It’s hard to believe that’s the year over already. I count 215 posts this year including this one – that’s a lot of papers! I hope at least a few of them sparked your imagination. I don’t think many people read 200+ papers a year (especially outside of academia) – and even fewer are mad enough to try and write up notes on them!! But by following this blog, you’ve still been able to take in a quick overview of a lot of research. It doesn’t matter if you don’t read every single post, but I do hope you dip into some of the papers I cover that are outside of your core areas of interest too.

Clearly many of you like to have The Morning Paper delivered to your inbox. I don’t have time to do much other than focus on reading papers and producing the content, but The Morning Paper email list still managed to double in size during 2016. The big investment in time is in reading the papers and writing the posts – sharing the wonderful stories contained in the papers with a wider audience comes essentially for free (well almost, MailChimp do like to charge me more as the list grows, but that’s a price I’m very happy to pay).

If you’ve enjoyed The Morning Paper this year therefore, please help spread the word to your friends and colleagues whom you think might like it. If we all introduced just one extra person to the joys of good CS research, we could double the dissemination impact between us!

The Morning Paper will be back for 2017 in just a couple of weeks, and I’m busy planning some of the content. In the meantime, here are some of the highlights from 2016 (5 per quarter):

• ARIES: A Transaction recovery method supporting fine-granularity locking and partial rollbacks
• No compromises: Distributed transactions with consistency, availability, and performance
• All file systems are not created equal: on the complexity of crafting crash consistent applications
• Not-quite-so-broken-tls: lessons in re-engineering a security protocol specification and implementation
• HyperLogLog in practice: Algorithmic engineering of a state of the art cardinality estimation algorithm
• The amazing power of word vectors
• Deep learning in neural network: an overview
• How to build static checking systems using orders of magnitude less code
• Gorilla: A fast, scalable, in-memory time-series database
• A survey of available corpora for building data-driven dialogue systems
• DBSherlock: A performance diagnostic tool for transactional databases
• Goods: Organizing Google’s datasets
• Flexible Paxos: Quorum intersection revisited
• On designing and deploying internet scale services
• On the criteria to be used in decomposing systems into modules
• Simple testing can prevent most critical failures
• Towards deep symbolic reinforcement learning
• When CSI meets public wifi: inferring your mobile phone password via wifi signals
• Kraken: Leveraging live traffic tests to identify and resolve resource utilisation bottlenecks in large scale web services
• Morpheus: towards automated SLOs for enterprise clusters

And oh so many more I wanted to put on that list!!

What were your favourite papers / posts of 2016 and why? I’d love to hear!