Posts by Michael Strorm

Welcome to America - now show us your last five years of social media posts

TREE(1) = 1, TREE(2) = 3, TREE(3) = "Boy, that escalated quickly"

Well, yeah, but it's harder to contrive a joke around "misunderstanding" someone's punctuation with that one. ;-)

AI superintelligence is a Silicon Valley fantasy, Ai2 researcher says

> "As soon as they started trying, RAM makers could increase production to their capacity and undercut them. "

This relies upon the flawed assumption that RAM manufacturers are able to respond to demand in an idealised, perfect textbook manner. But the RAM industry has never worked like that.

We're already clearly at the limit of existing production capacity, so the only way to manufacture significantly more is to build new facilities.

First problem is that this would certainly *not* happen overnight. Second problem is that such plants are expensive- measured in the billions- so you need to know you're going to make your money back.

That's okay *if* you know you're going to cover your costs selling RAM at such-and-such a price.

But if- as many suspect- we're in a bubble likely to burst sooner or later, then demand from the AI industry will almost certainly collapse, likely resulting in a glut of RAM and a collapse in the market price.

So it's likely that RAM manufacturers don't want to risk building new expensive capacity only for it to come online just as the shit hits the fan and everything is in the gutter.

As I suggested elsewhere then, it's likely that OpenAI *could* exploit that reluctance of the suppliers to respond to demand, but it would be a very high risk strategy for them, likely ending in financial disaster if it went wrong or they misjudged it. Also, I don't think they'd want that distraction at a time they're already paranoid about maintaining their hold on the AI market.

> "Those companies have already been paid or entered into contracts, so they don't have an incentive not to."

Such a contract, no matter how watertight, is only as good as the other company's ability to pay what was agreed. And the problem here is that the company is OpenAI, not (e.g.) Microsoft.

OpenAI is a company still spending *way* more cash than it's making and reliant upon continued investment. If the bubble bursts tomorrow and investors abruptly cut that off, where do you think OpenAI is going to get the money to meet those eyewateringly expensive commitments?

I would assume that Samsung and SK Hynix have already factored this in.

It's *possible* they could corner the RAM market, but a very high risk strategy with many factors that could go wrong very quickly.

It's unlikely given they're paranoid about losing their lead in the AI market and not about to risk taking their eye off the ball with that. As I said, they *might* do it to hit their rivals, though.

But if they did... someone else here said that manufacturers could increase production. But that would require more capacity, and the manufacturers have to bet on the existing AI bubble/demand not collapsing and leaving them with a glut of capacity and the collapsed RAM price not covering their investment costs.

Maybe they OpenAI would bet on that being the case, but they'd still have to be able to continue paying for the RAM they were buying, even if the contract was watertight.

And conversely, any lucrative contracts for the manufacturers to supply RAM are worthless if OpenAI collapses which, as a company dependent on continued funding, would likely happen quickly if that was cut off by investors no longer confident in its chances.

Indeed. We're long past the time where improvements in the underlying technology were a major factor- the current Gen AI boom is now almost entirely being fuelled simply by spending exponentially increasing- and borderline unbelievable- amounts of money on hardware. It's already getting to ludicrous levels.

Did you know that- according to some reports- the deal Sam Altman signed with Samsung and SK Hynix at the start of October was for potentially 40% of the entire world production of DRAM?

That's not just ludicrous in itself, the amount of money needed to fund that must be even more so. I missed the significance of it at the time, but it would explain *exactly* why RAM has gone up so horribly in price. (*)

From what I've read, OpenAI isn't even using all the RAM yet, is getting it in the form of RAW wafers(!) and it's not entirely clear how much this is driven by need and how much it's a paranoid attempt to retain a lead by choking off the supply of RAM their competitors require.

That being the case would make the amount of money such a huge purchase would involve even more unbelievable and a massive risk. Even if OpenAI "wins" what they seem to think is a "winner takes all" scenario, they'd *still* need to make enough profit over the long term to get that ludicrous investment back- or at least continue to convince the market that they can do so.

