amen zwa, esq.

I am an ordinary man, so I am unique up to isomorphism.
I am on MathsTodon not to preach to the choir, but to attend the concert.
I admire well constructed arguments, but I am not argumentative.
I always follow back valid accounts, but I do not chase down celebrities.
JD ⊥ MSCS ⏚ BSEE • GPS +38.89 -77.01 • CO₂ low 320s

Public Key

npub1pjrrj9n026ujgy6aeqagywwct7gsrt5wux2mjxvl7436jejprhcs5fay94

NIP-05 Address

AmenZwa@mathstodon-xyz.mostr.pub

Profile Code

nprofile1qqsqep3ezeh4dwfyzdwusw5z88v9lygp468wr9derx0l2cafveq3mugpz4mhxue69uhhyetvv9ujumt0wd68ytnsw43qg295k5

Publishing to

relay.mostr.pub

Author Public Key

npub1pjrrj9n026ujgy6aeqagywwct7gsrt5wux2mjxvl7436jejprhcs5fay94

Show more details

Last Notes

2026-09-26 01:01:59 UTC

At social gatherings, I have heard someone or another describe himself as an “amateur mathematician”. Whenever I hear that appellation, my mind instantly say, “Um, that sounds dangerous, like being an amateur neurosurgeon”.

2026-09-01 15:22:26 UTC

#AssemblyLanguage
• Set Theory is the assembly of mathematics.
• 𝜆-Calculus is the assembly of Computable Functions Theory.
• JavaScript is the assembly of the Web.
• PDP-11 instruction set is the assembly of the hearts and minds of a particular ilk of geeks.

2026-08-31 16:00:48 UTC

There is much wasted effort, resources, and talent in #OpenSource #software development today, due to the superfluous forking, the nonexistent coordination, the lack of appreciation for others’ ideas, the strong conviction in “My way is the best”, the typical coder ego, and all that.
The early progress, from the 1950s to the 1970s, in modern #computing—both hardware and software—were led by #academia, given that the industry did not yet possess the socio-economic relevance and prevalence it now enjoys. The self-regulated academic publishing standards were much higher in those days than they are today, and the academic research community far smaller. This common high standard of a close knit community served well its intended purpose: to select the best ideas, to promulgate them to the community, and to encourage improvements. The whole community understood that critique and refinement of each other’s ideas was more beneficial than divisiveness and replication.
There are those who claim that forking is a way of applauding or commenting. In fact, forking is like that other popular four-letter f-word. A superior form of accolades for an open-source project is to donate tangible resources to that team. A better form of communication with that team is to volunteer for their project.
The modern IT #industry needs an academia-style common ground, instead of the present disjoint territories ruled by self importance and defended by pecuniary interest.

2026-08-24 00:17:48 UTC

Back in the day, owning an 8-bit home PC was quite special. But working during the day on the university's 16-bit PDP-11/70 running UNIX 7ed, then coming home to the 8-bit PC—well, that was a bit of a downer.

2026-08-19 13:40:45 UTC

- reply

@nprofile…8jwj
I'm with you, mate, both chronologically and philosophically. Unlike my IT colleagues (in my own little corner of the field), I backed into IT, from engineering. Through the decades, I've tried to inject traditional engineering practices into the IT work, but to no avail.
What disheartens me is that in my own career, I've observed the decline in quality of work, intellectual effort, emotional investment, and ethical standard of the practitioners. One man's career, however long, is but a short blip, compared to the life of an industry. Still, the precipitous decline over this period is palpable. And today's "IT thought leaders" would rather pray to the AI Lord and lead blindly, instead of listening to the wisdom of the experienced.
What would happen, when our generation pop off, in just a handful of years....

2026-08-19 13:13:31 UTC

- reply

@nprofile…8jwj
Your succinct statement expressed much more truth than my long, circuitous attempt at a joke.👍

2026-08-19 12:14:36 UTC

- reply

@nprofile…8jwj
That was a sardonic statement about the present state of our industry. The full weight of a mathematical argument should not be brought upon to assail a mere joke.

2026-08-18 23:27:25 UTC

When I was an engineering undergrad in the early 1980s, #chaos theory and #fractal geometry were coming into vogue, thanks to #Mandelbrot's excellent book "The Fractal Geometry of Nature". What struck me about, then and now, about chaos processes is their description of how a simple, predictable, ordered process, under certain conditions, suddenly explodes into a wild, chaotic results.
Of relevance to the present quagmire of our #IT industry is the Verhulst population growth function: \(x_{t+1} = (1 + r)^t x_t\), where \(t\) is time and \(r\) is growth rate. The dynamic process emerges, when this function is iterated over time.
When \(r < 2.0\), the population dynamic eventually converges to a stable number. When \(r = 2.3\), the population oscillates between high and low numbers. When \(r ≥ 2.5\), the population dynamic transitions into a chaotic process.
I wonder if the present growth rate \(r\) of the IT industry has risen way above \(2.5\).

