From a 37c3 talk on an ethernet chip that requires to bitbang a clock into itself, through a multiplexer, rather that Broadcom fix the firmware.
]]>I took a slightly indirect walk to the event, so I could say hello to the river. When I arrived, I found a lovely outside seating area with water features and lots of sun.
The food was excellent, with vegan options and interesting things like toast with figs and mustard. There was plenty of chia, quinoa, coconut yogurt, interesting raw sugar mate, and generally all the hipster vegan stuff.
After introductions, we proposed topics to discuss for the first day. These are put on the wall to be arranged. The second day we would work on implementing something.
Perhaps the weather was too good for it though. :)
I talked about how I would like to have something like livejournal used to be - a place where you can talk about your day with your friends. This needs friends only posts and comments. I have a bit more work to do on the site to implement friends only posts, then I will see about comments. There seem to be a lot of hassles with trying to do private webmentions, and I came to wonder if it is trying to force something into a system that was really not designed for it. ActivityPub instead?
I was also interested in considering trust relationships and how to model these on the web, with a specific application in mind of couch surfing. I would like to make a standard for marking up couch offers and requests, and syndicate these with trustroots. I would also like to be able to search the social graph for people in places I am visiting, and see trust relationships for visitors.
Trust graphs could also be applied to credibility of web content, which is something the W3C Credible Web Community Group is working on. The credweb group is just starting up so I joined it and will try to follow what is going on at least. It would be good to have a trust network that can work in the IndieWeb.
We also talked about readers quite a lot. I would like to start using an IndieWeb reader - at the moment I, and I suspect many others, use twitter.
I also learnt a bit about web standards in general. I was wondering what happened to XHTML, and why I cannot just use RDF and XML and everything is simple and works? Why do we need to parse HTML and use class names as markup? I believe what happened is that people failed at writing valid XHTML. This meant browsers decided they had to try and work with what they got, valid or not, so they invented HTML5. They had to work out what to do with any sort of HTML like input anyway, so they decided to just write up what they have to do as the standard, rather than try to tell people to do it properly. W3C didn’t like this, so the browsers made their own standards body called WHATWG.
We also talked about live data on IndieWeb sites. I would like to publish my local situation - including for example weather, water depth, speed, other ships in the area - basically what is in the NMEA stream I get from the boat. I will probably put this information on a map. Aaron Parecki does some stuff like this, particularly location logging.
Most of this discussion happened while lying on the bank of the Rhine, or near the docks.
After sunset I went to a flat where I was couchsurfing (actually a place from trustroots) with an old snoring dog. :)
]]>tldr: Let’s Encrypt failed to fetch tokens from the webserver, but I could fetch them myself. IPv6 was not configured on the server but was in the DNS. My connections worked over IPv4, but they checked both, and did not provide a clear error message.
I was using http-01 / webroot authenticator.
Let’s Encrypt logs / verbose output reported in json:
Error getting validation data
,
"status": 400
,
"status": "invalid"
. The 400 apparently was not an http status code, although it looks like one.
It looks like, from this report, that it may only be a problem when there is an http redirect involved too, as there was in my case.
IPv6 configuration failed on the Debian server because of cloud-init interfering with manual IP configuration. /etc/network/interfaces was used manually. cloud-init replaces the main config file. I used a file in the config directory to avoid this, but didn’t know that it also ran some if-pre-up script that tried to run DHCPv6. This caused the manual configuration of the interface to fail.
ifup script causes:
RTNETLINK answers: File exists
This seems to happen most commonly when attempting to configure two default gateways, such as ones manually specified on two interfaces, or when one is already auto configured. This could be from SLAAC or DHCPv6.
]]>Last night at about 0115 I heard an explosion from my room in Piraeus, the port of Athens. I wondered if it might have been a ship crashing into the dock, but my first thought was correct - it was a bomb.
I went to the roof. I was concerned about fire, and assessing if we needed to move. Looking in the direction of the car alarm sounds, I thought I could almost see smoke from a nearby building, just past the closest church. I wondered if someone had taken objection to the very long and loud church ceremonies that had been broadcast on speakers for much of the previous day and night. There was light breeze from that direction. I didn’t smell any more pollution or smoke than usual, and saw no flames.
I decided to go out and investigate. There were already quite a few police cars and cops around the area of the building I saw smoking. One shouted at me to go away - there was a bomb. It seems the smoke was probably dust.
I met a couple of lads who asked to interview me for their youtube news show, which I declined. Otherwise there were surprisingly few people out.
I returned to my room and stayed awake for a while, listening and looking out for any developments. There were still blue lights flashing outside the window.
In the morning I went for a walk to have a look at the site. The entrance passageway to a tall building looked the most affected. A brick wall and the roof were damaged, but presumably nothing load bearing, as the building was still standing, and there were people on the first and higher floors tidying up, sweeping glass. The shop windows on both sides were gone. Some windows in what I presume are flats above on the 4th floor were also broken.
The cats in the park just outside the building were all loafing and seemed subdued. If they were there last night I expect they will have burst eardrums at least.
One of the things I find most interesting about this event is how uninteresting it seemed to be to anyone else. I searched twitter, mastodon, and youtube, and news sites throughout the night, in English and in Greek, and there was nothing.
The first news stories I saw were about 10 hours later in the morning, and they said almost nothing. A report from one of the largest news papers in Greece was clearly written by someone who had not even gone to the site - my favourite comment in this article was, “The large building houses a shipping company, law and insurance firms and shops on the ground floor, according to the authorities”. Apparently they needed authorities to tell them there were shops there.
The article also gave no context, or even speculation as to the reason for the bombing. Is it anything to do with the date, 1312 - ACAB? It’s also a week after the traditional 6th December riots in commemoration of Alexandros Grigoropoulos who was murdered by police on this day in Exarchia.
I speculate it could be directed at any of the companies in the building, for unpaid debts or business rivalries, or it could be an attack on the law firm in the building to disrupt some legal action, or it could be an anarchist group. This sort of thing has happened before, I found when I searched the web.
A very similar bombing happened in Piraeus in 2018. In 2009 the port was bombed, possibly targeting a shipping company office. In 2008, a bar in a Piraeus marina was bombed, probably by a rival business. There was another blast in Piraeus in 2006 - that one claimed by “A left-wing extremist group calling itself Revolutionary Brigade”. The ekathimerini article covering the latest explosion mentions the same building was bombed before in 2020, but I could find no more information.
Maybe five bombings in the area - and more more I didn’t list - is why the locals didn’t seem very interested. Business as usual here it seems.
]]>“one of the crew on board managed to dismantle an old battery and used the carbon electrodes to fashion new brushes for the generator”
I guess it would have been a zinc carbon AA(A) cell, which has a carbon rod down the middle.
]]>For some reason my friends seem to be getting more stressed and unhappy recently.
]]>The old world is dying, and the new world struggles to be born: now is the time of monsters.
– Antonio Gramsci (paraphrased)
You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.
– Buckminster Fuller1
I see two main schools of change amoungst activists, which I shall very grossly categorise as the hippies and the punks, because that’s how I first started to see it. You could just call it creation and destruction.