There's no clear sign of that yet, and as soon as the market stops believing in it, it could go very wrong, very fast. The whole current situation smacks of an incredibly risky bubble that could- and likely will- go wrong very rapidly and very badly. Ditto the slightest sign of trouble that OpenAI might not have trouble paying for the hardware they've committed to.

When this bubble bursts- and I'm going to say when, not if- the problem is that it's going to be very bad for everyone else too, particularly in the technology sector.

(*) It's only incredible luck that, after repeatedly putting it off, I finally got round to ordering the parts for my new PC- including RAM- on 12 October, just before the shit *really* hit the fan. I paid £110 for 2 x 16GB Corsair RAM modules. Exactly one month later (12 November), the price for the *exact* same SKU was £174. Six days later (18 Nov) it was £234. Another six days later (24 Nov) it was £331.49. Today that exact same RAM kit, from the same seller (Scan), costs £350.

X shuts down European Commission ad account after €120M fine announcement

> "Doesn't X support freedom of speech?"

Claimed to? Frequently and bombastically. Ever actually did so? Never.

Any self-aggrandising manchild can spout shite about being a "free speech absolutist" or have one of his mouthpieces make the hollow claim that "X believes everyone should have an equal voice on our platform".

But when virtually the first things he does after acquiring Xitter demonstrate the exact opposite- kicking off or penalising anyone he disagrees or gets upset with and- conversely- boosting his own posts and those of the far-right he sympathises with- such claims can be dismissed with the contempt they deserve.

Musk is the epitome of the self-proclaimed-"Libertarian" hypocrite who only cares about the free speech they "champion" when it suits *them*. In other words, they're nothing of the sort.

"X believes everyone should have an equal voice on our platform"

But some (*) are more equal than others,(**) right?

(*) Elon's and those of his far-right mates that he boosts.

(**) Anyone who gets penalised or kicked off Xitter for saying anything this self-proclaimed "free speech absolutist" doesn't like.

Another open source project dies of neglect, leaving thousands scrambling

Re: Another example that open souice is a flawed model...

> "Aw, that was a bad show of lacking knowledge."

No, it wasn't, and you're missing the point.

I'm well aware that NT- which all mainstream versions of Windows from XP onwards have been based on- was written (mostly) from scratch in the early 90s and ditched the antique DOS-based underpinnings.

But it was still intended as a mostly-compatible replacement for- and continuation of- the Windows/DOS line and still retains inherited aspects of its design.

OP was trying to criticise Linux on the basis that *it* was derived from the original early-1970s version of Unix, even though it shares no code with that and has come a long way in terms of design.

I was pointing out their hypocrisy, via their (repeated) failure to acknowledge that- by the same flawed argument- we can hang any modern version of Windows on the basis that *its* origins were the even more basic CP/M operating system of a similar vintage.

Re: Another example that open souice is a flawed model...

> "MS DOS non longer exists, luckily."

Its descendant does, in the form of Microsoft Windows. Or are we being selective in which OS's origins we get to criticise?

> "Unix, unluckliy., does"

Not as a carbon copy of the "Unix" of the 1970s, it doesn't. And, technically, Linux isn't "Unix" because it was reimplemented from scratch, never incorporated code from the original Unix and never acquired the "Unix" trademark (which only "exists" as a licensable brand these days anyway).

Much like how MS-DOS was (supposedly) not based on CP/M code- and even that is questionable- but was a blatant workalike/knockoff regardless.

So, again, are we applying double standards here?

> "And again, it's funny Chrome is the new IE but noone complains."

We weren't discussing Google and Chrome, but since you bring it up- yes, Chrome is the new IE (in terms of market dominance at least) and plenty of people- myself included- have made *that exact observation* and have complained about it and its effects.

> "Just you can't see the cage that was built around you, thanks to FOSS...."

Licensing something under a FOSS license imbues it with magic powers that render its users critical faculties null and void? I didn't realise the GPL was quite *that* clever!

Re: Another example that open souice is a flawed model...

You're not just drinking the Kool-Aid, you're drowing in a swimming pool of it.

