Skip to content

Thibault – Chapter 1

The Proportions of the Human Body, Related to the Figure of our Circle and to the Proper Length of the Sword

Chapter 1 is very dense, both with theory and the way that the theory is discussed, but has a number of key statements that need to be observed. It’s a pretty good illustration of his habit of stating something once, and expecting the student to remember it.

This chapter begins with a lengthy discussion about the ratios and proportions of the human body, calling back to Pythagoras, Plato, Vitruvius, Pliny, Hippocrates and the Bible as examples of previous proofs. This is a fairly typical way of constructing an argument at his time, where first the prior knowledge is laid out, and then the author’s own assertion posited as though another layer on top of the previous arguments. The truth of the argument can be tested because it has some logical connection to the previous arguments that have been accepted as true.


Thibault – Author’s Note

Since I last wrote, and started searching around, I have found a record of one sale through Christies of a copy that changed hands for around €19,500 in 2009. There’s some interesting details there around that physical copy, including the size, listed as 550 mm x 410 mm, which is roughly the size of an A2 page. As I said last time, this is a big book.

If you are looking to be able to leaf through the images, the National Library of Netherlands has most of the images up on line, and you may care to have a look at the Wiktenauer page for the book for some background.


Thibault – Introduction

I have in recent weeks began re-reading Academie de l’Espee by Gerard Thibault d’Anvers, for a variety of reasons. Sadly my virtually non-existent French means I am reading the 2005 translation by John Michael Greer. To my surprise and delight, re-reading after having left it alone for several years I find it opening up for me, and it makes much more sense than it did the first few times.

I believe there is value for me personally in taking notes as I go – essentially a gloss on the book – and so I may as well publish those as I go, hence this first of roughly 45 posts about the book. But first, some strong caveats for the reader. The most important is that this is my understanding, opinion and interpretation, and is in no way authorative or even necessarily correct. The second is that I am working from a translation without having sufficient French or easy access to one of the originals to verify the quality of the translation. I’ve heard different opinions about the quality of the translation, but the general consensus from people in the know is that it’s ok.

Since this post takes roughly the position in the book of Greer’s introductory notes, I feel free to make my own introductory comments, starting with a general disagreement with the tenor of Greer’s introduction. In my opinion, Greer focuses way too much on the ‘esoteric’ nature of the book, perpetuating – probably inadvertently – the myth of the mysterious Spanish circle.

I contend that there is very little, if no, estoreric mystery in Thibault’s book. If you consider the philosophical and intellectual spirit of the years in which he was writing it, there is zero surprise that he leans heavily on geometry and briefly relates the geometry of the body to the broader geometry of the universe – if you pick up any scientific or mathematical text of the early 17th Century, you will find this tendency because it was one of the primary threads of scientific theory. Indeed, one way to look at the history of western science through the 17th Century is to see this mapping of earlier less rational thinking onto a mathematical models, followed by a rapid refinement of mathematical technique as it became evident that simple geometry is not good with the messy perceived world.

Indeed, while it is framed by a very particular mode of discourse, I am startled by how clear and unambiguous Thibault really is. I do not think the decision to call the book “The Academy Of The Sword” was taken lightly, it is entirely descriptive: there is a very distinct didactic method, and he aims to take the reader on a journey from novice to competency. That is not to say he is a forgiving teacher. He definitely expects that the student is practicing everything that is in the book, and very seldom states a lesson twice – he expects the reader to retain everything that he wrote in previous chapters. Occasionally he will refer forward to later materials by saying “don’t worry about this, we deal with it later”, and occasionally he will call back to previous points, but in general it appears he expects the student to take on the responsibility of retaining knowledge from one chapter to the next.

I think that it is easy to believe that Thibault’s book is obscure or difficult, because it is a very dense text. There is a good deal of exhaustive detail, coupled with a system which appears to be unusually focussed on very precise situational awareness, and his explanations are inextricably linked to the illustrations in the book. It is this latter point that I feel lets the Chivalry Bookshelf edition down, as many of the illustrations are not well reproduced. However having handled one of the original copies, this is understandable – the original book is huge, and the original engravings are very finely detailed. The reproductions lose detail, sometimes significant detail, and wind up more separated from the text than they were in the original.