In this model, punks are people who are angry. They hate how things are and want to bring it down. The music is angry, the clothes are dark, and the attitude is aggressive.
The hippies also don’t like the ways things are, but their attitude is more of love for something better. Instead of fighting the system, they will build a new one. The clothes are soft and colourful, the music is peace or mantras.
On the destruction side, activities include:
On the creation side:
I tend to feel more on the creation side of things, but obviously both have their place, and are quite complimentary (in theory, even if there can be arguments in practice). For example, consider a squat. Who do you think cracks open the door, and who then does the cooking? :)
Something that bothers me about a lot of activists is that they seem to complain without offering an alternative, or even understanding what they are really asking for or how that would work. For example, in the film “The Corporation”, there is a section I found rather amusing in which some protestors visit the house of the CEO of Shell to do a banner drop. They end up having tea on the lawn and the CEO’s wife apologises that she has no soya milk. Other than being amused at the civility of this compared to recent protests in the US, it seemed to me that they had not considered what they were really asking.
Let’s say they got their way. The CEO realises his mistake, repents, and resigns. Then what? The next in line steps up and nothing changes. If he tries to do something helpful like stop oil production, he will be removed. The real solution is to stop giving money to Shell. That means there must be cheaper alternative energy sources before anything will change.
I don’t want to say that protest is useless. I think XR have succeeded in getting more people to take notice of climate change. In large enough volume, it might even have some policy influence. Politicians though can’t really do anything on their own. They are useless without engineers to actually build something better.
I know there is value to both sides, there is overlap, and punks do make things too. There is no need to fight me about an oversimplified model. Meanwhile, I shall probably continue to complain on the internet while failing to make anything much of use.
1. As quoted in Beyond Civilization : Humanity’s Next Great Adventure (1999), by Daniel Quinn, p. 137
]]>When I was in Mexico, at least six months ago, I was thinking about changing my name, and to River. I can’t remember exactly why, but I think it was because I was talking to someone about the importance of words and sounds. I was interested in seeing if a change in name would change how people saw me, and treated me, and how I acted and felt about myself.
I forgot about this idea while I was travelling, but was reminded of it again In Panama where I met someone who had given herself a hippie name and encouraged me to use my new name. I started to do so, in earnest, when I got on the boat to Colombia. I crossed the water and entered a new continent with a new name.
I only used it with new people though. I still wasn’t totally sure whether to completely change it. I knew I wanted to, but I thought I should check whether this was the right name - maybe there is a better one - and what people think about it. I asked a few people, but I really didn’t consider any other names. I think I came up with the name when I was thinking about flowing, and the idea of the bike and I flowing through the world. Now though it felt like there was not really a choice - this was just meant to be.
Someone sent me a song about a río. I went to sleep with the sound of the river outside my window. I changed my name.
It felt like a relief. I no longer had this pressure of thinking about who I was. I feel freed of the past. I was not particularly happy for most of the nearly three decades past. I am happier now, and don’t need to carry this any more.
I am already finding things I like about this name. I think of rivers differently - now they are brothers and sisters. :) I say hello to the river when I walk past, and to the bamboo, and the bananas, and the other plants that grow around it.
I went to one of the organic hipster veggie restaurants around here, to celebrate my birthday. I’m writing here now. After I sat down, I noticed that I can hear the river and see the bamboo from my table. The water dispenser has stones in the bottom.
]]>I know many languages to some extent but have never really practiced functional programming seriously. I am familiar with some of the concepts, which I have picked up from languages with functional aspects, such as perl (see Higher Order Perl) and javascript, from speaking to a more knowledgeable friend and of course from wikipedia. I also started to learn common lisp a few years ago and got as far as reading most of a book before having to do something else. (University work. It was quite good at getting in the way of me learning things.)
I decided to have another go at learning functional programming properly. It is an important programming style which I do not fully understand and I think it is important that I should. I do not like not knowing things and I feel stupid when talking to people who know more about programming languages.
I think Haskell is probably the best choice as a language to learn functional programming because, unlike lisp, it does its best to be purely functional, that is functions have no side effects. This should force me to do things properly. It seems to be becoming quite popular so there is a chance people may even start to use it for things other than research. Also, I studied at Glasgow, home of the Glasgow Haskell Compiler (GHC). I’m actually in Edinburgh at the moment, home of ML but if you have a look in lib.ed.ac.uk, you will see hunners (a Glaswegian term) of Haskell books, not ML.
Another good choice for learning functional programming would be Structure and Interpretation of Computer Programs, also called SICP and the Wizard Book, which uses scheme. I suspect it is worth following this course even if you do already know how to program functionally as it is generally highly thought of and I know of good programmers who have found it useful, despite it being designed as first year course.
I first saw Haskell in second year at Glasgow when I went to a few classes on it. I was studying electronics too though so couldn’t officially take the class and I didn’t get very far. It seems I have now ended up using the Glasgow Haskell Compiler for the first time here in Edinburgh. :)
Compilers and books then, which should you use? I am reading Thompson. This was used for the course in Glasgow and also fills the shelves of the Edinburgh University library. I noticed on the book website there is actually a third edition now.
A rather more trendy book is Learn You a Haskell for Great Good!. I am told this is less dry than Thomson but I’m not sure if soggy books are a good thing. It does have a nice picture of an elephant on the cover though and pretty colours on the website.
What about a compiler and interpreter? An interactive interpreter, or REPL (Read Eval Print Loop) as lispers call it, is certainly a good thing to have and learn with. Thompson uses HUGS, which I started with but the GHC also has an interpreter, GHCi, which I have now switched to. I note that the latest edition of Thompson uses GHCi too. GHC seems to be the standard compiler so I would recommend using it. See also Haskell Implementations.
Back to what I’m doing to learn Haskell then. I’m now on Chapter 14 of Thompson, Algebraic Types, but skipped most of the exercises, which I intend to go back to. I’m also doing Project Euler. These mathematical problems seem well suited to a mathematical sort of language like Haskell but perhaps this is making it too easy. There is the old question of whether functional languages are suited to real problems, and the fact that I won’t know how to print “Hello, world!” until Chapter 18.
I think I’ll leave it at that for now but I intend to say more in further posts.
]]>Firstly, the obvious. Voltage, (Vgs and Vds) must be within limits. The same for current, Id. There is also the power limit given by the maximum junction temperature. I calculated this limit given the Tj (max), the ambient temperature and the thermal resistance from junction to case, Rθjc. I then operated my circuit within these limits and could not understand why the MOSFET was dying.
When I looked at the safe operating area (SOA) graph in the datasheet I found values much smaller than expected from my calculations. I thought the SOA was limited only by the maximum voltage current and temperature as calculated with the Rθjc, but it turns out there are other thermal limits.
These are particularly notable in the transistor I was using, the IPP114N03L G.
Note the steep limits to the right, severely limiting current at higher voltages. This was the region I was trying to operate in.