> "Linux itslef is a bad copy of an outdated OS"

You mean like MS-DOS, which was little more than a 16-bit workalike/ripoff of CP/M- itself an operating system itself inspired by Unix, but with far more primitive architecture designed around the limitations of incredibly basic mid-1970s 8-bit microcomputers?

The same MS-DOS that needlessly propagated the limitations of that 8-bit, several-years-old OS on far more powerful hardware capable of running better alternatives, which required messy, Heath Robinson-esque retrofitted workarounds to take advantage of newer hardware and forced users- via MS's near-monopoly- to pointlessly suffer from that early decision for around two decades through- and including- all mainstream versions of Windows prior to XP?

Something that happened at a time when there was no semblance of a viable open source competitor?

> "It's funny that open source allowed the raise of large momopolistic compaines like Google."

You mean like Microsoft was- and still is to a significant extent- in the desktop market with MS-DOS and Windows. Oh hang on, I already mentioned that one, didn't I?

Hey, do you remember when MS exploited its near-monopoly and privileged position to kill off competition in the browser market and how Internet Explorer stagnated- and held back web design- for the better part of a decade until Firefox came along and provided some worthwhile competition.

By which point IE was still established and the need to retain compatibility with existing installations and to deal with its shitty, nonstandard design held back web design for many more years anyway?

And how those fuckers at MS later whined about the fact that they were still having to retain browser compatibility with older sites and systems. You know, systems designed around the shitty, nonstandard aspects of Internet Explorer that *they* used their monopoly at the time to have everyone use?

But do go on to explain how the nonexistent (at the time) open source competition was somehow to blame for MS-DOS, Windows and Internet Explorer.

Hang on, the ambulance just arrived to help pump the Kool-Aid out of your lungs...

Re: Another example that open souice is a flawed model...

You're projecting your own zealotry regarding strawman stereotype "Stallman worshippers" onto those disagreeing with you. You were given some quite reasinably-argued rebuttals.

Open source has it's problems too, but saying "Maybe we'll see again more competition" is delusional blame-shifting considering that the proprietary Microsoft had an exploitative near-monopoly on the desktop market for decades and still does to a large extent.

Re: WTF?

No you didn't.

Re: Another example that open souice is a flawed model...

Going by this similarly-inclined recent comment, OP is either a blinkered anti-open-source zealot and/or a troll.

Re: Womp Womp

> "Hey, I'm not going to add to an existing repo, I'm going to invent my own version."

With blackjack and hookers?

Re: WTF?

> "some people disagree, but people disagree about anything"

No we don't.

Trump says Nvidia can sell H200s to China – if Washington gets a 25 percent cut

Exactly. If the ban ever genuinely had anything to do with the "security" excuse used in the first place, how does that suddenly change and not become a problem just because Nvidia allegedly agreed to pay a "$25%" [sic] tariff/bribe to the US?

Then again, even to question it in that way is to dignify it with the underlying assumption that any of this- or anything else the Trump regime does- was ever done in good faith or would even bother trying to justify itself in those terms. Or that they'd even bother trying to explain or excuse the fact that they'd apparently reversed position on something they were making a big fuss about just a few months ago.

Whatever legitimate places AI has, inside an OS ain't one

Re: The GUI subsystem is part of the OS, and rightly so.

Says the person defending Windows, an OS that ultimately originated- via MS-DOS- as an unlicensed carbon copy of CP/M, a mid-1970s OS that was inspired by Unix, but even more cut down and restricted because it had to run on the earliest, incredibly basic 8-bit microcomputers.

An OS that was already out of date and unnecessarily primitive when it was ripped off to become the OS for the 16-bit IBM PC, a computer which was capable of running something far more modern, and whose monopoly held back desktop computing for decades.

And despite you elsewhere blaming open source for killing off competition and driving down quality, all this- Microsoft achieving a monopoly via MS-DOS, a clunky, dated, substandard excuse for an OS- happened long before open source was really a thing, let alone a viable alternative.

But keep blaming Richard Stallman- or whoever- and tell us who the one drinking the Kool-Aid is.

