also mean that the delegates are split, and as delegates at such a niche conference it can be very hard to choose which session to attend; I found myself popping in and out of the sessions to view alternate presentations and thus missing their starts due to different time keeping in the sessions. Parallel sessions also stopped delegates from engaging in more holistic debates, which can happen in single session conferences.

Conference Proceedings (Copyright D.Rush)
The high number of accepted oral presentations also had a negative effect on the presentation of the posters.  In Zurich for SiF 2012, those delegates with posters had five minutes to present their work, after which they were able answer more in depth questions over tea and coffee.  This format worked really well, but was unfortunately not adopted in Shanghai, probably due to the fact that more time was required to accommodate the larger number of oral presentations. This meant that, although the posters were accompanied by research papers, they were relegated to a consolation status, being presented in a short 40 minute session over tea and coffee.     

All the research papers were accepted for this conference on the basis of a two-page abstract.  This is not a full paper due to the desire of this conference series to include the most up to date research.  However, the 282 two-page abstracts submitted must have taken a lot of time to review by the scientific committee, which aims to have at least 3 reviews for each submission. In the 6 weeks between submission and notifying the authors of the decisions made, the 52 reviewers from on the scientific committee each reviewed on average 15+ abstracts on a wide variety of topics, collated all the reviews, and decided which abstracts to invite to present full papers. This is no easy task and not conducive to high quality control, and as in Zürich, some questions were raised by a number of fellow delegates over the review process and the quality of the oral papers being presented. The main issues worthy of consideration after this conference are the same as in Zurich, and to quote from my previous Blog after the SiF 2012 conference (none of which are unique to the SiF conference series):

1.  Too many papers lacked novel or significant context:
Many of the presented papers struggled to inform the listener of why they should care about the work being presented; there was little context to the work provided in most cases. Even the most engaging presentations and presenters frustrated delegates by not clearly saying at any point why the work was done and why one should care about it. The main question that needs to be answered is who benefits from the research apart from those doing the research?
2.  The lost art of presenting:
Too many of the presentations were difficult to follow and were easy to disengage from with which detracted from the discussions, as instead of collective expertise being used to further research and how to apply it in the real world, questions were being asked simply for clarification rather than for extension.  For the presentations that were presented well constructive discussions generally followed.  There were three trends that could be seen when comparing the good and bad discussions.
i.  Language barrier – Many of those who did not have a good grasp of English, the chosen language of the conference, struggled to communicate their work effectively. In some cases such presenters resorted to simply reading their slides. This problem is admittedly very hard to solve. The post presentation questioning was particularly problematic.
ii.  Hard to see the forest through the trees – many of the presenters had graphs or images, on which they would be making a comment, but rarely did they explain what the graph or image was showing or representing.  For those who aren’t experts in the area (which should be a large portion of the audience if the research being presented is at the cutting edge of the science), there would be no way of understanding what they were looking at.
iii.  Mathematical blindness – as was evident with the images, there were too many equations with not enough explanation, and in some case there were too many equations to even keep track of.  The effect of this was to turn the slide in to a blur of black and white with little apparent meaning.
3.    The works presented in the papers, for the most part, was not directly assessed by review:
With the papers being presented on the basis of only a 2-page abstract, the actual work presented in the 8-page papers was hardly critiqued at all, nor rigorously assessed, but likely only checked for editorial issues.  This led to the quality of many papers being presented being lower than it really should have been be for the SiF conference series. 

One thing that has changed for the better since SiF 2012 in Zurich is that this SiF conference did include a number of delegates from industry and consultancy, with the conference being sponsored by a fire protection paint manufacturer, International Paint Ltd. However most of the delegates and presenters were academics, myself included. Without delegates from consultancy and industry, those who apply the information in the real world and lead us to the critical and relevant areas of research, we fool ourselves that we are doing a good job because we lose perspective and context to our work and presentations. Our work becomes self-satisfying rather than industry leading, and we end up producing redundant, sometimes pointless knowledge, that in practice will never be applied.

So where could we go to improve the SiF Movement?

First, we need to increase the quality of the papers presented, which could be done by either papers selected and reviewed on full paper submissions rather than on 2-page abstracts, and/or increasing the number (or quality) of reviewers to reduce the workload on the scientific committee and improve the quality control. There should also be some guidance/template for the abstracts to help the authors highlight the relevance of their work. 

Second, we need to improve the quality of the presentations, and we could provide some guidelines for the presentations including, amongst other things, stipulations that if you are presenting graphs and equations that you explain clearly what the audience is looking at, and why they are looking at it.  If the quality of the work is good enough and the presentation of it clear, then issues of language are significantly lessened.

Third, to improve the discussions around the cutting edge research being presented, allow access to the proceedings (digitally would be preferable) a week before the conference so those who wish can read papers and prepare questions. We could also change the question time format and have a series of presentations around a theme and then a general discussion on all the presentations rather than each individually.

Fourth, we need to remain relevant to the construction industry.  One way of doing this is to invite our industrial partners to join us and invite them to speak on the subjects that they feel are of major concern to them.  They could be designers, consultants, materials suppliers, architects, practitioners, and we could discuss what matters to them and to the broader building design industry, rather than existing in an isolated academic bubble with few industrialists present.

Finally, as we grow we need the leadership to grow, and I believe that we need new members on the steering committee to help guide the conference in the future. 

If we improve the quality and directly indicate the relevance of the research which is being conducted, improve the way we present it and allow more time for more informed discussions, we can grow, become more relevant, and remain the world leading forum for Structures in Fire.

David Rush

No comments: