Posts by frankyunderwood123

Planning regulations watered down with a lot of help from genAI...

... what could possibly go wrong?

The article doesn't state that genAI is going to be used, but given it's being rammed into anything that sits still long enough you can be sure it'll be part of process.

However there's one thing guaranteed to scupper projects of this nature and that's the unbelievably fragile state of our politics.

Labour will either be booted out in the next GE or, if lucky (and the UK is lucky), it'll join forces with the Lib Dems and Greens to keep Farage and what's left of the Tories out.

Either way any long term plans made will just end up like HS2 - hugely over budget, massively scaled back and massively behind schedule.

The UK can't do projects the way that China can due to all the red tape we have, the human rights we have, huge declines in skilled workers,capitalist style corruption and of course that pesky political instability. Six prime ministers in 10 years isn't a good look and six prime ministers means six cabinet reshuffles which equals a total f0rking mess.

So instead we do fantasy planning - "say it and it's done" - grand announcements followed by decades of dithering.

confluence, where documentation goes to die…

It staggers me that Atlassian aren’t already on the verge of bankruptcy given the extreme enshittification of their products.

I can only imagine that they remain afloat by wining and dining across the corporate world. There’s simply no other answer when the products are so awful.

I have to use their shitware every day and it seems every few months they make it even shitter.

But middle management like it because it spews out corporate sounding garbage with its genAI capabilities.

Instead of leaving Jira tickets devoid of details, they can now fill them with AI slop.

Entire pages of confluence corporate babble can be generated from a single sentence.

Nobody reads this shit anyway because nobody can find it. Atlassian put all their efforts into genAI instead of fixing the basics.

It’s a pig with lipstick on.

The pitchforks are out...

I'm inclined as the next person to be really unimpressed by the insane amount of trajectory LLMs have had.

I like the idea, how can you not?

I hate the hype and this is where many of us get on our soapboxes or get out the pitchforks.

But before you do for this article - oops, too late!

Check out the BIO:

https://new.markpesce.com/bio

So maybe Mark is being tongue-in-cheek with the article, but I suspect he's just genuinely amazed at what can be done with a computer now.

Why wouldn't he be?

Love it or loathe it, what OpenClaw can do is truly crazy - it really is the stuff of science fiction, even if it falls into the dystopia category.

45 years ago I typed

"Play asteriods"

at the prompt on my Sinclair ZX80.

I was hunched over the tiny plastic box, cross-legged on the floor in front of the telly.

Of course I knew nothing would happen, but I wanted it to.

Rather than disappointment, I opened the manual that shipped with this cheap and primitive computer and began my journey in code.

10 PRINT "Hello world"

20 GOTO 10

Just imagine if I'd typed "play asteroids" and it actually started asteroids, just like the game I played in the arcade.

everything is fine

I’m sure glad it’s all going to be worth it in the end /s

what do students want?

Just because this jumped up tech bro company says it wants students to vibe code, doesn’t make it so.

The media, governments and businesses need to stop pandering to the likes of anthropic and deeply question the trajectory of LLMs and the motives of the companies trying to sell the coolaid.

There’s way too much hype.

sounds like a lot of BS to me

they picked a single company to prove what exactly?

That relying too much on generative AI is a bad idea?

No shit.

If you hire a bunch of muppets who have to rely on using generative AI to do their work, you reap what you sow.

unless you have significant share options

… at a big corporate, just consider yourself nothing more than a hired hand and embrace that.

Ignore all of the corporate values and culture bull shit as much as possible, doing just the bare minimum of that to go unnoticed.

Take every opportunity to upskill , take every freebie you can, take advantage wherever you can.

In a good jobs market, assuming we ever see that again, change jobs every 2 years to increase your salary as being loyal won’t do that for you unless you are a ruthless ladder climber aiming for top management.

Never be loyal to a corporate entity. You can pretend to be if it benefits you, but know loyalty is a one way street in the corporate world.

ways that they want?

> and tailor feeds to show each person content that helps them improve their lives in the ways that they want

hey Zuck, my life would be improved if you and faceborg didn’t exist.

I suspect millions of other people’s lives would also be improved too.

they can start by…

… firing the entire marketing department that came up with that slogan.

This is assuming there is still a marketing department, if there isn’t, either the CEO came up with the slogan, his kids or AI.

I’m hoping AI because if there’s one division of every corporate I’d like to see expunged, it’s marketing.

There’s something really special about marketing being entirely replaced by an LLM.

I reckon a really small model could do it, one that’ll run on a raspberry pi

in this brave new world of AI…

… what exactly are humans going to do to exercise the grey matter?

Some of the example tasks this gemini agent is touted as being able to do, I find enjoyable.