I think the problem is caused be high thermal gradients causing expansion and cracking of the die. This explains the quiet pop I heard when one failed - the silicon cracked. The thermal resistance seems like more of an average limit - if the whole die is at a similar temperature it applies. Temperature variation over the die seems to be higher in FETs designed for switchmode power supplies. They are optimised for on-off switching and can’t cope with operation in this linear region. BJTs suffer from second breakdown, which limits the SOA. MOSFETs are not meant to do this but this seems to be a similar limit.
Moral of the story: don’t trust the headline figures. Sometimes these seem to be made up by the marketing department. Consult the SOA and give the limits a wide berth.
I learnt most of this from my question on stackexchange.
]]>As a programmer, my first thought before doing something is to find out how it has been done before and to use libraries. I struggled to find a clear good answer. I also didn’t know much about it. Since then I have been asking people, reading and trying things out, so I think I can now make some recommendations to a beginner to the field.
SVG is a good choice of graphics format. It can be used directly on the web and made interactive or animated with javascript. There are various tools for working with it, from programming libraries to Inkscape and other drawing tools. You could also use postscript / pdf but that seems rather less convenient for most tasks.
As for libraries, I was expecting there to be some clear winners for this but there are not. I think the reason for this may be that good visualisations are often original. For example, the rose diagram, a favourite of Tufte which, by clearly showing deaths in the army caused by poor sanitation, persuaded the British government to improve conditions and saved lives. O’Reilly’s Beautiful Data also argues that beautiful visualisations must be original. Higher level libraries can restrict what you can make.
There are libraries for standard charts but even in these cases they do not really offer much. For example, drawing a bar chart is a matter of drawing rectangles. The scales can cause a bit more work, but it’s not a big deal really.
I did try some charting libraries: Google Charts for one. I found it inconvenient and limiting. I was trying to make something visually very similar to github’s punchcard, which is made with it. I started by copying how they had done it. Generating the GET request string was fairly unpleasant. Once I’d done it, it worked to start with but as soon as I added more data it complained the URL was too long. I found I could then use POST instead so I changed, with more fiddling. Then, having done all that, I found that the size in pixels of the output image was limited below what I required. I gave up on this and decided to do it manually.
Everyone I spoke to who got stuff done at the Culture Hack used low level drawing commands. It seems that this is the way to go. I used this method with the Perl SVG module to visualise source repositories. It works perfectly reasonably.
Some libraries worth looking at for SVG:
Other useful visualisation / data tools:
The first book I read on this subject was O’Reilly’s Beautiful Visualisations. I was initially rather disappointed but I think this is in part because of the nature of the subject. When I learn a new subject, I want a theoretical overview and introduction the main concepts and principles. I am used to this in more purely technical subjects. This book, and to a lesser extent some others I have read, did not do this. This book is mostly a collection of reports on various visualisation projects. It may serve as a source of inspiration but it does not provide any overview or principles.
I then read Tufte’s second book [2]. I was expecting this to be much better because Tufte seems to be considered the master of visualisation and his books are highly regarded. Unfortunately I found again a book of examples with little theory or practical guidance for designing my own visualisations. I also did not like how he seemed to jump from one subject to the next - it often felt like I had missed a page. This style I think was because he was really just making notes on the image on each page, rather than having some bigger theory or idea to explain.
There were some more general points though, for example the use of colour. A principle here is not to use large areas of strong colour but to have neutral background and highlight some parts with colour.
I then read a paper [1] which was much more what I was expecting. This outlined some basic methods of visually communicating quantitative values and evaluated some of them.
I am currently reading Tufte: Visual Display of Quantitative Information. This seems to try and introduce some principles but they are a poor attempt. For example, “Graphical Excellence” is defined pretty much tautologically, near enough, “Excellent Graphic is Excellent”. The book smells very nice though.
Some principles do seem useful though, particularly to maximise the information to ink ratio. This is similar to writing natural languages and code.
Vigorous writing is concise. A sentence should contain no unnecessary words [and] a paragraph no unnecessary sentences, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts. This requires not that the writer make all his sentences short, or that he avoid all detail and treat his subjects only in outline, but that every word tell.
— Elements of Style, William Strunk, Jr. - 1918
I would like to learn more about the perception of data graphics. For example, if you present some circles sized to represent some data, do people perceive the value as the diameter of the circle, the area, or something else? Should your graphics be adjusted to allow for human perceptual anomalies? (I write about these questions in a further post on area and circles.)
There is of course the artistic aspect to the design as well. If art is theft, books of good visualisations like I have seen should help in this regard. I would rather be able to produce something correct and functional first though.
In general, it seems best to use an API at the shape drawing level. Visualisation libraries are often more trouble than they are worth, and they will limit your ability to create an original visualisation.
Use SVG in general and for the web. For a paper, TikZ, a LaTeX library, can produce very nice graphics.
Read Tufte, but don’t expect much theory, it’s more a catalogue of ideas and critique. If you are interested in things like the perception of graphics you need to search the literature for research papers.
1. Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods
William S. Cleveland and Robert McGill
Journal of the American Statistical Association, Vol. 79, No. 387 (Sep., 1984), pp. 531-554
http://www.cs.ubc.ca/~tmm/courses/cpsc533c-04-spr/readings/cleveland.pdf
2. Envisioning Information
Edward Tufte
The choice is quite important and I have had problems with them before. On my last tour one of the bolts attaching the rear carrier to the frame fell out. I had to remove another one of these to fit the solar panel and almost couldn’t because it had rusted solid.
The brazeons on my bike are M5 threads. This is the same as used in my SPD shoes, which saved me when I lost one last time, so I’m standardising on M5.
I want something common so I don’t need special tools and can find tools if I need to. This means slot, cross / phillips, or hex / allen key. Slot and cross heads can be deformed too easily - hex gives much more torque and is harder to damage. This is important if the bolt gets stuck.
Some hex socket screws have a rounded off head and some are cylindrical. The later kind could be gripped with pliers if the socket was damaged so I favour these.
For strength I think it has to be some sort of steel. The options are zinc plated, galvanised or stainless steel. Hot dip galvanisation gives too thick a layer of zinc for threads of this size, so the real options are zinc plated or stainless.
The screws in my rack which have rusted solid were zinc plated. Clearly this is inadequate unless you constantly attend to them by oiling and replacement. That leaves stainless steel. The specific grade of stainless steel is probably irrelevant for this application.
There is a possible problem with stainless though. When dissimilar metals are in contact in the presence of water they form an electrical cell which oxidises the less noble metal. My bike frame is made of aluminium, which is less noble than stainless steel but more noble than zinc. Stainless steel risks corroding the frame whereas zinc plated screws would not. According to the British Stainless Steel Association, the risk of problems from a small amount of stainless in contact with a larger piece of aluminium is quite low. Additionally, galvanic corrosion is only a problem if the parts are wet, and particularly with salt water. Most of the time I expect to be touring in warm dry weather (this is an advantage of being able to move all the time) so the risk seems quite low. Nevertheless, I could reduce this risk my adding some sacrificial zinc somewhere. An easy way to do this would be to use zinc plated washers.
