Posts by Andy 73

SpaceX just about nails Starship test flight 13

I'm going to decide if "the progress rate is remarkable" when they have a working commercial service.

SpaceX guys are smart, certainly, but there's a whole load of organisational and architectural stuff in this project that is either radically new, or imposed by outside forces.

That's where you have to accept that Musk's dependency on government support, and tendency to gamble big may outweigh any technical expertise within the company. History may well overtake them, even as they're doing their best to deliver his slightly distorted vision.

Some of us are watching this with cynical hats on, not because we underestimate the complexity of the project, or the excellence of the engineering, but because we understand that SpaceX are having to sustain a high wire act for far longer than many companies have endured.

And more importantly, with Musk, we've seen this before. Self driving cars have been "nearly there" for over a decade. Smart solar roof tiles were ready to be delivered. Hyperloop was going to revolutionise cities. The electric semi was going to transform transport. Grok was... I don't know what Grok was meant to do, but I'm sure it was world changing.

The current huge shift in space has occurred because private companies have suddenly become economically viable without total government subsidy. SpaceX is leading the way, but it's now one of many, and still has a lot of work to do. The attitude on here comes from the fact that we've spent a very long time being told Musk was the golden boy, and this and other of his projects are allowing us to judge just how much of that is hype and how much is true. If that judgement sometimes seems harsh, maybe look to what the boss is doing to see why people are giving his companies a hard time.

Java was a three-day hotfix away from dying horribly on stage

Re: anti Java sentiment

There used to be a lot of snobbery from the greybeard C programmers who were threatened by a new language and (quite fairly at the time) viewed performance and memory management of a VM as a serious flaw.

These days, I suspect the main anger comes from a combination of exposure to sprawling estates of legacy applications, and the regularity with which complexity of online systems explodes in our faces. Neither of these are strictly a problem with the platform, but it's hard not to associate "what the hell is this mess?" with the language.

For people who just want to create an online "Hello world" program there are of course a thousand newer platforms that abstract away all of the complexity so that you can write the bare minimum and the rest is assumed. That's very welcoming for newbies and c-suite management, but of course isn't where the bulk of the Java ecosystem originates. When things begin to grow to a certain scale and level of complexity, that exact same ecosystem suddenly becomes a very powerful tool that can express key architectural concepts in a robust and repeatable manner. Is it perfect? No. Is it often quite verbose? Yes. Is that something a competent developer should be able to handle? Also yes. Are there tools and frameworks that manage this? Of course.

Re: Threading at the time was particularly new

Simplified for a broader audience and dramatic tension.

Re: Fawning drivel

The fact that vast swathes of the internet run on Java quite happily, and have done so for decades without the apocalypse arriving suggests that your favoured gotcha is not quite the showstopper you wish it to be.

Log4j was notable simply for being one of the very rare serious security lapses in the Java ecosystem. Considering how much banking, financial transactions, security access and online shopping run through Java servers, it's not as if the entire platform isn't under constant and very tight scrutiny from both black and white hats.

As it is, the herd of frustrated techies who enthusiastically embrace each new wave of language and platform with claims that "this one removes all of the mistakes of the last" seem to miss that a consistent and (on the whole) well managed ecosystem that has successfully evolved over decades of production use is an extremely good environment for delivering actual working product.

Lucky 13: SpaceX aims for July 16 Starship flight test

Agreed

Just a year ago you would have received dozens of downvotes for saying such things.

The reality of Musk's "jam tomorrow" machine has taken a decade to be recognised by professionals on this site - which suggest that ill-informed retail investors (who make up the bulk of the investment in Tesla and Space-X) have many years yet before reality sinks in. Long enough for Musk to "declare success" and hand over the sinking ship to someone groomed to make him look good. He'll retire with status intact amongst his fans.

Frame: A new X11 server – implemented directly in assembly

So...

... an incomplete window server that's not necessarily even maintainable by it's "owner"..?

