Taken from Brainstorming: Writing a PyCon Proposal. While this post is for PyCon, the US-based conference for Python developers, users, educators, and everyone with an interest in Python, this advice can apply to any language-centric conferences, even the topic suggestions themselves.
Hey you! Ever thought about submitting a proposal?
What? oh no, no no no.
What would I talk about? I have nothing to say!
So how about this:
Tell me what talks you went to at previous years’ PyCons that you found were pretty good (regardless of speaker performance)?
What about: what would you like to see at PyCon this year?
Ok. You want to see those topics?
Why don’t you write the talk you want to see?
Bam, you have an idea. (cheeky lil’ blogger, aren’t I?)
No seriously: if you want to see a talk, write it yourself. Or: take what it is that you do with Python (professionally, hobby-wise, side-jobs, whatevs), and write a talk on it.
There is significant interest in the “Python in the Wild” – such as, Python in (non-technical) corporations (product/services or used internally), government, education (CS degrees, high school programs, communities), and science (NASA, robotics).
There is always interest in popular topics, like deployment, big data, python packaging. Not only do people have short memories, but every year, it will be many people’s first PyCon.
What tends to be popular are talks that compare frameworks/libraries/packages, critique a well-known/used tool/framework, and extreme talks (e.g. we’re getting down to the nitty gritty details of Twisted). Topics that could do surprisingly well are security/cryptography, event-driven networking, and subjects that people should know about and probably default to industry standard, but don’t really know in depth. Very interesting subjects include alternatives to giants like Django (e.g. Flask, Pyramid), or where has Python failed.
There are important subject matters like accessibility within Python, and diversity & community building. Not all talks at PyCon are technical (my head would explode), but are needed and well-respected nonetheless.
When proposing a talk:
Submit early. You know how many proposals the Program Committee has to read through? Not too many right now. There is more time and patience to give feedback, give a second look after the speaker responds/edits. The committee also is not that tired yet. Submit on September 27th? Dead tired. Not really wanting to give feedback; just wanting this voting process over.
Write an outline. Yes just do it. It can be bare bones, but it helps a lot to see where you’re going with this talk, if there is enough meat behind this talk or if it’s too ambitious for the time slot. Don’t write out your whole talk. My n00b confession: I wrote a 2-page single-spaced essay for my proposed talk at OSCON. /facepalm
It also significantly helps the reviewer and you to associate length of time per bullet/subpoint/etc. The reviewer has a sense of where you’re going, if it can actually be a full talk or if it’s too long. It also helps you frame your actual talk when you come to flesh it out (because no one writes the damn talk before they propose it for the first time).
Nervous about actually speaking? Find a partner. Either someone that shares your beginner level of speaking (both hold the burden of being nervous for your first talk) or someone that is already a seasoned speaker (you can relax a little!).
Add links to the ‘about you’/bio portion, or more context in general about why you are the person to speak about this topic. Did a project with Google Summer of Code? Show the link. GitHub repo? Love to see it. Published a paper on the topic? Well it might not be read, the fact that it’s there would give assurance . Remember though, provide the reviewers with more context on why you should be speaking on this topic (not just some random article you wrote about how lame PHP is (but it is lame).). Note: it is not important that you haven’t spoke at PyCon or another conference before. But do prove that why you should now.
Think posters might be a better option for your first time presenting something at PyCon? With a talk, you have about 30 minutes. You can plan out what you’re going to say, how your talk might lead to certain questions (pro tip: leave some unanswered/unaddressed items from your talk, and look awesome when you know the answer if they’re asked). After the talk, for questions you don’t know, you also have the forgiveness of the audience:
“Oh actually, hmm, that’s a very good question. I’m not sure I can address that here right now, but catch me after the talk.”
(and then sprint away…:D)
With a poster, you are defending your PhD thesis (practically), standing around for hours with some of the audience’s expectation of:
“You know everything about this talk and I’m going to grill you because you’re stuck here.”
It can be pretty demanding. But it’s why posters are great for in-depth and/or unconventional topics. There also a second shot at doing something with PyCon if you’re talk doesn’t get accepted since they are due Jan 15th.
I think these Google hangouts that PyLadies did were productive; I got the feedback that they were helpful. I also got the “zomg such a good idea but I can’t make it tonight.” I’d like to see more, perhaps from the Program Committee itself (full disclosure, I am a part of the committee). We have 4 weeks until talk proposals are due. Let’s help you give it your best shot.