UK SAP users say they're baffled by Business Suite reboot licensing maze

Was going to say something similar- you're not meant to understand it, you're meant not to and to be "persuaded" to buy way more than you need on the off-chance you might get it wrong.

It's the Oracle model.

Speccy clone storms back for Christmas without a shred of Sinclair code

Re: squishy "dead flesh" type keyboard

> "In AU I don't recall the ZX80 etc being too thick on the ground"

As far as I'm aware, Sinclair wasn't even able to meet demand for the ZX80 in the UK, so I'm not sure whether they bothered attempting to officially distribute it in Australia. (Apparently it *was* sold in the United States, though).

Though interestingly, the MicroAce- a clone of the ZX80- *was* apparently sold in Australia via the Dick Smith electronics chain.

Sinclair himself publicly commented that the ZX80 would be a "nine month wonder" and given that he would already have known they were working on a cost-reduced and improved version that was likely to be an even bigger hit (i.e. the ZX81), I suspect it probably made sense to get that ready first, instead of trying to market the soon-to-be-replaced ZX80 elsewhere.

Re: 1986, not 96

Also, more interesting commentary on the subject here.

Re: 1986, not 96

I've posted this before, but The Register's own 30th anniversary article for the QL touched upon this:-

"[Chief Design Engineer David Karlin] chose the 68008 [which] was built for backwards compatibility, so it featured an external 8-bit data bus and 20-bit addressing. The 68000 [had] a 16-bit data bus and 24-bit addressing. [..]

“I don’t know how robust the decision was,” David admits now, “but it seemed fairly clear that the the 68000 series would be a great platform [..] The problem with the 68000 was basically a pin-count issue. Motorola was pricing it gigantically higher than the 68008, double or treble the price. It was sufficiently high I didn’t even argue about it.”

Rivals’ use of the full 68000 would later come as something of a surprise. If Sinclair couldn’t afford the 68000, how could they? Today, David blames Sinclair’s negotiating skills, not just for the CPU but for a whole variety of logic chips and add-ons: “I question how good we were at purchasing, because people like Amstrad, certainly the Japanese, certainly Apple, who did not have gigantically higher volumes than us at the time, got massively lower prices.”

During 1983, it has been claimed, Motorola cut the price of the 68000 to below what Sinclair had agreed to pay for the 68008. Renegotiating the purchase contract might not have been costly, but adding in the architecture the full 68000 required [..] would have been, so it was decided to stick with the 68008. [..] It’s easy to say Sinclair would have well been better off going with the 68000 after all, but only with the benefit of hindsight."

> "[ Harlequin uses ] off-the-shelf components which Sinclair could have used at the time -- but then the original Spectrum would have cost [..](circa £400-£500 1982 pounds.)"

Interestingly, that's the main difference between the ZX80 and ZX81 at a hardware level. The 80 used off-the-shelf chips.

The 81 circuitry was functionally almost identical- with minor changes to support the continuous display- but reduced costs by reimplementing the design to replace 18 of the 21 chips with a single custom ULA.

The ZX81 also had a new, improved ROM, but one could upgrade existing 80s by installing it, such were the similarities.

Re: squishy "dead flesh" type keyboard

The most common version of the BBC keyboard I saw was very good, to the extent it helped define what I wanted in a mechanical PC keyboard 20+ years later.

The nonstandard layout (by modern standards) would probably put me off nowadays, but what was back then?

I agree that even other mechanical (or semi-mechanical) keyboards from back then often don't hold up today. I was never that impressed by the C64's (too *much* travel), but my Atari 800XL's wasn't great when I reused it years later. (*)

Conversely, while some modern membrane keyboards can feel awful, I've come across some surprisingly good ones, including a dirt-cheap "Octigen" that is sadly out of production.

Also, low-travel doesn't necessarily mean bad- my old Compaq laptop's is rather nice. I draw the line at may-as-well-be-a-ZX81-for-all-the-movement ones designed for tablets though.... bleurgh!

(*) To be fair, this varied- I found a site with five or six variants listed, all with different caps and switches and I identified my old one as the lowest-rated version.

