Notes on Works in Progress Issue 16

You can read the full Issue 16 for free at https://worksinprogress.co/. See my notes on Issue 15 here.

Crispr Gene Editing.

Mathias Kirk Bonde provides a clear description of how gene editing works – possible before, but made much cheaper with Crispr.

Every time I read about Crispr I understand a bit more. I need to read / watch another eight articles/Youtubes to fully get it at a high level.

The piece made me appreciate is how long humans live relative to mosquitos. You can do Crispr on a mosquito species, rendering one sex infertile, and this quickly propagates. Were there to be a malicious edit of human DNA, there is more time to detect and rectify that edit.

Millions of people die due to malaria each year, so killing those mosquitos is interesting. It hasn’t been done yet. The question is, what is the profile of the knock-on risks?

Advance Market Commitments

An advance market commitment means guaranteeing there will be a buyer for a product if it is successfully developed.

This is more or less what the US government did to accelerate COVID vaccine development. They guaranteed that, were successful vaccines developed, the government would buy them – even if the vaccines were not needed (because, say, COVID went away by itself). This de-risked the development for pharma companies*.

Nan Ransohoff writes briefly about vaccines this but moreso about the concept of advance market commitments in general. Ransohoff is currently involved in leading Frontier – an advance market commitment initiative for CO2 removal from the atmosphere. In this setup – meaningfully backed by Stripe as a buyer – a range of companies pre-agree to buying CO2 removal from the atmosphere in various prices and quantities over time*.

The basic idea of an advance commitment is to bridge the gap between today – where a technology does not make sense to invest in – and tomorrow where there can be a viable market. More specifically, advance commitments are relevant where patents and trade secrets don’t work, namely where there aren’t well funded future private buyers (customers, companies). Rather, there is only a public or non-profit buyer (e.g. the GAVI Alliance, a public-private alliance for global health launched an advanced market commitment for pneumococcal vaccines).

What I liked about this piece is the focus on detail. Yes, the concept of advance commitments is a nice tool for public initiatives, but it requires diligent work to clearly define the guaranteed prices and volumes, as well as the required product standards. This is the coordination work Frontier provides as a service to the consortium of corporate “carbon removal” buyers.

*Is there a case to be made that pharma would have developed the vaccines anyway? Or, perhaps it was in the government’s interest to reach an early agreement, rather than have distribution through insurers or other channels? Or maybe it was a bit of both – pharma liking an early de-risked deal and government (and the population too?) liking central control of the vaccines?

** I don’t understand why we would remove CO2 from the atmosphere when it appears more cost effective to do solar radiation management. i.e. We understand how a doubling of carbon dioxide is increasing the net solar radiation by ~1% (each doubling has a log effect); We understand the physics of how we might counter this with small amounts of gases in the upper atmosphere, and the thermodynamics and economics of this seem more favourable.

Reading and Discovering Ancient Texts

This piece by Justin Germain covers the matter of ancient texts we a) know about and have copies of (or often, copies of copies), b) know about but don’t have copies, and c) had copies of but only got interested in recently.

Technology is helping with discovery and we can now digitally unroll manuscripts and piece together the text using AI – as was done in last year’s Vesuvius challenge to unroll scrolls from Mount Vesuvius in 79 AD.

Side-note: If you have to toss one out – get rid of Mark’s Gospel, it’s basically a copy of Matthew. As to why one should drop Mark, but not Matthew instead, I don’t know, yet.

Libraries of Objects and Materials

Virginia Postrel writes about libraries; not of books, but of materials – stuff you can feel. Apparently quite a few materials libraries require an (expensive?) paid subscription, unlike public libraries for books.

One remark that struck me was that these materials libraries might not convey much scientific progress over the last decades. As Postrel points out, this is because materials libraries aren’t equipped to handle the latest biology or microchips that do represent today’s advances.

On the topic of free libraries, I was reminded of a conversation at the Manifold conference earlier this year… A friendly chap explained to me how the cost public libraries comes out to $100 per resident per year in some cases. If you consider those costs on the basis of only those who use the library, you come to a very large cost per person. Generally a few free books and some free wi-fi strikes me as a worthwhile public service. Those numbers – even if off by 10x – made me question my views there a bit.

People are loving the fake diamonds

Apparently De Beers, the real diamond cartel, made it popular to give diamonds upon engagement – and set expectations for men to spend a few month’s salary on the ring.

De Beers have been deepfaked and a majority of engagement diamonds are now artificial. A majority of diamonds used for industrial purposes have long been artificial because they are larger and purer.

Some chap, Hall, applied huge pressures and high temperature to graphite and successfully made it into diamonds. More recently, it has proved cheaper to use chemical vapour deposition: You send in hydrogen and methane into a chamber containing diamond seeds; Microwaves then propel the reactions needed for deposition and growth of those seeds.

How Pour Over Coffee Got Good

For the impatient, footnote 3 in the piece has the list of where to find the best coffee in New York, San Francisco, London and Tokyo. I have to say I like Starbucks coffee myself – or any coffee sufficiently dark roasted for the beans to be uncriticisable.

If I were to write a recent history of the coffee shop industry, I would summarize it as follows: “How can we find ways to make coffee more slowly”.

Less automation sells I suppose? More ingredients (sugar) sells I suppose?

What about plain old high throughput? How is it we’re doing crispr for the price of coffee? These questions explain why I would be bad at owning a coffee shop.

Nick Whitaker is a good writer and the piece is an easy pour. He never gets distracted by frappucinos or steamed milk. He’s in there cranking his high end home setup with a refractometer on the counter… measuring TDS.

Where inflation comes from.

As an avid reader of Tipswatch, this piece by Carola Binder had to be my favourite.

From the title, you’re thinking this will be a three-way political skirmish between Covid, Russia and the finer points of price gouging.

Joyfully, it was not.

Instead, Binder charts a history of why and how prices are measured, and how they came to the definition of the Consumer Price Index (CPI) in the US today.

When the price of any one good goes up, consumers typically shift to buy other goods instead. This ability, called “substitution”, makes inflation a little less bad than what it otherwise would be. Interestingly, CPI doesn’t account for this effect. PCE (personal consumption expenditures), by contrast, allows for partial substitution. So the PCE tends to be a little lower than the CPI (maybe a quarter of a percentage point).

Now interestingly, in managing interest rates, the Federal Reserve focuses on PCE. But, for the purpose of social security / pension / welfare (and, for the adjustment of coupon payments on Treasury Inflation Protected Securities) CPI is used. And this misalignment, a beautiful source of lunchtime economist debates, costs the US billions (maybe tens of billions) per year.

And that’s it for Issue 16. Recommended.