Category Archives: Techie

Low code: The end of software engineering for apps?

Application development tools haven been improving for decades and are to the point where many applications no longer require the skills of a highly trained software engineer:

When I was starting my career in tech, software development was the preserve of skilled engineers. That paradigm is rapidly becoming obsolete, thanks to the rise of low-code dev tools that allow employees without coding experience (aka “citizen developers”) to create powerful apps using prebuilt templates and intuitive, drag-and-drop interfaces.

Source: The untapped potential of citizen developers

The “shortage” of software developers may disappear as fewer skilled software workers are needed.

Most application development, even when writing apps in Javascript or Python – which are not “low code tools” – does not require the detailed skills of computer science. Hence, six month “coding boot camps” and certification.

Software development becomes a combination of moderate skilled blue collar “boot camp” programmers, and low-code. The latter is more akin to the teaching of typing skills in high school 3 or 4 decades ago. It’s just a modern “typewriting” class to learn low-code tools.

Stop Calling Everything AI, Machine-Learning Pioneer Says – IEEE Spectrum

This is true – with a reminder that even as a brain injured idiot, I am a retired computer engineer and computer scientist:

Artificial-intelligence systems are nowhere near advanced enough to replace humans in many tasks involving reasoning, real-world knowledge, and social interaction. They are showing human-level competence in low-level pattern recognition skills, but at the cognitive level they are merely imitating human intelligence, not engaging deeply and creatively, says Michael I. Jordan, a leading researcher in AI and machine learning.

Source: Stop Calling Everything AI, Machine-Learning Pioneer Says – IEEE Spectrum

Much of our “AI” mimics some human behaviors while much more is sheer computational horsepower enabling us to create models with complex weighted input mappings which we refer to as “machine learning”. This is all good but its not the AI that many people think of when they hear the term.

And this is actually quite cool:

He says he believes that developments in machine learning reflect the emergence of a new field of engineering. He draws parallels to the emergence of chemical engineering in the early 1900s from foundations in chemistry and fluid mechanics, noting that machine learning builds on decades of progress in computer science, statistics, and control theory. Moreover, he says, it is the first engineering field that is humancentric, focused on the interface between people and technology.

DOJ sues Facebook for allegedly discriminating against US workers

Facebook is accused of hiding job openings from potential U.S. applicants so that it could recruit temporary foreign workers (i.e. H-1B visa workers) instead, for about 2,600 open job positions.

Source: DOJ sues Facebook for allegedly discriminating against US workers

Years ago, a law firm posted videos on Youtube of talks they had given at a conference – about how to legally not recruit U.S. workers. The videos were eventually deleted but not before others had saved them, of course. Their recommendation to employers was to meet the letter of the law by advertising a tech position in a very rural, small town newspaper – where no prospective U.S. tech workers would see the ad. Then, claim there were no U.S workers available.

Another trick was to over specify the job description – ads would list dozens of specific knowledge and skill area requirements – such that nearly no one would qualify – and probably not apply. Then the employer could say no U.S. workers were available.

Facebook is accused of employing “tricks” to meet the letter of the law in order not to hire U.S. workers. The DoJ says Facebook did this for 2,600 jobs – by hiding the open job positions so they would not be seen by prospective U.S. workers.

I saw similar discrimination from Facebook advertising.  In all the years I had been on Facebook, I never saw a job ad – not one – for a software developer. Considering I had a BS in computer science, an MS in software engineering, and an MBA, and had worked in Silicon Valley and for Microsoft, have two patents and written a dozen technical books, this seemed odd to me. I had 100% of the qualifications that were supposedly in super short supply and for which employers were desperate to hire.

Why did this happen? Probably because by the time I was on Facebook I was “too old” to work in tech and Facebook allowed advertisers to discriminate on the basis of age.  (Other research demonstrated that numerous U.S. firms were specifically running job ads targeting specific, young, age groups.) I completed my MS in software engineering at age 51 (GPA 4.0, thesis on Android and mobile power management). I have zero doubt that tech companies like Facebook play these games. No doubt what so ever.