As well as the carrier bolt which fell out, in testing my solar panel mount around town I lost another screw. I suspect this is because of the vibration, particularly from hard tyres I like to reduce rolling resistance.
The one which came out used a nyloc nut but the bolt was very short so it might not have reached the plastic. I think I will use nyloc again with the correct size of bolt and also use threadlock. I would like to use copper grease to avoid seizing but I’m not sure about how to use this and threadlock.
Someone told me you can use nail polish instead of threadlock. Presumably threadlock is better, but I don’t know. I also found the idea when searching about this to use nail polish to indicate movement of the bolt - put it across bolt and other part and it will crack if the bolt turns.
The value of lock washers seems to be debatable, so I think I won’t use them for now.
My standard bolt is an M5, cylindrical hex socket head, stainless steel. It should be used with threadlock, stainless nyloc nut and plain stainless or zinc plated washers.
]]>The plan was to make a usbpicprog and bootstrap it with a serial port based programmer using an old computer. I bought all the parts, including copper clad board and ferric chloride but never quite got round to doing it. This was in part because I was not very keen to have to work with ferric chloride and I didn’t want to dissolve the sink.
Meanwhile, I had been hearing about the Arduino more and more. Initially I ignored it as something for artists and people who didn’t know what they were doing. I had spent some time at university learning about and using real microcontrollers with none of this boot loader nonsense so I did not think I needed such things. Then I realised that being leet enough not to require simplification did not make quicker development a bad thing. :-) I decided I had wasted enough time and just wanted to get something done, so I bought an Arduino.
I actually chose the seeedunio, an Arduino clone. The design is open so this is allowed and in fact there are some improvements made in the seeeduino. Primarily I chose it though because it was the cheapest. I still don’t like the idea of paying decapounds for a single basic microcontroller and some soldering. I also ordered an ethernet shield, which is what an expansion card is called in the Arduino world. These arrived yesterday.
It took me longer than I would have liked to get started, with the blinking LED
hello world of microcontrollers, because there were no instructions about the
correct settings. After some googling and trial and error I determined that I
had to set the reset switch on the board to auto and, in the Arduino IDE, set
the board type to
Then I decided to plug in the ethernet board and run the example programs. I had to stick all the source files in the same directory and change the ip address and then I was pinging the board. It took only a few minutes. This I thought was pretty ridiculous. Then I used the web server program. It worked. I couldn’t really believe this - this did not seem like electronics at all. I would expect this to have taken weeks of debugging, loose wires, errors in PCB design, software bugs and frustration but here it was working in a few minutes with no test gear in sight. I have certainly confirmed my get things done faster conjecture.
It was not quite as good though when I wanted to connect stuff up to it. The board did not come with a data sheet and I could not find much at all online. In the end I went to Atmel’s web site and downloaded the datasheet for the ATmega328P on my board. Included below are some useful figures from it.
I am using the Seeeduino V2.21 which is also marked ARD128D2P.
It uses the ATmega328P.
This is the error I got initially:
avrdude: stk500_disable(): protocol error, expect=0x14, resp=0x00
The solution:
Deumilanove or Nano w/ ATmega328
.Operating Temperature | -55°C to +125°C |
Voltage on any Pin except RESET with respect to Ground | -0.5V to VCC+0.5V |
Maximum Operating Voltage | 6.0V |
DC Current per I/O Pin | 40.0mA |
DC Current VCC and GND Pins | 200.0mA |
The Arduino environment uses C but with some magic, see Arduino Build Process.
It’s ENC28J60 based. This is not the same as the official one, which uses the Wiznet W5100.
Bought from ekitszone. Looks well made and works.
]]>I went to a hacker camp organised by Chaostreff Flensburg. It’s a cosy meeting in a house on the beach, overlooking the fjord. Lots of sailing boats out on Saturday and Sunday. Clear dark sky, moon, fire, and a view to Denmark over the water. Back in the Baltic again. I should have brought the boat.
After this rather cold swim, there was plenty of reviving slush / ice to drink. Some of it later did have rum, specifically Tschunk slush (Club Mate and rum).
Flensburg actually looks like quite a nice town. Certainly more to it than Rendsburg, which rather put me off anything smaller than a city, and Schleswig-Hostein. There are a lot of Danish people living here too (the area used to be in Denmark), and they have their own bakery and church(es?). Still generally German prices for food though, which is nice.
]]>The streets smell of roasting coffee, grown by the coöps in the surrounding hills. I also smell chocolate and bread from the panaderias. There are punks on bikes - real bikes, not rusting Mexican style ones. Restaurants advertise vegetarian and vegan food. Colourful gnomes and greyed hippies sit in the streets.
There are anarchist and squatting symbols painted on the walls. If you wear an anarchist patch in Edinburgh it is ignored. Here, the town was actually captured by rebels. I don’t think it would be a good idea to cycle past the army base they built down the road on my own with Zapatista colours. There are armed police and soldiers around, which is a nice reminder of what is possible. The government is wary.
There are at least two places here devoted to permaculture, with a vegan community. Until recently there was a community bike workshop.
I wandered into the yard of some casa, following the sound of a woman singing reggae. There was a guy contact juggling, and silks hanging from a frame. The smoke was the pleasant smelling kind - not tobacco. A menu on the wall listed hummus and verduras.
I wandered on aimlessly, and reached a cafe. Now I’m drinking tea that tastes of the earth where it came from.
Mexico is sick and makes me feel sick. The world is sick. San Cristobal is a hill station for the recovery of humanity.
]]>I considered putting it at the front, but the back seemed to be the easiest and best place. In placing it I wanted to avoid shading from me or any radio antennas as much as possible, and it was very important to try and design to my principle of minimising faff.
I decided I needed to mount it above the rear rack somehow. Getting it higher also has the advantage of getting more sun if I am in the way at the time.
The simplest way to mount it was to bungee it on top of the other gear on the rear rack - my tent and roll mat. This was very easy to do, kept the weight low - so reducing forces from putting at at the end of a lever higher up, and it was well cushioned from shock by the gear. However, the problem with this setup is that it makes it harder to get at my tent and to fix and remove and other gear. One think I remember really hating when touring before is the time spent mucking around trying packing and unpacking things. I wanted to get the panel out the way.
A rack could be built in metal, as an addition to the existing luggage racks, or a custom rack for panniers and solar panel could be built.
I looked at rear racks that attach to the seat post - this would give an additional rack to the conventional one I had fitted. This would add even more weight though, and they did not go very high, so there would not be much space between the racks. I also wanted to save money on buying things, but given the time spent it turned out this was not a valid reason.
The idea is to build a custom structure that would serve as luggage storage and solar mount. This would remove the need for panniers. It could also incorporate spaces for electronics and batteries - for example giving an easily accessible slot for changing batteries.
One way I’ve seen to do this that looks very good is a cardboard core glass reinforced plastic structure. This can be easily built and hacked to any shape. This was the technique used by Steve Roberts for a similar project.
Another fibreglass method was used for a solar bike project:
I think the idea for this was:
So it doesn’t end up as a composite with the cardboard core as with the other method.
