Josef Průša is appreciated for his open-source fused deposition 3D printers, He is shown here at the Prusa Research 3D printfarm, in Prague, Czech Republic. Being open-source means that the designs are available for anyone to use or modify. A 3D printer is a typical example of an embedded system, a product with a microprocessor inside. Photo Josef Průša.
In this weblog post, the focus is on you – as a user and maker of products containing microprocessors and software programs, and ensuring that these products meet your specific needs. While people may not currently have all of the skills they need to make what they want, skills can be learned.
Throughout my teaching career, I have had the fortune to meet many disillusioned pupils, both at a conventional secondary school, as well as in a Norwegian prison. While I find these pupils challenging, they have most often not failed themselves, as they have been failed by an educational system that is determined to compartmentalize/ standardize learning experiences, and create cookie cutter automatons out of living and feeling human beings. Thus, I delighted in taking on their specific circumstances to ensure that they could create for themselves something meaningful out of the limited educational opportunity I could provide them.
In particular, I remember two young women who were totally bored with the academic program imposed on them. They just wanted to get through it, so they could take higher education, become nurses and productive members of society. Instead, they were having to spend hours a week studying Norwegian literature, and other equally boring subjects. They decided to enroll in my technology class.
Since they both worked (part-time) at the municipal nursing home, they were well aware of the needs of live-in patients as well as out-patients. In the end they designed a prototype of an automated pill dispenser, for out-patients living at home. This prototype used an Arduino microcontroller. It is my belief that this experience allowed them to preserve their sanity, that had been seriously challenged by, for them, irrelevant writers of a long past, and forgotten age.
One of the most important elements in a technology workshop is the role of design. The essence of design is to specify how something is to be made, even if one cannot make anything oneself,
Suggestion: Spend today, or even longer, thinking of what you need, that is not being provided, or – if it is – has so many defects, that an improved version would suit you better. When you are ready, tomorrow or next year or ???,
STOP the Press! A pandemic has struck, and much of the content written for this post has become irrelevant!
Pre-pandemic considerations
The focus was to be on you using community based, technological workshops. These have many names, including hacker/ maker spaces. The one I know best is the Teknoverksted ved Spiren, the Technology workshop at Spiren = the Sprout.
They may be located inside a school or library. Alternatively, they may a separate entity, that is publicly or privately funded/ operated. Regardless, they provide facilities for learning and making, that use anything from no tech through old tech to high tech tools. These spaces are typically open to youth and adults. In alphabetical order, equipment may include 3D printers, CNC machines, laser cutters, knitting needles, scissors, sewing machines, soldering irons, table saws and welders.
They provide. a collaborative working environment. People spend their time learning and making. Hopefully, they then go on, teaching and helping others to make. Thus, over time, the skill base of the workshop improves, people with experience make more sophisticated products, when these are appropriate.
One of the first of these workshops was Fab Lab started by Neil Gershenfeld at the Massachusetts Institute of Technology (MIT). It is a small-scale workshop offering digital fabrication, or “a technical prototyping platform for innovation and invention, providing stimulus for local entrepreneurship. It is also a platform for learning and innovation: a place to play, to create, to learn, to mentor, to invent.”One major problem with the Fab Lab concept, is the extensive and costly list of equipment these labs are expected to provide.
Pandemic considerations
During the pandemic, one main concern is mental health. Some people need more help than others to survive through it. Take the weaker sex, men, as an example. Compared with women, men live shorter lives, have worse health, suffer 70% of injuries, commit 75% of suicides, access health services less and delay seeking health services more, spend less time with doctors, focus on physical problems, avoid discussing mental and emotional problems and …. The Men’s Shed movement as a mechanism for resolving some of these issues has been discussed in a previous weblog post.
Single people are especially exposed. It may be hard keeping one’s sanity living with someone, but it is more difficult when one lives alone, and every effort is being made to eliminate social contacts at work and play. Zoom meetings are not a suitable replacement.
During the pandemic, there are also large groups of people who are suffering more than others because of poverty.
Teaching technical skills works best in small groups, ideally individually. Even if workshops are shut down, small groups can meet. The size of the group can vary from two to, say, five. Outdoor, socially distanced events should be possible, augmented with individual on-line meetings. Those who have the technical skills for making embedded systems, are encouraged to seek out those who are most impacted by the pandemic, and help them learn new skills, or provide them with relevant embedded products.
As noted below, the upcoming post, Tidepool (2020-12-15) is devoted to automated insulin dosing, much of it involving embedded systems designed and made by amateurs.
Post-pandemic considerations
As this is written (2020-11-10) an 90% effective C-19 vaccine has been announced by Pfizer. Its main disadvantage is that it has to be stored at extremely cold temperatures. Then (2020-11-16) Moderna announced a 95% effective C-19 vaccine, with significantly reduced storage requirements. This post will be updated with further information when it is , regarding a post-pandemic world.
Upcoming changes:
Originally, the schedule of weblog posts in this series was: 8. Visual impairment (2020-11-24); 9. Hearing impairment (2020-12-01); 10. Dexterity impairment (2020-12-08); 11. Mobility impairment (2020-12-15), and 12. Computing: A Summary (2020-12-22). This has been changed to the following: Sensory impairing 2020-11-24) – which combines hearing and visual impairment; Dexterity/ Mobility impairment (2020-12-01); Telemedicine (2020-12-08) – which looks at telemedicine generally; Tidepool (2020-12-15) – which examines automated insulin dosing.
Update: On 2020-12-13 at 12:30, Tidepool (2020-12-15) has been changed to Nightscout. It still examines automated insulin dosing.
This charming resident of Cliff Cottage, Hardy, could be taking on new responsibilities as a voice activated room controller. In this role, s/he would need to be fitted with a Raspberry Pi microcontroller, a Power-over-Ethernet circuit, microphone, speaker and other hardware,
Note: The content provided here is an updated, abridged and modified version of a weblog post titled, , that dealt mainly with the use of room controllers.
My preferred term for this subject area is not smart home, but domotics. Yet, not everyone appreciates Latin terms, when perfectly good Anglo-Saxon ones exist.
Part of the reason for my dislike of the term, smart home is the ascription of smart to basic electronic devices that are dependent on sensors and logical circuits, to actuate motors, lights and heaters. Another part of the reason is the use of home. For many people, this is an emotionally loaded term for a residence. I suspect some public relations/ advertising agency imagined that substituting home for residence, house or building would result in increased profits. Admittedly, there are other, more obnoxious phrases than smart home, such as the internet of things, a phrase I increasingly avoid.
Various dictionaries show the noun domus refers to house, home, family, household (with dependents), school (of philosophy). The adjective, domestic, is in common use in a variety of contexts. As a noun, it can refer to a hired household servant. Hopefully, in the future it will refer to a robot offering similar services, although dombot could emerge. If it does, expect puns contrasting dombot with dumbot.
MQTT Broker and clients
At the heart of any domotic device is communication. Andy Stanford-Clark and Arlon Nipper authored the first version of the Message Queuing Telemetry Transport (MQTT) protocol in 1999. The name is a misnomer, as there are no queues with MQTT. Typically, there are numerous clients, and at least one (and potentially more) message brokers. A broker receives all messages published by clients and then routes these messages to subscriber clients. In a household, a client is usually a micro controller, perhaps a Raspberry Pi, while the broker is typically software located on a Network Attached Storage (NAS) server.
In a home automation situation, publisher clients are attached to sensors, and they publish sensor data; subscriber clients are attached to actuators, typically motors, but also heaters, vents lights and much more, that they regulate, at the most basic level – turn on or off.
The two most common open source home automation systems using MQTT are: open Home Automation Bus (openHAB), a project started in 2010, with code written in Java; and, Homa Assistant, started in 2013, with code in Python. While Home Assistant has been praised for its privacy features, it has been criticized for its file-based setup procedure. Recently, it has become more user friendly and accessible, with a simplified, web-based graphical user interface. An iPhone or Android hand-held device, can also be used as controller. This is especially important when one wants to control features remotely.
Brokers and clients manage: lighting, temperature and humidity (indoor climate), audio and video (entertainment), unauthorized access, smoke/ fire detection (security) and related services. Here, it is only considered in terms of a residence, but almost all types of buildings can use these: barns, shops, manufacturing facilties, etc. While it might appear that micro-controllers are in charge, managing these services using algorithms, most systems can have their decisions overridden by humans.
Ethernet
Wireless communication, using Bluetooth or WiFi, should be avoided, if possible, for most domotic/ smart home automation hardware, because there will always be throughput and power issues with these technologies. The one exception to this rule is the use of hand-held devices, including mobile phones and tablets, that use apps to function as remote controls for the system. These apps can be downloaded from the appropriate device app provider.
