GianFabien

The Gaslighter in Chief wants to abolish the First Amendment.

... again

Didn't work too well last time. Will be amazingly effective this time </snark>

If it's anything like the "negotiations" where the other party gets bombed before even leaving the venue, it could be a trap.

What's the point? No matter how detailed and comprehensive the instructions and steps by the AI, you still don't get a PBJ sandwich to eat. You have to go to the kitchen and do it yourself.

Good on you for rising up to the ranks of Linux/BSD.

You just need to recognize that not everybody aspires to be competent with lower-levels of hardware and software that Apple makes that much more difficult. Most Apple users are content to use apps written by others and that is as far as their interest goes.

An analogy is the car market. Most people don't care how the car works, etc. They just want to get to places. If you only need to drive to the shops and do minimal errands, you don't even need a truck - a sedan will do just fine. Same with computers, lots of different market segments with distinct needs and expectations.

I spent a part of my career supporting IBM PC/XT, PC/AT and then Compaq and the many PC clones.

Microsoft, as its name implies, started out and cuturally was always a software company. By the time Microsoft started dabbling in hardware it was seen as a competitor by the many software licencees. Microsoft could never control the hardware side of their platform like Apple has.

The other factor is that Apple ditched the proprietary OS and went down the Unix/BSD and derivatives path. MacOS is effectively a FreeBSD derivative overlaid with a custom desktop.

Without a quasi-portable OS at the core, there is no way that Apple could have made the switches from Motorola 68000 to PowerPC to Intel to ARM.

Windows, on the other hand, is running NT which is basically a port of DEC VMS. Didn't work out well for DEC/Compaq/HP.

I don't use WASM as a replacement for JS. I never have a need to manipulate DOM, etc from WASM. JS is perfectly fine and performant for those purposes.

As I see it, WASM is used to augment the JS/WebAPI ecosystem. For example, when you need to do heavy bit manipulation, complex numerical processing. The round-trip JS->WASM->JS is an overhead. So the WASM modules should perform a substantial amount of processing to offset that inefficiency.

I frequently find that V8 optimisations yield sufficient performance without needing to delve into WASM.

IMHO if you want to write WebApps in Rust, you're holding it wrong.

With a miniPC you can always use a Raspberry Pico (or Arduino that supports Device Mode) for your GPIOs.

Perhaps you could be more specific.

For example, for architectural 3D modeling software, the operating the software is being the architect who visualises, designs and refines the building's design.

Even before recent AI capabilities, writing software was (now is) table stakes.

Deep domain knowledge and expertise is essential. Until you actually work at the coal face in a given industry you don't know the complexity nor the opportunities for improvements. Talking to the workers is good, but you never get the complete picture.

From a technical perspective: prices change constantly; prices are inconsistently expressed and the volume of products is massive. Scrapping the volume of data required and then processing it and updating it is a substantially difficult task.

From a business point of view an independent operation would require a very large investment. To recoup at least part of that requires advertising and affiliate revenue which in turn creates perverse incentives.

As a project lead, I have worked on dozens of projects and 100s of contractors. Contractors have ranged from brilliant and productive to struggling incompetents. In almost every case, I worked with teams I was given and had no say in the hiring process. A typical case when management thinks they know better than their tech lead.

I very much believe that comprehensive domain knowledge and technical proficiency are both essential. Actual code production can be mostly delegated. If AI produces better quality code than the contractors available to you, then it is the preferable option.

IMHO a small team of experienced engineers using AI is the optimal choice.

Vibe-coded startups without competent technical oversight is tech-debt on steroids.

Unlikely. There is just too large an installed base of software that will only run on Windows and is widely in use.

However, we have to acknowledge that if you consider Android as being Linux based and all the servers in the cloud, then Linux has already supplanted Windows.

When people receive any written material from me, it is I that wrote it. My thoughts and expressions. If they wanted to read some AI generated drivel, then they don't need me and my contribution.

My process is very simple. I just write from top to bottom. Of course, the first pass is rough. But the focus is on capturing the material in approximately the logical sequence. For really complex and lengthy materials, I might write an outline with mostly headings and snippets as they come to me.

For emails and other time critical writing, I go back to the top and edit / re-write. I don't need nor use AI for this. Then send. Rarely do I use a third pass.

