What is a good poster for a math conference? I'm going to participate to a conference and they ask me to do a poster on my research. I've never made a poster for a conference/seen a poster session in a conference. So what is important? What do you want to see in a poster (references, basic definitions and ideas)? What are the things that must not appear in a poster (completed proofs?)? What is a poster session?
PS : My subject of research is about PDE (theoretical point of view : existence of a solution, regularity, unicity) and I'm going to a conference about Numerical Analysis (mostly) and PDE.
 A: I learned in a course on scientific writing and communication that sometimes when someone tries really hard to make a poster aesthetically pleasing, the aesthetics end up being visually distracting from the main content.  An analogy I once heard for this is that a good poster should have a high signal-to-noise ratio.
A: Know your audience so that you can communicate to them most effectively.
Text should be large enough to be seen from 5 feet away.
The pieces should be organized in a way that leads the viewer through the display.
Make illustrations simple and bold.
The display should be self-explanatory so that you are free to talk.
Keep displays simple and text brief; a viewer should "get it" in 30 seconds. 
You can provide in-depth information in a handout.
A neutral colored poster on matte board is more pleasing to the eye than one on a bright colored background.
Organize your material and edit your content to eliminate distracting visual noise. 
When in doubt, edit out; make sure every item is necessary.
Take a note pad and pen for notes, extra thumbtacks, pins, tape or glue.
Here are some templates of posters. you can use of online latex to make a nice poster with Latex . https://www.overleaf.com/gallery/tagged/poster
And http://www.latextemplates.com/cat/conference-posters
Here is good advise to make a scientific  poster http://www.personal.psu.edu/drs18/postershow/
A: I recommend the highly entertaining "Guidelines on Graphics" on pp 91-97 of http://anorien.csc.warwick.ac.uk/mirrors/CTAN/graphics/pgf/base/doc/pgfmanual.pdf
A: According to my opinion, a poster should:
a) give rough ideas and hints rather than details.
b) be the starting point of a mathematical conversation. Make the people interested in your topic. In the best case there is an obvious good question that people will want to ask after scanning the poster.
c) make clear the take home message. What do you want people to remember about your work?
d) contain as little distraction as possible.
If not all conference participants know you already, then I think it is nice to include a photo of yourself. It makes you easier to recognize as the author of that particular poster.