In the previous millennium, it was not uncommon for people to wire their houses with Ethernet cable. People who did so, and have not removed them, are the real winners. The reason for this is that room controllers need to communicate, and to communicate they need power.
The various Ethernet cables were standardized for different speeds. Cat 3 provided 10 Mbit/s in a standard presented in 1990. Cat 5 increased throughput to 100 Mbit/s in 1995, and then to 1 Gbit/s in 1999. Cat 5e offered 2.5 Gbit/s in 2016, while Cat 6 increased it to 5 Gbit/s, the same year. The Cat6A standard, is actually ten years older, dating from 2006, but provides 10Gbit/s.
The advantage of Power over Ethernet (PoE) is that it eliminates the need to install separate power cables. PoE technology sends the above mentioned 10/100/1000 Mbps of data and 15W, 30W, 60W, and up to 90W of power to devices over Cat5e, Cat6 and Cat6A Ethernet cables for a maximum distance of 100m. Cat3 and Cat5 cable can also be used with restrictions, They supply of 48 V with a maximum current of 400 mA using two of the available four pairs of wires on Cat 3 or Cat 5 cable. While this appears to provide a maximum power of 19.2 W, system losses will normally reduce this to under 13 W.
MQTT Software and supporting hardware
The household’s MQTT broker is typically a software program, housed in a server. Wifi-based handheld devices connect to the broker using Wifi, while Ethernet-based room controllers connect to it through cables, and a switch.
A switch is a box that allows multiple other devices in a local area network to be inter-connected. A typical commercial switch may have up to 48 different ports for cables connected to 48 separate devices. Home oriented switches typically have 8 ports.
Business users commonly sell off their equipment once the warranty period has expired. Servers as well as switches can be purchased by private individuals inexpensively, because businesses won’t buy used (read: out of warranty) equipment, and most people don’t know what the equipment can be used for.
Cables usually run between the switch and other devices through the walls of the house, although there is nothing to prevent more visible wiring, along walls/ ceilings/ floors. If people are considering remodelling their house, adding Ethernet cable is no real issue. Whatever cable is used, it will ultimately pay to use the fastest Ethernet cable type currently available that supports PoE. Cat 6 is more readily available than Cat 6A. Using either means that the cable won’t have to be replaced anytime soon. Hopefully, the cables will last 30 years, or more.
Room Controllers
Control units, like all devices, needs power. Power over Ethernet (PoE) is an ideal way to provide power, since Ethernet connectivity is the preferred approach to wired data communication. PoE eliminates the need to install separate power cables. Each controller is provided with power from the switch itself.
Several different types of microprocessors can be used in controllers, including Arduino, ESP and Raspberry Pi. At Cliff Cotttage, Raspberry Pis are used.
Room controllers, and similar devices, are one of the main categories of devices that need PoE connections. A typical unit could use a Raspberry Pi Model B 3+, sensors connected to general purpose input-output (GPIO) pins. with a PoE hat (hardware on top, which provides extra capabilities) to provide power and cooling, and – in addition – a Pimoroni Automation HAT to provide support for actuators. Some room controllers would have a 7″ touch screen. Others would be fitted for voice communication with microphones and speakers.
A Pimoroni Automation HAR on top of a Raspberry Pi Model B (Photo: Piomoroni)
Controllers need to be placed in the following locations: 1) access control at entrance doors; 2) living room; 3) dining area; 4) kitchen; 5) bedrooms; 6) study, studio and workshop areas. A few people may want to have controllers in 7) bathrooms and/or laundry rooms, while most prefer to avoid this. In addition, there shouldl be 8) PoE access points for WiFi. This allows WiFi connections in parts of the house that are inaccessible for direct signals to and from the server.
Other switches, without PoE, can be used for other devices dependent on higher power levels. These include: 1) a home theatre connections; 2) a printer and/ or scanner; 3) clothes washer and/ or dryer; 4) dishwasher; 5) refrigerator and/ or freezer; 6) stove top and/ or oven; 7) microwave oven; 8) kettle; 9) hot water tank; 10) heat exchanger; 11) heat pump or solar thermal controller; 12) greenhouse controller. Not everyone has appliances with Ethernet connections, but they are increasingly available.
A front-door access controller will typically have an infra-red camera, proximity sensor and infra-red light connected to it, that will be activated as someone approaches. Video of each event will be sent to an external location, that could be located anywhere in the world. A room controller may have proximity sensors as well as other sensors to register temperature, CO2, humidity levels and more. Data gained from these sensors and others throughout a house, can be used to activate lights, or heating, display time, temperature and other data on a touch screen. It can even listen to verbal instructions with a microphone and answer using a speaker.
Recently, a Compute Module (CM) 4 was released. Unlike more conventional Raspberry Pis, the CM is totally flat, but uses the bottom edge for connection. In the future, I hope to acquire a Compute Module 4 Development Kit to develop suitable applications based on these modules. CM boards are used because they avoid unnecessary components, making products potentially cheaper and smaller.
Update of content in Printers, published 2020-03-24:
The Raspberry Pi (RPi) 400 Personal Computer kit. The keyboard contains a RPi board running at 1.8 GHz (faster than the 1.5 GHz found in the RPi 4 Model B). There is also a mouse/ rodent, a RPi operating system (OS) on a micro-SD card, a (UK, shown here) USB-C based power adapter, an HDMI cable to connect the video and audio output to a display/ monitor/ screen/ television, and an official RPi Beginner’s Guide. Price, about £/ $/ € 100. Photo: Raspberry Pi Foundation.
The pandemic has increased hunger and homelessness, and prevented some people from attending a physical school. This might be because of a legitimate health reason, but in some cases it might just be political. Affected students need digital equipment to access online schooling. Hopefully, this is being provided by the educational authorities, but where it isn’t an inexpensive solution may be needed.
The Raspberry Pi 400 Personal Computer Kit, may be precisely what the teacher ordered. Of course, the machine can also be used for work and entertainment purposes, in addition to education. At a cost of about £/ €/ $ 100 it contains everything one needs in a basic personal computer (except the display/ monitor/ screen/ television). As long as a proposed display supports a high definition multimedia interface (HDMI) connector, it can be used. Unfortunately, even an inexpensive machine like this will not help if the real problem is a lack of broadband infrastructure.
2 × micro HDMI ports (supports up to 4Kp60) for video display
Video decoders: H.265 (4Kp60) / H.264 (1080p60 decode, 1080p30 encode)
OpenGL ES 3.0
microSD card slot for operating system and data storage
78- or 79-key compact keyboard (depending on regional variant)
USB-C power connector
Dimensions: 286mm × 122mm × 23mm
Currently, the English (UK) keyboard (with UK power supply) is available as is an English (US) keyboard with a North American power supply, and a French keyboard with a European power supply. Soon, these will be augmented with German, Italian and Spanish keyboards, with an EU power supply. Beginner’s guides are provided in the same language as the keyboard type: English, French, German, Italian or Spanish. There is no mention of an official Nordic keyboard.
For the first time a RPi has an on/off power switch, activated by pressing Fn+F10 simultaneously. Pressing these for two seconds will turn on power.
Unusually, the mouse plugs into a USB-port at the back and to the left of the keyboard. This makes it ideal for left-handed people. This appears to be a design issue (read: flaw).
A wireless, Bluetooth mouse will not have any such challenges. These cost about £/ €/ $ 7. If this is needed, then one should probably buy all of the components separately. A universal USB-C power supply costs about £/ €/ $ 9. The keyboard module is about £/ €/ $ 67. The Beginner’s guide can be freely downloaded. This results in a total price of about £/ €/ $ 83.
The RPi 4 series is noted for having temperature issues. The RPi 400 contains a heat spreader to dissipate heat from the entire machine, so no part will be too hot. It is claimed that there is enough thermal capacity to overclock = run faster than the design speed.
The RPi’s GPIO header/ interface/ connector is often used for interacting with electronic components placed on a breadboard = electronics test circuit, such as sensors or actuators (motors, for example). This may be only for educational purposes, or it might even serve real-world needs. It is this capability that makes the RPi a much better machine for learning about computers than a laptop or desktop machine.
The smallest and cheapest RPi model, the Zero, at about £/ €/ $ 5, is being used to provide computing power to ventilators being produced from locally available parts in Columbia, South America. The RPi 400 could also do this, but its capabilities far exceed the computing needs of a ventilator.