For reports and papers, I tend to put the first pass aside for a day or more. When I return to it, I edit viciously and re-write. Depending on the importance of the writing, I might repeat this process 2 or 3 more times.

I see you having a unique opportunity. There is a scarcity of strong SWE skills in the earth sciences related industries/businesses. You could refresh your reasons, motivations and interests for choosing earth sciences for your PhD. Perhaps your thesis and research connections reveal further dimensions.

In any earth sciences based work, there would be requirements for either data sciences or AI related processing, improvements, etc. Those are the sorts of areas that you might want to peddle your SWE skills as delivering.

Awesome! I wouldn't have thought that it is possible to make ICs in a garage. Of course it requires a lot of knowledge, etc. But still, not a multi-billion dollar clean room with specialist equipment.

I think that long, tedious, vapid text is AI written. Pithy, deeply insightful materials I assume are written by humans.

These days, I just delete and move on if the "article" doesn't deliver interesting and meaningful content within the first half page or so.

The bloat. I prefer lean designs with plug-in modules for additional functionality. Not only do unused sub-systems take up memory, but they also increase the potential attack surface.

I think Deno's management have been somewhat distracted by their ongoing lawsuits with Oracle over the release of the Javascript trademark.

I started out with Deno and when I discovered Bun, I pivoted. Personally I don't need the NodeJS/NPM compatability. Wish there was a Bun-lite which was freed of the backward compatability.

Removing the licence and/or authors from a FOSS project would generally be a violation of the licensing terms. The tool(s) you use don't change the legal principles.

Of course, the big AI companies blithely ignore moral and legal issues.

RGB LEDs (e.g. WS2812) connected to an ESP8266 and running microPython are great for experimenting. I have several "installations" that respond to broadcast datagrams (UDP). Let your imagination run riot. Add sensors, even MIDI interface.

I do doubt that you could actually discern 16M colors. But even many thousands is entertaining.

Vibe coding in the hands of competent, experienced developers provides amazing leverage. For example, whipping together utilities to perform domain specific tasks. Of course, it could be done the conventional way but why bother when your AI "junior dev" can whip it up in minutes.

Instead of downloading various open-source utilities and finding none does exactly what you require, you can have AI build a bespoke program, just for your specific needs.

Is the consulting company actually named somewhere?

If you are a government procurement manager, you certainly would want to know where the biggest kickbacks come from.

Using Advance72 (made by Sonova) has BlueTooth which makes using a phone a breeze. The Hearing Remote app allows me to override the audiologist configured settings for volume, listening modes and equalizer.

IMHO rechargeable models are not worth the extra cost. With a typical 5 year replacement cycle the batteries will no longer hold the early levels of charge. 312 batteries are cheap and easy to carry spares.

I struggled with C++ after coming from C.

Then I read Alan Kay: "When I invented OOP I didn't mean C++". So I explored Smalltalk, loved the environment, hated that deploying apps is a pain.

Tried Go, because Google ...

Now mostly JS because it's everywhere and Python for the batteries.

I continue to find it staggering that managers in corporations have so little understanding of information technology. Hubris and the Dunn-Kruger effect have totally eclipsed any critical thinking abilities.

The silver tongued sales persons who wine and dine the decision makers continue to make out like bandits selling snake oil.

Yes, it is often the case.

BUT why not get a startup going and address the opportunity?

Amazon, eBay, PayPal, Uber, AirBnB, etc likewise saw problems as being opportunities in disguise.

When you look beyond office jobs, you see many real opportunities.

For example, there is a housing crisis. Not enough trades persons, building supplies, capital to solve that problem.

The unemployment statistics aren't detailed enough to show IBM, MS, Facebook, Amazon, etc laying off tens of thousands of employees a year, each. Last I read, over 500,000 staff have been laid off in the past couple of years.

I read all the comments. But I don't see anybody asking:

"Why do you think you need to share your code?"

"What benefit(s) do you expect to gain from sharing your code?"

It might be worse than just the AI investments. Oracle's client base are predominantly enterprises. When enterprises are cutting staff and expenses, they also look at reducing their licensing spends. Oracle has been very predatory with both Java and database licenses fees. I suspect that revenue from those areas is also likely to shrink.