I've looked at a few AI assembly programs, and almost universally they're two to three times larger (and that typically translates into two to three times slower) than a hand written piece. They all look pretty much like how they are - a series of stack overflow answers glued together in a shape dictated by another stack overflow answer.

It falls into a weird uncanny valley - none of the advantages of a compiled language for succinctness and expressivity, and none of the advantages of assembly for speed and size.

Given anyone with enough credits can "press the button" and announce they've written <X> in <Y> (pick well known app/tool/framework and programming language, Mad Libs style), the only real success being demonstrated is the ability to get the item into the news.

Microsoft is losing the battle to protect license lucre. It better get used to the feeling

Re: Great start, and then...

"My hypothesis is that software licensing will be dead, commercial and open source. Not completely, but mostly. Just because the build-or-buy decision will be massively slanted towards build."

There are infinitely more users than builders, even with AI allowing comparatively inexperienced people to create software. Again, the AI corporates would love everyone to start building their own software, because a million people all reinventing the wheel (but personalised mildly to their circumstances) will be paying a million subscriptions. However realistically, AI does not absolve you of the need and ability to reason about software (your example of replicating Erlang/Elixir demonstrates your level of experience - I suspect you overestimate other people's abilities). So we're in a position where users are not free to build their own tools, but the people with the skills and training are not able to afford to provide the work.

Licensing addresses that asymmetry - a handful of developers writing a tool for thousands of users.

Additionally, we want those skills, expertise and business drive as they will always be applied to new problems. The ability to build is great, but it is an incredibly inefficient use of our collective resources to devalue experts in favour of local handcrafts, however well augmented.

And note this isn't an argument specifically for or against patents. This is about the broader question of protection of intellectual property, ownership of "work" and individual rights to assert ownership. The current legal frameworks of patents, copyrights and trademarks were not devised with anything like the modern digital world in mind, let alone the wrecking ball of AI.

Great start, and then...

The starting premise is fine - subscription fees for software that can be bought once are a bad deal for consumers. Certainly we shouldn't allow Microsoft to artificially tilt the table in favour of subscriptions by making on prem purchases more of a short term loan.

However. the suggestion that AI could be used to take any software and recreate it as a legal "loophole" to ownership is.. problematic. In the short term, AI is not a "one click software duplicator" - and may never be, depending on who you ask. It therefore requires a reasonably competent team and some reasonable expenditure and effort to generate our imaginary "legal copy" of software. This may be great for groups that want to dismantle profitable businesses (or instantly compete with them), but it's less so for people who want to build new and innovative software. We're in a situation where it's expensive to create new applications and tools, and even more expensive to establish a market, but relatively cheap to reduce the value of that market to near zero. This destroys the potential reward for writing new software.

Suddenly we're in a world that looks like the early industrial revolution - once someone else has a good idea and proves it (the expensive part), it's incredibly easy to "rip it off" and basically remove all reward for the inventor. We quickly figured out that this caused chaos, and strongly disincentivised innovation... and patents were born.

(At this point, half of the people reading will stop reading and just go and downvote).

In the modern world, the rise of corporatism, legal warfare and overreach of patents has certainly caused a lot of problems. I can absolutely understand the desire to "rip it all up" and make the whole world free. However, the harsh reality is that a lot of evidence is pointing towards this sort of freedom largely benefiting corporate giants, and rarely benefiting individuals. The enthusiastic supporters of free software point to a very healthy ecosystem of projects, frameworks, tools and applications. What they don't point to is the people who have to work a day job to pay their bills, the projects that struggle for support, the immense effort successful projects have to expend on constant fund raising and support, or the burnout many maintainers experience.