2026-08-17 21:59:40 UTC

Great programmers do not pick #favourites amongst the #programming languages; they strive to be proficient in every language at hand.
————
Right so, these are my favourite languages:
• 6502 assembly '75
• C K&R '78
• Smalltalk '80
• Scheme R⁴RS '91
• Standard ML '97
Thug life of an ordinary programmer....😒

2026-07-17 18:45:21 UTC

Someone’s favourite #programming language can be likened to a knife. It could be a well made, but common, Swiss Army knife, or it could be a well made, and pricey, high carbon steel santoku hand forged by the last remaining blade master in Seki, Gifu Prefecture, whose ironworks date back centuries.
A knife user would not think of using his favourite knife, whichever type, to clean the toilet, to dig a trench, or to perform some other task for which the knife was not designed. And although he would use his favourite santoku in the kitchen daily, he would only pocket his SAK on a camping trip. The SAK is as generally applicable as a knife can be, but even this estimable knife has its limits.
Likewise, a programmer should select a #programming language that best fits a particular task. There is no such thing as the definitive GPL for all occasions. Every GPL is designed for a particular purpose with a particular axe to grind. A GPL is but a Turing-complete DSL in disguise.
It, therefore, behoves the programmer to learn, over the course of his career, as many languages from as many paradigms as practicable.

2026-07-17 15:41:44 UTC

- reply

Like who they are, whether you know any members, what prompted them, etc.
And are you guys aware of any other similar in-depth studies on the 6800, 6502, 8080, and Z80?
I know why I’m still interested in 8-bitters: their complexity is the right level for my mind to handle. But I’m curious about why, around 20 years back, this retro-cool revival suddenly erupted, round the globe. Was it the availability of information (manual archive) and technology (FPGA)?
@nprofile…j9f5 @nprofile…8qh7 @nprofile…45cc @nprofile…k6gw @nprofile…3zf2

2026-07-17 15:22:41 UTC

Mates, any of you know the details about this particular #ReverseEngineering and #Simulation of the #MOS6502?
@nprofile…8qh7
@nprofile…45cc
@nprofile…j9f5
@nprofile…k6gw
@nprofile…3zf2
https://youtu.be/fWqBmmPQP40?is=n36NgBS2l7HZGXX6

2026-06-23 00:02:15 UTC

#UNIX is not just an OS; it is the ultimate #IDE for #programmers. Arguably the best written introductory book on UNIX programming is "The UNIX Programming Environment", Kernighan (1983). This book covers UNIX V7, but I used it with 4.2BSD.
There is another jewel of a book of a similar kind, a free one to boot, by Prof. Ballesteros (King Juan Carlos University): "Introduction to Operating Systems Abstractions Using Plan 9 from Bell Labs" (2007). Prof. Ballesteros wrote this book to teach his CS students the fundamentals of good OS design.
Although almost no one uses #Plan9 today, it is alive and kicking. It even runs on Raspberry Pi. Anyone interested in OSs and likes the philosophical underpinnings of UNIX should take a look at Plan 9.
The other must-read book is his "Notes on the Plan 9 3ed Kernel Source" (2007). It is a Lions-style commentary.
Prof. Ballesteros has many Plan 9 related publications in PDF format.
https://lsub.org/books-papers/
PS—Those with a historical bent should also read Organick's "The Multics System: An Examination of Its Structure" (1980). Multix is the mother of UNIX, and Plan 9 is the grown-up UNIX.

2026-06-22 13:31:43 UTC

- reply

@nprofile…c5ug @nprofile…v2v2
James, I am aware that your question was addressed to David, and I'm just a little DSP-on-MCU guy who has never designed a CPU microarchitecture.
But I offer my own confined perspective. It is eminently possible to break the x86-C-Linux stranglehold of modern computing. Indeed, well into the early 1980s, that was how things were done. Each new architecture was incompatible with the one prior. Innovation was explosive, and the laws of Moore and Dennard were extent. Such exploratory divergence was inevitable. But once the industry stumbled upon the cost-performance-familiarity tripod in the shape of x86-C-UNIXen, it took the most steady-profit path.
I doubt, though, that today's profit-driven, risk-averse industry has the moxie to take on risky, groundbreaking innovations. No one dares to play chicken with a half-billion-dollar investment that would likely flop—no one would switch from their familiar environment.
Here is a wild thought. Everyone is obsessed with LLMs these days, using them indiscriminately to generate code they cannot understand. Why not just have AI generate Verilog or VHDL and make FPGAs for small-scale and ASICs for large-scale directly, and be done with architectures, languages, and operating systems? (Forgive me; at my age, I am prone to suffer from these frequent pangs of resentment.)

2026-06-22 02:42:06 UTC

This article, "C Is Not a Low-Level Language; Your Computer is Not A Fast PDP-11", by David Chisnall of Cambridge, is a critical self-assessment of C, modern CPU architectures that pander to C, and modern C compiler writers, like himself.
The intertwined successes of UNIX and C was inevitable, given the nature of the computing technology in the late 1960s. But that instant, meteoric success continues to demand backward compatibility through the decades, and that backward compatibility engenders much forward restraint on future advances.
As a long-time fan of C and PDP-11/70, I find Chisnall's critiques painfully true. The same could be said of UNIX, my all-time favourite operating system. And x86, too, followed a similar path to immediate success and perpetual dominance.
The power of inertia is terrifying: even after 50 years, the current score of the computer architecture and programming language game remains "von Neumann 1 v Backus 0".
https://spawn-queue.acm.org/doi/pdf/10.1145/3212477.3212479

2026-06-17 00:08:52 UTC

"The #C #Programming Language"
K&R C 1ed (1978)
I've owned this little treasure for over 40 years. I've reread it countless times through the years, just for inspiration. It is still my all-time favourite #technical #book.
https://media.mathstodon.xyz/media_attachments/files/116/762/495/755/048/296/original/64748f1db67f6d1c.jpeg

2026-06-02 00:33:11 UTC

- reply

@nprofile…hm2z
All the ills Ms. Gebru’s paper identified have come to fruition in an alarmingly short period, even when measured by the spasmodic cycles of the IT industry. In just a handful of years, the following troubling trends have emerged in the industry:
• AI has eroded the professionalism and ethics of the very practitioners who profess to be AI experts
• AI has replaced the in-mind knowledge repository with the by-chat information lookup
• AI has transmogrified itself from being just a tool of the intellectual to being the religion of the indolent
These might well be the signs that IT, as a field of practice, is no longer a human endeavour.

