Open Source is now 20 years old, having been officially started as a label on 1998-02-03 at a strategy session in the offices of VA Linux Systems, in Mountain View, California.
More information about the open source concept, including a time line, can be found at: https://opensource.net/
This weblog post was updated 2021/12/21. to eliminate Seeds from the title. This post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, updating the text to a block format and other minor formatting changes, the text above this paragraph remains as it was before. Any significant content changes are found below this paragraph.
Within a year we should have fiber broadband to the house. Today, 2018-01-24, we have to make a decision and sign papers ordering products.
Currently we have a «Bredbånd 5» ADSL subscription from Telenor. It offers 0.2 – 6 Mbit/s (down) and 0.1 – 0.6 Mbit/s (up). We pay NOK 358 per month for this. In addition, we pay NOK 196 per month for telephone, for a sum of NOK 554 per month.
Our new broadband supplier will be NTE. They want to supply us with «100% trønderfiber helt inn til husveggen» = 100% Trønder [an adjective referring to people and things from our county] fibre right to the walls of the house. This presents a conundrum, since the fiber is being sold under the brand name Altibox, which is being used by over 35 local Norwegian and 6 Danish FTTH (Fiber to the House) networks, and was originally set up far from Trøndelag county by Southwestern Norwegian multi-utility firm Lyse Energi in 2002 under the name Lyse Tele. It became Altibox in 2009. Since 2002, over 360 000 houses have been connected, the majority self-install (over 80 per cent).
At the top of the information sheet provided by NTE is their blurb about fremtidens tv-løsning = futuristic television solution. I didn’t even know that there was a future to television. Personally, I am very happy to decide what I want to watch, and when to watch it. So, we won’t be watching television, or buying any of the packages that cost NOK 1 099 or 1 599 per month, [providing storage, television options and standard 500 or extra 1 000 Mbit/s up and down, respectively.]
After having consulted with our children, we decided to buy the lowest speed product available: 50 Mbit/s up and down. Here is a breakdown of the costs, compared to the standard package. At a 30% income tax rate, NOK 6 588 per year after taxes is equivalent to earning NOK 9 411.
Product
50/50 Mbit/s Internet
Standard Package
Monthly cost
549
1099
Annual cost
6 588
13 188
Startup charge (NTE)
4 900
2 400
Connection charge (Inderøy)
12 500
12 500
First year costs
13 988
28 088
First year cost savings
4 100
Subsequent year costs
6 588
13 188
Subsequent year cost savings
6 600
This weblog post was updated 2021/12/21. to eliminate Needs from the title. This post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, updating the text to a block format and other minor formatting changes, the text above this paragraph remains as it was before. Any significant content changes are found below this paragraph
At the time of this update, we were paying NOK 659 a month = 7 908 a year, for 80 Mbit/s up and down. A speed test was conducted to confirm this, which it did. This is no longer available as a new product, but continues to be provided to those who opted in at an earlier date. Broadband, without television, now provides 150 Mbit/s for NOK 719 per month, or 8 628 per year. So, today, we ordered this, which required a telephone call. About five minutes after the order was placed, a second speed test was conducted. This confirmed that the new up and down speeds were available. The price for the standard and extra packages deliver the same content (storage, television channels and broadband) as before, but now cost NOK 1 229 or 1 729 per month, respectively.
In my retirement, I am currently a denizen of an analogue world that that roughly approaches my teenage ideal. In that world, Plywood, marine plywood especially, was the material I preferred to shape. The preferred shape being that of a hard-chined sailboat. The radial arm saw was, unquestionably, the most exalted workshop tool. Yes, Roy Henderson had one that occupied a central position in his workshop. When I think carefully about it, that is where my idea of a line of tools, Machine Alley, has come from. He had few options, as an under-used recreation room occupied most of the basement. It was in the rec room that his son, Grant, spent his time, building and painting plastic model cars from kits.
Roy’s shop was one of four that has influenced me. The second was a commercial workshop run by English immigrants building hard-chine, plywood hulled Enterprise sailboats from kits along the shores of Blind Bay, on Shuswap Lake, British Columbia. The third was the school workshop at Vincent Massey Junior High School, where I learned to use assorted woodworking tools, and found that mastering the jack plane was harder than mastering the band saw.
The fourth was the unloved workshop at my parents house in New Westminster. Its tools seemed to be from a previous century, and many probably were. They had belonged to my father’s Uncle, Thomas McGinley. He was the same uncle that had participated in the Klondike gold rush, but had otherwise worked as a carpenter. These tools were all rugged and heavy, designed for work on ship’s timbers or log cabins, rather than more delicate objects. I never saw my father use any of the tools. I’m not sure if it was from a lack of skills or a lack of interest. When my parents sold their house in 1972, these tools were disposed of.
These days I am more moderate in my opinions, but more excessive in my purchases. I am fortunate in being able to buy the tools I want. Yet, I hesitate to buy the best quality. I am buying the equivalent of Craftsman tools: Good, but not great. I don’t mind the challenges of working with imperfect tools. The fact that I may have to use extra time to adjust the table saw’s fence rather than have it snap into a precise position is a challenge with its own reward.