So, excuse me for rambling, I will try to be more focused in later posts on this topic and hope to demystify Thibault.

Academy of the Sword
Gerard Thibault d’Anvers, 1630, trans. by John Michael Greer
The Chivalry Bookshelf, 2005
ISBN-13: 978-1-891448-40-9
ISBN-10: 1-891448-40-4

On Git Submodules…

Git Submodules. Just Say No. Not Even Once.

Docker and Consul and DNS, oh my

I’m still trying to wrap my head around networking when it comes to Docker and related technologies – I think because a lot of the documentation and examples around are either not quite correct, or are subtly out of date. I’ve noticed too that a lot of the writing out there around setting up Docker and/or Consul hand waves away the trickiness of the networking. Particularly egregious is the blithe insistence on just specifying host networking for all containers, something that the Docker project itself frowns upon.


Lovely Rita Meter Maid…

The saga of the non-functioning SourceLondon / PodPoint units in Woolwich Arsenal continues, with the lightning pace usually associated with continental drift, and the rise and fall of mountain ranges.

We had a period of about six weeks where one of the dead units would periodically and arbitrarily come back to life for enough time for us to get a charge. This was great for us, as we finally got a chance to really explore the benefits of the PHEv, but abruptly came to a halt after the charging unit was vandalised. The timing of this did roughly coincide with when my contact at SourceLondon/BluePoint/Bollaré (Maryline Marilly) had indicated that they would totally be replacing the units, so I once again began to pester her.

This revealed there would be at least some progress: a meeting between Bollaré and the Head of Estate at Berkeley was scheduled for 4th August. This did not deter me from beginning to shotgun annoying messages out on Twitter about the Arsenal units, and other broken units I pass on my commute, which resulted in shaking loose a direct conversation with the bloke from Berkeley, one Simon Challen, who took time out to have a chat with me on the 5th about the meeting with Christophe Arnaud (someone who I had begun to think was a myth).

I don’t know why it has taken months for the two organisations to reach the same understanding of the situation that I was able to determine from the outside: TFL pushed Greenwich Council to install the units, who did some undocumented deal with Berkeley to get the units onto the Arsenal, and then walked away leaving the responsibility with PodPoint. As it stands, Challen did confirm the understanding that I had arrived at: none of the organisations have a record of who paid for the units, who owns them, and who will fix them.

Hilariously – from my point of view, Challen did know about the work done by some navvies to cement the broken unit back upright, but admitted he had no idea that none of the units were working. Less hilariously, and frustratingly for me, he did not seem to get my point that it’s not a good look for Berkeley and the Royal Arsenal to have the local management company (Rendall & Rittner) unaware of the ownership and responsibility for infrastructure in the Arsenal.

There was some hopeful news though: Berkeley and Bollaré have agreed that they will draft an agreement for Bollaré to take on responsibility for the units. Of course, this is still at the stage where lawyers are discussing the agreement to make an agreement, so I can anticipate that sometime in 2018 they may discuss the agreement having agreed to make an agreement. Simon did say though that Christophe would “do something about fixing the units at BluePoint’s risk” before the agreement to make an agreement was concluded.

So, we will see. Today it is 208 days since I began trying to report the broken units. The units have been broken for much longer. Let’s see if it can take less than a year.

SSL Made Easy

Time for a shout-out to DreamHost, who have partnered with LetsEncrypt to make using SSL with this website very, very easy. DreamHost have always aimed to make many actions against the site push-button, with sensible defaults, and clear documentation, and generating and attaching the certificate was a walk in the park.

I was a little surprised to see the certificate expiring so soon, but LetsEncrypt’s rationale is very sound: re-rolling certificates can and should be automated, and limiting the life time of a certificate automatically limits the exposure if the certificate is subverted. It is very much in line with a core idea that they have: the default for HTTP traffic should be across SSL, or in some other way encrypted.

For me, the process was as simple as pushing the buttons on the DreamHost control panel, then do a bulk find-and-replace on my site to update any http links to be https. I will probably have to chase around the interwebs to find where I’ve published the old URL, but I’m pretty sure I’ve found and updated the important ones already.