The RPi 400 PC kit is following a tradition that emerged in the early 1980s, with Acorn’s BBC Micro (1982), Sinclair’s ZX Spectrum (1982), Commodore’s Amiga 1000 (1985) , and others that integrated the motherboard directly into the keyboard. There was no separate system unit or keyboard cable. Just a computer, a power supply, a cable to a display, and (sometimes) a mouse.
The first RPi was launched 2012-02-29. According to the RPI Foundation, this new RPi 400 model, launched 2020-11-02, is about 40 x more powerful than the original. On launch day, Chris Barnat, on his Explaining Computers YouTube channel, made a video describing the RPi 400 in detail. Other single board computer (SBC) influencers, released similar videos.
The Fisker Ocean will be contract manufactured by the Canadian owned, Austrian located, Magna Steyr facility in Graz, Austria. Photo: Fisker, Inc.
While many Americans will be focused on their presidential election taking place today (2020-11-03), this observer is awaiting the result of the Massachusetts Right to Repair Initiative (2020), a referendum appearing on today’s Massachusetts general election ballot. This could update the state’s right to repair laws to include telematic electronic vehicle data. This was specifically excluded on the 2012 referendum that passed with 86% of the vote.
It comes as no surprise that Elon Musk is opposed to the Massachusetts Right to Repair Initiative (2020), and is actively encouraging people to vote no. Right to repair legislation is generally supported by consumers, independent repair/ after-market companies and associations. It is generally opposed by original equipment manufacturers (OEMs), such as Ford or GM, and dealerships.
The Clean Air Act of 1963, is a United States federal law that with the purpose of controlling air pollution. It has been amended several times since then. The 1990 amendments required all vehicles built after 1994 to include on-board computer systems to monitor vehicle emissions. The bill also required automakers to provide independent repairers the same emissions service information as provided to franchised new car dealers. California further passed legislation requiring that all emissions related service information and tools be made available to independent shops. Unlike the Clean Air Act, the California bill also required the car companies to maintain web sites which contained all of their service information and which was accessible on a subscription basis to repair shops and car owners.
Today, microprocessors control operation-critical vehicle systems: brakes/ ignition (on internal combustion engine (ICE) vehicles) / air bags/ steering/ and more. Repairing/ servicing requires computer diagnostic tools. At the same time, OEMs have taken on gatekeeper roles to control information and parts necessary for service/ repairs. Control, in the above sentence, is particularly aimed at restricting access.
Most ICE vehicles use a controller area network (CAN bus) to manage microcontrollers, smart sensors and other devices to communicate with each other without a host computer. Each of these components is referred to as a node, with a hierarchical structure in relation to each other. No two nodes are equal, one always ranks above or below the other. The network features a message-based protocol. When two or more nodes transmit simultaneously, it is always the highest ranking node that is allowed to continue.
The electronic control unit (ECU) is typically based on about 70 nodes, each featuring, say, a 32-bit, 40 MHz microprocessor with about 1 MB of memory. This is orders of magnitude less powerful than those used in laptop or desktop computers.
Each node has to be able to handle a large set of processing tasks. These include: Analog-to-digital converters (ADC) – where a physical property usually measured in volts is converted into a digital number; Digital-to-analog converters (DAC) – provide an analog voltage output to drive some component, with a digital number telling the system what analog voltage to supply; signal conditioners make adjustments to input or output data so that it aligns more correctly with real-world needs; communication standards are implemented capable of sending appropriate signals to other nodes. The CAN-bus communication standard allows for speeds of up to 500 kilobits per second (Kbps) using two wires.
The CAN-bus, and similar devices, simplify vehicle wiring through the use of smart sensors and multiplexing. In ancient times (prior to about 1990) a wire ran from each switch to the device it powered. The circuit was completed by grounding one terminal of the battery to the chassis.
Smart sensors are integrated components, that include not only the sensor, but an ADC and a microprocessor. This allows it to read a voltage, make compensations for temperature, pressure or other factors using compensation curves or calculations, and then send digital output signals onto the CAN-bus.
With multiplexing a microprocessor monitors sensors in one area of the vehicle, such as a door. When that a specific window button is pressed “downward”, the microprocessor will activate a relay that will, in turn, provide power to the window motor so it moves downward.
Among the parts carmakers buy assembled from external suppliers are instrument clusters. These are designed by the supplier to the vehicle maker’s specifications. This is advantageous for both for the maker and the supplier. However, it also takes power away from the OEMs, and gives it to suppliers, such as Bosch or Continental.
Some of the nodes include: Battery Management System (BMS); Brake Control Module (BCM) which may also incorporate an Anti-locking an Braking System (ABS) and Electronic Stability Control (ESC); Door control unit (DCU); Electric Power Steering Control Unit (PSCU) or a Motor-driven Power Steering Unit (MPSU); Human-machine interface (HMI); Powertrain control module (PCM): which may combine an Engine Control Unit (ECU) and a transmission control unit (TCU); Seat Control Unit; Speed control unit (SCU);Telematic control unit (TCU).
Confusingly, ECU is also used as an abbreviation for the Engine Control Unit, which is one specific node. Here, and in many other circumstances to avoid confusion, it will be referred to as an ECM = Engine Control Module. It uses closed-loop control. Depending on the intended usage of the vehicle, the ECM will optimize specific goals: maximum torque, maximum fuel efficiency, minimum emissions, etc.
The CAN-bus allows module to communicate faults (errors) to a central module, where they are stored, then sent onwards to an off-board diagnostic tool, when it is connected. This alerts service personnel to system errors.
With electrification already a reality, and autonomous driving becoming one soon, the CAN-bus methodology will be unable the flow of data. Tesla uses a dual (read: duplicate/ redundant) artificial intelligence (AI) based, Samsung produced microprocessor system, running at 2 GHZ, to control vehicles. Compared to the CAN system, these are extremely powerful,
Volkswagen’s ID3 is going the same route, where it is using high-performance computers (HPC) supplied by Continental for control purposes.
Some vehicle designers do not have the capability to set their designs out in life. A notable example is Fisker. Danish-American Henrik Fisker (1963 – ) has made some exciting vehicle designs, but not all of the businesses he has started have survived. The latest manifestation is Fisker Inc., which was started in 2016. It has presented a SUV EV, Ocean, and a pickup proposal, Alaskan. With the Ocean’s design finalized, it is outsourcing vehicle production of its Ocean to Magna Steyr, a Canadian-Austrian contract vehicle manufacturer. For Fisker, this will reduce manufacturing complexities and costs, in contrast to building and operating its own factory. Magna’s electric vehicle platform, Partial payment for this will be in the form of (up to) 6% stake of Fisker Inc.’s equity, currently valued at $3 billion.
Returning to the Massachusetts Right to Repair Initiative (2020), a yes vote can have dramatic consequences for the computing equipment put on vehicles (ICE as well as EVs) in the future. Starting with the model year 2022, all vehicles with telematic systems, sold in Massachusetts (but more likely throughout the United States, if not the world) will have to be equipped with a standardized open access data platform.
On 2020-10-15, Foxconn, the Taiwanese multinational electronics contract manufacturer, responsible for production of an estimated 40% of all consumer electronics sold worldwide, announced its MIH open platform for electric vehicles. If Tesla is the iPhone of electric vehicles, Foxconn wants to be its Android. Foxconn has been involved in automotive manufacturing since 2007.
Currently, according to Foxconn, the battery pack accounts for 30 to 35% of the total production cost of an EV; powertrain = 20 to 25%; Embedded Electronic Architecture (EEA) = 15 to 20%; body = 13 to 15%; otheto develop and establish an open industry standard for automotive electrical-electronic (E/E) architecturer, including wheels & tires = 10 to 12%.
The MIH platform would be prepared for 5G and 6G, comply with AUTomotive Open System ARchitecture (AUTOSAR) and ISO 26262, and be ready for OTA (over-the-air) updates and V2X (vehicle-to-anything) communication.
AUTOSAR has been in operation since 2003 Its founding members include: Bavarian Motor Works (BMW), Robert Bosch GmbH, Continental AG, Daimler AG, Siemens VDO (until its acquisition by Continental in 2008), and Volkswagen. Later members include Ford Motor Company, Groupe PSA, Toyota Motor Corporation (all 2003), General Motors (2004). Thus, it represents a very large proporttion of the automotive industry. Its objective is to create/ establish an open and standardized software architecture for automotive electronic control units (ECUs). Other goals include “the scalability to different vehicle and platform variants, transferability of software, the consideration of availability and safety requirements, a collaboration between various partners, sustainable use of natural resources, and maintainability during the whole product lifecycle.”