You would think Republicans would be in support of using H-1B workers and the Democrats would be opposed. Yet Trump issued rules making use of H-1B workers harder, while the Clinton/Gore administration dramatically increased the use of H-1B workers. By the year 2000, 1 in 4 information systems jobs in the U.S. requiring a Bachelor’s degree were filled with temporary foreign workers.

Consequently, while this DoJ effort is initiated under the Trump administration, it might be dropped under a Biden/Harris administration. We have no idea where this case will go.

What the pandemic has taught us about science | Climate Etc.

How should the public begin to make sense of the flurry of sometimes contradictory scientific views generated by the Covid-19 crisis? The only way to be absolutely sure that one scientific pronouncement is reliable and another is not is to examine the evidence yourself. Relying on the reputation of the scientist, or the reporter reporting it, is the way that many of us go, and is better than nothing, but it is not infallible. If in doubt, do your homework.

Source: What the pandemic has taught us about science | Climate Etc.

Communications: California loses cell phone service in power cut offs

Up to 1,000 or so cell sites went off line due to the extended power shut down in northern California:

Even as California burns, the cell phones of many residents have gone mute, preventing them from giving or getting emergency information.

Source: Kincade Fire: Why you can’t make cell calls in this emergency

Most cell sites are configured with battery backups to keep running during short term outages but are unable to continue running without power for days. Some also have generators but due to the power shut down and wild land fires, they were unable to be serviced with additional fuel.

Some have thought that with communications becoming a commodity, we would no longer need distributed emergency communications from ham radio. Instead, the more communications we’ve built and the more it has been used, the larger the impact of its loss. Amateur Radio emergency communications remains a vitally important component of emergency planning and service delivery.

#Intel intros new notebook processors, high performance, lower power, integrated 802.11ac

From the spec sheet, the new CPUs provide high performance at low battery power consumption, and integrated Gigabit Wi-Fi (802.11).

See Intel’s page on the new processors for more information.

Meltdown and Spectre are the names given to a security vulnerability that takes advantage of speculative branch execution to capture the side effect of cached data, thereby revealing protected data. However, just one of the two new processors addresses Meltdown while the other does not. Neither addresses Spectre.

Intel launched new eighth-generation processors slated for laptops this week: Ultra-low power 15-watt Whiskey Lake U-series chips and extremely low power five-watt Amber Lake Y-series chips. After the launch, Intel was asked if these two processor families include hardware fixes for Meltdown and Spectre.

Source: Intel’s New ‘Whiskey Lake’ CPUs Have a Hardware Fix for Meltdown | Digital Trends

University of Washington: upcoming Internet accessible lecture on #IoT #Wearables Technology

Next Thursday at UW CSE or view remotely:

Computer Science and Engineering

SPEAKER:   David Kotz, Dartmouth College

TITLE:     Amulet: An Energy-Efficient, Multi-Application Wearable

DATE:      Thursday, December 1, 2016
TIME:      3:30pm
PLACE:     EEB-105
HOST:      Tadayoshi Kohno

Wearable technology enables a range of exciting new applications in
health, commerce, and beyond. For many important applications, wearables
must have battery life measured in weeks or months, not hours and days as
in most current devices. Our vision of wearable platforms aims for long
battery life but with the flexibility and security to support multiple
applications. To achieve long battery life with a workload comprising apps
from multiple developers, these platforms must have robust mechanisms for
app isolation and developer tools for optimizing resource usage.

We introduce the Amulet Platform for constrained wearable devices, which
includes an ultra-low-power hardware architecture and a companion software
framework, including a highly efficient event-driven programming model,
low-power operating system, and developer tools for profiling
ultra-low-power applications at compile time. We present the design and
evaluation of our prototype Amulet hardware and software, and show how the
framework enables developers to write energy-efficient applications. Our
prototype has battery lifetime lasting weeks or even months, depending on
the application, and our interactive resource-profiling tool predicts
battery lifetime within 6-10% of the measured lifetime.