2026-05-26 18:39:42 UTC

#SpaceX-ploded, as usual.
https://apnews.com/video/spacex-rocket-ignites-into-flames-after-landing-in-the-indian-ocean-fffe9fcd3ce24c589383eeef56da62f3

2026-04-30 03:41:18 UTC

- reply

@nprofile…k3ke I can only surmise, as oil market is not my area of expertise. Oil is an essential commodity, speculative investment opportunity, policy control mechanism, and political lever, all wrapped into one untidy bundle. Put all that on a worldwide gambling platform that is the stock market, and we have a witch's brew.

2026-04-30 00:01:23 UTC

A well written #novel sucks down the reader, until he becomes hopelessly enmeshed into the fabric of the story.
A well written #textbook hoists up the reader, until he could see far beyond the edge of the subject matter.
But people are no longer #reading and #writing, these days. So, who cares?
Ah, but #LLMs care.
How did society get here....🤦‍♂️

2026-04-28 15:17:02 UTC

https://apnews.com/article/stock-markets-trump-oil-iran-war-1901470c64a6055c80656fad64f863e5
As #oil stocks rise, #AI stocks sink.
All of Modernity is still soaked in oil, an ancient kind of #energy.
All of Universe is but conserved energy in various guises.
All of Life is mere squabble over means of energy conversion.

2026-04-07 15:00:19 UTC

#Computing Lifecycle:
• 1930s—mathematicians conceived how to mechanise computation
• 1940s—physicists and electrical engineers (EE) designed and built electronic computers
• 1950s—business computing (IT) split off from business management
• 1960s—computer science (CS) split off from mathematics
• 1970s—statistics became mechanised (SAS, MINITAB, etc.)
• 1980s—computer engineering (CE) split off from EE
• 1990s—web technologies swamped IT
• 2000s—data science (DS) split off from CS and statistics
• 2010s—IT, DS, and CS succumbed to AI
• 2020s—AI began its campaign to eradicate all those computing-related fields
• 2030s—no one on earth will understand computing anymore

2026-03-28 00:35:28 UTC

“I need to discuss science vs. engineering. Put glibly:
In #science if you know what you are doing you should not be doing it.
In #engineering if you do not know what you are doing you should not be doing it."
― Richard Hamming, "The Art of Doing Science and Engineering: Learning to Learn" (1997)

2026-02-21 01:43:21 UTC

#Learning is vital to society, just to forestall that quick slip-back into the darkness. By "learning", I mean both the formal and the informal varieties, but with a strong emphasis on self-guided, lifelong pursuit of knowledge to an adequate depth.
A common rule-of-thumb is that 10,000 hours, or 10 years, of sustained study and persistent practice are required for an ordinary person to become proficient in a substantive endeavour, art of science.
A typical career lasts about 50 years, often longer. Hence, it is not unreasonable to say that an ordinary, motivated person could—I say, "could"—become proficient in 5 different worthy pursuits. These specialisations may be orthogonal (neurosurgery and jazz). They may be intersecting (corporate law and tax law). They may be entangled (computability theory and complexity theory).
By all means, a practitioner must be competent in the primary area of specialisation; rules of professional ethics demand so. But, nowadays, most practitioners in many fields—especially those that are not regulated, licensed professions—tend to hyper-specialise, to the detriment of both their careers and their industry. Of late, in #IT, hyper-specialisation has reached pandemic proportions.
In general, hyper-specialists are not sua sponte lifelong learners. It must be extremely taxing upon the mind to just lift-and-coast, for 40 years straight.
Loads of folks talk about learning something new in retirement—as if!

2026-02-20 22:58:51 UTC

Possible Study Paths in #QuantumComputing for Undergraduates
—— #EE, #Physics, and #Maths ——
• PROB—"Grinstead and Snell’s Introduction to Probability", Doyle
• LALG—"Introduction to Linear Algebra", Strang
• QM—"The Feynman Lectures on Physics, Volume 3", Feynman
• QC—"Quantum Computation and Quantum Information" (Mike & Ike), Nielsen
—— #CS and #DS ——
• PROB—"Probability: For the Enthusiastic Beginner", Morin
• LALG—"Basic Linear Algebra", Blyth
• QM—"Introduction to Quantum Mechanics", Griffiths
• QC—"Introduction to Quantum Computing: From a Layperson to a Programmer in 30 Steps", Wong
—— Optional reading for all ——
• VISUAL—"3D Computer Graphics", Watt
—— Arguments against QC ——
• https://youtu.be/pDj1QhPOVBo?si=P4h0GaPceLCxSuye, Yoganathan
• https://youtu.be/MukMOZ0J-Ww?si=WOIzDT-b7sdKqvkP, Hossenfelder
—— My Personal View ——
As of 2026, I see QC as just another Wall Street hype. But this technology has come a long way since its 1982 inception, at least on the hardware side. On the software side, though, progress is almost nonexistent over the past thirty years. And forget about generating QC code using ChatGPT, at least for now.🤣
Still, from the perspective of an intellectual pursuit, #QC concepts have much to teach the #STEM undergrads. Moreover, #QM is one of those studies that squashes a young, blobby mind then remoulds it into a mature, structured mind; it is an eminently worthwhile pursuit. But it is a subject universally reviled by STEM undies, except those in EE and Physics. If QC does nothing else but to induce some of those kids to pursue QM, that is good enough for me.