No - I'm not advocating patents. Or supporting Microsoft. I'm pointing out that people have real world physical needs - food, shelter, comfort - that must be met. If we use (corporate supplied) AI to make ownership of software rights unenforceable, the only people who benefit are the corporates who charge subscriptions for AI tools. We need to ensure that creative works, innovation and even simple toolmaking provide fair reward to the people that do them. That's the challenge of our age, and anyone telling you there's an easy solution is selling snake oil. No matter how wonderful the promises of the AI giants sound, we are not living in an age of abundance, just rapid change. Up until AI puts food on our tables for us, we have to protect individual workers' rights - and that includes the rights of protection for their work. Sure - they can choose to waive those rights if they want to. Many people in the wealthy West can afford to do so (or could, whilst we had a booming economy fuelled by endless cheap credit) - but no matter how much we might want to punish Microsoft, forcibly removing rights without option hurts all of us.

TLDR; the vision AI hypesters sell of infinite free software is a mirage, and the legal loopholes they wish to encourage do not magic it into existence - instead they harm the workers and individuals who actually drive our industry. An industry shorn of legal protection is one that is at the complete mercy of corporate giants far worse than Microsoft.

MPs tell Brit government: Sort out your tech sovereignty or get left out in the cold

Re: Hold on...

It's nice to see a fellow cynic with a firm grasp on the reality of the situation.

For those of us in the trenches (as you say, about half the people who read this forum) it's become increasingly clear that the idea of democracy, or even meritocracy in these technical issues is somewhat disproven by a few decades of watching smart people discuss the issues whilst our beloved leaders studiously ignore them and continue to repeat the same mistakes time and time again.

Hold on...

Hold on, wait... the government has a strategy? For anything?

I thought we just waited to be told what to do by the shiny American suits? That's why we let Plantir run the NHS, Comcast run two national broadcasters, Oracle, Google and Amazon run most of our public IT infrastructure and Musk run our military communications network?

Europe's new import rules are coming for your bargains

Law of unintended consequences

Yet another example of governments announcing "we will do <X> to stop <bad thing>", where X and bad thing are only tangentially related.

Some people will accept this at face value. Others will look for the unintended consequences.

The per-line-item charge will certainly harm small businesses wanting to sell into the EU (an intended consequence, but justified by implying that it's targeting Temu, not mom and pop hobby shops), and increase prices to consumers (because, well everything increases prices to consumers).

One probably unintended consequence is that it is likely to also entrench the very companies it's claimed to stop - Temu, Amazon et. al. are more than capable of setting up IT systems and logistics that avoid the tax. Niche businesses that grew out of the sudden availability of global trade for everyone are likely to struggle. For products that rely on a diffuse community of supporters worldwide, this is another nail in the coffin - generic Temu tat will ultimately win.

AI search could kill the web without new quality signals and revenue models

Well, duh

We were pointing this out in the comment section a year ago when Google was claiming that search was "better than ever", and got shouted down by people who couldn't understand why keeping them on the search page might be the goal after all.

UK government wants 'trusted' news sources promoted above the social media noise

Hmmm..

Given this Government's enthusiasm for control over the media, allowing them to 'promote' favoured news sources over other is a bit worrying. Even more so when we look at other political parties here and overseas who have abused such controls for their own benefit.

Add in the noticeable decline in journalistic quality at some of the mainstream media outlets, and the ability to speak truth to power becomes rather important.

My only suggestion would be to address some of the root causes for the existing distortions in media - that global corporate ownership of media favours near monopolies that are more concerned with share price than quality of output. It's pretty clear that a very small and financially incentivised group having control over a wide range of media in the West leads to some 'blind spots' in reporting, news and educational content.

And the sooner the BBC can stop doing the decades long dance to prove it's worthy to a revolving door of Governments, the better. It has become a weaker broadcaster for having to jump through constantly moving hoops to justify its existence.

A modest proposal: Reformat everything to make documents more palatable to AI

Astonishing admission

AI is "great at programming", "able to come up with novel algorithms", "capable of spotting patterns in data without being told"...

... and yet apparently incapable of reading structured documents?

It's clown shoes all the way down.

Angry devs vow to flee GitHub Copilot as metered billing takes hold

Small question...

Am I allowed to find this funny?

Git is unprepared for the AI coding tsunami

Hmmm..

