droidcon NYC CFP Insights

droidcon NYC Call for Papers ended on June 10th and our Program Committee has been hard at work voting on talks. While they’re busy with that we thought it’d be interesting to analyze stats and let you know interesting insights, as well as answer some commonly asked questions about CFP. Huge shout-out to Nitya Narasimhan for crunching all the data and visualizing it for us to see!

New record for number of talk submissions!

This year we got 347 talk submissions — a record and about 30% more than last year.

How long does it take for Program Committee to vote on talks?

If you spend 5 min on every submission, that’s 1,735 min, which is over 28 hours. Even spending 1 min per submission, it’d take almost 6 hours to go through them all. It’s a LOT of work, huge respect to our Program Committee for doing that!

Key topics this year

Top words used in proposals (must have appeared 5 times or more to make the cut). Gives a really good idea of what the key focus topics were this year.

What topic levels were submitted?

Understandably (or maybe because everyone wanted to play it safe), most of the submitted talks were for intermediate level — 59% of all the talks submitted.

Introductory and overview — 110 talks (32%)

Intermediate — 205 talks (59%)

Advanced — 32 talks (9%)

Is it better to submit intro proposals on new things or deep-dives on old concepts?

It really depends. New stuff is interesting, but there’ll probably going to be other talks submitted for the same thing (e.g. Navigation Controller this year). Deep dives can be great, but you have to go deep and pick something that’s going to have relevance/general appeal.

In general, submit multiple talks and in a range of newness/deep dive/skill level/topics. Don’t put all eggs into one conceptual basket. In fact the data shows that two-thirds of submitters sent in only one proposal which effectively had them compete in only one category (introductory, intermediate or advanced) — by contrast many seasoned speakers (e.g.,DroidCon alumni) tend to fall into the multiple-proposal category.

No of people submitting one proposal vs. multiple proposals

There were about 216 unique authors (where if there was a pair of presenters, they were treated as a single unique author) — and of these 66% submitted just one proposal. Two folks submitted 7 proposals each that were all related. And about 60 people submitted 2–3 proposals, which was kind of interesting.

Distribution of the length of title

4–7 words in title seems to be the norm, with 3 or 8 being kind of the edges to that. 2-word titles like “Advanced Kotlin” or “Functional Android” would probably need folks to read the abstract to learn more.

Gender breakdown

Sessionize doesn’t ask for gender so to get gender breakdown Nitya manually annotated gender based on name/photo and this basically says 56 papers (16.2%) had a female author. It’s slightly misleading in that this does NOT mean 56 women submitted — rather there were 33 unique women authors, many of whom submitted multiple proposals.

Submission timeline

Submission flow into the system is remarkable and a true indication of human nature. You’d think there would be a huge bump after Google I/O where folks figured out the new things — but there wasn’t (well just a small one) — instead the last-minute submission spike is a sight to behold. 95 proposals in 1 day!

We get asked a lot if submitting early or late make a difference. The truth is, it doesn’t, and if you look at the chart below you’ll understand why — most of the people wait till the last moment anyway, so there’s no point in starting voting early. All voting starts after CFP ends so there’s really no difference when you submit.

That’s all folks! Stay tuned for first batch of speakers to be announced!

Don’t have a ticket to droidcon NYC 2018 yet? You can get it here. Early Bird ends June 27th at 11:59pm EST.