Book review: The Code Breaker by Walter Isaacson

code_breakerFor my summer holiday reading I have The Code Breaker by Walter Isaacson, the author was recommended by a friend. It is the story of CRISPR gene editing, and Jennifer Doudna, one of the central characters in the development of this system and winner of the Nobel Prize for Chemistry in 2020 with Emmanuelle Charpentier for this work.

CRISPR is an acronym for "clustered regularly interspaced short palindromic repeats", a name derived from the DNA sequences that prompted its discovery. CRISPR are the basis of a type of immune system for bacteria against viruses. The CRISPR repeats form a fingerprint which matches the viral DNA and the associated system of enzymes allows a bacteria to snip out viral DNA which matches this sequence.

Whilst CRISPR is interesting in itself, it has applications in gene editing as a cure for disease in humans. CRISPR simply requires a short piece of RNA to match the target DNA in a gene to carry out its editing job. Short RNA sequences are easy to synthesise making CRISPR superior to earlier gene editing techniques. In addition there is potential to use CRISPR as a diagnostic tool for identifying infections such as covid and even as a cure for viral diseases. The Code Breaker does a good job of explaining CRISPR to a fair depth.

There is a section of the book on gene editing in humans and the moral issues this raises. Perhaps central to this is the story of He Jiankui, the Chinese scientist who led the work to carry out germ line edits to add a gene protective against HIV. Germ line gene edits mean editing the genes in an early stage embryo which means that all the cells in the child it gives rise to have the edit, including reproductive cells, hence the gene edit will be passed on to descendants. This is considered more radical than somatic cell gene editing where the changes stop with the person treated. I must admit to having some sympathy for He Jianku. Principally Western scientists had made a great show of considering the moral issues in germ line editing eventually deciding that the time was not yet right, but going against a moratorium or regulation in the area. This seems an ambiguous position to me, and the associated comments that Jiankui had done his work for publicity is a bit rich from a group of scientists who have been so competitive in the research over CRISPR. Jiankui conducted his research with the approval of his local ethics board but was subsequently disavowed by the Chinese authorities and then convicted.

Coronavirus is woven through the book because the work on CRISPR is very relevant here from a scientific point of view, and the key characters including the author are involved, as we all are! As far as I can tell Doudna et al have been involved heavily in conventional covid19 testing and have done research on CRISPR-based diagnostic tests which have great potential for the future – essentially they would allow any viral illness to be definitively tested at home (rather than a sample being sent off to do PCR test) – but are not yet used in production. Similarly there is the potential for CRISPR-based vaccines but these are not yet been deployed in anger. The Pfizer and Moderna vaccines are based on RNA but use older technology.

A chunk of the book covers the patent battles over CRISPR principally involving Doudna and her co-workers and Feng Zheng, scientist at the Broad Institute. The core of the patent dispute is how obvious the step from understanding the operation of the CRISPR system (which Doudna’s team demonstrated first) to applying it to human cells (which Zhang did first) is. I think my key learning from this part of the book is that I’m not very interested in patent battles! Tied up with the patent issue is the question of the great science Prizes which similarly give a winner takes all reward to a small group. The Nobel Prizes have a limit of three on the number of winners, so do more recently instituted prizes. Science simply isn’t done this way, and hasn’t been for a long time. There’s a group of at least a dozen scientist at the core of the CRISPR story and probably more, singling out a couple of people for a reward is invidious. It made me wonder whether the big science prizes are really about the prize giver rather than the winner.

The book is written in the more journalistic style that has arisen in scientific biography relatively recently, that’s to say there is a lot of incidental detail about where Isaacson met people and their demeanour than in older scientific biographies. I must admit I find this a bit grating, I’ve tended towards collective biographies recently rather than single person biographies which have a bit of a "great man" feel to them. However, I’m starting to make my peace with this new style – it makes science feel like a more human process, and makes for a more readable book. It’s fair to say that this is in no way a "great man" biography, although Doudna and her life and personality are a recurring theme other people get a similar treatment.

Book review: Guitar Looping – The Creative Guide by Kristof Neyens

creative_loopingFor completeness I include my review of Guitar Looping: The Creative Guide by Kristof Neyens. This is in the same series as Guitar Pedals by Rob Thorpe. These are both quite short books but I’ve found them useful.