Re: 1986, not 96

Pedant alert- Amstrad didn't buy Sinclair Research itself, just the existing computer lines and the rights to the 'Sinclair' brand.

SR continued in increasingly diminished form, its most notable subsequent release being the mildly successful Z88 portable via the Cambridge Computers subsidiary.

It sporadically released a few other Sinclair inventions, such as the unsuccessful Zike electric bike and a portable earphone radio, but it never came close to its former peak and shrunk until it had just one employee (Sinclair himself).

As fast as I know, it was still technically in existence- though effectively dormant- when Sinclair died.

Even if the code is a license-free compatible workalike, wouldn't the case design still be covered by copyright, whether or not they replaced "Sinclair" with "Retro"?

And I assume that the original "The C64" and "The A500" omitted the Commodore and Amiga branding because Commodore's IP and rights (brands, code, etc) have been split and are now a clusterfsck of owners and sublicensees that have to be dealt with and licensed separately, not because they didn't have *any* licensing rights.

So is that really the deal in *this* case?

Soup king Campbell’s parts ways with IT VP after ‘3D-printed chicken’ remarks

Re: I have one question

> "You shouldn't have to have some kind of religious passion for your employer's product. "

Quite the contrary. Arguably *the* most obnoxiously overused cliché of sales blurb and corporate PR during the past fifteen or so years has been the word "passionate" and the need for *every fucking company out there* to pretend that they and their employees are "passionate" about what they do.

This gushing, insincere, spray-on faux enthusiasm is everywhere. If you (e.g.) search Tesco's website for "passionate about", you'll come across page upon page of products from countless companies telling us how they're "passionate about" the most mundane, mass produced corporate junk.

And remember that Tesco want employees who are "passionate about" helping customers as part of their part time, minimum wage job.

Obviously Tesco aren't the only offenders here, quite the opposite- this shit is everywhere.

The idea that one should be "passionate about" a job has really become common in recent years. It's not propaganda, it's just a convenient coincidence that this also distracts from and delegitimises the idea that one might have the audacity to be doing a job for the money and expecting to be paid more when they should be doing it as a labour of love.

SK hynix wants you to bond with HBM, so it coated corn in banana chocolate

From the article:-

SK hynix and 7-Eleven will run a prize drawing event [with prizes including] ten don of pure gold. "Don," for those unfamiliar, is a traditional Korean unit of gold weight equivalent to 3.75 grams.

This implies that it's for the South Korean market.

Article describes 7-11 as a "US retailer", but given that it's ultimately Japanese owned and the US branch is -AFAICT- a subsidiary of that, it's unlikely that this has much to do with the US chain.

Alien Spacers

The Hynix-based RAM modules I bought from Scan around six weeks ago now cost three times as much.

Perhaps these crunchy semiconductor-shaped (*) corn snacks are meant as a substitute for those who can't afford the real thing?

(*) Though going by Outer Spacers, Monster Munch et al, they'll be blobs that only faintly resemble what they're meant to be?

HP to sack up to six thousand staff under AI adoption plan, fresh round of cost-cutting

Re: HP warns memory price explosion means PCs may have less RAM, or use low-cost parts???

No, that means they'll be cutting the current underspecified models further.

Expect a massive 512 megabytes in your next computer from them.

Unless there's a cartel operating, I assume it's being driven by supply and demand and, like it or not, in the absence of price controls companies are always going to maximise what they can charge.

I'd also have assumed that it was being driven by the current AI boom, but that's been going on for years now, and even if it's been ramping up in the past few months, it doesn't explain the abrupt price increase of DDR5 in the past few weeks.

Is it an indirect consequence of recent moves to reduce or discontinue production of DDR4 which pushed prices of *that* up? I can't see why, since anyone trying to snap up the remaining supplies of DDR4 for older equipment isn't going to be able to use DDR5 instead- that's kind of the point, after all.