I decided to make a metal frame myself. This would put the panel where I wanted it, and it should be possible to make it much lighter than a seat post rack. Fibreglass looked like it could give a very nice structure, but it would also be messy to work with and I didn’t have anywhere I could do that work. It would also be expensive and I suspect very time consuming.
The final design. More detailed photos are on flickr.
I had never made anything like this before and didn’t know anything about working metal. I didn’t even know where to buy it. One day though I found what I think might have been a bike rack for a car dumped beside a bin. I recovered the smaller tubes and started working out what I could do with them.
The design I ended up with was mostly determined by the material I had available. I used aluminium tube, which seemed like a good choice because it was light and probably strong enough.
The most difficult part was bending the tubes. Bending a tube by hand can cause it to collapse. I researched online and read that filling tubes with sand first would work. I did this, but one of the bends still flattened out to the point I was concerned it might break. It survived a little testing round Edinburgh though so I kept it.
The tubes I attached to the bike at the front by screwing onto the rack mounting points, and at the back with Jubilee clips (hose clamps) padded with inner tube. The panel is attached to the rack with p-clips, which are horrible to work with because they are so difficult to bend into the right shape. It worked though.
It worked until an awful road in Mexico. Update to follow…
]]>“The Corporation”, a film, argues that corporations are psychopaths. It reaches this conclusion by considering the diagnostic criteria and the behaviour of various companies. I think the film is right, but further I think that corporations are demons.
What is a demon? There may be other kinds, but I am going to talk about a kind which I think can be understood and believed without any special spiritual belief. Demons are something that you believe in, probably fear, and can hurt you. In a sense they are imaginary, but that does not mean they are not real. It’s quite possible for thoughts to hurt you and others.
What is a corporation? It is an imaginary person. By law, companies are defined as people. They are allowed to act in the world as people, for example by owning things and making contracts. Despite being called corporations, they have no body. You cannot kill them and see the corps. They are not real, they are just made up and only exist in the mind. They are however real in the effect they have. If you tell people you don’t believe in companies (or money for that matter), they might think you are mad, even when they would deny the existence of demons, which are of the same nature.
A company is not just the sort of demon that lives in one person though - they are collectively created. They are created and sustained by many minds. In occult terms, this creature is called an egregore.
The influence of a company extends beyond the members and directors. They gather energy in the form of mental attention throughout society by use of magickal symbols which they fight to display everywhere. By means of this mind control they extract money, tolerance, and regulatory favours. To be clear, they have logos, which can also be called sigils, and advertisements - a sort of black magick designed to control the victims mind. These tricks insert the company sigils and servitors into your unconscious mind, controlling you even without you knowing it. If you think this is nonsense, you need only research the effectiveness of advertising.
Some companies gain almost all of their power directly by collecting and controlling information and feeding it to people. Thankfully these creatures are still somewhat limited by their ability to project out of the computer networks and into our minds. For now, they are mostly seen through the scrying mirrors they have convinced so many of us to carry. They have pressed into homes with microphones and cameras to monitor us, and speakers to talk. The fact they have tricked so many people to accept this when it was not long ago written of as a horrific dystopia shows how much power they have. They still feel limited though by the barrier between internet and human minds. The promotion of the metaverse is an attempt to break the thinning veil and give them unlimited access to our psyche. Beware.
Corporations are created by prescribed incantations in the arcane language of law. (A profession in which, in England anyway, the higher degrees of initiation are granted by societies such as “The Honourable Society of the Inner Temple”.) Their names and motivations are inscribed in grimoires maintained by countries (a sort of arch-corporation) which support them by use of violence against humans who do not serve them. Company formation also requires an initial sacrifice of humans called directors who are bound to serve them.
Directors may be seen as having power over the company. This is partly true, but they are also bound by it. They are part of a control loop, and also controlled by other interests. If you think the CEO is in charge, imagine what would happen to the boss of Shell if he decided to order a halt to oil extraction because he doesn’t want environmental collapse to hurt his children.
Corporations are maintained by recruiting or enslaving people by tempting them with money and power. These people so incorporated give the company agency in the physical world. With growing physical power, they do not even need to offer a good bargain anymore, because they can simply force people to work for them by controlling access to food and shelter - that is, life.
How are they banished? Fighting them directly may be a distraction. They run on your energy, so do not give them any. Do not feed them or display their sigils. Instead, use your energy to build communities and create change that you want, not what you have been told to want. This requires the use of imagination, and hope. These things have likely been sucked from you, but you can rebuild resources when you realise what is happening and how to escape. If you wish to create a shared entity, be careful to give it a good spirit. Create an angel of loving grace.
Bibliography: This post was partly inspired by https://twitter.com/pee_zombie/status/1451696805612228608, The Corporation (film), and knowledge from the underworld
]]>The bluetooth adapter that comes with the Thinkpad x230 is bad. The range is barely enough to reach from desk to ears. Also, I had trouble with the device disconnecting when I pressed the case wrong. It’s held in by one screw on a precarious connector.
Largely for this reason, and for bonus faster wifi, I decided to upgrade the wifi card to an Intel AX210, which is about as new and fast as it gets. The card also has bluetooth and uses the main antennas for this, rather than the poor PCB trace on the original USB connected one.
The Lenovo firmware does not allow using this card, so it was also an opportunity to flash Coreboot to the machine. This was not quite straightforward, but not terribly difficult. The installation instructions did not correspond to the reality very well, but I managed to work it out. I also needed to upgrade the linux kernel for support of this card.
The other major problem I had was after using the SOIC clip, the machine would not boot. The power light around the power button flashed three times, but otherwise no sign of life. It turned out this was because I had broken off two resistors which are close to the flash chips. Searching the web I found I was not the only one to do this.
Luckily, I was able to find a circuit diagram for the thinkpad and I had suitable replacement chips. Note: the diagram was found easily on google - the first result in fact - but duckduckgo returned only some terrible spam and malware bait links.
For reference, the resistors in that area are:
R122: 10k
R430, R431, R1375, R1376, R1377, R1378: 33R
R432: 3k3
All 5% 0402.
I also replaced the fan with what seems to be a genuine AVC one, which was rather difficult to get hold of, but I deemed necessary after a terribly noisy cheap fake Delta one.
The machine is working well again now. Bluetooth works with my headphones from desk to the garden. It’s quiet. I am pleased.
]]>I suspect that this board was designed as an adapter for the previous PS/2 model, and the case has not been changed. It’s stuck on with a sticky pad and glue, and converts USB to PS/2 which links to the main board.
]]>3 of 3 cumari germinated. 2 of 4 Chocolate Habaneros, which is much better than the 1 of 6 at the last attempt. I had also treated them with bleach though, as some went furry last time.
There is also a weed that came with the compost. I’m growing it to see what it is.
]]>I bought an ultrasonic humidifier from Aliexpress. I was looking for one for improving the environment for my chillies, which seem to prefer higher humidity than is normal here.
Obviously I bought the cat one. There are probably more efficacious choices, but this one makes me happy. It is a Totoro Cat, watching over the plants and making them grow.
