Thursday, June 8, 2017

What about conferences? What should we (CS area) do?

A friend of mine asked me this question on Twitter:
What about conferences? What should we (CS area) do?

I do not think there is a single answer to that question. Rather than that, I will try to organize a few thoughts, personal and subjective thoughts, about it. Please bear in mind that these are guidelines for myself, I do not intend them to be general rules.

First of all, what is a scientific conference?

For me, they it is a meeting where people who work in a similar field gather to discuss what they are doing and/or what they intend to do.

Just that?

Yes!

It may sound trivial, but not for me. Most of my academic trajectory can be explained by high-impact encounters I had at scientific meetings.

How many of these meetings are considered "high impact" by part of the Brazilian CS community? None. Zero.

Let's go back to my previous humble definition of a scientific conference, and concentrate on people gather to discuss. This is core to my line of thought.

What kind of research results I find worthy of presenting at conferences? Preliminary results, even seminal. I go to conferences to discuss face-to-face, to gather feedback. Such feedback is fundamental to identify flaws, and to have insights that may lead to an improved research product.

I consider journal papers are my research products, not conference papers.

But why?

The whole process of selecting papers for the conferences I enjoy going to is completely different from that of the Journals I aim at publishing with. Let's take IGARSS 2017 as an example of the former, and IEEE Geoscience and Remote Sensing Magazine as an example of the latter.

IGARSS receives submissions in the form of extended abstract of at least two pages. Each paper receives very brief reviews, and a decision. About 80% (approximately 2000) of submissions are accepted, and those rejected tend to be out-of-scope, disorganized beyond hope, or illegible. Some editors reach 90% of acceptance.

This may sound totally unacceptable for those who advocate for highly selective conferences. People who think like this measure the quality of conferences by the rejection rate: the higher, the better. Fine with them, but not for me.

I do not care how many papers are rejected out of those that are submitted, as long as there are enough good papers to make the conference attractive. As said, the face-to-face experience is what I am looking for in good conferences.

I do not choose by selectiveness either when I submit my manuscripts to a journal. I choose by the visibility it will give to my research, and this visibility is formed by a number of factors, among them:

  • quality and number of reviews
  • academic reputation
  • quality of editorial preparation
The IEEE Geoscience and Remote Sensing Magazine is an example of excellence in all these, and many other, criteria. Do these criteria make a journal selective? Most likely yes, they do, but selectiveness is not the criteria, it is the consequence of quality.

So, my dear friend, after this long detour (that may increase in the future, provided more ideas, time and patience come at hand), I will conclude summarizing my point:

About scientific conferences

I love them, I go to a few of them as often as I can, but I do not aim at publishing in them. I aim at participating in them. What I send for presentation, my conference papers, are preliminary results that I want to show to a big qualified audience. The feedback will lead me to either discarding the approach, even the research, or to an improved manuscript that I will eventually send to a journal

About scientific journals

They are the venue for consolidated (not definitive, as I practice empirical science most of the time) research results. They are the memory (dynamic, not set in stone) of scientific research. What is published in them has passed a typically careful (sometimes long, sometimes painful, most of the time enlightening) process of improvement. I admit that only ISI journals interest me, because they have quality measures that help me choose where to send my works.