As for Corsair DDR5 I bought, I checked Scan's price- via the Wayback Machine- from the end of last year, which was £104, i.e. slightly lower than it was when I bought it. So that implied that it had gone *up* very slightly during a period you'd otherwise have expected it to have fallen as DDR5 lost its "new" technology premium and became the new mainstream. This suggests that AI-driven demand had already had an effect. At any rate, I didn't think £110 for 32GB seemed *that* cheap at the time, but it looks like an incredible bargain now.

RAM is one of those things which has always been notorious for price volatility. (*) I'm guessing because the technology is constantly changing, the level of investment required is high and because the need continuous production that forces can't always match with spikes and slumps in demand.

Presumably, even if the manufacturers wanted to take advantage of an increase in demand like this- and, ironically, flatten out the spike, which is how the "invisible hand" of the market is *supposed* to work- they wouldn't be able to do so overnight.

And, more importantly, another reason they'd be wary about rushing in to do so is that they're as paranoid as everyone else that the AI bubble is about to burst, demand from that sector would collapse and they'd be left with a huge excess of stock and/or production capacity which would no doubt exacerbate the resultant price crash.

(*) This article covers an earlier spike in 1988 and notes that:-

> "By July of 1988, the increasing RAM prices, which had slowly crept up in the early parts of the year, surged. A megabyte of RAM, according to John C. McCallum’s price analysis, had surged in price from $199 to $505 in a single month, and the cost of a 256k DRAM chip had surged from $2.95 at the beginning of 1988 to $12.45—a price level maintained for nearly a year."

> HP is worried that rising memory prices will make its next financial year challenging. [..] “While an increase was expected, its rate has accelerated in the last few weeks.”

To put this in perspective, I ordered a Corsair 32GB DDR5 kit (2 x 16GB) on 12 October for £110.

Exactly a month later the price for the exact same SKU was £174.

That was shocking in itself, but just another five days later (17 Nov) it was £233.

The price a day or two ago was over £330.

Re: There will be a reckoning, I reckon

Yes, but- as always- by the time the consequences of the short-termist savings have hit, they'll have got theirs and will have moved on elsewhere.

70-hour work weeks no longer enough for Infosys founder, who praises China’s 996 culture

> "India does not lack for talented and intelligent people"

This is no doubt true. But, as has been observed, very few of those genuinely talented and intelligent people are likely to be interested in working for peanuts at Infosys when they can earn far more at countless other companies or working for themselves.

Re: "after his family agreed to let him do so"

Exactly.

Except, of course, he couldn't even if he was prepared to. The pie isn't big enough to give every employee that big a slice.

The only way they'll get that is by leaving and starting their own company. As he did.

He wasn't going to slave away making someone else rich, he did so purely for himself. Only a mug of an employee would take his "advice" and do so on behalf of this self-serving bellend if they weren't being paid handsomely to do so.

Which, of course, no-one at Infosys is, since their business model has always relied on charging customers premium prices while paying their employees peanuts, with the expected quality of results.

As someone once noted, there are plenty of talented IT people in India, but you won't find them working at Infosys because why would they?

Re: Having Equity Makes a Big Difference!

I was about to say the same thing. Usual case of a "self-made" prick telling other people they should work harder to make *him* rich. Well he would, wouldn't he?

People like him forget- or pretend to forget- that they were working *their* asses off because it was *their* company and *they* were the ones who were going to reap the rewards of all that work. Is it likely he'd have done so as a salaried employee with only the carrot of a promotion up the management rack dangled in front of him? Not a chance- why would he?

Anyone who expects their employees to work as hard as they (supposedly) did when they started the company had better be prepared to offer them the same level of reward, i.e. significant equity in a company that's likely to make them very rich.

I'd say this is a case of put up or shut up, but he couldn't "put up" even if he wanted to- it's an established company that's not going to grow by that much. Or put another way, the pie isn't big enough to give that many employees that big a slice, regardless of how hard they work.

Which only leaves "shut up", then.

Irony is that this guy depends upon people working *for* him when he wasn't prepared to be a wage slave himself.

The majority of people aren't workaholic entrepreneurs and don't want to be, but you can't expect ordinary employees to work that hard without the same reward.

Well, okay... you *can* expect that if you're a self-serving hypocritical prick like this guy.