2026-02-12 17:55:36 UTC

People who had a brush with near-#death have reported fantastical “out-of-body experiences”, a preview of death, as it were.
I wonder if those #AI-powered, self-driving #Crashla cars have “auto-body experiences”—a preview of their impending bodyshop visit, so to speak—at the time of their inevitable wrecks.🤔

2026-02-10 01:13:57 UTC

- reply

@nprofile…llgf 👍👍

2026-02-09 19:16:19 UTC

Doctors do not use the stethoscope and the scalpel everywhere, but they employ countless diagnostic tools and intervention techniques to perform their duties. Lawyers have many substantive and procedural tools, reserving law suits as the last-resort option. Engineers, EEs for instance, use many tools and techniques besides the multimeter and the FFT. In all these regulated, licensed professions, the professionals use both their skills and their knowledge.
In #IT however, #managers are indentured servants of LLM and #programmers are but afternoon snacks for Python.🤷‍♂️

2026-02-09 18:11:35 UTC

\(GPL = \neg DSL\)
• All Turing-complete #GPL are splendid DSL in some contexts: Fortran for scientific parallel computing, Cobol for high-throughput transaction processing, C for low-level system programming, Haskell for categorical programming, and so on.
• Some Turing-complete #DSL are poor GPL in all contexts: Matlab, Mathematica, PL/SQL, Verilog, Forth, and the like.
In pragmatic terms, a GPL that happens to perform well in some specific application domain can reliably be put to many different use cases, albeit with varying fitness for the purpose. But a DSL that offers Turing-complete programming facilities is usually a poor choice for implementing general-purpose applications.
But in modern #programming practice, the distinction between GPL and DSL has become almost as pointless as that between compiled and interpreted. C began as a GPL, but it is now used almost exclusively as a low-level system programming DSL. JavaScript began as a web page animation DSL, but it is now used as a web-borne system programming GPL.
Indeed, some languages—including C++, Java, Haskell, Rust, TypeScript, etc.—have Turing-complete type systems.
Is this the emergence of convergence?
Or is it the convergence of emergence?

2026-02-06 21:52:07 UTC

#probabilistic #programming languages
https://github.com/automata/awesome-probabilistic-programming

2026-02-05 22:46:19 UTC

- reply

@nprofile…8qh7
See my reply to your earlier post about the use of Laplace transform in fintech. I would venture to say that, at a sufficiently sophisticated (complicated) level, the non-linear, time-variant, dynamic models in control systems, cybernetic systems, financial systems, economic systems, social systems, etc., share similar behaviours. But I don't have that advanced mathematical background.
These are the relationships between the three commonly used transforms in ordinary engineering work:
• Laplace transform: on the complex plane, where \(s = \sigma + j\omega\)
• Fourier transform: on the \(\sigma = 0\) axis of the complex plane
• z-transform: on the \(r = |z| = 1\) unit circle on the complex plane, where \(z = re^{j\omega}\)
@nprofile…72p6

2026-02-05 22:30:59 UTC

- reply

@nprofile…8qh7
What I'm about to say is at the outside edge of my mathematical competence. Also, although I've worked in financial institutions on the systems side of things, I've never worked in fintech, specifically. So, please forgive me, if I muck it up.
In routine EE work, we primarily use the Laplace transform with linear, time-invariant (LTI) systems, and in control theory, specifically, we use this mathematical tool to analyse LTI system stability. A large portion of engineering systems are LTI. Of course, there are non-linear, time-variant engineering systems. But the design of such complex, dynamic systems are beyond my competence.
Financial and economic systems are, by nature, non-linear, time-variant systems. I am aware that economists use Laplace transform in their stability analysis work. I just don't know the details of their techniques.
Thomas, what say you?

2026-02-05 21:23:33 UTC

- reply

@nprofile…8qh7 You’re quite right. Where this instability, chaos ensues. Where there is chaos, opportunities erupt for those who are willing to turn a blind eye and to reap the profits.

2026-02-05 21:19:50 UTC

- reply

@nprofile…8qh7 Baggers can’t be choosers.🤣

2026-02-05 21:14:32 UTC

In earlier times, whenever I mentored my clients' in-house #IT staff, they always wanted to learn about not only the practice (the prescriptive, the how), but also the theory (the descriptive, the what) and the philosophy (the ethics, the why) of their work. In other words they, most of them, were learners, not merely doers.
But in recent times, the trend is toward a deaf, blind focus on the how, and an utter disregard for the what and the why. The indiscriminate use of AI to perform technical work exacerbates this trend. The scary bit is, these people are the ones pre-training and fine-tuning these #AI models, in their own image.
To use an analogy from engineering Control Theory, this modern trend is a closed-loop non-linear feedback system with a high (> 1.0) loop gain—a perfect formula for #instability.

2026-02-05 16:39:28 UTC

- reply

@nprofile…8qh7 On the flip side, police being police, they’re more like to be worshipping that lot than to subject them to search-and-seizure.