Having a little time up my sleeve, and a need to be off my feet for as much as possible… wait, did I mention that? Somewhere in the last six weeks I’ve done something undefined to my feet, which are painfully sore to walk on. I think that I managed to sprain one or more of the muscles that usually wiggle my toes, and as a result walking has felt like I’ve had stones in my shoes. Since I had a few unexpected extra days off, I elected to sit on my butt as much as possible and bang away at a little project that I’ve had hanging around for ages: PropertySource, which is a simple abstraction for finding “properties”. The code and README are there in GitHub, and there’s pointers on use from the README.

I’ve had two motivations for this: first, defining key/value properties can be done in a number of different ways, and there’s little consistency across projects in how or where that will be done. By wrapping this up in an abstraction, then it becomes much easier to not have to think about how or where the properties are defined, while at the same way providing an easy way to have a hierarchy of default and override properties.

Second, I’ve become interested in the use of things like Consul and etcd as distributed dictionaries for configuration, and wanted to put together a framework into which I can easily integrate using these as property sources as well, while hiding any complexity from the user.

If you use this, please let me know how you find it, and never hesitate to suggest extensions or modifications. I do intend to keep fiddling with this, first by adding support for different distributed key/value stores, and secondly by making it simple to specify the order of hierarchy when trying to resolve a property from a variety of sources.

And as an aside… my mind is a sieve at the moment, so I’m awfully glad that I documented the process I came up with for doing Maven releases with Git!

Feeling blue

It has been said somewhere… and therein lies the state of the art when writing something that sounds profound on the Internet in the first half of the 21st century. Somewhere or other I read, unattributed, or with forgotten attribution, something vaguely like what I’m about to repeat without attribution: Greater London as a city does not really exist, instead it is dozens and dozens of small villages that have expanded until they have grown into each other.

Flying over it, or using Google Earth or Google Maps to virtually fly, the eye can be tricked into believing that the sea of buildings within the M25 is some sort of cohesive entity. After all, the buses and tube go all over the place, the road rules are the same everywhere, and there is a Costa on every corner. The reality is far more complicated, and it is frankly a small daily miracle that the conurbation does not collapse into rioting and arson as an entire nation’s worth of people wade through a morass of niggling annoyances resulting from subysystems of the city not… quite… working…

I had the pleasure last week of meeting Maryline Marilly, Stakeholder and Partnership Manager for Bluepoint London, who generously took the the time to answer some of the issues and concerns I’d raised around Gill-Hank.

When Bolloré, or rather their sub-entity Blue Solutions, purchased the car charging network set up under the SourceLondon scheme by TfL, they did so with an unabashedly and transparent commercial intent: they really want to get their electric car club system up and running inside the M25, and on the back of that promote and push for more widespread adoption of electric vehicles. Particularly their electric vehicles, using their battery technology. They certainly were aware at the time of purchase that the charging network was a shambolic, malfunctioning mess, but saw it as a good entry point into the infrastructure.

We don’t perceive the infrastructure that exists to support the use of cars reliant on fossil fuels, any more than a fish perceives water. The infrastructure is pervasive and unremarkable, and any driver in London unthinkingly and correctly assumes they are only a few miles from a fuel source, which will be open, functioning, and trivially easy to use. I have no idea how much it costs to build a petrol station inside the M25, but at a guess would expect it to be a six-figure sum. Buying the land, sorting out the legals, constructing the building, installing the specialist equipment for dealing with fuel… not something you pay for with the change you find down the back of the couch. And yet, companies still build them, even though the income from selling fuel, energy drinks, fluffy dice and crisps will probably take decades to cover the construction cost. Of course, the actual cost is amortised over all the other stations that the company owns, and it’s not a huge financial gamble as long as their pockets are deep enough to deal with unexpected costs.

The infrastructure for electric cars is in many ways technologically much simpler, and largely boils down to “somewhere to plugin and charge”. There are of course some caveats there – the electrical network the vehicle plugs into needs to be fairly robust, and capable of carrying currents a bit above your suburban house light circuit, but this is not a huge ask. The charging stations themselves also have a small footprint, requiring much less space and supporting tech than a petrol pump, but the crunch is that they need to be placed where the driver can park for a non-trivial amount of time. Current vehicles largely need to be able to charge over 4+ hours to attain full charge, and that brings all sorts of complications: publicly accessible charging stations necessarily are in public spaces, such as road sides and car parks, which immediately means that the provider of the infrastructure has to negotiate with a startling array of third parties before they can plop a station down and advertise it’s presence.