ISO 26262, Road vehicles – Functional safety, was defined in 2011, and revised in 2018.
The MIH platform can accommodate wheelbases from 2 750 to 3 100 mm, with tracks from 1 590 to 1 700 mm, ground clearance from 126 to 211 mm. Three battery packs will be available. Vehicles can be rear wheel drive (RWD), front wheel drive (FWD) or all wheel drive (AWD). Motors on the front axle can be: 95 kW, 150 kW or 200 kW. Motors at the rear can be: 150 kW, 200 kW, 240 kW, and 340 kW. This allows a range of vehicles from a FWD with 95 kW to an AWD with 540 kW.
Part of the MIH strategy is to use mega castings. Foxconn cites one example, where they reduced 7 front suspension body panels to a single cast part and 27 rear longitudinal rail components to yet another single cast part, using a 4.2 Gg = 4 200 Mg (commonly called a ton) die-cast machine.
This post will end with a rhetorical question: What is a vehicle device? There may be many answers, but there are three I would like readers to consider. The first, is that there are subcomponents on a vehicle that could be regarded as devices. Second, the vehicle itself is also a device. Indeed, unlike a so-called mobile phone, which is a hand-held device, a vehicle is a true mobile device. Other potential members of this category include robot lawnmowers, electric airplanes and exoskeletons that are sometimes used by people with mobility issues. The third, is that the production platform is the device.
A QNAP TS-251D A Network Attached Storage Server, suitable for most families. It can hold 2 x 4 TB hard drives. (Photo: QNAP)
Many families do not recognize their need for a server. In this weblog post, this need will be explained.
In almost every residence there is always a need for more storage space, physical as well as digital. A server can increase the effective use of the digital storage space. Some users need/ take more space than others. With a server, users do not need to get allocated the same set amount of space. Instead, the space can be pooled allowing greedier users to use more space than their more restrained counterparts.
People have many ongoing commitments, so it can be easy to forget something that is dependent on human brain power. Perhaps the best example is backup of data. Backup can be automated in several ways. It can involve the act of plugging in an external drive into a laptop, with software activated to upload new/ modified materials to that drive when that occurs. This will work fine until the user forgets to plug it in, or it can’t be found because someone else is using it.
Another approach is to store data on a cloud = someone else’s server, located externally somewhere in the world. Amazon, Google, Microsoft and many others are willing to provide cloud storage. Unfortunately, these services can be expensive, and one will always be uncertain as to how far the cloud service provider will allow others (unauthorized) access to documents. This can be especially important with respect to both personal and commercial documents. Sometimes people working at the cloud make a mistake that destroys all/ some of the stored content. This has recently happened at Adobe, where some users discovered that their entire content had been permanently deleted in error.
A server can be set up to automatically backup content found on other devices, mirroring changes made locally. As soon as a change is made to a document or folder, this change is implemented immediately on the server.
Even families need to collaborate. A server encourages people to work together by allowing users to access common documents, that are held on a server. Software is available that allows group editing of documents, without have to resort to commercial products, such as Google Docs or Microsoft Office 365.
An important benefit of using a server, is that it incorporates software options that allow for remote access, this is, beyond the local area network (LAN), so that users can access files everywhere.
Further considerations
The successful use of a server is dependent on having a useful data/ document management system.
Some documents are very private, and should not be shared. Others may be slightly less private and can be shared with specific individuals. Then there are those that can be freely shared with friends and family. Sometimes there are even public documents that one could potentially share with everyone in the world.
Every user should have their own password protected (and potentially encrypted) area where they can store personal documents.
Common areas are often better organized by file type. There can be text files (including e-books), images (including photographs, reproductions of drawings, paintings and other artworks), videos (including feature films, television episodes and documentaries), audio (including music), and games.
As an example, music tastes can be very personal. Not everyone in our household appreciates Finnish rock/ Suomirock music. Yet, there is no damage done putting Leningrad Cowboys, Nightwish and Sonata Arctica in a common area. Those not wanting to listen to it, can simply ignore it.
A server can also work as a media center / home theater facility. This server is connected to the internet. In addition, it includes a user interface that allows users to navigate through what amounts to a digital media library, search for, and play back media files.
Some other capabilities include: Play, catalog, and store local hard drive, flash drive and memory card music. It should also be able to display digital pictures, including CD album art. With some caveats, it should also be able to stream video and audio files, including those from Netflix, Youtube and Spotify over the LAN. Similar caveats apply to games. It is usually unable to play content found on CDs, DVDs or Blu-ray disks.
Most of the time, individual users will rely on their own personal devises to browse the Internet, check email accounts and access assorted social networking sites. However, these could be made accessible using a server, functioning as a media player. Participating in video conferences can be enabled by connecting a webcam and microphone.
Most families would be adequately served with an off-the-shelf Network Attached Storage (NAS) server, such as a QNAP TS-251D with 2 x 4 TB of storage capacity (eg Toshiba N300), plus 2 x 4 TB external disks (eg Toshiba Canvio Advance) to be located outside the house, and provide “insurance” in case of fire or other event that physically or otherwise destroys the NAS. Setting up and running this device is relatively easy, using a web-based interface.
Cheaper servers can be made using Raspberry Pi based equipment. However, these require a skill set not every family has.
Typically a NAS uses Redundant Arrays of Inexpensive/ Independent Disks (RAID). This is fine for storage up to, say, 10 TB. Beyond this capacity, another approach is more appropriate.
OpenZFS (ZFS) is a file system that provides is an advanced open-source storage platform. It includes protection against data corruption, support for high storage capacities, efficient data compression, content snapshots and copy-on-write clones, continuous integrity checking and automatic repair, and remote device content replication.
Most ZFS system implementations are complex, and require trained personnel. However, a TrueNAS Mini is suitable for families, with a (relatively) easy-to-use FreeNAS/ TrueNAS web interface accessed from any computer or mobile device in the LAN. Note: 2020-10 is a transition period for TrueNAS. The FreeNAS software on these machines is in the process of being replaced with a TrueNAS CORE release. Once this is updated, machines will be able to run TrueNAS.
The ZFS system has been covered previously in more detail. At Cliff Cottage, the rack based Mothership NAS has a storage capacity of about 100 TB, where 1 Terrabyte = 1 trillion (short scale) bytes = 10¹² bytes. Currently 4 of 12 bays are used, with 8 TB disks = 32 TB. For most people, with the exception of some extremists – such as this writer, this is far beyond what is needed.
This television set made by Tandberg in Oslo, using the Radionett brand, is a Grand TV kabinett, from 1961. The CRT screen is bowed, as was typrical for the time. The varnished wooden cabinet is mounted on removable legs with white plastic wheels. It features stereo speakers, one on each side. Brass is used decoratively on the feet and around the screen. There are white plastic buttons under the screen, and black knobs on the side, to control the machine. “This is the TV model in greatest demand, with a 23″ large picture tubes, and two large concert speakers”, according to a Tandberg ad from 1961. This machine is in the collection of the Stiklestad National Cultural Center, in Verdal, a neighbouring municipality to Inderøy.
In 2020, a media player is a redundant piece of equipment – if one has a appropriate network attached storage (NAS) server, the topic of next week’s post. Yet, it is its own separate topic because of three different, but related, challenges.
First, people do not understand humankind’s physical capabilities and limitations when it comes to normal seeing and hearing. Normal is the operative term here. Visual impairment will be discussed in its own weblog post: 2020-11-24. Hearing impairment will be the topic one week later: 2020-12-01.
Second, this limited understanding results in the purchase of peripherals, such as speakers and screens, that don’t match user needs. On top of this they will purchase media players or a NAS that are neither suitable nor likeable, from hardware/ software/ operational perspectives.
Third, people tend to make evolutionary, rather than revolutionary, purchasing decisions. They replace components, rather than rethinking the components they need. For example, they may be used to having a television set with an attached recorder for watching video and the accompanying audio content. At the same time, they will have a separate stereo system for reproducing music. Unfortunately, such thinking belongs back in the twentieth century, not the current century that is already more than 20% over.
When we (along with the rest of our local community) updated to broadband, effectively at the beginning of 2019, a number of cable television options were available as a package for about NOK 1 500 a month, with an internet speed of 500 Mbits/s. We declined to have this, and opted for a 50 Mbit/s speed, without cable television, for NOK 600 a month.
A main purpose of a media player is to take content either from an online-source or from a file stored somewhere, including personal handheld devices aka mobile phones, and to send the video content to some form of display/ projector, and the audio content to some form of loudspeaker/ headset. Handheld devices, can be used to control the media player. In essence, they are the 21st century’s remote control.