As I approach 70 years, I realize that the time I have to use analogue tools is limited. Yes, I am focusing on analogue woodworking tools. I am more comfortable working with wood than metal, or textiles or plastic or clay. In five years time, the worst of my infatuation with band saws, sliding compound mitre saws (UK)/ chop saws (US), spindle moulders (UK)/ wood shapers (US) and lathes should have eased. That is tomorrow. Today, I want to master this analogue world around me.
Because it is so many years since I used analogue tools seriously, I have to rebuild my skills. At the same time the workshop is being formed. The wisdom of what I had hoped would be a single line of stationary tools along a wall, Machine Alley, is being questioned. The table saw, an essential tool for transforming plywood, MDF and even OSB into useful components is demanding a more central placement. Already now the as yet un-purchased lathe has been repositioned in Machine Alley. The prudence of purchasing a separate thickness planer, rather than one in combination with a jointer, is being questioned. While tools are cheaper now, it doesn’t mean that they are easy to come by. I regret Norway being outside EU’s Customs Union. It makes my purchasing decisions more complex and expensive. Because importing goods is an expensive and bureaucratic hassle, Norwegian tool retailers and importers can ignore people like me, and just offer a selection of popular tools. All of the tools that I want, but cannot find in Norway, can be found stocked in Ireland.
At this point I would like to comment on my feelings in relation to my fate. It is complex, combining regret with acceptance, even contentment. Yes, I regret never having built my own house. Yet, I am sure that I could never build one in Norway, in a way that I would like to build it. Norway is a country without building inspectors, that allows each trade to police itself. A loose canon, such as myself or anyone without trade qualifications, would never be given permission to build such a major undertaking. In Canada, anyone can do anything, but it has to be inspected, to ensure that it meets the standards. I am equally sure that I would find it equally frustrating to build a house in Canada. I am not sure that I could regress to 24″ from 600 mm.
A lack of house building means that my workshop activities have limited scope. The workshop will ensure that improvements are made to the various rooms of the house, including the kitchen and living room. A minor addition or a shed will be added. Siding will be replaced, possibly with stucco. Furniture will be built. That could take up to five years. What will happen after that? Unfortunately, many makers do not plan for their future. They see their activities proceeding linearly, forever.
What I do see happening is that at some point old age will demand a transition away from analogue tools. My eyesight will worsen, and some of my skills may degenerate. Yet, hopefully, working in the workshop will keep my strength up. I am giving myself five years to accomplish my analogue goals. Everything has to be finished by 2022-12-31, although that date may be extended, health permitted.
Let me repeat that comment about my feelings in relation to my fate. “It is complex, combining regret with acceptance, even contentment.” What I am looking forward to is replacing analogue with digital, working more with workshop automation, home automation and robotics, including robots for the elderly.
I am not quite sure what this world will look like. Yet, in my later retirement years, I am looking forward to being a denizen of a digital world that is vastly different from my teenage ideal. In this future world, I may still be using plywood, but my table saw will be replaced with a CNC machine in the centre of the workshop. It will allow me to work with different processes, simply by replacing a head. It will work with different materials, some currently unknown.
Yet, transitions have to be planned. An analogue workshop will neither appear nor disappear by itself, a digital workshop has to be planned and implemented.
This weblog post was updated 2021/12/21. to eliminate Seeds from the title. This post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, updating the text to a block format and other minor formatting changes, the text above this paragraph remains as it was before. Any significant content changes are found below this paragraph
Version 2: Includes the conclusion left out of Version 1, and some corrections.
Does the world really need another meaningless abbreviation: DM&D? Probably not, but abbreviations are cheap, and give the impression that there are many users embracing the term, and the term is used so often that it is necessary to abbreviate it. The University of Buffalo, through Coursera (The MOOC organization) is offering courses in “Digital Manufacturing & Design”. They referred to something called opendmc.org (where dmc appears to stand for Digital Manufacturing Commons). This site only provides cryptic error messages, until one finds www.portal.opendmc.org, after which it is indeed possible to explore some of the site and meet a bunch of dead ends. Finally, one stated: “Our platform is currently in a closed beta.” So much for the openness of opendmc.org.
Now, the main reason I actually visited the site was to find out what distinguishes DM&D from CIM, Computer Integrated Manufacturing. This latter term has gradually won favour in all sorts of environments. It has been in continual use since 1973, with the publication of Joseph Harrington’s book, Computer Integrated Manufacturing. It has become the preferred term since 1984 when computer-integrated manufacturing actually began, promoted by machine tool manufacturers, and CASA/SME or the Computer and Automated Systems Association and the Society of Manufacturing Engineers. So why change?
In a quest for greater insight and illumination (in the more abstract sense of the term) I turned to Wikipedia, and their article on Digital manufacturing: “Digital Manufacturing is an integrated approach to manufacturing that is centered around a computer system.” This sounded suspiciously like CIM, just with a more abstract digital replacing the more concrete computer. Yet more enlightenment followed, “Overall, digital manufacturing can be seen sharing the same goals as computer-integrated manufacturing (CIM), flexible manufacturing, lean manufacturing, and design for manufacturability (DFM). The main difference is that digital manufacturing was evolved for use in the computerized world.” One could only ponder. Does computer-integrated manufacturing only exists in some non-computerized world? Perhaps CIM is only some form of primitive virtual reality. Readers are left to cogitate: Digital Manufacturing is Computer-Integrated Manufacturing evolved for use in the computerized world.