2026-02-05 16:22:34 UTC

Today, the #FoundationModels (Transformer, LLM, etc.) of #AI predominate every field of human endeavour. Perhaps a grand-sounding moniker, like the following, is appropriate for this much-revered technology:
Regnal Transformative Foundation Models
And recent studies in various fields had shown conclusively that quick access to #information in the short-term erodes the drive to acquire and retain #knowledge for the long-term. So, this estimable label's #RTFM acronym, which harkens back to that famous, 1980s tech aphorism, is equally appropriate.

2026-02-05 14:03:02 UTC

The competence of those who wish to operate a car are subject to the licensure regulation by their state. Older drives are additionally subject to medical scrutiny, by a doctor whose privilege to practice medicine is, itself, regulated by that state. The point is, just to be able to operate something as mundane as a car, the person must submit to a hierarchy of independent scrutineers whose conduct is governed by a hierarchy of regulations and laws. And these control points exist for very obvious and sensible reasons.
But such reasonable #regulation of and independent #scrutiny of #competence is utterly absent, when it comes to #gun ownership and #WhiteHouse occupancy.

2026-02-04 20:23:54 UTC

- reply

@nprofile…8qh7 🤣 No more matrices, no more GPUs. Let's go!

2026-02-04 18:04:10 UTC

- reply

@nprofile…8qh7 Hey, no worries. Loads of people follow like sheep those Wall Street finbros and Silicon Valley techbros. And their uneducated, irrational instincts drive these two industries. But that drive keeps us gainfully employed.😆

2026-02-04 17:58:36 UTC

- reply

@nprofile…8qh7
Why are you bringing logic, mathematics, and physics into a conversation on the AI religion?
😆🤣

2026-02-04 15:14:37 UTC

Since #ChatGPT came out in 2022, the whole of the #IT industry, and indeed #CS academia too, had gone topsy-turvy.
These days, people are trying to use the #LLM for every problem large and small, even ones that could be solved using well known, traditional mathematical methods. Worse, they are now tossing problems they have never heard of and, hence, do not grasp the true nature thereof, into the gaping mouth of LLM, hoping that this miracle machine could make head-or-tail out of something humans were too lazy, or too incompetent, to analyse and understand.
This is the classic case of the hammer seeing nails all around. It is a waste of resources—humans, machines, electricity, space, money, time. In my own career, I have seen this irrational exuberance several times: the LISP rule-based AI boom; the BP connectionist AI boom; the Web boom; the DotCom boom; the Cloud boom; and now the Transformer boom. Up next, the Quantum boom, maybe?
By all means, surf the Wall Street money wave, if that is one’s wont. But intellectuals must always strive to observe the patterns that repeat, analyse abstractly, synthesise concretely, not succumb to the hammer-nail myopia, and above all never abdicate the duty to #think.
“What has been will be again, what has been done will be done again; there is nothing new under the sun.”
— Ecclesiastes 1:9

2026-02-03 23:03:32 UTC

- reply

@nprofile…f4r9 All technological advances have both up and down side, of course. But if a technology is being used to erode freedom and that technology is also readily available for use in defending freedom, why not use it in a way that advances humanity instead.

2026-02-03 22:59:02 UTC

- reply

@nprofile…8qh7 Prejudice is innate all humans. Sad.

2026-02-03 20:53:46 UTC

- reply

@nprofile…8qh7 Mate, I was born and raised in Burma! I know well Don’s “crimes”.🤣

2026-02-03 20:37:52 UTC

- reply

@nprofile…8qh7 Humans 🚬, so why not that which they created in their own image, AI.😆

2026-02-03 20:28:07 UTC

#biometrics
Face recognition is not the only means of identifying criminals and securing our liberties against them; fingerprints, voiceprints, earprints, iris, interpupillary distance, locomotive gate, and many other forms of biometrics exist. Indeed, since COVID, #AI can now perform facial recognition in the presence of masks.
Beware, murderous thugs hiding behind face #masks.

2026-02-03 20:00:18 UTC

- reply

Jimmy interviews Don about being put on ICE.
https://youtu.be/tW0UKsrxjyA?si=a-6Pn986JK9keU5G

2026-02-02 22:31:17 UTC

We used to be a nation of rescuers of stray kittens. But in just one decade, we turned into a nation of people who hunt and kill innocent humans for sport.
What happened to civility, empathy, humanity, religion, law, shame?

2026-01-30 05:32:59 UTC

I know of many things that I do not know but ought to know. That #knowledge of “known unknowns” is the earnings, the profits, the benefits of age.
Life is short. #Learn!📚

2026-01-28 16:21:45 UTC

- reply

@nprofile…llgf
Ah, the classic name conflict between the theoretical “G-machine” and the practical “G Machine”. I was sloppy; I should haves used the “-“.

2026-01-28 16:09:31 UTC

- reply

@nprofile…llgf The is the lot in life for FP, I’m afraid….

2026-01-28 03:52:37 UTC

The period between the mid 1970s and the mid 1980s was arguably the most exciting time for #FP language design.
The period between the mid 1980s and the mid 1990s was decidedly the most exhilarating time for FP language implementation.
Every new FP or FP-like language that had emerged thereafter is simply rehashing great ideas from those heady days, often shamelessly slapping new labels on old concepts, just to sound fresh and original.
Modern FP #programming languages…. 🤷‍♂️

2026-01-24 06:02:42 UTC

- reply

@nprofile…8qh7 🙏 Much obliged, mate. Ronan's back on campus. And guess what—his first class, Advanced E&M was in that Physics building where Brown campus shooting took place. Oh, Lordy!🤦‍♂️ But he's fine.
The nearly-defunded NOAA is predicting that we in the DC area will get about 61 cm of snow. My house is up on a steep hill with a driveway about 120 m in length. So, we'll see how things shake down, in about 12 hours.