As pointed out elsewhere, this is a lot of people selling solutions to the general frustrations with Microsoft's tenure of GitHub - much less problems with Git itself. Yes, there are issues, but some of the biggest pain points are being addressed in 3.0 and for most sanely organised projects, that will be enough.

However:

If you find Git "difficult to understand" - then letting AI agents loose on your codebase to pebbledash it with random changes and inexplicable alterations is not going to end well.

And if you already have difficulty with merge hell, concurrent development and conflicts, then this points to organisational and discipline issues that are absolutely not going to be made better by throwing autonomous bots at the problem.

As ever, the new shiny comes along offering a way to make all of the complexities of software development go away. This week it's AI, last week it was Rust etc. etc. The reality is that getting this stuff right combines problems of architecture, decomposition and organisation long before we talk about which lines of what language we want to commit. It's notable that most of Microsoft's troubles with GitHub are of that nature, not so much the actual git codebase itself.

The interesting thing about the AI revolution is that it's returning us to the times when the number of lines of code written was seen as a major metric for how well your project is progressing. I remain unconvinced this is a good metric to chase.

Minister gives Palantir's NHS platform a clean bill of health

Owned by American grifters....

... is that a misremembered song title that just sprang into my head?

Palantir CEO: 10 percent of the world 'professionally hates us'

Hmmmm

For an organisation that prides itself on analysing data to get accurate statistics, claiming it's only 10% looks extremely embarrassing.

Talking of which, nothing gives "I live in my Mum's basement" quite as much as referring to plush offices staffed by sociopathic kids as the "factory floor". Indeed this whole statement is full of.. something: "On the factory floor side, the demand on the defense industrial base to ramp production and sustainment has been so acute that we have surged resources from our commercial business,"

Sustainment?

See, I'm hating them outside of office hours. They'd have to pay me for the professional hate. And no, they can't afford my "anything to do with Alex Karp" rates.

Vintage chatbot lives in the past like an elderly relative

Re: AI doesn't "think"

The continued determination of otherwise intelligent adults to ascribe thinking to a statistical model is extraordinarily frustrating.

I would phrase that in more modern parlance, but I suspect the post would be censored.

Musk bets Tesla's AI future on Intel node that isn't finished yet

Watch..

He's looking (and sounding - check out the unedited audio of the earnings call, it's painful) more and more like a washed up gambler, resorting to bigger and bigger bets to try and claw back all of the money he's thrown around.

He had a (nominal) trillion dollars to build a world-class car company, and is now in the process of abandoning it for an entirely unproven business model based on entirely non-existent technology that *might* turn up some time, just as FSD has turned up after ten years of promises... oh.

Forget call centers, local energy prices mean Britain's latest offshoring wave is AI projects

Beware the lobbyists

This is a perfect opportunity for lobbyists from every energy provider to spread further FUD around pricing and the cost of sources - so beware before quoting received wisdom around the solutions to a crisis that has been decades in the making. Vested interests are keen to benefit from the current panic.

That said - it shouldn't take the new shiny of AI datacentres to finally understand that high energy prices strangle a nation. Even the most un-sexy industry and smallest homeowner has to pay the bills, and the lack of commitment to research, new technology and larger projects is deeply embedded in our political system. Slogans about being green don't cut it when the actual policy is (and has been for years) inaction and handing over responsibility to unreliable and expensive external suppliers. Add in a completely broken planning system for infrastructure projects and.. well here we are.

I vibe coded a feed reading web app. It was enlightening and uncomfortable

Re: Lowering the Bar

The problem is that people want someone to blame either way - they anthropomorphise the corporation as maliciously setting out to destroy local shopping, when it's an "unintended consequence" of growing a business with a narrow definition of efficiency and value.

So it turns out both consumers and corporations do things that are against our collective best interests. Both go out of their way to justify those actions - big oil by running campaigns, consumers by telling you how they don't have time to shop around and make those ethical decisions that are more expensive.