What could be better than cloud computing, except cloud-based manufacturing? The same Wikipedia article on Digital Manufacturing, states: “Cloud-Based Manufacturing (CBM) refers to a model that utilizes the access to open information from various resources to develop reconfigurable production lines to improve efficiency, reduce costs, and improve response to customer needs.”
That quoted text contains any number of insights (although the most probable number is 0). Unfortunately, this reader lacks the ability to understand what the text actually means. Could women and men of insight please help me understand this text? I would be eternally grateful. Yet, inside of me, I know there is nothing to understand. These are simply empty words.
The major challenge with texts about computer/digital manufacturing is the role to be played by people. Dark factories want to prohibit people from even entering them, at least during the manufacturing processes. At the other extreme, there is the growing field of collaborative robotics which in some way wants to hook up (as it were) humans and robots in the workplace.
As expected, trade unions are pressing for a humanized working environment. Tim Page writes in The Fourth Industrial Revolution: a breakthrough that must be humanized, ” So we must put people at the heart of digital manufacturing. The German engineering union IG Metall has developed some clear priorities for the introduction of this production revolution. Alongside Industrie 4.0, the German name for digital manufacturing, IG Metall have called for Arbeit/Work 4.0. This should include:
Job security and fair remuneration
A reduction of workload
A revaluation of activities
Better professional development and learning opportunities;
More time sovereignty
Informational self-determination
Involvement and participation on an equal footing
The introduction of digital manufacturing must be accompanied by the relentless quest for new jobs, better jobs, empowering jobs. The German approach, introducing this with the full involvement of the future labour force, is the right approach. It means working constructively with trade unions and other civil society organisations. ” http://touchstoneblog.org.uk/2016/11/fourth-industrial-revolution-breakthrough-must-humanised/
These are all very nice sentiments, but the pathway from “Industrie 3.9” (or where ever we are now) to 4.0 is unclear.
Martin Ford in Rise of the Robots: Technology and the threat of a jobless future, sketches a new economic paradigm in his tenth, and last, chapter. He writes about diminishing economic returns from education, cites Nicholas Carr’s The Shallows, which Ford regards as anti-automation. He then writes warmly about a basic income guarantee, especially from Friedrich Hayek’s perspective. This warmth continues as he writes about markets as renewable resources. Many other proposals are taken up, but in the end Ford presents no other solution than a basic income, bread and circus for the 21st century. Even though I gave it a 5 on Goodreads, the last chapter of Ford’s book was a depressing read.
There seems to be no need for yet another phrase (digital manufacturing) to replace Computer Integrated Manufacturing. Yes, manufacturing processes have matured, or at least aged, these past 45 years, but that is no reason to discard a perfectly good term. We still call a 2017 laptop a computer, even if it differs significantly from a Digital Equipment PDP-11/20 mini-machine from 1970.
Be thankful that today’s microcontroller is not your grandmother’s microcontroller.
In the early to mid 1980s, one could use any microcontroller one wanted, as long as it was a Motorola 6800 EPROM (Erasable Programmable Read Only Memory). Thinking about those experiences almost 35 years ago is painful. Sensors required their own army of hardware connections for analogue to digital conversion. Relays were needed to connect any form of actuator, even a light bulb. Devices were programmed in Assembler. Any and every mistake required the removal of the microcontroller’s top so UV light could flood the chip to erase code. Each erasure took about 15 minutes. It also required everyone to use UV glasses during this process to prevent eye damage.
The situation didn’t change until 1993, when Microchip Technology introduced the PIC 16C84 EEPROM. Those two Es stand for electrically erasable. EEPROMs don’t need UV light to be erased, which meant that code could be changed instantaneously. The other main difference was the chip’s architecture. Most computers, use von Neumann architecture, storing programs and working data in the same memory. PIC controllers used Harvard architecture, which separated programs from working-data, which allowed programs to be stored in cheap read-only memory. This resulted in a massive decrease in price, and a massive increase in the number of processes that could use microcontrollers.
About the same time, in Trondheim, Norway, Alf-Egil Bogen and Vegard Wollan developed a RISC (reduced instruction set computer) processor, hence AVR, that also featured Harvard architecture. This concept was sold to the Atmel corporation, who began producing AVR microcontrollers, such as the ATmega8.
The main challenge of using a microcontroller, was that the user was expected to know what they were doing. Even getting the software to program a microcontroller could cost almost USD 1 000. Microcontrollers at this point were not for amateur hackers. Parallax, Inc.defused this issue by launching the BASIC Stamp, a microcontroller with its own BASIC interpreter. It was especially popular with electronics hobbyists. With this the programmer was essentially free, but only worked with the BASIC Stamp. In addition, BASIC Stamp microcontrollers were expensive.
Wiring, an open-source electronics prototyping platform composed of a programming language, an integrated development environment (IDE), was developed by Hernando Barragán, starting in 2003. It was built on Processing, which offered a similar environment for multimedia, initiated by Casey Reas and Benjamin Fry. Although the Wiring IDE is written in Java, it comes with a C/C++ library that makes common input/output operations much easier. Wiring programs are written in C/C++.