Inside there is a water tank, and a cotton rod which wicks water to the transducer. It looks like a piezo, but must have microscopic holes in it because water goes through the middle. A spring holds the wick to the disk.
The tank is specified as 200ml, and it takes some hours to get through it. In a small room, it makes little difference to humidity sensor readings unless it is spraying directly on it. Maybe increases by 5% RH, but hard to tell with the noise. I think something more powerful is needed for a room, when I want to raise humidity from around 40% to 70%. The tank size is also a limitation.
I wanted to automate the control with a humidistat. I thought I could just switch the USB power, but there is a switch on the back of the cat’s head which needs pressed to turn it on. I had an though idea.
I opened the cat and found, as expected, only the power lines of the USB connector are used. I found the switch grounds an MCU pin. I connected the switch pin to the D+ and D- lines, so now it could be switched remotely. I used a 2k2 resistor for the connection for some protection. It would be better not to short the data lines, but it was hard enough to solder to the pins even without trying to just connect one of them. I think USB hosts are meant to be protected against any lines being shorted, but I shall avoid testing it. If I want to use it with at computer as power supply, I can do so with the supplied cable, which only wires the power lines anyway. With a full cable, I can connect it to my custom controller.
The driver circuit is very simple. There is a microcontroller marked 6280AOH. It looks like it’s probably a Sino MCU MC30P6280. This outputs a square wave to a transistor, connected to a coil and then the transducer, forming a boost supply. It also rainbows an RGB LED, and reads the button. That’s about all there is to it.
It switches at about 114kHz. Peak voltage on transducer 140V. I was concerned about voltage on USB lines, but it looks like the ripple is only 1V p-p, despite a lack of filtering.
There are much more powerful looking humidifiers which run on 24V, don’t look like they have a switch, and are submersible. These would be a better choice for more humidity, but they are significantly less cat.
The resistor is actually soldered to the wrong switch pin here. I fixed it after the photo. The controller for the other end of this cable will hopefully be the subject of a later post.
The cat: https://www.aliexpress.com/item/1005003720730550.html
]]>Topics include boat building, gardening, cats, Catnip for Humans?, What’s inside a microwave?, and How to Make a Witch or Wizard Hat.
]]>Some possible solutions are:
Add a real time clock to the pi. These are available in the form of a small module with a coin cell battery. Once the time is set it will then remember it. These cost about $2 for aliexpress.
Use a GPS. GPS receivers know the time very accurately. The pi can extract the time from NMEA messages if it is connected to some GPS receiver on the boat. Alternatively, you can add a GPS module to the pi itself. These are surprisingly cheap, and on a GRP boat you probably don’t need an external antenna. If you can afford to splash out about $5, many of these also come with a real time clock and backup battery, so even without GPS signal they do the same job as an RTC module.
Set the clock from another computer, like your laptop. You can configure NTP on the pi so that instead of asking servers on the internet what time it is, it can ask your laptop, which probably knows the time well enough from when it was last on the internet. Obviously this only works if your laptop is on and connected to the pi on your boat’s network.
I won’t describe how to do this here. I’ve not tried it, and there are already tutorials on the web which explain it well.
The only advantage I think there is to this solution, compared to a GPS module, is that it will use less power. It’s probably also easier if you don’t like configuring linux systems.
To set the time from an NMEA stream you can use ntpd (whether or not you also want to use it to set time from the internet). The NMEA data can come from the boat’s network or a locally connected GPS. ntpd knows how to read the time from gpsd, which reads data from a locally connected gps. I use kplex, which can read NMEA from the network or locally, and can be configured to look like gpsd so ntpd can also read the time from it.
gpsd is simpler but more limited than kplex. It will only look for a locally connected gps device. I won’t describe the setup here because I don’t use it. If you want to receive NMEA data from a network over tcp for example, or from USB or serial GPS devices, kplex is very good.
If you are using kplex, you can add the following to your /etc/kplex.conf
to
provide something that ntpd will recognise:
# Output time for ntpd
[pty]
mode=master
direction=out
filename=/dev/gps0
baud=4800
# Output filter on this interface - only output the needed message with time
# in it. I am using GPGLL but you should use GPRMC or GPZDA to get the date
# too if your GPS outputs that.
ofilter=+GPGLL:-all
You must also have setup kplex to receive NMEA data from somewhere.
If you pi is running another variant of NTP software such as chrony or systemd-timesyncd you should disable those to avoid fights. Here is my ntpd config file:
restrict 127.0.0.1
driftfile /var/lib/ntp/ntp.drift
# If you have lines like this in the default config for your operating system
# it is polite to use the servers provided. (i.e. not debian if you are using
# something else.)
server 0.debian.pool.ntp.org iburst
server 1.debian.pool.ntp.org iburst
server 2.debian.pool.ntp.org iburst
server 3.debian.pool.ntp.org iburst
# Change <laptop> to the hostname or IP of another computer on your boat
# network that knows the time.
server <laptop> iburst
# NMEA GPS Input
# The IP address 127.127.20.0 is just a kludge ntpd uses to mean local gpsd /
# read time from NMEA stream. It's not actually something on the network.
# My NMEA data comes from kplex, pretending to be gpsd.
server 127.127.20.0 minpoll 4
# fudge 127.127.20.0 flag3 1 flag2 0 # This config is for PPS, which I don't have.
# Set stratum very low so it won't be used unless nothing else available. I do
# this because without PPS, the delays and imprecision in the serial NMEA
# stream probably make it worse than network time.
fudge 127.127.20.0 stratum 14
Time sent over NMEA messages will not be exact because NMEA is slow and there is a delay in the message getting from the GPS to your computer. For exact time, GPS receivers may have a Pulse per Second (PPS) output, when switches on the second each second. If precision within a second is good enough you don’t need to worry about this. A marine GPS receiver will probably not have this anyway. If you need it, you can buy a GPS module with PPS to connect directly to the pi. You almost certainly don’t need it.
An actual problem might be lack of date. The GPRMC and GPZDA NMEA messages (can) contain the GPS date and time. On my boat, I only get the time but not date from my receiver.
A GPZDA message that contains only the time (19h 46m 44s). The date would be between some of those commas.
$GPZDA,194644,,,,00,*42
I could set the date manually, but I actually use NTP.
On the pi, you can use the same ntpd configuration file as the above used for GPS, but comment out the GPS server line. You can also use any other ntp software and add your laptop or other local machine that knows the time as an ntp server.
Then you need to configure the laptop as an ntp server so it will respond to time requests from the pi. I won’t go into the details of that because it depends on what software you are using on that machine. You may need to configure it to allow requests, possibly only from the local network, and also allow connections in on your machine’s firewall.
]]>• 214 PAGES yet NO REAL WORLD USE found.
• Wanted rules anyway for a laugh? We had a tool for that: It was called COLREGS.
• “You might think I’m not normal because I’m under 20m long, but I can carry 12 people and put a yellow cone on my head so you know I’m really normal.”
• “Red flag? Anywhere else at sea that means dangerous goods, but I’m just using it so you know I’m a narcissist according to Article 3.17”. Statements dreamed up by the utterly Deranged.