But it turns out most corporations have a limited shelf life, and people's values shift. Cancel the dozen streaming subscriptions you have, stop buying everything from Amazon, and reject Temu tat - those are all Boomer obsessions that make us miserable. Recognising that is what changes both consumers and corporation behavior. Maybe not overnight, but eventually..

Palantir trial plugs into UK financial watchdog's data trove

Just no.

How clearly do we have to say it?

On security grounds, on digital sovereignty grounds, on cost grounds, on grounds of sanity... there is not one single justification for letting those politically compromised hacks anywhere near our public services.

Elon Musk wants to build 50 times more chips than the world currently produces, using 'new physics'

So....

Straw poll of Register commentariat seems to suggest no-one's buying what Musk is selling.

It took too long to get here, but at least we're beginning to see the end of the acceptance of outright lies coming from him and his cohorts.

US state laws push age checks into the operating system

Embedded tracking

Note that the secondary requirement then has to be that the person using the device or operating system is robustly identified and that identity is communicated to the service they are using.

This may not be in the legislation as written, but to enforce it, we inevitably get to the point where users are personally identified, and that identity is shared with any online system that requests it.

Suddenly we have state mandated tracking of all computer activity.

That'll go well....

UK mobilizes lawyers to keep report on Gatwick 'drone' chaos under wraps

Re: The drones never existed

Whilst a full-on conspiracy is perhaps far-fetched, the essence of the story is pretty close to reality. A number of well known large corporates were heavily lobbying governments at the time to carve out legal air-space for the drone delivery services that they were convinced were going to be the next big thing. As part of that, getting hobbyists out of the air space was an openly stated objective, and there was a great deal of FUD spread around hobby pilots being dangerous, unregulated and unsafe. The campaign was similar to the one that established jay-walking as a crime in America.

The FAA, EASA and CAA were all pushed heavily to 'commercialise' low level airspace for drones, and specifically to regulate potential competitors and individuals out. They did very little to push back.

On top of that, various police and public service agencies saw drones as an exciting new tool, and were very keen to establish their credentials. This resulted in a number of relatively inexperienced public service operators presenting themselves as experts and in the process going to great lengths to disparage the hobbyists that had both more experience and deeper technical knowledge.

Add in the newspapers who are never able to let a good bit of hysteria go to waste, and we had the perfect storm for demonising drones - despite the ever growing evidence that safety was never the earth shattering issue being claimed. We're still in a position where politics and commercial interests are pushing increasingly onerous regulation on operators - and in the UK the lack of support and clarity from the CAA and government has effectively killed any domestic industry.

UK tech hit by double trouble: Fewer foreign techies amid skills squeeze

Note to Employers

You don't HAVE to grapple with AI. It's a tool like any other, and for many businesses has absolutely zero relevance to the work you do, or minimal benefit at best.

You didn't HAVE to grapple with grammar checkers in Word, which for many businesses is pretty much what AI can offer - a tool to automate some communication tasks.

You didn't HAVE to grapple with website chatbots, which many users find next to useless when it comes to actually solving their problems.

Sure, AI may eventually improve things, but early adopters are being bitten hard when it turns out services are over sold, under constrained and cost far more than the benefits they provide.

And as far as the employment stuff is concerned, can we talk honestly about the many challenges facing the workforce and employers at the moment, quite a few of which are the responsibility of the government, which seems to be asleep at the wheel at the moment as businesses and individuals struggle with rising costs, poor infrastructure and lack of support.

Altman: You think AI is wasted energy? Try raising 100 billion humans

In other news

Man jumps shark.

UK council faces data breach claim after mishandling trans complaints

Oh yes, free speech

..of the "I'm just asking questions that I know will offend people, that aren't entirely relevant to the subject at hand and that I'll suddenly act all innocent about when challenged" variety.

You then have to wonder who's PR department is working so hard to take the bickerings of a local council and turn it into national news.

£111M later, frictionless post-Brexit border dream 'brought to early closure'

Re: Found a project consultant

I sell products worldwide.