At the Interaction Design Institute Ivrea in Ivrea, Italy, students initially used the BASIC stamp, but these units were expensive for students to buy. The Arduino project was started in 2003. It was a fork of Wiring that supported an ATmega8 microcontroller. The initial Arduino team consisted of Massimo Banzi, David Cuartielles, Tom Igoe, Gianluca Martino and David Mellis. but Barragán was not invited to participate.
Like the BASIC Stamp, the Arduino allowed any idiot capable of plugging in a printer cable and pressing download, to get a microprocessor to run. The printer cable was not only a power supply, it was also a mechanism for uploading code to the microcontroller.
The main difference between the BASIC Stamp and the Arduino board was that the Arduino was designed not only to give users simplicity, but to give the board longevity. The circuit had protective devices to prevent damage. My teaching experience shows that this is not foolproof, but a major help. Pin 13 includes its own LED, allowing the famous Blink program to run visibly, even without a LED fitted to the pin. It was a superior product.
Arduino has gone through some dark times. While in early 2008, the five co-founders of the Arduino project created a company, Arduino LLC, transferring ownership of the Arduino brand to the company. Boards continued to be manufactured and sold by external companies, with Arduino LLC receiving a royalty. This is a problematic situation for a company claiming to be open source. At the end of 2008, Gianluca Martino’s company, Smart Projects, registered the Arduino trademark in Italy (and thus the EU). When Arduino LLC tried to register the trademark in other areas of the world they discovered that it was already registered in Italy. Negotiations with Gianluca failed to bring the trademark under control of Arduino LLC. In 2014, Smart Projects began refusing to pay royalties. Its new CEO, Federico Musto, renamed the company Arduino SRL and created the website arduino.org.
By 2008, innovation had stopped at Arduino with the introduction of the Duemilanove board. Since then, there have been many different boards using various processors and the IDE code has been tweaked. However, there have been no revolutionary improvements. The Uno is an enhanced Duemilanove, and has been Arduino’s main board since September 2010.
One of the challenges with working with an Arduino come from the design of its pins. These pins are located on the top of the board, using female 0.1-inch (2.54 mm) headers. While plug-in application shields are available that use male pins that fit into the female headers, this arrangement does not support effective use of breadboards. With the development of the Arduino Nano, male header pins on the underside of the board allow it to plug into breadboards. It lacks a DC power jack, and uses a Mini-B USB cable. This minimal board measures 43.18 mm x 18.54 mm.
Perhaps the reason for the popularity of the Duemilanova and the Uno (and the relative failure of the Nano) was their use of the first two in Arduino kits, that were often made and supplied by third parties. The first ten kits using Duemilanovas were purchased for Leksvik Upper Secondary School and came from Oomlout . ( http://www.oomlout.com ) who described themselves in the following way: “What We Are We like to describe ourselves as a plucky little design house. We focus on producing delightfully fun, open source products. Where We Are A scattered organization we are based in Vancouver, British Columbia (manufacturing). We have satellite offices in Leeds, Yorkshire (design), and Point Roberts, Washington (shipping). There is no sprawling corporate campus to offer tours around, yet. But if you are in any one of our neighborhoods, drop us a line and we’ll see if we can arrange something.”
The kits came with a plexiglass plate that held the Arduino board, and a half-size breadboard in place. There was also a box with sufficient hardware to allow at least ten to fifteen projects to be made. These projects showed the versatility of the microcontrollers in general, and the Arduino in particular, in many divergent areas. It was a winning formula.
MKR boards have their own form factor 61.5 mm x 25 mm, while claiming to use the minimal format found on the Arduino Nano. This means that the boards can be fitted directly onto a breadboard. The main difference is that the MKR boards use 3.3 Volts, and they will fry if supplied with 5 Volts. With low power consumption, MKR boards are targeted as battery-powered IoT “edge” devices, to be programmed via the Arduino IDE software.
There are now five Arduino MKR boards, each made for to satisfy a particular connectivity issue: MKR ZERO, MKR1000, MKR FOX 1200, MKR WAN 1300 and MKR GSM 1400. The person responsible for the MKR boards obviously has a defective keyboard, since the names use upper case letters exclusively.
MKR ZERO distinguishes itself by having an on-board SD connector with a dedicated SPI interface (SPI1) that supports music files. Two libraries also support music applications: The Arduino Sound library can be used to play and analyze audio data, while the I2S library uses the I2S protocol to connect digital audio devices using the electrical serial bus interface.
Arduino MKR 1000 is similar to the MKR ZERO adding conventional Wi-Fi connectivity but lacking an on-board SD and SPI interfaces.
A LPWAN (low-power wide-area network) allows long range, low bit rate communicationbetween battery-driven sensors and other connected objects. LPWAN contrasts with a wireless WAN designed to connect users carrying large amounts of data. The primary advantage of a LPWAN is that it allows the deployment of sensors without investment in gateway technology.
The MKR family has two different LPWAN boards.