A looper pedal is a simple recording device which is started and stopped using a footswitch. A loop can be built up by making successive recordings, or layers, one on top of another. Typically loops are only a few bars long at most but modern looper pedals can record for tens of minutes.

I bought a looper pedal a year or so ago (reviewed here) and, to be honest, it has languished a bit on my pedalboard. I think the problem is a lack of education in the right format. Also I probably should have started with the simplest looper available, the author uses a tc electronic ditto rather than a step up (my Boss RC-3).

In common with Guitar Pedals, Guitar Looping contains lots (117) of short examples annotated in normal musical notation and guitar tab notation with accompanying audio files downloadable from the website. There is a brief text introduction to each example. I find these nice exercises in ear training, it’s good to be able to follow along with the tune.

The author is quite fond of the volume swell as part of a loop, this has got me thinking I need a volume pedal – previously I couldn’t see the point of them. This presents a problem because I’ve run out of space on my pedalboard!

Aside from the technical skill of starting and stopping loops at the appropriate point, there is also the skill of controlling the volume of your play within a layer and also getting the volume of different layers right. The loops illustrated often contain a percussive layer made by playing with strings muted, a rhythm/bass layer and a melodic layer which may be single notes or simple chords. Neyens talks about providing both harmonic space and dynamic space in layers. That’s to say there is no point in recording a layer loud and filled with sound because there is nowhere to put additional layers. This means that individual layers can sound quite simple and sparse. To get harmonic space you might play low notes with an octave pedal, on the lower three strings and melodies on the higher three strings, further up the neck.

The other useful piece of information I picked up was how to make your guitar sound like a clarinet! You pick the string 12 frets from where you are fretting – so if you are holding down the low E string on the third fret you need to pluck it and see. Try it and see.

After reading this book I’m using my looper pedal a bit more, there’s a lot of ideas in here and perhaps the most important thing is a stimulus to play around a bit – it doesn’t cost anything!

Book review: Data Pipelines with Apache Airflow by Bas P Harenslak and Julian R De Ruiter

data-pipelinesMy next review is on Data Pipelines with Apache Airflow by Bas P Harenslak and Julian R De Ruiter. The book was published in 2021, and is compatible with Airflow 2.0 which was released at the end of 2020.

Airflow is all about orchestrating the movement of data from sources such as APIs and so forth into other places, it originated in Airbnb. It is designed for batch processing, rather than streaming data, and for pipelines that do not change much.

Data pipelines in Airflow are represented as "directed acyclic graphs" or DAGs which are defined in Python code using "Operators" which carry out tasks. A graph is a collection of nodes (tasks in this case) with "edges" between them. The "directed acyclic" bit means tasks have a definite order, the edges between them are "directed", and the graph cannot have loops or cycles because that would imply having to finish a set of tasks before you could start them. Simple data pipelines would just be a linear set of tasks that always follow one from another, a more complicated pipeline might bring in data from several sources before combining them to produce a final data product.

The Operators are strung together using expressions of the form "operator 1 >> operator 2" or even "[operator 1, operator 2] >> operator 3". 

Operators do not have to use Python, they can invoke code in other languages such as the BashOperator, or interact with other systems such as databases or storage systems such as S3. It is relatively easy to write your own operators. Alongside operators that do stuff there are branch operators which select one or other path in the DAG, and there are also sensors which detect changes in filesystems and trigger work and hooks which form connections with external services. Dummy operators can be used to simplify the appearance of DAGs.

As an orchestration system the intention of operators is that they should not contain a great deal of code to process data, that function should be off-loaded to libraries or systems elsehwhere.

The Airflow system is comprised of a web server which allows you to observe / trigger execution of DAGs, a scheduler which is responsible for the scheduled running of DAGs and workers which do the actual work of the DAG. The Airflow system loops over the tasks defined in a DAG, and tries to execute tasks which depends on the tasks upstream of the task in question, if they have been successfully completed then a task can execute.

A basic implementation runs DAGs locally using a simple queue to schedule work, and a sqlite database to store metadata. A production implementation would use something like Postgres or Amazon RDS as the metadata store, schedule work using Celery and run tasks in Docker containers marshalled using Kubernetes.