It does feel a little that the vehicle manufacturers have a vision that the prime charging location is at home – mum and dad and three kids and a dog with the car in a garage overnight before Dad drives off to his Work with a full charge there and back. Looking around London, the reality is necessarily different. It has proved to be interestingly difficult to find overall what proportion of private cars are parked on-road. I did find this research by TfL on Residential Parking Provision in New Developments showing that 80% of Londoners live in some sort of flat or unit, 45% are renting, and 56% are parking on the street in new developments. The proportion parking on the street for all of London must therefore be somewhere north of this figure. This is not a city where the vehicle owner is probably going to be plugged into their domestic supply. Even if the building owner provides off street parking, it is not likely the renter will be able to access power in the car park (or talk the building owner into providing it).

Like it or not, during the transitional years before electric cars are the norm, Londoners are dependent on the charge points rolled out by the Source London scheme, and similar.

Which brings us back to BluePoint. They know they can’t make a profit on vehicle charging, or indeed I suspect come close to recovering the costs directly from charging, but they are willing to wear the cost. If the chargers are available, Londoners will use them, and if more people start buying or using electric cars, then more of the electric cars will be ones using Blue Solutions’ battery tech. They are happy to give away the safety razor handle, as long as they know they’ll sell a hell of a lot of razor blades and shaving cream.

One thing that Maryline admitted to was that BluePoint were not prepared for just how complicated it would be to deal with all the local authorities, and all the paper work. As she put it, when Paris decided it wanted a charging network, there was a single local authority, a single Mayor, who could just make it so across the entire conurbation. London has got 32 boroughs, each of which has to be bought to the table individually, each with variations in parking regulations, variations in how the pavement can be used or modified, variations in how power cables can be run to the charging units. Describing the time frames required for the negotiations they were going through, her frustration was palpable, and understandable.

None of this should be this hard. Oil prices will rise steeply some time soon. The pressures on consumers to move away from internal combustion engines will ramp up. BluePoint’s proposition is a sensible one: build the infrastructure now rather than doing it later when it is urgent and expensive. Oh, and if they wind up in a very nice commercial position as a result, well, good for them…

Coming back to Gill-Hank, Maryline was able to fill in the last pieces of the puzzle. It seems that PodPoint went somewhat off-piste when they replaced the older SourceLondon unit that was there (although it’s not clear why that happened, as nobody involved appears to have any records of who asked for it to happen). As a result the unit literally dropped off the radar of Source London, and it seems that Greenwich Council never had a particularly clear idea of which Source London units they had committed to. My previous rant managed to rattle the right cages, sufficient for Greenwich to lean on Berkeley to talk to BluePoint, and for an agreement finally made to replace the two dead units in Cadogan Road. The units are indeed technically on private roads on private land owned by Berkeley, even if at some point in the past they were Greenwich Council’s financial responsibility under the SourceLondon scheme. With the new arrangement stitched up by BluePoint and Greenwich, the unit reverts to BluePoint along with the rest of the SourceLondon network in the Borough, and Berkeley blesses BluePoint turning up with pick axes and cement mixers to replace the dead units.

It’s still going to be 6-8 weeks before the units are actually replaced, which is bloody frustrating, but I can understand why it is going to take that long. So, all is well that ends well. Except I am going to ring fence the new unit with traffic cones once it is put in place, and will go on the warpath for the next lorry driver that backs into it.

While they waited and listened in awe…

Erect and sublime, for one moment of time,
In the next, that wild figure they saw
(As if stung by a spasm) plunge into a chasm,
While they waited and listened in awe.

Benighted PodPoint Gill-Hank is now standing proudly and boldly erect!

Yes indeed, workers from Berkeley were out over the last few days hauling the corpse upright and repairing the pavement to keep it that way. Not, of course, that the unit actually functions.

When quizzed by Delia, the workers told her they had been instructed by Berkeley to make it look better…