Arduino MKR FOX 1200 provides SigFox connectivity, and includes a two year subscription to the SigFox network. Sigfox uses proprietary technology for communication on the ISM (Industrial, Scientific and Medical) radio band, with a low energy, wide-reaching UNB (Ultra Narrow Band) signal that passes freely through solid objects. Sigfox is a network operator, so users have to wait for them to deploy, and then pay a subscription fee for each device (about 1€ per year).
Arduino MKR WAN 1300 also provides LoRa connectivity, but by using a Murata Lo-Ra module. This uses the CSS (Chirp spread spectrum). With this form of LoRa, anyone can deploy a network to just cover a local area. It can be done at any time, and there is no subscription fee.
Arduino MKR GSM 1400 provides global 3G communications. It does not come equipped with a SIM card.
This weblog post was updated 2021/12/21. to eliminate Seeds from the title. This post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, updating the text to a block format and other minor formatting changes, the text above this paragraph remains as it was before. Any significant content changes are found below this paragraph
I have totally given up on Arduino as a suitable microprocessor for any projects that I work on. Currently, I am using Raspberry Pi Picos, but would also consider using Teensy microprocesssors for audio projects. For more complex projects, Raspberry Pi currently offers even more powerful boards, including compute modules.
If communication is a time thief, then connectivity is a communications prison where inmates are compelled to waste their lives communicating with other inmates, who have no opportunity to listen. In my previous life I was expected to attend a lot of meaningless meetings. I did so reluctantly. Fortunately, I was also given the opportunity to interact with a smaller group of young people, which was much more meaningful. Unfortunately, I also chose to engage in a lot of trivial Facebook communication, until I cut out 95% of it. This post attempts to explain why social connectivity, through social media, such as Facebook, is doomed to fail.
This post is divided into three sections, labeled assumptions, consequences and justifications, respectively. The first looks at a few important characteristics of social connectivity. The second attempts to explain what is happening, and is yet another pedantic homily. The third examines why, from a quasi-mathematical, semi-logical perspective.
Assumptions
Assume that there are different levels of communication. Keeping to the nearest order of magnitude, these are divided into four levels (A to D).
Level A. There are 10 people in the world where every individual wants regular contact. This probably includes members of their immediate family, possibly a few good friends. It feels natural to communicate with these people on a daily basis.
Level B. There are 100 people in the world that every individual feels inspired to communicate with.
Level C. There are 1000 people in the world that every individual feel some sort of an obligation to communicate with.
Level D. These are the zillion other people populating the world, with whom every individual will have almost no contact, and feels no obligation for more than fleeting contact, at best.
Assume each person is prepared to spend one hour a day communicating using social media. Further, assume she would want to spend most of her time communicating with people belonging to level A, and least with level C, and only incidentally with level D. For the sake of argument, let us put some values on the various types of communication. Assume that writing a message takes twice as long as reading and reacting to a message. Assume that a standard message takes 2 minutes to write, and 1 minute to read. Second, Assume that every person in each group acts precisely the same way.
Assume people would be willing to spend 30 minutes a day communicating with level A people, twenty minutes writing and ten minutes reading. For level B, they would only be prepared to spend 15 minutes a day, ten minutes writing and 5 minutes reading. For level C, they would probably not be prepared to spend much time at all, but here the same values as for level B will be used, ten minutes writing and 5 minutes reading.
There are two ways in which information can be disseminated: privately, directly to one specific person, or publicly, broadcast to everyone at that level. Private dissemination corresponds to sending an email, a private message or an sms. Public communication corresponds to writing on someone’s Facebook wall or publishing a blog post.
Consequences
There is one important consequence of connectivity that I will attempt to express in three different ways (1a, 1b and 1c). In addition, there are two less important consequences (2 and 3).
1a. Facebook understands, and acts in accord to its understanding. Algorithms matter.
1b. Facebook consumers don’t understand, and through their actions show their lack of understanding.
1c. There is no way for people to engage in meaningful two-way communication, with a thousand other people. In fact, it doesn’t work with one hundred. It barely works with ten.
2. There are mechanisms to broadcast messages, but normally one wants to restrict communication to specific demographics. If I look at the ads I have been targeted with in the past few days, I seem to be in the market for: a) an older, single Russian woman; b) a Jeep Renegade from an Ottawa car dealer; c) windows and doors from a Norwegian supplier. Note: this ad was in The Independent, and I actually clicked on it because it was relevant! Pinterest seems to do the worst job of targeting me. Almost all of the pins they think I am interested in are related to soccer (that’s association football, for some of you readers).
3. There seems to be a gradual understanding that social media are using people for their own purposes. Facebook is not really interested in helping me connect with long-lost girlfriends from a previous century; they want me to focus on buying products and services that they have mediated. They find these by examining what I read and write on their media.
An aside. At the dawn of the computer age, in my student days when IBM 360 machines were the rage, I was encouraged to save myself time and money by filling in a form (and investing money) to get the names and telephone numbers of three women who, on the basis of a computer analysis, were ideally suited for me. What I hazily remember in the fifty years since then, is that I only actually met one of these ladies, spending half an hour drinking coffee at Brock Hall. It struck me that she was the least appealing woman I had ever met. She admitted, that she had been given over thirty names, and she was working her way through the list.
