2020-12-23 DEC

Trends, hopes and fears 2021

> LIVE <

State of the episphere Union

@/all I’ve been thinking about our conversation yesterday and thought I thought I’d share with you a couple of foundational references by Turing awardee (the Nobel prize of CS) Fred Brooks on the management of sofware engineering. I think you’ll appreciate them since you all see many of these challanges in your daily work. I thought this may as well be this year's state of the union address (just like last year we are due one ;-) ). Please have a look and we’ll touch on this at C4B this coming Wednesday.

The first one is The Mythical Man-Month, around the verification that adding manpower to a late software project makes it later. I’m sending the wikipedia link because that entry is so well organized, maybe better than the original book.

The second one is a paper by the same author that came out in IEEE Computer No Silver Bullet – Essence and Accident in Software Engineering. It’s a bit philosophical but I think you’ll relate to it quickly: software engineering has been evolving from being bogged down by low level accidental complexity to being raised to high level (architectural, algorithmic) essential complexity.

You see this making full circle a few years later with the work of another Turing Awardee, John Backus, when in 1978, Can Programming Be Liberated from the von Neumann Style? A Functional Style and Its Algebra of Programs. I.e. John Backus reminds us software engineering is tendentially (essentially would Brooks say), and by necessity increasingly formally, a type of functional mathematics. That brings us back to Alonzo Church Algebra of Computing (Lambda Calculus). He was of course the PhD mentor of Alan Turing and the circle of this narrative is nicely closed :-) .

So this is my end-of-year little speech. Innovation is an exercise in experimental mathematics which empowers us to both say no to efforts that drag you into accidental complexity, while proposing avenues that say yes to raising constructs backed by working less and thinking more. Amongst other things we are a lot faster that way, and we'll go places.

Please feel encouraged to comment on it here ! We have quite a few accomplishments (and papers !) in 2020, that empowers us to be bold about 2021.

User Research

slide below

Neural Network Norms

FAIR Symbolic Regression and numeric federation of Biostatistics?

Going Serverless

the wider argument by FORBES

Serverless Workflows (thank you Daniel!)

We knew this was coming, well it is here :-)

https://cloud.google.com/workflows

https://youtu.be/Uz8G8fTwwXs

Causal inference

https://www.selectiveinferenceseminar.com

https://sites.google.com/view/ocis/home

Github-MS

They want to mend their open source ways ...

pre-holiday Open Mic

...

muda, mura, muri

...