If people use a particular processor or operating system on their laptop or desktop machine, they will often want to use the same one on their media player. This is understandable, for it keeps them in their comfort zone.
Unfortunately, there are still hardware issues with respect to Intel processors, and – to a lesser degree – those of Advanced Micro Devices (AMD). This design flaw has been known since the beginning of 2018. People are encouraged to use ARM or AMD, rather than Intel processors where these are available. Apple, has recently abandoned Intel and gone over to ARM.
An aside: In 1990 ARM stood for Advanced RISC Machines, a company that started life in 1983 as Acorn RISC Machine, referring to the processor in the Acorn Archimedes computer, used by many schools. From 1998 ARM is simply Arm, not an abbreviation for anything.
Yet, for example, Windows users will often complain about the bloatware on Windows 10, but not do anything about it, either on their personal computer, or a media player. More specifically, they will seldom explore open-source software, including operating systems, as an alternative to the commercial products on offer.
At Cliff Cottage on 2019-04-11, an Asus PN 40 computer was ordered to replace an Asus Tinkerboard, similar to a Raspberry Pi (RPi), as a media player. It is equipped with an Intel Celeron processor, effectively demonstrating that I don’t follow my own advice. The hardware on the PN 40 is more than adequate for its role as a media player. It includes: a Western Digital Green 3D Nand M.2 2280 triple-level cell (TLC) solid-state drive (SSD) providing 240 GB; and, 2 x 4 = 8 GB matched Corsair Vengeance SO DD4 2400 MHz, CL16, 1.2 V DC, non-buffered, non-EEC RAM cards.
The device is low-powered and fanless, but capable of playing videos and music stored on Mothership, the family Network Attached Storage (NAS) server. It is equipped with a LibreELEC = Libre Embedded Linux Entertainment Center, operating system, described as “just enough Linux for Kodi“, an open-source media player, which is also installed.
If the purchase of a media player had been delayed to now (2020-10) it would have been an Asus PN 50, a similar, low power (read: minimal watts), yet a machine with adequate processing throughput. It would be equipped with an AMD Ryzen 3 processor. At the time of purchase, these were not available in Norway. Cliff Cottage now has an Asus PN 50 with a Ryzen 7 4700U processor, but it is being used for other tasks.
An even cheaper, but perfectly adequate, media player can be constructed from a RPi 4 B with anywhere from 1 to 8 GB of RAM. Because of its tendency to overheat, cooling must be used. One approach is to place the RPi in a Flirc case.
Vision
The human eye is not impressive. For eyes to write home about, examine assorted species of mantis shrimp.
Boomers probably experienced cathode-ray tube black and white televisions in the 1950s with PAL providing 625 lines, and NTSC providing 525 lines, essentially 480i (interlaced). This was then upgraded to colour television in the 1960s.
In the 1990s, flat-screen televisions emerged and now dominate. High definition (HD) = 1920 x 1080 pixel resolution, became a standard with the high definition multimedia interface (HDMI), defined in 2002. A decade later, in 2012, ultra high definition (UHD) = 3840 x 2160, four times the pixel count of HD (hence, 4K) was defined, and in 2020, is now dominant.
Early flat-screen televisions used plasma technology, then came liquid crystal displays (LCD) that needed backlighting. LCD usually refers to the use of cold cathode fluorescent lamps (CCFL) for backlighting. LED screens are LCDs that use light emitting diodes (LED) for backlighting. This technology is actively used today. Organic light emitting diode (OLED) technology uses diodes to generate colour and light, like plasma screens, but they are smaller and thinner than LED-lit panels, and are capable of producing the best black levels available (highest contrast ratios). OLED screens are considerably more expensive than other screens.
A television in 2020 has 4K resolution. HD is out of date, while 8K is not only too expensive, it lacks content. A video refresh rate of 60 Hz is acceptable, 120 Hz is better, while gamers maintain that 144 Hz is best. Yes, this is a real consideration, if one is to avoid having multiple screens in a household. High dynamic range (HDR) refers to a colour standard that provides a more extensive range of colours, greater contrast levels and increased brightness, compared with the colour rendition on standard HD and 4K sets. Some users feel HDR offers more realistic colours. Others disagree, describing them as artificial. Most regard them as an improvement. OLED TVs typically provide better colour rendition than standard LCD sets, but are considerably more expensive. QLED TVs, where Q = Quantum, referring to photo-emissive particles, are a more affordable middle ground. Many reviews state that it is better to have four or more HDMI ports, with one or more in HDMI 2.1 format.
Currently, at Cliff Cottage there is a Samsung 40″ LED HD screen, made in 2010 but purchased used at half the new price (NOK 2 500 vs NOK 5 000) in 2012. We have never used more than two HDMI inputs. One is attached to the media player, the second has an HDMI cable permanently attached, that can be plugged into a laptop. When this screen ceases to function, it is uncertain if it will be replaced with another screen/ monitor. A 4K 55″ television now costs about NOK 5 000. A 4K 65″ is about NOK 7 000.
Projectors vs Screens
One main question is if the quality difference between a screen and a projector outweighs their respective ecological impacts. A large television/ screen/ monitor consumes many kilos of electronic components, that have to be recycled when the novelty/ economic/ physical lifetime of the product is over. My estimate is that a projector weighs < 10% of a television, although I have not undertaken any research to substantiate this claim.
Thus, we will be examining 4K projectors carefully, over the next few years. When our family was living in Molde in 2002-3, we regularly used a projector, borrowed with permission on weekends from the school where we worked. At that time, bulbs were expensive, used a lot of power and had to be replaced regularly. Today, most projectors are either LCD-based, or Digital Light Projection (DLP) based with a laser, LED, or LCOS (liquid crystal on silicon) light source.
If the projector route is taken, a Philips PicoPix Max provides a minimalist solution. It is a wireless native HD projector, running under the Android OS, with a control touchpad on top. Its built-in battery lasts for three hours of projection. It has Wi-Fi, Bluetooth and USB-C connectivity. Its LED projection source should last for 30 000 hours, and project up to a 120″ (3 meter) image. A more expensive alternative is the Aaxatech 4K1 mini-projector, that costs US$ 1 000. It offers 3840 x 2160 (4K) resolution, 1 500 lumens and 30 000 hours of illumination. Another choice is an Xgimi H2 that uses DLP with four channels of LEDs to beam content. Providing 1350 ANSI Lumens of brightness, requires an effective cooling system. This increased light reduces the wash-out of colours and allows daylight viewing. With its own camera, it is able to adjust focus automatically. It also provides auto keystoning. In terms of sound it provides 16 W through two speakers for treble and mid-range, and a diaphragm for the bass. One major problem with many manufacturers is their repeated attempts at vendor lock-in.
Personal comments: 1) For many years, I had hoped that the Gigabyte Brix Projector would evolve into something useful. Currently, it offers a machine with 75 ANSI lumens, which is inadequate for all but the darkest of localities. 2) The use of a projector like the Xgimi H2 will probably mean that the Asus PN 40 can be repurposed, as the projector is its own media player.
Sound
The human audible frequency range stretches from about 20 Hz to 20 000 Hz (20 kHz). The (Harry) Nyquist (1889 – 1976) – (Claude Elwood) Shannon (1916 – 2001) sampling theorem states that sampling must exceed twice the maximum frequency. This means that the rate has to be 40 kHz or more.
In practice, two sampling rates are used. The first, 44.1 kHz used on compact disks (CD) was inherited from pulse code modulation (PCM) adaptors used to transfer data from recording studio tape to CDs, with a sampling depth of 16 bits/ sample. This capability was developed by Philips and Sony, starting in the 1970s. The second, the digital audio tape (DAT) format of 1987, uses a sampling rate of 48 kHz, which is the standard for professional audio.
The HDMI TV standard (2003) allows both 44.1 kHz and 48 kHz standards, although DVD-Video and Blu-ray Discs use only 48 kHz. Most PC sound cards contain a digital-to-analog converter capable of operating native at either 44.1 kHz or 48 kHz. The average human has no need to exceed these capabilities.
