Data science is a world that doesn’t seem to ever stay put. In the blink of an eye dozens of open source projects are launched, companies develop new data products or find new use cases and practices for machine learning. Many awesome, and exciting, meaningful, and groundbreaking, or just really fun projects are shared by data enthusiasts everywhere daily. Every day, experts deliver knowledge constantly at conferences and meetups around the world. So how can you keep up?
Because of this constantly growing flow of data news, keeping up has become an art. Data experts everywhere collect articles, share them on sites and compile digests every week to get everyone their regular data fix. Keeping up with data newsletters and medias in itself is becoming a job in itself!
I know this first hand because I spend a few hours every week putting together our curated data science newsletter: Banana Data News, along with it's very own Tumblr and Twitter account. Even if I'm lucky to have an awesome team at Dataiku helping me find great content every week, it's a lot of work. So here are a few tips we've put together after spending a few months trying to keep up-to-data!
First Thing's First, What Kind of Data News Are You After?
One of the main challenges is realizing that there are lots of different types of data news out there, and not all of them relevant to what you’d like to know about. You won’t look in the same places if you want to read interesting data journalism articles, be updated on the advancement of open source projects, find creative data visualizations, download research papers on advances in machine learning, or get articles on how to successfully build a data department.
Finding relevant information also depends on what level of information you’d like. You won’t go to the same sources if you want long analytical essays on the impact of artificial intelligence on society, or if you want to find fun projects you want to be the first to share with your colleagues.
Basic Tools to Get Your Data Science News Fix
After these preliminary considerations, let's dig into tools to to get your data news fast.
Google Alerts are useful for collecting EVERYTHING. Just set them up with your favorite keywords and watch them come in daily or weekly. The fact that you make no selection means you don’t miss out, but you have to make the selection yourself.
Specific threads on reddit (aka subreddits) are also a goldmine, if you have the patience to go deep (check out data is beautiful, machine learning, or even data science).
Twitter lists are going to be a great way to constitute live streams of information, from companies or experts who specialize on communicating on these subjects.
You can look for lists of influencers to follow in the fields you’re interested in, and once you’ve gotten one guy you like, check out who he retweets, or even who he follows to find more and more sources. You should also search for hashtags linked to the fields you like (like #dataviz, #dj or #ML for example). You can also trust the Twitter algorithm and check out its follower recommendations. This is pretty obvious stuff, I know. The point is, once you’ve gotten a hold of a little bit of interesting thread, the ball of information yarn rolls out on its own!
If you want to make this process more efficient, you should look into setting up a few recipes in If This Than That. For instance, automatically creating a list from people who’s tweets you’ve liked is very easy to set up!
And speaking about algorithms, never undermine the power of pocket or Noosfeer to suggest relevant or surprising articles to you. Because they do make things easier.
Going Deeper With Data Science News Sites
Another solution of course is to go to data science news sites. There are several of these specialized media sites, like KDnuggets or Data Science Central, that repost successful content or publish original articles. O Reilly has fewer articles as well, but good ones. Smart Data Collective and Inside big Data are also great sources. The equivalent for data visualization are sites like Flowing Data or Information is Beautiful.This is a also great one for artificial intelligence and machine learning projects: Creative AI.
Here’s the thing to keep in mind though. These popular sources are just that, popular. This means:
- a lot of people read them so you you’ll be getting the same news as your colleges,
- they often repost content from other sources, a little later, so you won’t get the freshest news,
- they are businesses, who sometimes make money by publishing sponsored content, and that’s fine because companies write great content as well, but you should be aware of it!
These reasons don’t make these sites any less valuable of course. You’ll always find great reads there!
Data Newsletters and Playing With Feeds
The next step is to subscribe to these guys’ newsletters, as well as other specialized newsletters. You’ll actually find a good deal of data newsletters out there, centered around different topics. This article is a list you can start with, even though it’s a little bit outdated. My personal favorites are data elixir, and of course banana data news! But there are lots of smaller ones you'll stumble upon once you start looking.
this way, you’ll get lots of great stuff, straight to your inbox! Honestly newsletters are the best. Signup to all of them, and only keep the ones you like.
Quick tip though: if you’re using a gmail account, signup with youremail+datanews@gmail.com and create a filter to send these to a specific file. Okay so maybe everyone already knows that tip, but hey, you never know.
If you want to go even further, feedly is a great tool. It plugs into rss feeds so you can get content from pretty much everywhere and organize it. There are some pre-existing data feeds by serious data guys that you can add. These are great because you go straight to the source, and track what lots of small blogs are putting out there! You can also set up your own feedly, with the guys you liked on twitter, and the sources of articles from your favorite newsletters (sneaky).
To wrap this up, the best tip I have really, is to keep your ears open, and discuss things with your friends and colleagues. That's where I always find the coolest data news, my awesome teammates!
If you have other tips, I’d be extremely happy to hear them: alivia(at)dataiku.com