The minimum requirement for computer dating to work, is to have an approximately equal number of participants on each list. This just doesn’t happen. The Canadian dating service company, Ashley Madison, is infamous for its data breach. Annalee Newitz, Editor-in-Chief of Gizmodo, stated that there were only 12 000 regularly used female accounts, out of 5.5 million at Ashley Madison, but she has subsequently disavowed this analysis. Adultery seems to have gone out of fashion at Ashley Madison. Its motto has changed from “Life is short. Have an affair.” to “Live the Moment”.
Justification
Here are the results formulated as a matrix:
Level
A
A
B
B
C
C
Method
private
public
private
public
private
public
People
10
10
100
100
1000
1000
Messages out
10
10
5
5
5
5
Messages in
10
100
5
500
5
5000
Read
100%
10%
100%
1%
100%
0.1%
Time to read all
10m
100m
5m
500m
5m
5000m
Messages out
10
1
5
0.05
5
0.005
Days to wait
1
1
1
20
1
200
What the results mean
In any situation dominated by private (one-on-one) communication, the number of communiques in, is equal to the number of communiques out. This applies at all three levels (A to C). There is no problem sending and receiving private communiques.
Difficulties arise when mass communication is used, such as writing a post on a wall, or sending out an email to everyone in one’s address book. Level A: With 10 people, 1h 40m has to be devoted to reading these public messages. Level B: With 100 people, reading increases to 8h 20m. Here one is sending and receiving messages at half intensity compared to level A. At the same intensity as level A, it would take 16h 40m. Level C: With 1 000 people, reading increases to 83h 20m. At the same intensity as level A, it would take 166h 40m. There are 168 hours in a week.
If the number of messages was to be reduced to correspond to the intended reading rate, then only one public message a day would be allowed at level A. This is reduced to one message every 20 days at level B, and to one message every 200 days at level C.
There is just no way in which everyone can engage in mass communication. This fact of life brings us back to reality. If mass communication is to occur, then algorithms have to be employed to limit who receives which communiques.
Behold, my lifetime favourite keyboard is the Logitech K380, in blue with teal highlights. One currently resides on my desk. I am contemplating buying a few to hold in reserve, to replace the current one when it wears out. I first came across a live version of the K380 in a Staples store on a trip to British Columbia in 2016, and knew immediately that this model was for me.
One important reason for its appeal is its small size, best demonstrated here in a photo by J A Watson. While the keyboard is compact, the individual keys are relatively large, and with appropriate spacing, at least for me, a person with large hands. One can feel when a key has been depressed, but there is no audible click. It is a relaxing keyboard to use. (For A J Watson’s impressions of the keyboard, see: http://www.zdnet.com/article/hands-on-logitech-k380-compact-multi-device-bluetooth-keyboard/ )
Three different Bluetooth devices (computers, tablets, smartphones) can be connected to the keyboard. To pair the K380 to a device, press and hold one of three highlight colored keys (F1/F2/F3) on the keyboard’s top left for three seconds. When the associated status LED above it starts to blink, the keyboard is ready for pairing. It stops blinking when pairing is complete. Bluetooth 3.0 allows connection to a wide variety of devices. The same LEDs used in pairing, indicate which device is active, if any.
Its small size, light weight and multi-device capability ensure portability. It is powered by two AAA batteries, which Logitech claims should last two years. A power on/off switch is positioned on the upper left side.
Did I consider other keyboards, before buying the K380? Yes, I examined both the K480 and the K780, but rejected both.
The apparent advantage of the K480 is its ability to hold a smartphone or tablet at an angle suitable for reading. At the same time, this adds to the keyboard’s dimensions and mass. If the K380 were not available, I am sure that I could live with a K480.
Like the K480, the K780 can also hold a smartphone or tablet in its grasp. If both the K380 and K480 were unavailable, I would probably still reject the K780. I do not, and never have, understood the appeal of numeric keypads. I make an exception for those unfortunate people who have to enter large quantities of numeric data regularly, and who may find a keypad user friendly. However, that involves very few people. For most users, and from an ergonomic perspective, keypads require the movement of hands away from the keyboard, a positioning of hands on the keypad. At the end of its use there is a similar need to reposition hands once again on the keyboard. Personally, I have no issues with entering numeric data using the number keys on a K380. (Yes, I typed all three of those numbers in without repositioning my hands, or making any errors.)
This weblog post was updated 2021/12/21. to eliminate Deeds from the title. This post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, updating the text to a block format and other minor formatting changes, the text above this paragraph remains as it was before. Any significant content changes are found below this paragraph.
For forty years vendors of computing equipment and their attendant programs have used power to sell products. A year on, and a revised product is unveiled as yet more powerful. At the same time, these devices are diverting energy from their primary task to run fans and other cooling equipment, in an attempt to mitigate the negative effects of their energy usage, notably the production of heat, that creates unbearable working environments for people, not to mention silicon.
Power architecture refers to IBM’s RISC microprocessors, promoted by power.org, used in the PowerPC and the Apple PowerBook. Power software was an IBM operating system enhancement package. Not to be outdone, Microsoft has PowerPoint, its slide presentation software, and PowerShell, a task automation and configuration management framework. Power is pervasive.