Direct experience of the way this and previous governments choose to serve only companies like Amazon who can afford to lobby them suggests that there is no interest in helping domestic businesses and understanding the needs of traders. You remember that "they work for you" thing? Sure, if you believe government is only there to keep consultants in business, let's see how far that takes us.

Re: Not Brexit.

Anyone who's traded within the EU knows that cross border trade is still a thing, even if it appears wonderfully seamless from the consumers' point of view. And that's before we look at the wider supply chain.

It's a complete fiction that being within the EU eliminated these needs, and wilful ignorance to pretend that our trading systems are not woefully outdated. Membership of the EU doesn't change that.

That's the point. The problem is nothing to do with your (or my) politics.

Not Brexit.

Can we not conflate the fate of this project with Brexit?

In common with a lot of cross border trade and taxation, this area is still a mess, and still an unnecessary drag particularly on small businesses and individuals. It is a regressive (lack of) policy that benefits global traders like Amazon whilst making domestic providers jump through hoops.

Regardless of Brexit, sorting out the accumulated legacy systems and streamlining the rest would be a huge benefit to domestic businesses. A government that cares about trade and business growth should be making this a priority. Of course, because we get the knee jerk "It's Brexit innit?" stuff from people (on both sides of that particularly toxic debate), neither the current government nor any of those in waiting are going to prioritise work like this.

So we're in the ridiculous position of people cheering on the fact that we're quietly going backwards and yet again demonstrating a complete inability to actually improve government services.

Elon Musk paints exodus of xAI co-founders as 'evolution'

Re: Moonage daydream

That there is NASA funding is the whole point. He needs money, so for a few dollars he'll ditch his "vision" and promise another one. It's a grift.

The astonishing thing is that professional business people believe him, despite year and years of evidence that he's running on empty.

AI spurs employees to work harder, faster, and with fewer breaks, study finds

Hmmm..

New toy that has a built in dopamine loop ("I just need to tweak my prompt a little...") encourages workers to play with new toy.

News at eleven.

Brussels drafts blueprint to spot and swat rogue drones

Sigh

We've had massive consumer adoption of drones for about ten years now. What do the stats tell us?

What they don't show is any significant harm from consumer ownership - despite millions of drones being sold.

The whole idea that an "evil nasty terrorist" will obediently fit a transponder to their drone (and totally ignore the optical control solutions so widely used in Ukraine) is patently ridiculous.

So we've got an imagined problem - consumers aren't harmful, terrorists aren't stupid - being used to drive a massive transfer of funds around the quangos of Europe. The usual people who profit massively from these programmes will rub their hands and take the money, whilst a couple of companies that would have done the actual work anyway will... continue to do the work.

I can't wait for the UK government to introduce a second rate knock-off of this proposal. 10/10 no notes.

Containers, cloud, blockchain, AI – it's all the same old BS, says veteran Red Hatter

React

The whole movement to an 'application' being a bunch of cats fighting in a sack, with weak dependency management, wildly inconsistent project structure and a state that defines any rational analysis or testing.

Though you could argue that a lot of the worst excesses are reactions to holes we've already dug ourselves. React as a response to the crude mashup of JavaScript and HTML, Containers as a reaction to dependency hell of Python-like environments, Agile as a response to unmanageable requirements and estimation demands etc. etc.

SpaceX wants to fill Earth orbit with a million datacenter satellites

It's not about Space, or Sentience...

By now we should all understand that Musk is playing the three cup game, and juggling unprofitable businesses around by wrapping them up in nonsense 'visionary' shells.

SpaceX, through its deep connection with government contracts, is possibly the only reliable money maker in his whole empire (so long as it is tolerated by the American President and other governments using Starlink for strategic communications). Burying X and xAI in there hides the failures. Adding a nonsense vision to replace the previous nonsense vision of cities on Mars is designed to distract from the massive delays (and possible failures) of the Starship program (the Cybertruck to the F9's Model 3).