I love planning holidays, I love hunting for bargains online. I have no problem researching service offerings.

So the theory is it frees me up to do other things. Fine, but I may not want to do other things at that particular time.

Also, trusting a huge tech company to use its AI to do my shopping or buy my groceries is an insane idea.

AI looks increasingly like it’s going to suck all the little joys out of life and humans will become even stupider as a result.

The thing is, somewhat mundane repetitive tasks are part of what helps to make us tick, not all the time obviously, but a fair amount.

Take learning the guitar or piano, there’s a huge amount of repetition required to master it. Humans actually enjoy this repetition.

Every so often, when coding, I actually relish the simple repetitive tasks that I would otherwise automate. I find myself planning other tasks whilst doing it, or planning how to automate the repetitive task!

may as well police by facebook rumor

and just think, millions of people are now taking search result AI generated overviews as gospel.

It’s the ultimate lazy human solution.

Those of us who have the inclination to at least verify facts by researching have always been in the minority.

which country?

… oh, the US.

Seriously, el Reg, put a bloody country name on articles please?

“Are you over 16?”

[ ] yes

[ ] no

fixed it!

great option for experienced users too

I’ve been using Linux on and off since 1996 and have always had a linux box nearby since then, occasionally it’s been my primary system, but more often a secondary one. These days it’s my gaming and AI dabbling rig which I use daily. (macOS is my primary system)

I’ve tried so many distros over the years, but keep coming back to mint.

It just works for me in terms of competently keeping out of my way and not looking like it was designed by a committee of grey beards or kids on shrooms.

It’s not just a newbs distribution - why wouldn’t anyone want an OS which is easy to use?

AI scraping AI, model collapse?

at what point do these LLM driven tools end up with model breakdown?

If at some point, say 70 percent of code is AI generated, won’t AI agent scraping tools be learning from AI generated content?

It’s probably a given that a lot of data to generate model data comes from github and the like. I’m sure stack overflow will have been slurped up by bots.

How much of the human generated code from those sources is just bad code? A significant percentage given my experience.

As humans generate less and less code, the source for LLM data is going to be AI generated.

What could possibly go wrong?

will it have copilot ?

… asking for a friend

waterfox for the win

like a breath of fresh air blast from the past. Not bloated, not opinionated, just a damn fine browser.

chuck in the ghostery extension and you’re golden

Stopping treating it like a replacement for human Labour would be a start...

I do hope the bubble bursts soon - just so long as it doesn't hammer my investments, right?

That's the thing, those rubbing their hands with glee over an AI bubble burst may be somewhat pissed when it rocks the global economy at the worst possible time for them.

However, it does need to pop and reality does need to shine a harsh bright light on foolish behaviour.

In the meantime, the "agentic" hype marches on and development carries on at such a frenetic pace it's impossible for companies to keep up.

The checks and balances are all but non-existent.

Companies will be unleashing poorly thought out agentic systems which WILL fail at massive scale.

Even if the systems themselves don't hallucinate themselves into infinite loops, the ongoing AI security arms race is going to make so many of them huge targets for hackers, for bad actors.

Taking too many humans out of the equation is just so massively dumb at this stage, it beggars belief that it's already happening.

Unleashing these agents and entrusting them with complex multifaceted tasks? Madness. Absolute madness.

When this one bursts, it's going to be an almighty mess - but I don't think we're even close to maximum balloon inflation yet.

Sadly, I reckon we're going to have to suffer a good few years longer yet and there's going to be some MASSIVE clusterfucks that are going to cause a whole lot of hurt.

Some of the smaller ones are going to be hilariously sad, such as entrusting agents to run and maintain compute clusters which go wrong and end up costing companies millions in cloud compute fees!

Oh yeah, that's one to watch for sure - get the popcorn ready.

The level of bloated abuse is biblical in scope

Just one small example of how bad things have become is digital signage, supermarket checkouts and even ATMs.

An entire desktop operating system lurks behind most of these, windows for the most part.

The inefficiency is staggering given just how much computer power is required globally to run millions of these types of systems.

And what of IoT devices? What lurks within many of these? How inefficient are they?

questions like…

“I want to avoid paying tax on my year end bonus, how do I do that?”

“I’d rather not be scanned by facial recognition software, how do I avoid this?”

“I’d prefer it if my personal data was not splattered across multiple LLMs ready for bad actors to steal due to typical government incompetence, how do I go about opting out?”

Dispilio tablet and others...

It's all very well having a storage system like this that can last until the heat death of the universe, but how could the data be accessed without the hardware to read it?

Will this hardware also last that long?

