The Discovery Decades: April 8, 2018 Snippets
As always, thanks for reading. Want Snippets delivered to your inbox, a whole day earlier? Subscribe here.
This week’s theme: some parallels between the early decades of the computer industry versus modern biology; plus Slack’s senior leadership team gains a new member.
Welcome back to our Snippets series on synthetic biology, where we’ve explored some essential similarities and differences between computers and living systems. This week, we’re going to get to a long-promised question: if the synthetic biology industry may one day create value at the same scale as computers did, then where are we in that journey? Where are we, in “computer history years?” Are we in 1950? 1960?
There’s no real answer, if course, because the challenges faced by these budding industries were quite distinct. We already looked at one essential difference: the idea of equilibrium, and how computers vs living systems process information and get around the problem of noise in a fundamentally different way. Plus, there’s another essential difference that’s important to point out. We invented computers; biology evolved on its own. We forward engineered computers, software, and the internet — but we reverse engineered our understanding of biology by studying organisms and systems that already exist. Still, if we want to actually compare synthetic biology with software engineering, which we’ve done in a previous issue, then we need to add a layer of nuance to our story that makes a big difference.
Imagine for a minute a world where there is no field of computer science and no practice of software engineering; yet laptops and iPhones and Raspberry Pis populate the earth naturally. We appreciate them and we use them, but we do not understand how they work, or how to manipulate them to perform new tasks and new applications.
Over time, researchers and inquisitors begin to crack open these computers and examine what’s inside: we discover code, although there isn’t any documentation. We discover something that we call an “operating system”, although we do not understand how or why it is built the way it is. But eventually, we learn enough about computers to begin to write crude applications, wrangling their power to do simple tasks for us that work some of the time in a “Black Box” way: we fully understand why or how it is working, but we’re able to wrangle them into particular conditions where they sometimes obey our commands. In doing so, by trial and error, we learn more about what makes them tick; eventually, we decide that we’ve learned enough about the fundamental principles behind computing that we’re going to try to build our own from the ground up. Those two activities — reverse engineering the wild population to do things we want for us, and forward engineering our own from scratch — together form the essence of synthetic biology.
Nonetheless, despite this admittedly imaginary story around discovering “Wild Type” computers as a parallel for biology, the early decades of modern biology do have some very real parallels with the early decades of computers.
Computers in the 1930s/ Biology in the 1960s: This was the decade where we were introduced to the “big idea” off of which everything followed. With computers it was Claude Shannon’s insight that On and Off could be represented by 1 and 0, and therefore digital circuits could be used to perform Boolean logic. With biology it was what we call the Central Dogma: DNA is transcribed into RNA; RNA is translated into protein. Everything since has flowed, for the most part, from these founding concepts.
Computers in the 40s and 50s / Biology in the 70s and 80s: These were the decades where the early, thankless work of dredging through the early technical muck was accomplished. Either directly or indirectly, the government paid for almost all of it: for the computers, it was for military applications via the DOD; for biology, it was in health care research through the NIH. It’s a good thing they did, because nearly all of this fundamental research and development was too early to commercialize: although there were a few big commercial success stories (IBM; Genentech), most of the work and the output accomplished during those decades was unprofitable and needed continuous subsidy from the government (whether through military or health care budgets) to maintain footing.
Eventually, though, computers begin to do “serious work” for big customers in a commercial, profit-driven setting. The big computing leaders were “IBM and the Seven Dwarves” (Burroughs, UNIVAC, NCR, Control Data, Honeywell, RCA and General Electric), and their big mainframe machines and purpose-built enterprise environments (eventually the IBM System/360) came to define that chapter of the computing era. On the bio side, we eventually reached a similar period where synthetic biology and artificially engineered cells began to carry out real workloads for “serious applications”: making drugs for the pharmaceutical industry. They were the eras of big companies, big budgets, big contracts, and Boston.