2026-01-22 01:20:03 UTC

- reply

@nprofile…8qh7
😆 No kiddin'.
And given the past history of rampant cheating in F1, I suspect that the 2026 Racing Bull car is hiding a massive turbojet engine under that ugly body work.😀
Honestly, I'd rather look at the Brabham BT46 than this new monstrosity.

2026-01-22 01:11:29 UTC

#ChatGPT made the 2026 #RacingBull #F1 car even "cooler".
https://media.mathstodon.xyz/media_attachments/files/115/936/083/354/982/270/original/db98b98d82c24c14.png

2026-01-21 20:49:30 UTC

- reply

@nprofile…8qh7
Great points!

2026-01-21 20:02:41 UTC

- reply

@nprofile…8qh7
Ya know mate, tRUMP's nearly-defunded NOA said it's a "weak" storm, but Mother Nature may have a different idea.
In any case, I have you and most of my former classmates living and working in in the Great State of California. Keep me apprised, mate.
Best of luck to y'all!

2026-01-21 19:06:43 UTC

- reply

@nprofile…8qh7
💕Love that old stuff!

2026-01-20 21:07:45 UTC

- reply

@nprofile…8qh7 🙏👍

2026-01-20 20:50:57 UTC

- reply

@nprofile…8qh7 🤣 Not at all. By “learned”, I don’t imply “book learning” exclusively. I meant “life learning”. We've all met (and have had to work with) "on-paper learned" folks, who are markedly different in character from the true, "in-life learned" folks.
In that sense, every experienced bloke has those same duties.

2026-01-20 20:09:09 UTC

An enlightened practitioner of a field owes himself and his field the duty to endeavour to become #learned. And by becoming learned in his field, he assumes the ethical responsibility to mentor those who are still #learning. And he further assumes the moral responsibility to be humble himself, but never to humble those whom he mentors.
In #IT today, many practitioners never reach even the first stage of enlightenment, because they are too busy praying to the Almighty #LLM and ritually sacrificing their young mentees at Its high altar.🤷‍♂️

2026-01-19 23:38:46 UTC

- reply

@nprofile…8qh7
“Just what do you think you're doing, Dave?”

2026-01-19 23:31:12 UTC

We need devices—phones, computers, toasters, toilet seats, couches, cars, ...—equipped with a switch labelled "#AI-free".

2026-01-18 17:51:29 UTC

- reply

@nprofile…8qh7
And often in IT, those so-called "thought leaders" like to name things that are removed from practice in the field. It's all a mess.

2026-01-18 15:15:48 UTC

- reply

@nprofile…8qh7 This is an odd field that came about due to the increasing complexity of the deployment environment. But I’ve seen in the field all kinds of mix-and-match: coders doing their own software deployment and maintenance; non-coding testers taking on that role; dedicated devops who Perth’s role and report to software lead; and other variants.🤷‍♂️

2026-01-15 15:28:04 UTC

- reply

@nprofile…a895
Once the code gets as close to the hardware as C or WASM, getting down to assembly is but a small step, by comparison to the effort required for the front end compiler stages. So, someone might inject pluggable assembly generators in the future.

2026-01-15 15:25:52 UTC

- reply

@nprofile…a895
Try Gemini instead.

2026-01-15 15:24:54 UTC

- reply

@nprofile…a895
Yes, especially on complex processors with long pipelines, large caches, delay slots, and so on, reading LLM generated code or LLM explanations thereof is a more efficient way to learn assembly.

2026-01-15 15:22:18 UTC

- reply

@nprofile…a895
True. But I would say that learning assembly on a massive modern CPU from a compiler generated samples far more arduous than learning straight from a book in the subject.

2026-01-15 15:15:59 UTC

- reply

@nprofile…a895
F* doesn’t generate assembly. It generates WASM.

2026-01-13 20:51:30 UTC

- reply

@nprofile…a895 😍😍😍Thank you for the link mate!🙏

2026-01-13 19:33:59 UTC

- reply

@nprofile…8qh7 Now, that’s the balanced approach, “Middle Path”, as Buddha put it.🤣

2026-01-13 15:59:06 UTC

The practice of law is about #judgement based on information and knowledge—of the plaintiff in deciding to take legal action, of the attorney to counsel his client, of the judge in reaching the ruling on his case, and up the appeal chain. This responsibility to exercise judgment is evident also in medicine, engineering, computing, art, music, and other intellectual endeavours.
So, I fail to grasp why young #IT practitioners are now willingly abdicating their ethical duty to exercise judgment and sacrificing their minds at the alter of the Almighty #LLM.

2026-01-12 22:53:35 UTC

- reply

@nprofile…a895 See Backus’s paper I mentioned, just for chuckles.😁

2026-01-12 22:52:50 UTC

- reply

@nprofile…a895 True. But you must prioritise.

2026-01-12 13:45:04 UTC

- reply

@nprofile…a895
Landin’s “Next 700 Languages” (ISIM) paper is hard to read for modern readers for a number of reasons: it uses classic terms to describe classic ideas; it talked about FP when the world was still rejoicing the PP ascendancy; it talked about SECG graph reduction evaluator only FP languages use; it tried to reconcile with PP; it talked about the offside rule; it uses a precise mathematical language in common use in academic papers of the era. It was truly a far ranging, forward looking paper. But it isn’t easy to read like the run of the mill modern arXive papers. This is one of the reasons why modern CS folk should read classic papers like they read historical texts in other disciplines.