And then, with a lot of hype (look! money! visions! sentient suns!) we move towards an IPO before anyone questions why xAI and the Optimus programme are in separate companies, and how chaotic the Artemis project has become.

Focus on the cups. Round and round they go. Where has the pea gone? Is it under here? No, silly - Musk has taken it.

Want digital sovereignty? That'll be 1% of your GDP into AI infrastructure please

Good grief

1. Yes, as observed, AI is not sovereignty

2. First solution: wait a couple of years until the AI hype collapses and a bunch of data centres become surprisingly cheap.

3. Second solution: Invest in actual digital sovereignty without Gartner (an American firm) being involved in any way whatsoever.

Note that the "cost" of digital sovereignty is likely to be offset by the savings of not sending all of your money to foreign owned companies, and the benefit of having domestic talent nurtured by a healthy industry that develops solutions more accurately targeted to local needs.

Tesla revenue falls for first time as Musk bets big on robots and autonomy

Re: I knew this day would come

He's moved away from cars because he cannot compete with China - or even the legacy manufacturers. The longer he stays in that business, the more his company looks like any other niche car manufacturer. The danger then is that it would be valued as such.

It turns out neither the cars themselves nor the factories that make them are all that special. Nearly every other manufacturer can do the same, or better. The one competitive advantage he had (no legacy software stack, no legacy tooling) has been lost. He should have been starting on at least two new mainstream models five years go to retain that lead, but instead went down the rabbit hole of the Cybertruck vanity project.. The rest of the world just caught up with him.

American auto manufacturing was lost when the roadmap that ended with the Model Y and the people behind it were abandoned.

So....

The cars show he cannot compete on manufacturing - despite having once claimed to have the greatest knowledge of manufacturing on the planet.

His work in Twitter and on Grok shows he cannot run a software development project - remember all the claims that X would be the "everything app"?

Boston Dynamics are three generations ahead of him in robotics.

Solar City is moribund (remember the fake solar tiles?). The energy division essentially repackages and sells third party batteries. The trucks are... limping along.

He's taken twelve years (to date!) do deliver the "just one year away" self driving promise - showing that he really doesn't understand autonomy. The demonstrations he's shown (always just before financial reports are due) have consistently been shown to have been faked.

And... there is absolutely no defensive moat whatsoever around either the AI or Optimus projects. That's why there are at least twenty Chinese companies that will sell you a humanoid robot right now (that doesn't have autonomy).

Anyone believing his predictions, or that he is unique in his vision is clearly not paying the slightest bit of attention to the reality of the business and leadership behind the over inflated stock bubble.

Musk distracts from struggling car biz with fantastical promise to make 1 million humanoid robots a year

Pertect..

"In my opinion, believing that this will happen any time within decades is pure fantasy thinking." - Perfect for Musk, who has built businesses entirely on fantasy thinking.

'Ralph Wiggum' loop prompts Claude to vibe-clone commercial software for $10 an hour

This is satire...

This is satire... right?

British Army's drone degree program set to take flight

First kill it, then flog the horse

The UK Government went to great lengths to kill the nascent drone industry - against all advice from suppliers and users - over the last ten years.

Knee jerk reactions to media, a complete failure to seriously engage with companies working in the sector, regulatory ping-pong and years of uncertainty have killed off early adopters and innovators, and ensured that companies struggling to exist have fallen behind developments in the rest of the world.

We were asked (hello Baroness Sugg and others) and gave clear answers that could have delivered skills and technology that are suddenly proving vital. The advice was roundly ignored, as the decisions appeared to have been made before the consultation provided a thin veneer of democracy.

And as the Western world suddenly needs drone technology from trusted nations, and sees incredible advancements in Ukraine - the UK is left spending risible sums of money on training after the fact - training that will almost certainly involve using other nation's products and support to deliver a second-hand experience. That's before we look at the loss of opportunity in the commercial and retail sectors.

Amazon CEO Andy Jassy goes wobbly on AI bubble possibility

Re: Rot economy not denied by enshittifier-in-chief

Important to remember that politicians are usually about 18 months behind techies in the hype cycle.