Sound reproduction
There are two major ways of getting sound into the ears of listeners: loudspeakers and headsets. The main advantage of loudspeakers is that they distributes sound to multiple people throughout a venue/ house/ room. This is precisely its main disadvantage. There are numerous types and quantities of loudspeaker systems available. Flat-screen displays often come with built-in speakers that are too thin to give a dynamic range people want. Thus, it is common to augment/ replace these speakers with soundbars, mounted either above or, more typically, below the display, to improve acoustics. They are easy to set up, and are usually less expensive than other stereo sound systems. Some soundbars have left, center, and right speakers plus detachable rear-left and rear-right speakers, and a sub-woofer, to provide surround sound. Premium soundbars come equipped with equalizers, which allow the sound to be tuned to compensate for the shortcomings of room dynamics. Despite these augmentations, some people are critical of their small size and/ or position, and opt to use speakers placed throughout the space.
The main advantage of headsets (earphones/ earbuds) is that sound can be individually adapted for each listener. At the same time, people who are not interested in listening to something can avoid it. Perhaps the greatest disadvantage is that every person must have their own headset. Headsets can be wired or wireless. The main advantage of wireless systems is that they allow a person greater mobility. The main disadvantage is their cost.
Appendix: HD vs 4K Displays
Guidelines from the Society of Motion Picture & Television Engineers recommend viewing at a distance where the screen fills up about 30° of the field of vision. Unfortunately, this recommendation does not take screen resolution into account. If one sits too close to a screen, the individual pixels become visible, detracting from the viewing experience. This is also dependent on the size of the pixels. With HD (1920 x 1080), a given size screen requires a person to sit further back, because it is the pixel size that has to be optimized. With UHD/ 4K (3840 x 2160), it is the field of vision that has to be optimized. .
Type/ limiting factor
40″
45″
55″
65″
HD/ pixel size
1.56
1.74
2.14
2.50
4K/ field of vision
0.71
0.80
0.98
1.25
Optimal viewing distance in meters proposed by Rtings
People are creatures of habit. If they are used to one particular viewing position, they will try to keep that constant, preferring to change the size of the screen, rather than moving closer. This means that consumers are often willing to spend more money to buy a larger screen that optimizes their field of vision, when pixel size issues are resolved.
Other considerations include refresh rates which should be 60 Hz for most activities, although gamers prefer 144 Hz. There is no need for anything beyond this. Note: HDMI 2.0 is needed for 4K video at 60 Hz with 24 bit/px colour depth.
The author’s laptop, an Asus VivoBook with a Ryzen 3 processor, and running Linux Mint 20. This weblog post is being edited in WordPress.
When one acquires a laptop or desktop device, the machine should be considered as part of a system, and not a independent machine. For example, if one is part of the Apple tribe with (or seriously contemplating) the acquisition of an iPhone, it may be appropriate to purchase other Apple products. There is considerably greater leeway with Android devices, so that people can purchase larger machines that have Microsoft Windows, Google Chrome operating system (OS) or a Linux OS. Yet, even here families might want to opt for one of these operating systems, including a specific distro = distribution = brand or variation of Linux (if that is chosen) so that competence is directed to that specific OS, rather than being dispersed over a multitude of types.
Because Apple is a “gated city”, its prices are higher than equivalent machines running Windows, Chrome OS or Linux. Windows machines require “more and larger” than machines running Linux. This means that people on low-income, using older and/ or less powerful machines, should probably use a Linux distro.
It is of the most importance that the chosen OS and device meet the real needs of the user. Three useful categories are: casual, administrative and power users. Casual users use machines mostly to access the internet, and to perform a minimum of other tasks. Administrative users need to input and access data. Power users are usually interacting with large quantities of graphics – games, videos or similar content. Of course, a user may have more than one machine, for different purposes.
The first decision for casual users involves determining if one wants a machine with, or without a keyboard. Tablets are devices without keyboards. These are more portable than a laptop, but less portable than a handheld device, aka [also known as] a cell phone. They use a touch screen for input. These screens vary in size from 7″ to 13″, and it is important to visualize their size, to determine their suitability. If one cannot physically try them out, a full scale two-dimensional template will offer at least some insight into the machine size. For many people, a tablet is an adequate solution. However, they are less suitable for people with issues with their hands and/ or wrists.
People wanting a keyboard can choose between three different device styles: a laptop with its clamshell design has a relatively small display, but usually larger than that found on a tablet, along with a keyboard. Screen size is a matter of taste. For laptops, some prefer 15.6″, others something in the range 13″ – 14″. Displays below this size, can be difficult to use. A desktop machine uses a separate keyboard and mouse for input, and has a large screen, typically 24 – 27″. In addition, there are all-in-one machines, which puts the computing components inside a relatively large screen. These are typically 24 – 32″.
Laptops have a keyboard fitted by the manufacturer. Quality varies but can be as good as average. Once again, this means that they can be unsuitable for people who have issues with their hands and/ or wrists. However, they may be a compromise solution for people who are mobile and need a computer in the many locations they visit, including different rooms in the same building where they live or work. Another reason for selecting a laptop is a lack of space for a desktop display and keyboard on a table or desk.
Most laptops let the user plug in an external display, a USB or Bluetooth keyboard, a mouse, headphones, Ethernet adapter and external drives for storage.
Desktops may be preferred, if one has the space and there is not the need for portability. They are generally faster than laptops because they: are more durable; use faster and hotter processors; do not limit display size or keyboard characteristics; are easier to expand, update and repair; allow more memory, larger drives and faster graphics cards; potentially more ergonomic, and thus healthier.
All-in-one computers are re-packaged laptops (in terms of technology), with a focus on ergonomics, with their bigger displays. They are also less cluttered than desktop PCs, especially with a wireless keyboard and mouse. Thus, they conserve space and are more portable. They are more difficult to repair and expand.
A very small computer is often referred to as a mini-PC. These can be placed directly on a desk or its equivalent or be attached to the back of a display using a VESA mount. These are typically low-power, fanless machines with low noise levels, and many of the same characteristics as a laptop.
Component characteristics: a. Processors, both brand and type. AMD is often preferred over Intel for performance and security reasons. A Ryzen 3 machine can be good enough. To compare processors, look at each chip’s score on the PassMark benchmark. Chips with better scores generally run faster. A score of around 2 500 is the minimum, with anything over 5 000 more than acceptable. Processor power has been an issue since 2005, when Intel decided that its primary goal was to increase performance per watt, rather than maximizing raw performance. b. Storage capacity varies considerably. Some machines only have 64 gigabytes = GB = 10⁹ or 1 billion bytes. This can be inadequate if one wants to store feature-film length videos, modern games or other graphic works. A size between 120 and 250 GB is sufficient, for many users. Yet, even 500 GB is affordable. At Cliff Cottage, Samsung EVO 860 units of 500 GB each, are preferred. If more space than that is needed, one should consider alternative arrangements using a NAS or other form of mass storage. c. RAM: Many users find 4 GB of random access memory too little. 8 GB is usually sufficient, except for people involved in graphic intensive activities, such as gaming or video editing. They may want to increase this to 16 GB, or even more. Memory is relatively inexpensive. Previously, it was easy to upgrade memory to a larger size, by purchasing new memory modules, and installing them. This is becoming increasingly difficult, as many manufacturers solder memory modules in place, preventing their replacement. d. Display: If possible it should be 16: 10 aspect ratio WUXGA = Widescreen Ultra Extended Graphics Array (1920 x 1200 pixels) for administrative tasks. This allows two pages to be displayed. The 16: 9 aspect ratio, is less versatile, but frequently used. FHD or HD 1080 refers to 1920 x 1080 pixels.
Fall Back Solution
A Mini-PC can work as a fall back solution in case of computer failure. If these are set up in advance, they can be operational in minutes, as long as the data needed is on external drives or online. An older laptop can also be used. If you have the economic means to do so, it is often less stressful to replace a laptop at regular intervals, for example – every three years, rather than waiting until a machine falls apart. The replaced machine is then kept in reserve, while the previous reserve machine can be given away to others who are less fortunate.
A Fairphone 3+ comes in a box with the phone, mini screwdriver, quick-start guide and two year warranty. If a USB-charger, USB-C cable or modular earphones are wanted, these can each be purchased separately.
Ubiquitous computing names the third wave in computing, just now beginning [when written in 1988]. First were mainframes, each shared by lots of people. Now we are in the personal computing era, person and machine staring uneasily at each other across the desktop. Next comes ubiquitous computing, or the age of calm technology, when technology recedes into the background of our lives. Mark Weiser
Mark Weiser (1952 – 1999) invented the term ubiquitous computing in 1988. Its principles are:
The purpose of a computer is to help you do something else.
The best computer is a quiet, invisible servant.
The more you can do by intuition the smarter you are; the computer should extend your unconscious.
Technology should create calm.