The computer industry is not the only sector to be obsessed with power. Vehicle manufacturers are worse. Not only do vehicles come equipped with factory air and cassette tape decks, there is an endless supply of power products, including power brakes, power steering, power windows, power seats and the more generic power accessories. People unable to understand kilowatts, can even use horsepower to express themselves, 1 HP = 746 W. Even the Swedish Amcar magazine is called Power.
Give me adequate power, but nothing more than that. Purchasing computers always involves compromise. Along with numerous Gigabyte Brix models, another desktop computer I considered before purchasing an Asus VivoMini VC65, was the CompuLab Mint Box Mini. It comes with Linux Mint pre-installed, but with the Mate desktop. It has 64 GB internal SSD storage, compared to the VC65’s 128 GB SSD and 1 TB HDD. While the Mint Box is fanless, the Asus retains a fan. I decided to purchase the Asus because it offered the best compromise, and was the only computer on my short list to have a DVD-reader.
Youtube Vlogger Joe Collins, in his Top 5 Mistakes New Linux Users Make, has several recommendations regarding equipment. Several of them are broken in the Mint Box Mini, including his advice to use Intel processors, and avoid AMD graphics cards. The Mint Box is powered by a 1GHz AMD A4 Micro-6400T 64bit Processor (Quad-Core), 4GB DDR3 Ram, AMD Radeon R3 Graphics and Realtek HD Audio.
I am truly thankful that the age of fanless computers has arrived. Miniaturization without excessive heat. Silence. I am equally thankful that the age of electric cars has also arrived, and an age of autonomous vehicles is on the horizon. If not silence, at least less noise. I will not mourn the disappearance of Harley-Davidson.
This post presents Marmot Chirp, an Intelligent Workshop Assistant (IWA) to be used at the Unit One workshop. It is an implementation of Home Assistant that could find its physical expression on a Raspberry Pi. However, there may be issues that require other hardware to be used. Thus, a more likely alternative would be a Gigabyte Brix EKi3A IoT unit with:
Intel 7th Generation Core i3-7100U Processor
Fanless 0.76 liter volume design: 180(W) x 117(D) x 36(H) mm, Weight 978g
1 x COM port (RS232)*
2 x USB 3.1 (1 x USB Type-C™), 2 x USB 3.0
HDMI 2.0 plus Mini DisplayPort 1.2 Outputs (Supports dual displays)
2 x SO-DIMM DDR4 Slots 2133 MHz, Max 32GB
1 x M.2 SSD (2280) slot
IEEE 802.11ac, Dual Band 1×1 Wi-Fi & Bluetooth 4.2 NGFF M.2 card
Gigabit LAN
Headphone/Microphone Jack
VESA mounting bracket (75 x 75mm + 100 x 100mm)
Regardless of the computer used, it will have to be fitted with additional equipment: speaker, microphone and light. The light will be off (indicating system is not functioning), steady on (system is in use), pulsing on (system in standby mode).
Why Chirp?
An intelligent personal assistant needs a name, so it knows when it is being activated. Female names are often used, eg Alexa and Siri. I pity the poor girls with those names, or anyone living in proximity to them, with such assistants. Similarly, an IWA has to be addressed, so it knows when users want something done. One of the most important characteristics in naming an intelligent assistant, is that the name is unique, and not used in normal conversation. In a workshop, calling the IWA Plywood or Lathe is just asking for trouble. Here, the name used is Chirp.
Dialogue
There are a lot of different situations in a workshop where a user would want to interact with a IWA, and other situations where the workshop owner, through the IWA wants information from a user.
Much of the terminology used would be based on voice procedure used in other situations.
1. Visitor arrival
The IWA has proximity and other sensors to detect the presence of a visitor.
A. Known user.
IWA: Welcome to Unit One. I’m Chirp, an intelligent workshop assistant. Who are you?
Brock: Chirp, I am Brock
IWA: Welcome Brock. Your personal safety equipment is in box <A/ B/ C/ C/>.You are logged in at <time> on <date>.
B. Unknown user.
IWA: Welcome to Unit One. I’m Chirp, an intelligent workshop assistant. Who are you?
Cynthia: Chirp, I am Cynthia
IWA: I don’t recognize you, Cynthia. Please use the computer to log in or register yourself.
…
IWA: Thank you for registering, Cynthia. You are inside the Unit One workshop, where people use technology to make the world a better place. In front of you is box <A/ B/ C/ D> containing your personal safety equipment, including safety glasses, hearing protection, gloves, and respirator. Please use the equipment you need. Please note that there is a first aid kit and a fire extinguisher near the entrance, and a second fire extinguisher at the end of the workshop, near the Annex. These are to be used when needed. If you need help, just call me by my name, Chirp. You are logged in at <time> on <date>. Please remember to log out when you leave.
I have heard some interesting titles used by an IPA to address one or more users. The one in particular that I would want to avoid is, “Yes, Master.”
2. Other Events
A. Audio (video?) contact with house, referred to as Fjellheim, meaning mountain home, in Norwegian:
User: Chirp
IWA: Yes?
User: Chirp, Patch me through to Fjellheim.
Event: Telephone rings Brock or Trish, depending on who is home.