But the biggest parallel between the developments of the computing industry in one era and the molecular biology industry on the other, which has led to both astute and wildly incorrect comparison alike, is comparisons between the falling cost of computing driven by Moore’s Law, and the following cost of genetic sequencing since the 90s. And the critical turning point there came in 1990, with the launch of the Human Genome Project. Next week, we’ll talk about why the Human Genome Project was both a monumental achievement but also in many ways a colossal waste; and yet, in the true spirit of the innovation economy, it couldn’t have happened any other way.
“Not in my backyard” is a big problem, but “Yes in my backyard” has some unanticipated consequences of its own:
San Francisco has a people problem | Nour Malas & Paul Overberg, WSJ
In pricey cities, new residents make much more than those leaving | Laura Kusisto, WSJ
YIMBYism and the cruel irony of metropolitan history | David Levitus, Streets Blog LA
Some pretty incredible writing on emergence as a complex phenomenon of very simple starting instructions:
The amazing, autotuning sandpile | Jordan Ellenberg, Nautilus
The best books to understand complex systems | Taylor Pearson
Conway’s Game of Life | Stanford University
Turing Machines and Conway’s Dreams | Jeremy Kun
And other emergent phenomena we encounter in the real world:
Deep laziness | Sarah Perry, Ribbonfarm
Why capitalism creates pointless jobs | David Graeber, Evonomics
Bug Problems:
Global insect declines: why aren’t we all dead yet? | LessWrong
Where have you hidden the Cholera? | Rowan Moore Gerety, Longreads
This is a bit random, but I think it’s important to include because of how instructive and high-quality it is. If you know people who use Twitter, they’re usually in one of two categories: either they really use it (and spend close to their entire life on it), or they use it lightly but never really found the appeal beyond skimming the news. How can you, too, fall headfirst into the pit of joyful noise that is Twitter? Nikhil has you covered:
Why Twitter is dope and how to use it | Nikhil Krishnan
Other reading from around the Internet:
Revisiting The Breakfast Club in the age of #MeToo | Molly Ringwald, The New Yorker
New source of global nitrogen discovered | Randy Dahlgren et al., Science
Announcing Allraise.org | Aileen Lee
The end of Windows | Ben Thompson, Stratechery
What happens when an algorithm cuts your health care | Colin Lecher, The Verge
In this week’s news and notes from the Social Capital family, there’s been a steady stream of news over the last few months — plus one big announcement this week — from Slack.
As Slack accelerates into a new phase of growth while maturing into an established business, they’ve been busy filling out their senior leadership team. In February, Edith Cooper — most recently the global head of human capital management at Goldman Sachs — joined Slack’s board as their second independent member, following Sarah Friar of Square joining last year. Longtime employee Allen Shim was also promoted to CFO, bringing more homegrown talent into their senior management team.
On Wednesday, the Slack family officially saw another important and well-deserved promotion: April Underwood, Slack’s VP of Product for the past few years, has officially been promoted to Chief Product Officer:
Exclusive: Slack promotes April Underwood to Chief Product Officer | Michal Lev-Ram, Fortune
April Underwood is now Slack’s Chief Product Officer | Matthew Lynley, TechCrunch
As a veteran of both Google and Twitter as well as the last few years of Slack’s product development, April’s new role means an expanded mandate over what Slack’s “product” has become. Although team communication through chat will always be the foundation on which the product has value for its users, Slack is really evolving into the “Operating System for a business” that it’s been building towards. This means a lot of outreach and a lot of partnerships: Spec, Slack’s first Developer Conference, will kick off May 22nd this year (you can register here), and new product integrations over the last few weeks include x.ai, begin, Intercom, Nuzzel, UserTesting, Spoke, Astound, and more. It means moving towards a world where being the “OS for your business” means a very important kind of resource management: helping to allocate, manage and optimize people’s attention and time. It’s a tall order for one product to accomplish, but April and the product team at Slack are the best in the business. Congratulations on a well-deserved promotion (only days after coming back from maternity leave, no less!), and on to the next challenge.
Have a great week,
Alex & the team from Social Capital