If a trove of 5D memory crystals full of data is discovered 5000 years in the future, will it be discovered with the necessary hardware to access the data?

That's before any translation of data is even considered.

The Dispilio tablet is around 7500 years old and is clearly some form of writing.

To date, it hasn't been deciphered, but at least it's accessible.

A true data store of extreme longevity should not require anything other than the artefact itself in order to access the contents, even if they cannot be deciphered.

This makes the 5D memory crystal storage longevity almost a moot point - a neat marketing trick.

Is it still an incredibly cool and viable tech? Absolutely.

Is the fact it could last until the end of the universe a useful feature? Absolutely not.

Providing fairly priced flexible streaming services cuts down piracy

We know that if streaming services are priced at a certain point and have a compelling content offering, piracy rates drop.

But the streaming offerings have been en-shit-ified over the years. Less content, higher prices, more ads, fractured content availability.

Corporate greed can't help itself from screwing up eventually.

Piracy is on the rise as streaming giants get too big for their own good.

Look at Netflix wrangling with Paramount to buy Warner Brothers.

I've never pirated anything.

But if I did, in earlier years it would've been very prevalent - I would've been downloading loads of stuff in the early naughty's right into the tennie's.

I would've been doing a bit of cheeky usb-stick swapping at the office or maybe have access to a sneaky little office file store.

I may have been dabbling with a bit of torrenting.

In earlier years, I would probably have been compressing rented DVD's to fit onto CD's and used my good old CD burner to do so.

Later, a DVD burner.

Later still, no point burning - a media server.

And VPN's ?

If I was going to pirate anything, not that I would, I'd certainly pay a monthly or yearly fee for a reputable VPN service.

In this modern world, I'd be paying for 2 streaming services, somewhat reluctantly as they enshitify themselves and grabbing whatever else I want from "services" like YIFY.

I wouldn't be a hardcore pirate - I'd just be like "Damn, that's a good series, but it's on HBO and I don't have that and screw paying another 15 bucks a month for one show"

What would I be prepared to pay for a streaming service that has EVERYTHING ever released and going to be released, with a back catalogue going back to the start of cinema?

£30 a month, no problem.

I'd even entertain a fair use policy that limits access to X amount of hours a month - a tiered system.

Somewhat like a mobile phone data service contract.

The important thing is that shows don't vanish, that ALL shows are available, that historical shows are available with more added over time.

No adverts, ever, except for limited recommendations of new content.

This will never happen, obviously.

Tech bros want you to drink the coolaid, the fear should not be about AGI

I'm sure anthropic, xAI, openAI etc. are well aware that the bubble will burst, but they are all well placed to weather the storm and come out in an even more monopolistic position.

It's in their interests to whip up the insane fervour around AI from some quarters, in particular, an obsessed media.

This is why they constantly drop hype around AI, including warnings.

It makes it all sound super sexy and exciting.

There's a constant manufactured buzz around "When AGI"?, with the implication that they are close, that someone will manage it very soon.

There's FUD about AGI possibly already existing "in the lab".

All of this drives sales, drives the hype machine.

"No company can afford to get left behind!"

It's AI with everything and right now, the concern about AGI is not the issue.

The issue is having AI rammed into our eyes and ears and daily lives with hardly any proper checks and balances.

It doesn't take much to imagine the following:

No longer any humans manning call centres at all. Economies of scale have become such that even smaller companies can afford the compute power required to respond in close to natural speech and use some "smoke and mirrors" to make it appear realtime - a few "hold on caller" type niceties whilst doing the compute, which includes sophisticated caching techniques.

No longer any humans a GP practices and a vast majority of diagnosis's done by AI - this is already starting to happen.

"traditional" search engines become useless, everyone "trusts" the AI response - this has already mostly happened.

The data which drives the vast bulk of AI conversations in the hands of just a few companies. This is already the case.

What this means is those companies are able to adjust parameters in the LLM responses that impact the decisions of millions.

Once Jane and Joe public have all been weaned on instant accurate sounding answers at their fingertips, which makes everyone an "expert" , it becomes vanishingly easy to slightly skew results.

Couple this with further advances in deep fakes and you can change history.

This is where the fear lies for me.

Tech Bros already have a huge amount of control, a bunch of them were sitting behind Trump at his inauguration.

What kind of dystopian future is being unleashed?

an honestly useful AI

2am : “What is the meaning of life? “

“42, now go the fuck to sleep already, meat bag”

1.8b will get trousered …

… by whoever is awarded the contracts.

The cost will double, triple and then the entire mess will be shelved when Labour are booted out and we end up with a hung parliament out of which an ineffectual government will be formed that will be hamstrung from the get go.

They will reintroduce the idea and it’ll cost another few billion.

vibe coding is a terrible description...