For some reason reading this I was reminded that big projects like Airflow are just other people’s code, and if you look too carefully you’ll find something nasty. This is both comforting and mildly scary. I think the issue was that Airflow uses jinja templating to inject parameters into code which feels wrong but is probably a pragmatic and safe why to do it, these shenanigans are not required for Python operators. Also discussed are issues with code dependencies, which the authors suggest are best eliminated by putting operators into Docker containers each of which contain their own code dependencies – allowing otherwise dependency incompatible libraries to work together. 

Alongside the material on Airflow there are moderate chunks on Python modules, testing, Docker and Kubernetes and logging so you get a well rounded view not only of Airflow but also of the ecosystem it sits in. The book finishes with deployment into various Cloud environments. I found these parts quite useful since the most complicated work I do in my role is trying to get things to work in AWS! The data science part is easy…

The book finishes with some short chapters on Cloud deployments, mentioning first fully managed services such as astronomer.io, Amazon MWAA and Google Cloud Composer before going on to talking about implementation of one of the demos in the book on AWS, Azure and Google cloud services. I considered skipping these chapters but they turned out be quite interesting in highlighting the differences between services and perhaps the preferences of the authors of both the book and of Airflow.

I found this a readable introduction to Airflow with some nice examples, and interesting additional material. Useful if you are thinking about using Airflow, or even if you are working on data pipelines without Airflow since it provides a good description of the moving parts required.

The code repository for the book is here: https://github.com/BasPH/data-pipelines-with-apache-airflow

Book review: Guitar Pedals by Rob Thorpe

guitar_pedalsAnother brief sojourn with a guitar related book, this time Guitar Pedals by Rob Thorpe. It has the lengthy subtitle "Discover How to Use Pedals and Chain Effects to Get The Ultimate Guitar Tone", and the front cover continues with a range of other promises as to content. This isn’t intended as a criticism, it just struck me as an usual stylistic effect.

For those not familiar with electric guitars, an guitar effect pedal is a little box of electronics, around the size of a cigarette packet but rather thicker, with a socket on one side to take input from a lead from your guitar and a socket on the other side to send the modified signal out to your amplifier. On the top face of the pedal is a switch to turn the effect on and off, and one or more knobs to configure it. Guitar pedals are usually grouped together on a pedalboard which will hold up to 10 or so of them, chaining their effects together. They introduce effects such as distortion, reverb, delay and so forth.

If you watch videos of live music you’ll most likely notice the guitarist and bass player with a pedalboard on the floor at their feet, occasionally poking it with a foot to change the sound of their guitar.

Guitar pedals are a cheap and easy way of changing the way your guitar, I have a couple of more expensive Boss pedals which cost about £100 and a couple of Donner pedals which were under £40.

Guitar Pedals runs through chapters describing a bunch of distinct effects, talking first about the background of the effect before going through some short examples of the effect in different contexts with different configurations (these appear as written guitar tabs, and accompanying downloadable audio files), and finishing with some examples in real music.

Since reading "The Birth of Loud" by Ian S. Port it struck me that much of the development of the electric guitar and its ecosystem has been the story of electrical equipment abused. Particularly so with distortion /overdrive pedals described in the first chapter – the original distortion pedal made by Gibson in 1962 (the Maestro FZ-1) attempted to replicate the effect Link Wray achieved in Rumble by stabbing his speakers with a screwdriver! Jimi Hendrix was a fan of the Arbiter Fuzz Face but quality control was so poor he would buy a bunch of them and pick the best (or even get his guitar tech to cobble together a pedal from the parts of multiple examples). Purple Haze is an example of Fuzz Face in action. Overdrive is what you get when you turn the volume of your amplifier right up – pedals can achieve the same effect without making a really loud noise.

Next up is a chapter on delay – essentially an echo effect which was originally implemented on tape. I’ve always thought of delay effects and reverb being related with reverb the more important of the two. Reverb and compressor effects each get their own chapter but Thorpe sees them as more production effects than pedal effects per se. Tracks like Beautiful Day by U2, King of Zion Dub by King Tubby and Country Boy by Albert Lee use delay.