LOOK at what UN committees have been demanding your Respect for all this time, with all the committee rooms we built for them (This is REAL regulations, done by REAL UN Committee members):
I like this quote from Alan Watt’s in “Tao: The Watercourse Way”, as an explanation of how “anarchy is order” (especially suitable for rivers).
Wu wei - anarchism for the inactivist.
]]>Mostly Bulgarian carrot 🥕 chillies with a few mild red ones.
]]>Hacker, or more specifically, “esr” question culture is laid out in ESR’s How To Ask Questions The Smart Way
This culture is predominant in free software and generally on the internet I think. The main point is that you must research the question yourself before daring to ask it.
Implied Principles:
I don’t know what to call this but the word human came to mind as it is sort of the natural way people do it, and it is part of human relationships. In this culture, it is expected and valued that you ask questions even if the answers have been written somewhere already. It realises that questions are not just about the knowledge content in an answer, but about relationships. Asking a question is an opportunity to communicate with someone personally. The question does not just ask, it can tell someone:
This may take more time for the questioned and save it for the questioner, which is why some people don’t like it. This is especially true if there are few with the knowledge and many asking, as is the case for a free software project which has a few developers and many users. In this case hacker questioning maybe the only practicable way for most people. It does come with a cost for the questioned though. They might miss good questions, and lose potential contributors and relationships.
Human questioning probably makes more sense in smaller communities with ongoing relationships. In a group, saving the questioner time saves the group as a whole time. On an anonymous internet there can be a practically infinite pool of questioners who disappear with apparently no return to the group.
]]>This sounds familiar. It is approximately impossible not to get it everywhere.
From a Wildlings Sailing video, a youtube boat channel.
]]>The way you see yourself is not the same as how others see you, and I think this is especially true as your appearance changes in gender transition. Hormones can change the shape of the face and the quality of the skin, but this usually happens very slowly. That makes it hard to see the difference if you see someone everyday, although it would be clear in a photo from years ago.
I think the change is especially hard to see in yourself, because you have a model in your head of what you look like and who you are. Human perception does not show you an objective vision of what is out there - your eyes are used to update a model in your mind. This is complicated by dysphoria amplifying the older aspects of yourself and generally distorting things. The effect is you cannot see yourself as others do.
People are a sort of mirror - they try to tell you who you are. For trans people especially they are often wrong. To break out of the matrix, Neo has to pass through a mirror - a fluid and warping one. On the other side he finds he has new eyes which he has never used before.
In psychology, apperception is, “The process by which new experience is assimilated to and transformed by the residuum of past experiences” (Runes). So what do you see in the mirror? It’s, at least partly, a residual self image.
One of the basic facts of existence noted in Buddhism is anicca, which means impermanence. By meditating one can observe thoughts and sensations arising and passing away. This really extends to everything.
Everything is always changing. The idea that there is normal is an illusion; a trick people play on themselves to deny and repress the fear of the unknown. You can’t read the same river’s blog twice.2
People tend to not notice change, especially when it is incremental. What you call a phone is actually a pocket computer and camera mostly used for messaging and sharing photos and videos. You might not even use it to make phone calls. A phone used to be a speaker and microphone attached to some copper wires in your house.3
Conservatism is a fear of change. It is denial of reality. Strategies which worked before don’t work now. You can’t live the way your parents did. Maybe if you are rich you can force an approximation for generations, but not forever.
As I was searching for something else, I came across an article in a 19084 edition of the BMJ5. The author was talking about the growing numbers of cars in town, and how they were unpleasant, the exhaust smelled, and speculated about the danger to health from the fumes. This was only 100 years ago. Cars are not normal! They are a new, and hopefully brief, blight on the planet. When people say they can’t live without them, remember that homo sapiens have existed for about 300,000 years.6
1. I had the thought after writing this that maybe normal here could be interpreted in the sense of normative, that is a desirable state, rather than a typical one. In that case, I wish that people realise that the world can be better than it was two year ago.
2. Heraclitus - https://en.wikipedia.org/wiki/Heraclitus#The_River
3. This particular example stolen from Vinay Gupta, but I can’t remember from exactly where.
4. 1908 is the year production started of the Ford Model T.
5. POISONOUS MOTORS, in Br Med J: first published as 10.1136/bmj.2.2495.1302 on 24 October 1908.
Also of interest in the same edition, “ROAD MAKING”, talks about the new requirements placed on road design and signage by motor vehicles, and “STORAGE OF GLYCERINATED CALF LYMPH” is interesting to compare to mRNA freezers.
6. I am aware disabled people exist, and technology can help. This is not why most people use cars.
]]>“In May 2017, Colombia’s Constitutional Court granted legal personhood to the Río Atrato”
]]>It’s sold as an investment book, subtitled “How you can thrive when oil costs $200 a barrel”, but the majority of it is a general overview of the oil situation, with a few pages of specific investment advice at the end.
I found it interesting to read as a book about the environment, resource consumption, war, and survival. It got me thinking about some possible strategies to reduce planetary damage.
Mostly I hear about the climate crisis - the global heating caused by CO2 from burning fossil fuels. What I don’t hear people talking about is the problem of what happens when the oil runs out. Regardless of planetary heating, no more oil is a big problem. I think the main idea of the book is that oil is a finite resource. Basically all the oil to be found has been found, and demand is only increasing, particularly from developing countries. The effect on price is obvious.
Following are notes and quotes from the book, and at the end further thoughts provoked by it.
There are gargantuan energy-income sources available which do not stay the processes of nature’s own conservation of energy within the earth’s crust ‘against a rainy day.’ These are in the water, tidal, wind, and desert-impinging sun radiation power. The exploiters of fossil fuels, coal and oil, say it costs less to produce and burn the savings account. This is analogous to saying it take less effort to rob a bank than to do the work which the money deposited in the bank represents.
– Buckminster Fuller quoted from “Utopia or Oblivion”
Governments can’t do long term. They are limited to short term thinking - the next election at most.
Authoritarians:
Groupthink ruins ability to make decent decisions or even admit to seeing the facts.
Asch Conformity Experiment. People were asked to compare the length of lines. When most people in a group were instructed to give an obviously wrong answer, 70% of subjects said the medium line was the longest.
Milgram experiment. People will give someone painful or possibly fatal electric shocks when told to by someone in authority.
Conclusion seems to be: we have leaders in a structure which cannot deal with the problem. This is human nature. Therefore, we must not trust or submit to undeserved authority, and take action ourselves.
China and India are growing. Whatever grows the fastest will come to dominate. They already use 90% what US does, by PPP (Purchasing power parity), i.e. physical stuff and services. 35% world population.
(Appears to be 36% in 2018 by my calculation from UN data).
Gas prices in the UK have increased 250% since January, largely because China is buying it all. (BBC)
Historically, oil price increases did not reduce demand (in the 1970s). Maybe this shows people are forced to buy oil if there is no alternative. Need to develop better tech / infrastructure.
He links oil usage, and generally energy usage, to economic growth. Restricting supply causes industry shut down and job losses. Conservation of oil would cause unemployment.