Conventionally, computing is always interested in maximizing throughput. Thus a 2020 AMD Ryzen Threadripper 3990X processor can perform 2 356 230 MIPS = millions of instructions per second, at 4.35 GHz. For Weiser this was an uninteresting value. He was more concerned about instructions per joule = watt-second, that emphasized low-power portable computing.
Not everything in Weiser’s life focused on calmness. He was also the drummer for Severe Tire Damage, with its motto “Don’t Back Up!”. It was the first band in the world to perform live on the Internet, 1993-06-24.
Many people attribute the iPhone, and smartphones more generally, to Steve Jobs (1955 – 2011), but like the Lisa and Macintosh computers and other personal computers before it, the iPhone and the smartphone had their origins at Xerox PARC = Palo Alto Research Center, where Weiser worked.
Sometimes, handheld devices are referred to as mobile devices. Yet, this alternative term perpetuates a misconception, for it is not the device that is mobile, but its user.
There are many contenders claiming to be the first handheld digital device. Some may want to include cameras, but most of these used film, excluding them from contention. The first commercial electronic single-lens reflex camera, the Nikon QV-1000C, released in 1988. Because of its quirks ( as seen from today’s perspective), others want to refer to the Fujifilm FUJIX DS-1P, as the first fully digital camera, because of its capability to save data to a semiconductor memory card. It also came out in 1988.
Another contender is the PDA = Personal Digital Assistant, where the first one, a Psion Organiser, was released in 1984, had limited features. Its successor, the Organiser II was launched in 1986, combined an electronic diary and searchable address database.
An earlier contender is the handheld game console, which Mattel started to make in 1976, when it sold Auto Race. Yet, before this was a pocket-sized electronic calculator. The first one was the Busicom LE-120A “HANDY”, from 1971. Personal note: Our family’s first hand-held device was a Casio FX-82 electronic calculator, bought in the early 1980s. We owned several versions of these over the years, but – like the slide rule they replaced – they are only kept out of historic interest. Currently, an FX82ES Plus is available in Norway for NOK 215.
Starting in 1950 New York City physicians could pay $12 per month to carry a 200-gram pager that would receive phone messages within 40 kilometres of a single transmitter tower. The system was manufactured by Reevesound but operated by Telanswerphone. This probably makes the pager the first hand-held (or belt attached) device.
Today, there is no need for anyone to own a separate game console, calculator or camera, while both pagers and PDAs are obsolete. A smartphone is capable of doing almost everything any dedicated handheld device can do.
Apps
The apps people choose to have on their devices will vary considerably. My most extensive use of a hand-held device is learning French and Swedish with Duolingo on a daily basis. This can take from 10 minutes to half an hour a day, or 2.5 hours a week. Photographs are taken on a daily basis. The amount of time spent on cell-based telephone conversations will also vary. Mine take up only a couple of hours a week, using mobile networks. However, in addition there are app based calls (most often outside Norway) that use a Wi-Fi connection. Reading (newspapers, weblogs, etc.) can occupy a much larger portion of my time, but divided between a hand-held device and a laptop. Then there is a need to listen to podcasts or watch videos.
A personal note
Since January 2019, we at Cliff Cottage have used a Xiaomi Pocophone F1, as our primary handheld devices. It is probably unnecessary to mention specifications, but it uses a Snapdragon 845 (10 nm) processor with an Adreno 630 to handle graphics. It has 6 GB of RAM, and 128 GB of memory. It features a 6.18″ IPS LCD capacitive touchscreen with 1080 x 2246 pixels. There are dual 12 MP rear cameras, and a single 5 MP front camera. It has a 4 000 mAh battery, that is adequate for our needs. In terms of software it runs MIUI 11 based on Android 10. In contrast to most users, who replace their phones after two years, we hope to keep these phones for at least four years (until the end of 2022).
The Confessional
In a perfect world, I would contentedly use a Fairphone. The Fairphone would also work, it would feature state of the art components and not need to have its parts replaced. Bas van Abel, Tessa Wernink and Miquel Ballester started Fairphone as a social enterprise company in Amsterdam in 2013. In 2017, van Abel admitted that it was impossible to produce a 100% fair phone, but said the phones were fairer. The Fairphone 3 has been on the market since September 2019. It was updated to a 3+ in September 2020, where a major change was in the cameras. This is because improving the camera, is one of the most important reasons people change their handheld devices. The 3+ is made with 40% post-consumer recycled plastics, while the 3 managed 9%. It has a modular, repairable design, is constructed out of responsibly-sourced, conflict-free, and recycled materials (where possible). It is essentially a budget phone, sold for a €/ $/ £ 200 premium.
The Fairphone 3+ has the following specifications: a Snapdragon 632 (14 nm) processor with an Adreno 506 to handle graphics. It has 4 GB of RAM, and 64 GB of memory. It features a 5.65″ IPS LCD capacitive touchscreen with 1080 x 1920 pixels. There is a 48 MP rear camera, and a 16 MP front camera. It has a 3 000 mAh battery. In terms of software can run on an /e/ Foundation open-source version of Android 10.
Repairability is the essence of a Fairphone. Unfortunately, some users have expressed the opinion that it needs this ability more than most other phones. It is constructed out of seven modules. The rear of the phone can be removed without tools. This done, the battery can be lifted out and replaced. A Phillips #00 screwdriver can be used to remove the display. Other modules are held in place using only press fit sockets. The motherboard, containing the system on a chip, RAM and storage, can also be easily removed (though the motherboard’s individual components would not be easily replaced).
At this time in the history of the world, we are trying to promote more local products, if they are suitable. This means hand-held devices should be made in Europe as a first priority. In terms of Asian countries, Japan, South Korea and Taiwan are my priority countries. This means that our next such device may not be a Xiaomi or other Chinese model. It is probably the camera technology that will determine the model purchased, and will determine if it is a Fairphone, or something else. There can be many changes before then.
A Canon i-Sensys MF633cdw colour laser printer with scanner suitable for family use. (Photo: Canon)
Conclusion: A printer is no longer the necessity it once was. Frequency of use is part of the equation in determining what type of printer (if any) to buy. Often alternative sources can provide all of the printing resources needed. At other times, it may be prudent to invest in a colour laser printer with scanner.
Before rushing out to buy a printer there are several questions that should be asked, and answered.
Who in the household needs to use a printer?
What do each of these people want to use the printer for?
What will their monthly production/ consumption be of premium colour print, regular colour prints and black and white prints?
How much will each person need to scan objects?
What alternative printing and scanning sources are available? at what price?
Unlike most of the products discussed until now, the answers to the above questions will show that most families will probably not need their own printer!
Ink in an ink jet printer will dry out, if not used at least once a week. Thus, most people with low usage requirements and an alternative printing source, will find that owning a printer is an unnecessary expense. While ink tanks attempt to make printing cheaper, they are only a solution if one is printing a large number of sheets regularly. Ink jet printers have a problem with drying out, regardless of how the ink is provided. Using an ink jet printer at home – say, once a month – will be frustrating and potentially expensive, as one has to cope with, and repurchase, dried out ink cartridges.
Laser printer toner cassettes do not dry out. They can be used minimally without problems. This is one reason why the author of this weblog post has purchased a laser printer, and disposed of the previous ink-jet printer. The laser printer is used mostly at the end of the month to make a paper copy of accounting documents, ingrained from long-forgotten accounting courses, that required people to keep a paper trail of transactions. Other people may not feel this need. Laser printers are more expensive to purchase than ink jet printers.
While many product reviewers are keen to emphasize the existence of monochrome (read: black & white) printers, these may be offering false economy. An inability to print a document in colour, may make the entire printing process irrelevant. A colour printer, can always print in black & white. Standard printing size is 8.5 inches by 11 inches in North America, and A4 in Europe. There is no need for most households to deviate from this. If there is, time should be spent sourcing where appropriate printers can be found, and used.
There are essentially three types of printing: Premium colour printing on glossy paper, for photographs and renditions of artwork, regular colour printing and black & white printing.
There are many brands of printers available. In my working life I have been subjected to too many HP printers that have had a number of technical issues, such as jambing. In my personal life, Epson has provided the most toxic experience, with its incessant requests to change ink cartridges to those that are Epson branded. In Norway, vendor lock in of this sort is illegal. I have used unofficial inks with Canon ink jet printers without problem. Currently, we are on our first set of original toner cassettes on our Canon i-sensyn MF633cdw printer, so this situation has not been tested, although unoriginal replacements have been purchased.