(Postive response) IWA: Contacted Fjellheim
(Negative response) IWA: No one is at home at Fjellheim.
With a positive response, a dialog would then follow. The IWA would be listening to the entire conversation waiting for its name and new instructions. The conversation would end with:
User: Chirp, terminate.
IWA: Conversation terminated.
B. Audio contact with emergency services:
User: Chirp
IWA: Yes?
User: Patch me through to <emergency services/ambulance/ fire department/ police>.
Event: IP telephone contacts appropriate service
(Positive response) IWA: Contacted <service>
(Negative response) IWA: Unable to contact <service>
C. Turn on electrical circuit for a particular machine (but not the machine itself, except dust control and air).
Negative IWA: Unable to disable <dust/ air/ drill/ band saw/ mitre saw/ table saw/ planer/ router/ sander>
3. Visitor Departure
Person: Chirp
IWA: Yes?
Person: <name of person> is leaving.
IWA: Goodbye <name of person> Please put your personal safety equipment in box <A/ B/ C/ C/>.You are logged out at <time> on <date>.
What other events need to be controlled? What other comments do you have about an intelligent workshop assistant?
Update: 2021-12-20 at about 21:00.
This weblog post was originally published 2017/10/29 at 05:49, which is four years and fifty-two days ago. This is a trial to see if eliminating Needs the title, has any consequences. That is because, at the time of publication, this post formed part of a Needs, Seeds and Weeds website that belonged to my daughter, Shelagh. In addition, other things are also out of date, or my opinions have changed. Apart from the title, the text above remains as it was, content changes are found below.
Internet connectivity. The workshop is in theory connected to the server (Mothership) with a fiber optical cable. Theoretically, inside the workshop the cable terminates at a switch. In practice none of these installations have taken place, yet. The cable has been purchased, tubing has been laid between the workshop and the house, but work is needed to get the cable from the outside of the house to the server.
Inside the workshop the switch could be something as simple as a Unifi Flex Mini, with five Gb Ethernet PoE connections, one in, and four out. For about ten times the price, 10Gb Ethernet PoE connections can be supplied using a Unifi Flex XG unit. Wifi should not be a problem if a Unifi 6 Lite access point is connected to one of the ports.
As discussed in a recent post about Forth, the machine controller would be programmed in Forth. This could be a single unit attached with an Ethernet cable to the switch. At the present time, this unit could be a Raspberry Pi 4B.
For design, engineering, administrative and other purposes a relatively high powered workstation computer should be available. An Asus Zen AiO machine was purchased used to be a kitchen controller. This machine was rejected, and could be taken into use in the workshop. It comes with a touch-screen.
With an access point, machine controller and workstation in use in the workshop, three of the four output ports of the switch are in use.
Rather than sticking with Raspberry Pi, the innovators at Cliff Cottage are using Asus Tinker Boards. The Tinker Board is simply a better machine. Not perfect, perhaps not even good, but better. A similar sea change may be happening with the Arduino.
The Fishino Uno is yet another Arduino Uno board, in term of connectivity and size. However, it is more than that. It has many features that the Arduino lacks: internet connectivity, large storage capacity using a micro-SD card and an on-board RTC (real-time clock) with battery backup. There is a minimal change to the Arduino form factor, caused by a 7mm wifi antenna overhang.
These improvements make Fishino better suited for home automation systems, than the original Arduino. The WiFi module allows the Fishino to be used as a WiFi station and/or access point, and allows smartphone control even without a WiFi connection.
In summary, some of the characteristics of the Fishino that make it better than the Arduino.
Fully compatible with Arduino Uno
WiFi module on board, that can be uses in station mode, access point mode or both
MicroSD slot on board
RTC (Real Time Clock) with backup lithium battery on board
Increased current capacity on 3.3V supply section
Connectors compatible with breadboards
For further information see: https://www.open-electronics.org/fishino-arduino-become-wireless/
Shield
A shield is a printed circuit board that fits on top of the main Arduino (or Fishino) board, offering additional attributes to the system. Here we will mention two that can be used in home automation.
Fishino Octopus
PWM – For the next few minutes please don’t ask what PWM means. Pretend that it refers to Portland International Jetport, or Peter Wallace Myrick or anything else … Just accept that there is something called PWM, and that it can be important.
Some motors, including those on sewing machines, require variable speed. In the past, a rheostat was used to adjust the electrical current flowing through it. This wastes power because it uses the same amount of energy regardless of the motor speed. What isn’t used to power the motor is released as heat.
What was needed was an efficient power adjusting tool. PWM solved this problem. On an Arduino / Fishino Uno there are only six PWM outputs. Many times this is far too few.
The Octopus expansion shield allows 16 PWM outputs and 16 additional digital inputs or outputs. Up to 8 of these can overlap, allowing a system to manage 128 PWM outputs and 128 digital inputs or outputs.
Fishino Colibri
Engineers do not always have the greatest understanding of how the natural world functions. Colibri means hummingbird, and surprisingly, it is not a fish. An octopus isn’t a fish either, but we won’t force the issue.
The Colibri is used to provide power to RGB LED lighting. These use 4 channels to provide a full range of colours using PWM control signals.
By the way, PWM stands for Pulse Width Modulation.