What is economic growth though? Does it mean an increase in real wealth, or just that the numbers keep going up? It seems to mean bigger numbers, but maybe we can have better lives with better quality stuff and less waste from throwing away broken things.
“Two-hundred-dollar oil will challenge the government’s ability to manage the economy through fiscal an monetary policy.” Expect stagflation as in 1970s, he says. We have not got to $200 yet, but it is high, and we have the additional problem of covid. People are demanding higher salaries to come back to work after a long period of being under paid, governments have been printing money over lockdowns, and may do so more to subsidise gas bills, so I expect it’s going to get worse.
“economic growth has become an absolute necessity in today’s leveraged, high-debt world”. “a debt-heavy economy cannot tolerate a slowdown in growth.” See also “Money as Debt”, a documentary film that can be found on youtube which explains the requirement for growth of money supply to pay off debt.
the transition to a zero-growth society would likely bring about a host of problems. In the first place, it would mean considerable changes to our culture. We would have to learn to see higher consumption as immoral and measure progress more in terms of quality and efficiency.
We need to change what we want.
renewable forms of energy carry a negative stigma because people associate them with hippies. Oil, by comparison, is as much a part of mainstream, conservative America as ten-gallon hats, fast cars, baseball, and Thanksgiving turkey. It is what “real” men living in the “real” world make “real” money from.
Luckily I think the stigma of renewable energy is reducing, but the challenge remains of how do you get a Texan to stop driving and eating beef? Hm, what if pollution causes sexual problems for men? :p
These are vague notes I’ve not thought about much yet.
Sell climate change action to people by what they care about: money and power. Convince people oil prices are going up, so you need to switch to something else for profit. Certainly with gas in the UK, price increase is now obvious, but it is amazing how people are able to deny inevitability until it is here.
How do you make solar power cool like Musk did for electric cars? What would Saudis think of building / installing solar panels? I guess laugh, but they have so much sun!
Reliance on oil is reliance on foreign countries that have it. Develop local manufacturing of solar power and electronics or you are fucked by China. Also China has 80% of world lithium supplies, I read somewhere, and demand for that is huge with a switch to electric vehicles. Interesting the US is sending some new nuclear submarines that way. AUKUS
The book says: “The effort required will be huge - on the scale of fighting a major war.”
Consider energy shortage a matter of national security. Thereby get the DoD to wage war on energy shortages and climate change. Not by stealing oil though, as they generally have specialised in so far. Put the budget into developing alternatives, and resources into mitigating climate disaster, building the climate refugee camps, evacuation logistics, etc.
Make it a matter of pride like the space race. USA is better than Russia because we have the best fancy new green tech.
Make renewable energy more profitable. Subsidies?
Make solar panels feel like oil and meat for Texans.
Mentally bring the climate threat forwards so even short sighted people can see it and panic. How? Link current events to climate change, …
Offer individual advice which also helps society.
Helps create some sort of climate of possibility of government listening to people maybe. Most people ignore hippies though, and protesters are automatically hippies / punks / anarchists / other bad words. Try to rebrand protest I suppose.
Show direct, specific, real threats. People don’t listen to something that might happen in the future. You need to make it personal and immediate. Also it should be directly related to the problem, not something with a vague link to the people you are complaining about.
Subsidies for oil industry are unamerican. They are unfair to small businesses. They go against free markets.
Pollution is unhealthy and costs NHS / insurers money. You tax money must pay for the damage.
]]>One of the first things I noticed was the missing components. The filter inductors and capacitors on the mains input are missing. These components are important in a switching power supply to avoid transmitting noise over the HF spectrum. Probably what has happened is they were put in to meet regulator requirements, then once the design had been approved they were removed to save money - most people would not notice. This is really nasty and poor quality, and probably illegal.
There is no DC converter chip or microcontroller - it is made from discrete components and a comparator IC.
There are some more photos on flickr.
Does the charger correctly implement a charging algorithm which is good for the battery?
I investigated the output by loading it with resistors. I powered the charger with the DC “car charger” input rather than the mains.
The data shown above are here.
The charger supplies 0.57A until the voltage reaches somewhere between 7.35 and 8.22 volts and then eventually settles down to a constant voltage of 8.39 volts.
The battery I have is 7.2V, 1200mAh (more than the original battery). 0.57A gives a charge rate of 0.78C. Between 0.2 and 0.7 C is recommended by [2] so this seems quite reasonable. I don’t have a datasheet for the battery so I will assume the typical peak voltage of 4.2V / cell, 3.6V nominal. This means the maximum voltage from the charger should be 8.4V, so it is within a +/- 50mV tolerance[1]. Within the normal charging range, it seems to do the right thing.
However, typically when a cell is discharged to below about 2.5V, it must be charged at 0.1C or less until it reaches 3V [3]. The charger is happy to supply 0.78C at 4.93V, or 2.47V / cell. It seems the charger could damage batteries which are severely discharged, perhaps by self discharge from being stored for some time.
The other difficulty is when to terminate the charge. This should happen when charging current drops to something like 10% to 3% of the peak charge current, depending on which source I believe. That is 17mA minimum anyway, and I did not measure that low, so I don’t know what the charger does. It is still charging at 42mA though. Some sources recommend a 1h timer to limit the constant voltage phase. It looks like the charger doesn’t have this. One way to simplify the termination requirement could be to use a voltage less than the peak, which this charger seems to do by 10mV. This might avoid overcharging but not fully charge the battery. I don’t know for sure though.
I also note that the charger does not make use of the temperature terminal on the battery.
At 14V in and 0.57A, 7.32V out, efficiency was 68%. That’s rather disappointing. I think 95% should be possible, or at least 90%. I got a rough idea of this from the datasheet for the MAX1556 - it claims around 95% efficiency at 600mA, 5V in, 3.3V out.
The charger is probably adequate if the battery is not discharged too far, it is supervised, and the battery is not left in the charger for too long. I will use it for now but would really prefer something better.
cf. subvertising, tobacco advertising bans.
]]>I suppose the common theme here is a sense of connection and being part of a system. In many cases though there is also a sense of distance - looking down on the world and observing. The cat on a shelf keeping an eye on things from a safe distance.
I suspect this feeling is not specific to infrastructure, but is probably triggered by other things for other people. Maybe this is just the engineers’ way of insight of god.
I remembered when writing this also a time in a house, high up on a magic island, where the tap had a sign saying “Holy Water”. The intention was to avoid waste, and it was normal tap water, but we agreed that it was indeed holy, as it is something that keeps people alive.
Infrastructure is generally like that. The civil engineer can probably see the beatific vision in a sewage treatment plant.*
The Temple of the Purification of the Dishes
* This idea is stolen straight from Alan Watts and his beatific ashtray.
]]>I used the parts I had, which for the 3V rail were 100nF and 10uF ceramics, and a 470uF Rubycon. The voltage reference also got a 10uF. The replaced cap was a 100uF Chong.
Autoranging is still frustratingly slow, but I think it is probably a bit faster once it’s in the right range.
]]>Streams at https://streaming.media.ccc.de/rc3 IRC: #rc3 on hackint
]]>