2026-01-12 13:36:35 UTC

- reply

@nprofile…a895
Grok is a bit off. It meant “dynamic scoping”, which is something only the LISP 1 programmers adore, even if they themselves did not manage to tame the beast. It wasn’t unit 1975, when Steele and Sussman invented Scheme with lexical scoping (from ALGOL) that LISP grew into its modern “static scoping” form.

2026-01-12 02:10:19 UTC

- reply

@nprofile…q7ly
By "addictive", I meant it as a complement to the Arduino, the concept, the implementation, the ecosystem, the product line, the educational implications, the lot.
I use STM32 most of the time, but I do admire the ESP32 and the Arduino.
Oh and, FPGAs are getting hotter, too.
We are living in the golden age of low-SWAP.

2026-01-12 00:54:27 UTC

- reply

@nprofile…q7ly
The biggest problem with Arduino, not just the Uno, but the entire line, is that they're addictive.

2026-01-10 23:48:01 UTC

- reply

@nprofile…a895 🤣 👍

2026-01-10 14:53:59 UTC

- reply

@nprofile…8qh7
Love those MIP displays, too!

2026-01-09 20:05:35 UTC

- reply

@nprofile…8qh7
I've never heard of CLINGO. This will be a sweet read for this weekend. Thanks so much, mate!
@nprofile…38ae

2026-01-09 19:41:48 UTC

I have met a few experienced #IT practitioners who are also well read in the classics. Some even know Greek or Latin. But even these erudite, senior folk balk at my advice to show due regard for the "classics" in #computing by reading early seminal papers (Gödel, Church, Turing, Kleene, von Neumann, Shannon, etc.) and by studying early pillar #programming languages (an assembly, Forth, ALGOL, SIMULA, LISP, ISWIM, etc.).
Most IT thought leaders today dismiss those computing classics as being too old for, too distant from, too irrelevant to modern IT practice. By "modern", they mean #AI/#ML, of course.
If one opts to allot a significant portion of his scarce free time to reading Plato or Virgil, one ought to allocate at least an equal amount of effort to studying the classics in one's own field of practice.
Perhaps these modern thought leaders are right; I may have become stale and crusty, mired in my tired old ways. But I still fervently believe that a conscientious practitioner should be a lifelong learner who never deems superfluous the knowledge of his own field's foundation and history.
Classics—their utility is not pecuniary, but perspective.

2026-01-08 21:29:11 UTC

- reply

@nprofile…8qh7
Legal Construction 101—Hey, the text said "well-regulated Militia" but there's nothin' on "poorly-regulated Militia". So, let's go blow some shit up!

2026-01-08 21:23:20 UTC

#SecondAmendment
A well-regulated Militia, being necessary to the security of a free State, the right of the People to keep and bear Arms, shall not be infringed.
#TenthAmendment
The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the People.
————
The 2A grants the People their gun rights in the context of their participation in the State Militia in defence of their home State. Against whom? Read on.
The Constitution, on the whole, expressly limits the Federal powers by granting only a few specific powers. These are called the "enumerated powers".
In contrast, the 10A reserves to the States those broad "general powers", so long as those general State powers do not conflict with the enumerated Federal powers.
When these contemporaneous amendments, #2A and #10A, are read in the context of the encompassing Bill of Rights, the textual tenor of the Constitution, and the historical setting, the original intent of the Framers of our #Constitution comes shining through: the Framers expressly chose to limit the #Federal powers so as not to impinge upon the #States' powers, and the regulated, armed Militia manned by the citizens of the State were to provide an armed defence against encroachments upon their home State's rights by the aggressive neighbouring States and by the overbearing Federal government.
Now, reread the current events in the news, under this light.

2026-01-08 10:24:53 UTC

- reply

@nprofile…8qh7 @nprofile…38ae
"Multi-paradigm" and "sophistications" and "modernity" were exactly the things I was trying to avoid. My goal was "simplicity" and "foundational".

2026-01-08 10:18:20 UTC

- reply

@nprofile…8qh7
Well said, as usual.
@nprofile…38ae

2026-01-08 05:02:18 UTC

Every young #programmer should master these #programming languages, at the earliest opportunity:
• C
• Smalltalk
• Standard ML
These are the pillar languages in PP, OO, and FP paradigms and, collectively, they cover a broad spectrum of abstract thought in #software. It is no exaggeration to say that every modern language currently in popular use is either a rehash or a remix thereof. Moreover, these are arguably the simplest languages in their respective paradigms. And through them, the programmer can master these three foundational paradigms, thereby gaining the ability to think effectively and efficiently about any problem from diverse perspectives.

2026-01-03 16:05:43 UTC

- reply

@nprofile…a895
Great points, as usual, mate. I skirted away from LISP-style prefix and Forth-style postfix syntax, because they are orders of magnitude simpler to parse than ALGOL-style syntax. My intent was for the students to work their way up to language processing tools like Lex/Yacc, ANTLR, Xtext, and so on.
By "call", I don't mean for the student to implement the "call" instruction; I meant for him to implement the procedure call activation record (preparing the stack frame). This code will become the procedure-call parser action, when Yacc comes into play, later.
Any language that has unconditional jump (goto), conditional (if), and the recursive functions (call) is a Turing complete language. This so-called high-level, human-readable language is almost identical to assembly language, so the backend can translate it down to any target architecture, with relative ease. Hence, these three language constructs.
The reason why I didn't suggest "loop" is my FP penchant. Moreover, implementing FP languages is lots harder than PP languages, in general. I want them to start thinking about FP as early as possible. Of course, this means they would have to tangle, early on, with inner functions, enclosures, and other mess. So, you're right; it's probably way easier for them to just implement "loop". But that is how much of the world is stuck in PP. And even with "loop", there must still be procedures.
Lastly, the main reason for choosing such a simple language over much more complex starter language (subset of C, Haskell, etc.), is for the students to see how computing goes from 0s and 1s to assembly to LISP/Forth/C to OCaml to ....