(Featured image: Seattle photo from University of Washington web site at

Continue reading University of Washington: upcoming Internet accessible lecture on #IoT #Wearables Technology

Zigbee low power wireless for the Internet of Things #IoT

Most everyone is familiar with Wi-Fi (IEEE 802.11 standards) and most are likely also familiar with Bluetooth wireless. Far fewer are familiar with some of the standard wireless technologies used for Internet of Things applications. These technologies include Bluetooth Low Energy (Bluetooth LE), Zigbee and others.

This post is about Zigbee.

Zigbee is a low power, very short range wireless technology that has been around for more than a decade, although it has continually evolved to support new requirements. Zigbee 3.0 is the current version of the specification.


Source – Zigbee standards group.

Zigbee is a set of protocols that sit on top of the IEEE 802.15.4 specification. 802.15.4 specifies a short range (less than 10 meters), low speed (up to 250 kbps maximum) wireless link operating in the 2.4 Ghz band (or alternatively the 868 Mhz band in Europe and the 902-928 Mhz band in North America and potentially in other forms including Ultra Wide Band).

While used for peer to peer communications, protocols add support for mesh networking, where nodes in the network can forward packets on to others in the network. In this way, a message can be delivered over a much greater distance than the few meters of a specific link. Zigbee specifies how devices join the network, plus how security is implemented.

Key to Zigbee is its design for extreme low power operation. This means a device can be powered by a small battery for very long periods of time (such as a year or many years) or the device can even use “energy harvesting” to obtain sufficient energy from a wireless field (similar to how RFID works). A Zigbee device might operate on 1/100,000 to 1/1,000,000 the power required for a typical WiFi connection.

Zigbee’s design is oriented towards applications of Zigbee, such as a switch remote controlling a light bulb, or a dimmer controlling a light. Zigbee devices are intended to be easily installed “plug and play”. Contrast that with setting up an Internet connection just a few years ago – where end users had to enter router and DNS addresses and possibly specify a subnet mask. The complexity was absurd in terms of the the end consumer. The consumer just wants to purchase a solution, plug it in, and it works.

The device should automatically discover its surrounding support network, plus, automatically adapt in the event the environment changes. In other words, each device may be part of a mesh network that forwards packets – but if devices within the mesh go offline or are blocked, a new path can be automatically identified.

As you can see, Zigbee is a low power, wireless communications standard that is designed for a world of small, battery powered, Internet of Things devices. Zigbee is not the only standard available – but it is certainly an important one.


Hacking embedded systems: Camera hacking the Canon PowerShots

Source: CHDK Wiki | Fandom powered by Wikia

If you have a Canon PowerShot camera, a great and fun software hack is available – for free – called the Canon Hack Development Kit or just CHDK.

CHDK is software that runs on your Canon PowerShot camera to add additional features and capabilities; which features are supported depends on which PowerShot camera is used.

When I had a Canon PowerShot SX1, I used CHDK especially for its motion detection feature. This hack added a feature to detect motion in a scene and then fire the shutter – which was perfect for photographing lightning. Yes, its detection is so fast that you could use it to photograph lightning bolts.

In addition to a set of features added by CHDK to the PowerShot cameras, CHDK also adds “scripting”. This is a feature that let’s you write a set of commands (similar to writing a program) to use and operate various camera features.

The hack is installed by copying files to a specially prepared mini SD card. When the camera is turned on, the hack software is pre-loaded, together with the camera’s own, original software.

I sold my SX1 (a great camera for macro shots due to its macro feature and small sensor size). Since then, I have missed being able to play with CHDK. I am thinking about buying a used Powershot with a larger 1/1.7″ sensor so I can play with CHDK again 🙂