Just as the tech works is beginning to talk openly about a bubble, they're getting excited about this exciting upcoming AI tech.

Open source's new mission: Rebuild a continent's tech stack

And round and round we go...

Repeat after me: clever is not the same as useful.

This is the problem Cambridge has faced for decades - yes there are lots of clever people and lots of smart solutions, but none of that makes a great business. And the key here is that providing this stuff to institutions - whether they're European, American, the UK or anywhere else in the world - is a business, not a technical exercise.

The EU can attempt to regulate this into existence, but so long as there are no service providers with call desks, support sites, 24/7 callout, migration tools, consultants for integration and all the other bells and whistles, there is no reason to move away from platforms that have all of those things. Sure, some council or university might have a vague desire to 'decolonise' it's software, but when it comes to signing a purchase order, the usual suspects win every time because they have the salesman who can promise all of the bells and whistles that aren't software (or even hardware).

I'd argue the way to make this stuff happen is to address the impediments to business, not mandate software. If a new tool can be made and sold across the continent, without additional barriers at every border, without regulatory burdens designed for global corporations yet applied to one-man businesses, without complicated tax and reporting requirements that change with each new country, then there is reason to adopt or write software that can be sold. Innovation comes when there is a customer, and for now America provides the single easiest customer base on the planet, with the least state intervention and least punitive financial environment, so that's where innovation goes.

And the same applies to the UK, that *still* fails to address the business environment that would allow local innovation to flourish.

Open source is a tool, not a end goal in it's own right. We want people to be able to use those tools to achieve goals, not for the sake of using the right tool.

Maker fight! SparkFun cuts ties with Adafruit in harassment dispute

Soap opera

I've read the article twice, followed some of the threads on a number of forums and... still don't know what's going on.

Both companies are small-ish and very dependent on the passionate support of a niche hobbyist community. I've got every sympathy for them as I also sell kits (in a much smaller capacity) and know how precarious things are, and how challenging it is to be heard. Their dependency on each other appears to be a consequence of the difficulty in building sustainable businesses with a small customer base of very opinionated and demanding techies. No-one buys electronic project kits to impress their neighbours.

It seems that the Adafruit team are working hard to put their side of the argument out there, which doesn't speak to innocence or guilt, but does leave me somewhat confused. Neither company seems very well able to deal with criticism or challenges to their deeply held beliefs.

Before buying from either of them in future, I'll be looking carefully for alternatives that aren't likely to drag me into someone else's argument.

UK urged to unplug from US tech giants as digital sovereignty fears grow

Re: Not enough

Sure, the atmosphere has been more convivial, but that doesn't mean we were doing the right thing by handing everything over to our very nice friends.

The turn in politics has made everyone nervous, but the problems have been embedding themselves for years - like the heart attack victim blaming the attempt to run a marathon and not the three decades of eating cake beforehand.

Re: Not enough

...and before anyone gets too excited blaming recent events and world leaders, it should be pointed out that this has been the state since the eighties when we relaxed the rules about corporate ownership, and handed over most of our computer industry to America.

Not enough

This needs to be more than a wish for open source technology. This needs to be a presumption in favour of domestic suppliers, physical infrastructure and home-grown solutions.

Europe has been following these policies for decades - particularly when it comes to foreign ownership of companies - and are at least resisting the heavy lobbying and commercial pressures of US influence. We cannot claim to be so resilient.

The UK doesn't have a meaningful industrial strategy in this area, and has watched as chips, cars, drones, manufacturing, robots and many other areas where we could excel have been 'offshored' to America, China and now India. Without a serious change, we're essentially in a state of managed (or unmanaged!) decline.

Imagine there's no AI. It's easy if you try

Re: Hmmm...

That doesn't really make anything cheaper though. It means sites have to be redesigned, highly customised sites being deeply challenging, and additional provision has to be made in content.

Adding more work is not "cheaper"... it's more work.