2026-01-02 19:37:23 UTC

#EE undergrads interested in #CPU and #compiler designs may wish to explore the following:
• Pick a classic 8-bit CPU architecture, 8008, 8080, Z80, 6800, or 6502, and learn the instruction set architecture and the assembly language for that CPU
• Implement, exclusively in C, a minimally-functional assembler for that instruction set
• Adapt the assembler to use Lex and Yacc, in order to gain familiarity with these language tools
• Implement a compiler frontend for a barely-Turing-complete high-level language, maybe supporting only if, goto, and call/return, and implement the backend that targets the chosen instruction set
• Add a strong, static type checker, along the lines of Pascal
• Expand the language by switching to language workbenches, like Spoofax, Xtext, or MPS
• Implement a simplified version of the chosen CPU architecture in Verilog or VHDL, perhaps even digging down to a cycle-accurate version
An average student can get through the first two steps in the first two years of college, perhaps even in the first two semesters.

2025-12-31 16:56:46 UTC

- reply

@nprofile…00p2 I just can’t tolerate the “range anxiety” on these smartwatches, smartcameras, smartcars, and all things smart. I resisted switching to smartphones in the early days because of the battery life. Smartwatches with bio tracking should have at least 7-day power reserve with a 1-hour recharge capability. But g’luck with that….
WW2 Elgin 24-hour cockpit clock in the F-4U Corsair had 8-day power reserve. Today, manual-wind mechanical wrist watches have similar power reserves. It’s time the electronics and the battery technology catch up to that.
I bet there will be a Nobel physics prize for the bloke who invents that new battery technology.

2025-12-31 14:36:51 UTC

- reply

@nprofile…pmx6
My, my, lookie how time flies….
Yes, this book is “old” by today’s kids’ standard. But I still use the Haskell-98, so….🤣

2025-12-31 03:32:54 UTC

"Write Yourself a Scheme in 48 Hours: An Introduction to Haskell through Example", Tang (2007)
https://upload.wikimedia.org/wikipedia/commons/a/aa/Write_Yourself_a_Scheme_in_48_Hours.pdf
I'm well and truly miffed—until this moment, I'd never come across this little jewel of a free ebook.🤦‍♂️
This book teaches practical #Haskell #programming (like monadic parsing) through an implementation of a large subset of R5RS #Scheme interpreter, all in just 138 pp. An experienced FP #programmer, even one who is unfamiliar with Haskell and Scheme (although I’ve not met such a person), can plough through this text, in one sitting.
It's brilliant!

2025-12-31 03:09:11 UTC

- reply

@nprofile…00p2
Wise words, as always.
Ditch the ephemeral smartwatches, return home to the perennial G-Shock. The classic Square shape now has models with MIP display, Multiband 6, Bluetooth, and solar. And you can get a handful of G-Shocks for one iWatch.
Ya caint beat ded!

2025-12-31 02:10:09 UTC

- reply

@nprofile…00p2 🤣Good one.
And seriously, it’s a damn good idea for a fun project. This trend towards hardware solutions is growing, of late, perhaps due to lower costs and lower barriers to entry.

2025-12-31 02:07:15 UTC

- reply

@nprofile…00p2
I regret that Forth is no longer the dominant tool for MCU programming today.
As you know, learning a language is never the hurdle. It’s the time and opportunity to reach mastery that gets tricky, either due to the choices we make in life or those that are made for us by others.
I learned Forth a long time ago. But I never had the chance to use it on anything other than small personal projects. And I can’t use it in good conscience (technically, managerial, financially, ethically) in the kinds of work I do now.

2025-12-30 17:38:08 UTC

I have many regrets in my life, and the longer I live the more those regrets pile up.😀 Were I to construct a priority heap of my regrets, my lack of mastery of the #Forth #programming language would be near the very peak.
I haves worked with DSP and DIP on #microcontrollers, since the very beginning of my career. But by the time I became an #EE in the 1980s, C was the “Python of the Era”, used on supercomputers, mainframes, and desktop, all the way down to 8-bit controllers.
Truth be told, I would not attempt to implement a relatively complicated DSP algorithm, say FFT, in a language like Forth that is but a veneer atop assembly. But my predecessors had implemented entire operating systems in assembly, so….🤷‍♂️

2025-12-27 16:52:16 UTC

- reply

@nprofile…pmx6 I regret never having had the opportunity to use the CDC-6000 line. You are one fortunate man, my friend.👍

2025-12-27 16:44:57 UTC

- reply

@nprofile…00p2 The size bloat of Haskell 2010 isn’t due to HM unification algorithm, which was in ML and Caml, too. The bloat is mostly due to Haskell’s innumerable “standard features” and the way they interact with the non-strict evaluation strategy.
But all FPs, even “little” ML, produces a massive “Hello, World!”, because of the GC and the runtime. But once the size of programme grows to a normal “system program” level, the overhead all but disappears.