The chapter on modulation effects covers phasor and flanger effects, where part of the signal is phase shifted and mixed with the original signal. Shine on you crazy diamonds by Pink Floyd is an example of a phaser in use, and Barracuda by Heart uses a flanger. Also included are chorus effects (where part of the signal is delayed) and tremolo (where the volume is modulated). The first chorus pedal, the Roland CE-1 started life in Roland’s Jazz-Chorus 120 Amplifier. A background in physics is quite handy here, vibrations and waves are at the heart of any physics degree, as are operational amplifiers – pedal effects are these things in action! Come as you are by Nirvana is a good example of the chorus effect, and How soon is now by The Smiths demonstrates the tremolo effect (for this performance the tremolo effect comes from the amplifier rather than a pedal).

My wah pedal is my favourite, and it gets a chapter largely of its own. Think Voodoo Child by Jimi Hendrix (watch his left foot at the start of this video) or the theme from Shaft by Isaac Hayes. A wah pedal is an adjustable band pass filter, in the same why that the tone knob on any manner of audio equipment is an adjustable low-pass filter. The wah pedal is unusual in that you adjust it during play – the position of the pedal controls where the band pass sits, other pedals have their configuration set before hand and are simply switched on and off. All I can say is it’s great fun to play with!

Next up are octave pedals and other pitch shifters and harmonisers, I’d assumed the point of an octave pedal (which plays a note one octave above or below the note you are playing) was to emulate a bass guitar, but it seems not.Jimi Hendrix’s Octavia pedal added a tone an octave above what he was playing, on tracks like Fire and Purple Haze. Jack White of The White Stripes uses octave effects to add notes both an octave above and below the played note to give a "thicker" tone – try tracks like Ball and a biscuit and Blue orchid.

Originally effect pedals contained simple analog electronic circuits (or even liquids) which did one job, now with digital processing a single pedal can emulate many different effects. I must admit I find multi-effects pedals a bit overwhelming – it’s no fun trying to navigate 50 or so effects, and their configuration on a one inch display with a couple of buttons.

The book finishes with a chapter on ordering of guitar pedals, and how this can change the sound made and finally there are some interviews with professional guitarists, and how they arrange their pedals. A point that both Thorpe and one of his interviewees makes is that tone, the sound of the guitar, depends a lot on the player and how they play. Chasing after a tone by buying the same pedals as your heroes is a losing game.

Guitar Pedals is a short book, it doesn’t have the high production values of the Rikky Rooksby but it carries much of the style – embedding the example riffs in the chapters works really well for this book. Online guitar courses tend not to cover effects pedals, this book fills the gap pretty well.

Book review: The Goddess & The Bull by Michael Balter

the_goddess_and_the_bullI like to vary my reading, so from my previous review on guitar riffs we go to Neolithic archaeology, specifically the archaeology of Çatalhöyük in The Goddess and The Bull by Michael Balter which carries the subtitle "Çatalhöyük – An Archaeological Journey to the Dawn of Civilization". Çatalhöyük featured in a recent TV programme "Raiders of the Lost Past with Janina Ramirez" which I recommend if it is still available – this is what prompted me to get this book.

Çatalhöyük, in Turkey, is a prehistoric township which was active between 7500BC and 6200BC, it reached a peak size of some 5000 residence and as such it notable for being one of the largest of the early Neolithic settlements.

The Goddess intertwines several themes, it provides some biographical detail of the key archaeologists involved in the dig, it talks a bit about the evolution of archaeological methodology, and it talks about what was found at Çatalhöyük and the wider human activity in the Neolithic period. This is laid out in chapters that proceed chronologically with each chapter mixing elements of biography, methodology and the Neolithic period.

Any story of Çatalhöyük starts with James Mellaart, the first archaeologist to dig the site in the early 1960s. It’s fair to say he was a bit of a character. When he arrived in Turkey in the 1950s the Neolithic remnants were little known and he discovered hundreds of sites in his tour, on foot, of the area. These were largely in the form of "tells", large mounds built of successive layers of settlement made one on top of another. Çatalhöyük stood out because it was large, and it was early Neolithic from top to bottom. Many of the tells were occupied over very long periods of time so the earliest archaeology was obliterated by the later.