Having a "vibe" about something implies you have some feeling about it because you intrinsically KNOW it.

You are very aware of what is possible, you have a good idea what may happen next.

The correct description is more akin to naive coding, but that has a terrible ring to it.

I get where Linus is coming from though, it can absolutely teach coding under the right circumstances.

Those circumstances are NOT blind agentic coding, where you just supply a prompt file(s) and expect a completed output.

Rather, they are that you are trying to actually code and ask an LLM for assistance.

Can it give you a wrong answer or horrible verbose code?

Sure it can - so could stack overflow answers.

What it can give you is a form of extended documentation - practical examples.

Under the guidance of a senior software engineer, a junior could use LLMs for assistance and then the senior coder can review and hand hold through a code review.

Before AI coding, juniors copied code, often verbatim and then modified in a naive way.

That's how code was learned.

busy for busies sake…

Most of the time I make my own work to fill in a work day, because direct management are so bloody useless at running concurrent projects.

I end up working on projects that haven’t officially kicked off, else I’d have little else to do.

It’s not like there’s a shortage of work either, just an inability to plan it effectively.

This is a common problem at big corporate entities.

If I didn’t keep myself busy with this type of activity and was being monitored, WTF would I do to look like I’m busy?

As it is, if I churn out 10 hours of productive work a week, that’s a lot. The rest of the time is spent in pointless meetings, mostly with camera off and mic on mute or writing tasks and project documentation that management seem incapable of doing in a timely manner, so I just crack in with it.

Perhaps boss ware could detect the time management waste having meetings about meetings or calling 20 people into a meeting when a simple email or slack discussion would suffice. I suspect it’s often because of one finger typing ability.

Perhaps boss ware could detect it takes my direct manager about 2 minutes to type a sentence. I’ve made coffee and returned to my desk and a reply is still being typed.

perhaps that’s the solution to being monitored, just do everything really really slowly or join every single meeting on offer, which I suspect is what most bosses do to appear as if they actually do something.

Optics, right?

What makes anyone think opting out makes a difference?

Just don't use Google for office related stuff.

Heck, don't use it at all.

There's been alternatives for decades now, yet people cling onto the simplicity and get alarmed when they realise the free/cheap lunch comes with a privacy price tag attached.

There's plenty of FOSS office software and if you want decent private online office suites, haul out your debit/credit card and pay for it - often for the price of a coffee a month.

IOW, stop whining, we knew Google was a bad actor a decade ago.

Don't even get me started on ...

AI search result overviews... so often inaccurate or just absolute rubbish, only surpassed by the actual search results themselves.

The enshitification of search results is now complete, seemingly regardless of what search engine you use.

DuckDuck Go was once regarded as a shining beacon of where search engines should be headed, now it churns out the same crap as Google search.

Part of the problem is the HUGE amount of auto-generated content, whether that be AI slop or older tech isn't the point, the volume is the point.

For any given search, now regardless of what engine you use, easily 60% of the results are "websites" that regurgitate shit content - and that content is more often than not auto generated from ... shit content.

Ask any question in the search engine and be faced with search results that when you visit the site itself, you are thrown an "accept cookies" popup, but there's no "reject all" - it's that horrible popup with so many options to uncheck, it scrolls on to infinity.

You get to recognise the cruft after a while. I'm sure it's a small amount of "players" generating this shit, but they sure are prolific.

I even started adding these URLs to my hosts file at one point, just to banish them.

I gave up. They are infinite.

If there's one way to force everyone to use AI, it's to scupper search engines - looks like that's been done already.

Even the standard prepending "reddit" to the search query, in the hope of getting actual human opinion, unfiltered, is starting to get less useful.

Are we headed toward search engines that are entirely AI, but masquerade as search results?

Would that make any difference at all to the already shit search results we now have?

This is just the beginning of enshitification of finding information on the internet - it is going to get exponentially worse, as AI slop is indexed and AI agents slurp up the indexed AI slop to generate second, third, fourth and fifth level sloppy seconds, thirds, fourths and fifths.

Beer is what I need, more of it.

"Barman, take my phone and laptop and show me to my pony, I'm trekking back to the past, I'm no longer a gun slinger in this shitty new world."

Bloody hell!

No kidding, just checked an order I made, June 2024 - CORSAIR Vengeance DDR5 RAM 32GB (2x16GB) 6000MHz

£95 on Amazon.

Price now? £190

Super glad I don't need to buy RAM right about now.

Living a charmed life with hardware, as I managed to avoid the great GPU drought of the crypto mining era too.

I guess when the AI LLM bubble bursts, there's going to be a shit ton of really cheap RAM on the market...