Many printers are classed as multi-function machines, a misnomer, as they usually add just a scanner. However, this ability to scan documents may tip the scale towards printer ownership. Once a document is scanned, it can be sent as an email attachment. Some printers can even offer faxing as an option. In the circles I frequent, faxes are cult items in the same category as telex machines. Both are irrelevant technologies. However, this is not the situation in all parts of the world, where faxes are still used.
A household printer should be connected to the household computer network of handheld devices, laptops, desktops and other computers. Normally, there is absolutely no need to have more than one printer in a house. There are several ways you can go about installing a network printer. However, two are most common. It can be part of the Wi-Fi network using an infrastructure controlled by your router, or it can be wired using Ethernet. Many printers are also able to read memory sticks, using a USB Type-A port. Not many printers are equipped with USB-C, yet.
There are several alternatives to owning a printer/ scanner. If one attends school or works for a living, there may be formal or informal arrangements for printing/ scanning documents. Some may require payment, others may not. A public library may provide an inexpensive service for patrons. Some office supply stores may also provide this service, charging for each page printed. Then there is FWB, a friend with benefits, which in some contexts may refer to something completely different, but here refers to a friend willing to let you use her/ his printer/ scanner. This may be precisely the benefit you need from a friend.
Another approach is that you become that FWB, and offer a printing service to people you know. This is not a money making proposition, but could be a way of reducing costs provided you are prepared to charge people for the services you provide. Collecting money is easy with an apps developed for this purpose: Venmo provides smart phone apps for payments in the US, Fooi does the same in Canada, and there is Vipps in Norway. Note: There may be more appropriate payment services that I am unaware of. If so, please let me know.
Below are links to spread sheets showing the calculations in LibreOffice format and Microsoft Office format, respectively, that can be downloaded and modified to suit your needs.
Here is some of what he writes: “HP’s latest gambit challenges the basis of private property itself: a bold scheme! With the HP Instant Ink program, printer owners no longer own their ink cartridges or the ink in them. Instead, HP’s customers have to pay a recurring monthly fee based on the number of pages they anticipate printing from month to month; HP mails subscribers cartridges with enough ink to cover their anticipated needs. If you exceed your estimated page-count, HP bills you for every page (if you choose not to pay, your printer refuses to print, even if there’s ink in the cartridges). If you don’t print all your pages, you can “roll over” a few of those pages to the next month, but you can’t bank a year’s worth of pages to, say, print out your novel or tax paperwork. Once you hit your maximum number of “banked” pages, HP annihilates any other pages you’ve paid for (but continues to bill you every month).
Now, you may be thinking, “All right, but at least HP’s customers know what they’re getting into when they take out one of these subscriptions,” but you’ve underestimated HP’s ingenuity. HP takes the position that its offers can be retracted at any time. For example, HP’s “Free Ink for Life” subscription plan offered printer owners 15 pages per month as a means of tempting users to try out its ink subscription plan and of picking up some extra revenue in those months when these customers exceeded their 15-page limit. But Free Ink for Life customers got a nasty shock at the end of last month: HP had unilaterally canceled their “free ink for life” plan and replaced it with “a $0.99/month for all eternity or your printer stops working” plan…
For would-be robber-barons, “smart” gadgets are a moral hazard, an irresistible temptation to use those smarts to reconfigure the very nature of private property, such that only companies can truly own things, and the rest of us are mere licensors, whose use of the devices we purchase is bound by the ever-shifting terms and conditions set in distant boardrooms. From Apple to John Deere to GM to Tesla to Medtronic, the legal fiction that you don’t own anything is used to force you to arrange your affairs to benefit corporate shareholders at your own expense. And when it comes to “razors and blades” business-model, embedded systems offer techno-dystopian possibilities that no shaving company ever dreamed of: the ability to use law and technology to prevent competitors from offering their own consumables. From coffee pods to juice packets, from kitty litter to light-bulbs, the printer-ink cartridge business-model has inspired many imitators.
HP has come a long way since the 1930s, reinventing itself several times, pioneering personal computers and servers. But the company’s latest reinvention as a wallet-siphoning ink grifter is a sad turn indeed, and the only thing worse than HP’s decline is the many imitators it has inspired.“
In this weblog post several different types of external drives will be examined for different purposes.
Hard Drives
The main purpose for having external drives is to have off-site backup of files stored on a family Network Attached Storage server(NAS). It is to ensure that irreplaceable data is not lost by storing it in two separate locations. There are five drives with historic data that are permanently kept at a second location. This is because all that data is already on the NAS. Then there are six drives that are divided into two sets of three disks each. One set is taken regularly to the second location. Once there, the other set is returned home, and used to update files on a regular basis for the next couple of weeks. In a worst case situation, a catastrophic event would result in the loss of up to two weeks data.
Currently, hard drives offer lower price than solid-state drives when mass storage is needed, particularly in the range of 1 – 5 TB. Western Digital My Passport products are inexpensive, but offer good performance. This is due to using the USB 3.0 protocol, with its cable terminating in a Type-A connector. In addition it comes with an excellent disk controller. There is 256-bit AES Hardware Encryption available. The maximum transfer read speed is 174 MB/s , and the maximum write speed is 168 MB/s. In general it offers acceptable performance and reliability.
If one selects the Western Digital My Passport Ultra, there are two main differences. First, it uses a more robust metal enclosure. Second, the USB 3.1 protocol is used, with its cable terminates with a USB-C connector, but includes a Type-A adapter.
There are different manufacturers of HDD units including Samsung, Seagate and Toshiba. I have mainly kept to Western Digital, because none of my eleven My Passport drives have failed. In contrast, one of my two Seagate Backup Plus drives failed after minimal use. The second, remaining Seagate drive was quickly given away. Others are not so enthusiastic for My Passport drives. Slant ranks two versions of Seagate Backup Plus as the best drives. I have also considered purchasing Toshiba Canvio drives because of their compact physical size, but their smaller 3TB capacity makes them less practical for a person with large quantities of data to store.
Solid-State Drives
Solid-state drives (SSD) become expensive with large capacity disks. The Western Digital My Passport SSD is considerably more expensive, with storage capacity ranging from 256 GB – 2 TB. There is a USB 3.1 Type-C port and AES 256-bit hardware encryption. At the moment, the main disadvantage of these drives is price, which is more than four times the HDD equivalent.
Transfer speed is up to 540 MB/s. This speed increase is because SSDs access data electronically instead of electro-mechanically. This also improves their mechanical ruggedness. There is continued debate about the reliability of SSDs, with some finding them more, others less, reliable than HDDs. SSDs are sensitive to sudden power interruption, which can result in the loss of a drive.
While there are wireless drives, not all devices are equipped with Wi-Fi. Almost without exception, these devices are equipped with USB ports. Thus, there is no need to invest in wireless external drives.
Price for assorted 2 TB My Passport models (2020-02-08) in USD before taxes and shipping charges.
HDD
Ultra HDD
Wireless HDD
SSD
Wireless SSD
$65
$80
$150
$300
$600
USB flash drives aka memory sticks aka thumb drives
Flash drives are used for various purposes. Perhaps the most common use is for the transport of files when a person will not accompanied by their own laptop or other machine where the files are normally used. They are especially useful if one has to upgrade a number of machines to several different devices.
For these purposes one wants a fairly rugged and durable piece of equipment. Personally, I use a Kingston DataTraveler Elite G2 64GB flash drive. For the security minded, the main drawback of this drive is that it is not encrypted. If necessary, one could password protect sensitive individual files. One reason for selecting this drive is its metal casing, which is shock- and water-resistant. It reads at a maximum of 180 MB/s, and writes up to 70 MB/s.
The more security minded may want to consider a SanDisk Extreme Pro USB 3.1 available in 128 and 256 GB varieties. Its read and write speeds are extreme, up to 420MB/s and 380MB/s, respectively. It has an aluminum casing and comes with 128-bit AES encryption software. Its price is also extreme.
One of the main uses of flash drives here, is the storage of assorted Linux operating system distros. Each distro is placed on a new Kingston DataTraveler G4 16 GB flash drive. At one time, these were upgraded as new distro versions became available. However, after a couple of drive failures one learned that this was false economy. 16 GB drives are used not because one needs so much space, but because they are cheaper than 8 GB drives!
There are many times one needs to give people files on a flash drive. This has to be regarded as one of the normal costs of having relationships with people, in much the same way that serving visitors (invited or not) a cup of coffee is an investment in smooth social relationships. These G4 drives are the standard drives given to people. They are usually bought online, ten at a time.
SD Cards
While this section is labeled SD cards, some would argue that microSD cards would be a more appropriate heading. The original SD format is far too large for handheld devices. If a port on a device only takes original or mini-sized cards, passive adapters are available.