After digs at Çatalhöyük between 1961 and 1965, Mellaart was banished by the Turkish authorities following accusations of facilitating the sale of archaeological artefacts. His case was not help by the "Dorak Affair" in 1958 in which a mysterious woman showed him a range of artefacts from the Neolithic which he dutiful wrote up in various articles – these artefacts never saw the light of a museum and the Turkish authorities suspected Mellaart in their disappearance. The mysterious woman has never been found. Over the rest of his life Mellaart continued to publish on Çatalhöyük, later articles becoming somewhat fanciful.

I found it striking how much of the archaeological work done in the Middle East was done by British, American and other Western European archaeologists, often with their own institutes in-country. It feels like a water-down version of the bad old days of the 19th century where Europeans pillaged the countries of the ancient world for artefacts, although it is clear these 20th century ventures were much more under the control of the "home" countries.

Also relating to politics, the book highlights how much modern archaeology is funded as rescue work, during the construction of roads and railways in the UK, and frequently dams and hydro-electric schemes in the Middle East. In the nineties phase of digging at Çatalhöyük, Ian Hodder, the director of the work, spent a lot of time fundraising from both public and commercial sources.

From a methodological point of view, Mellaart’s first archaeological digs were based on the vertical stratigraphic approach borrowed from geologists which had been made popular in the forties and fifties by Kenyon and Wheeler. Here layers of a site are stripped back successively to establish a chronology, aiming for depth rather than breadth. This replaced the 19th century approach of broad area excavation where discovering the horizontal extents of a site was the priority, as was the discovery of "treasure", I suspect. Following Mellaart’s excavation the "New Archaeology" arose which became the "Processual" movement in which the emphasis was on highly detailed digging and analysis with a view to testing hypothesis. Earlier schemes being more interested in cultural artefacts. In a nutshell, the processual view saw different forms of stone axes representing different uses, whilst the previous view saw different axes as representing different cultures.

Ian Hodder represented a further evolution in methodology, post-processualism which included the intensive specialist sampling of the processualists but added more context from sociology and anthropology, and even the subjective feelings of the archaeologists as they worked. Pierre Bourdieu, a French sociologist and a bête noire in our household, is mentioned! Hodder took over the excavation of Çatalhöyük in the earlier nineties, and continued through to the completion of this book in 2006, and beyond until 2017 when digging seems to have stopped at Çatalhöyük.

The core question around the Neolithic era was what made Palaeolithic hunter-gatherers settle and take up farming, and how did they do this. There is some evidence that settling, living in permanent villages occurred some time before farming started. But why did farming start? Why did settling start?

There isn’t really a conclusive answer to this. Çatalhöyük, as a substantial Neolithic settlement, is pre-dated by a short spell by sites such as Jericho. It is comprised on many relatively small dwellings with common features: evidence of a ladder and an oven on the south wall, indicating access from the roof, numerous human burials under the floors. There is no evidence of specialist or communal buildings or a social hierarchy. The walls of the buildings were decorated, sometimes with bulls’ skulls or vulture heads inside them. It seems that buildings were ritually burnt at the end of their lives, typically after one hundred or so years of occupation. The people ate cultivated crops, and domesticated sheep and goats and some wild cattle, more of the wild cattle during what appear to have been ceremonial meals.

One theme from the sixties epoch of excavation that didn’t make it through to the nineties was the idea of Goddess worship, in the sixties there was some enthusiasm for the idea that Çatalhöyük represented a matriarchal society which worshipped a goddess. This idea does not seem strongly supported by the archaeology, although there are a number of "goddess" statues discovered they are all small and not found in particularly salubrious situations. There is more evidence for the idea that the bull was venerated – I wonder about the links between this and the position of cows as sacred animals in Hinduism, and also its role as a pictogram that evolved into the letter "a".

I really enjoyed The Goddess and The Bull, having approached it somewhat sceptically because it was not the recommended book by experts (James Mellaart: the journey to Çatalhöyük by Alan Mellaart) and it wasn’t clear whether it was an intensely academic volume. The biographical material of the archaeologists is sometimes a little grating but it makes the book more readable. I felt I learnt a lot about archaeology and the Neolithic. There’s a website www.catalhoyuk.com, where you can see the latest developments – including annual reports up to 2017.