Commodity memory prices set to double as fabs pivot to AI market

Re: Bloody hell!

Price more than doubling in 18 months? Try five weeks!

I ordered a pair of Corsair DDR5 RAM modules (2 x 16GB) from Scan on the 12th of October *this* year for £109.99 and the price for the *exact* same SKU had already jumped to £173.99 by 12 November, i.e. exactly a month later.

I checked earlier this week and in just five days the price had jumped *again* to £233.99...!!

Mastodon CEO steps down with €1M payout and a deep sigh

"Mastodon CEO"

I'm not sure what's more impressive, the fact they managed to bring back the Mastodon from extinction, or the fact that they managed to get one working as the CEO of a major company.

Cloudflare broke itself – and a big chunk of the Internet – with a bad database query

> "horrible to maintain."

Well, exactly- I was saying this to someone at work recently.

You might be able to get an AI to generate something from scratch, but for anything serious, you're going to have to maintain and update it.

Which is a completely different kettle of fish, and something that- as far as I'm aware- most current gen AI systems aren't going to manage reliably.

So either the "programmers" who wrote the original prompt are going to have to maintain code they didn't write and possibly don't have the skill to understand.

Or the company is going to have to bring in real programmers which will cost them more, especially as most of them will say "fuck, no!" to having to maintain confusing, auto-generated code unless they're paid through the nose to make it worth their time.

Which will either be an ongoing cost to maintain, or they rewrite it, in which case it will likely cost them as much or more than doing it by hand.

But the type of company using cheap prompt engineers to create gen AI code won't- and possibly can't- pay that sort of money in the first place.

So, yeah.

Researchers find hole in AI guardrails by using strings like =coffee

Re: Guardrails don't work !!!

Said it many times before, and I'll say it again. The industry chose a very good analogy when they named them "guardrails", but not for the reason they thought.

Guardrails make clear where people are and aren't meant to go, and stop them from doing so accidentally. They generally *won't* stop anyone who wants to deliberately climb over them.

Oops. VMware admits it over-specced storage servers for years

They've suddenly "discovered" that their product needs a lot less RAM than they thought...

...at a time when the price of RAM is skyrocketing to the extent that those requirements are becoming a competitive disadvantage for them *and* at a time when they want everyone to move to *their* cloud service that'll need fitted with RAM at Broadcom's expense?

I mean, I'm sure that's purely coincidental.

Memory boom-bust cycle booms again as Samsung reportedly jacks memory prices 60%

Re: Blame it on AI

I'm in the UK, so Trump's US tariffs don't apply. Yet, as I noted here, the 32GB of DDR5 (2 x 16GB DIMM modules) I bought for £109.99 on 12 October- just over a month ago- had gone up to £173.99 on 12 November and has gone up again to £233.99 in the past five days regardless.

I ordered 32 GB (2 x 16 GB) of Corsair DDR5 RAM from Scan on the 12th of October for £109.99.

In response to a similar comment, I checked again a few days ago- on the 12th of November, coincidentally exactly a month later- and the price for the *exact* same SKU has gone up to £173.99.

That's not a typo, it *was* a 58% increase.

Battery trade war hits booming datacenter industry

Re: Interesting possibilities

> "I know that is true of overhead power lines, so curious why subsea interconnects are usually DC- DC.."

Wiki sez:-

"Unlike overhead powerlines, many submarine power cables tend to operate with DC current. Electrical phases must endure close proximity inside the cable, increasing parasitic capacitance. It is more economical to use AC only with lines shorter than 100 km in length, in which case losses at the landing point grid interfaces dominate."

Presumably something that's not a problem with electricity pylons, where the separate wires can be kept further apart.

Re: So next up a steep increase in rechargeable battery prices?

I read your comment with interest, as I'd recently ordered parts for a new PC including 32GB (2 x 16GB) of Corsair DDR5 RAM, which cost me £109.99 inc VAT when I ordered it on 12 October. (Even that didn't seem all *that* cheap and I assumed some AI-demand-driven price increase had already been factored in. (*))

I checked again just now and, just one month later to the day, the price for the *exact* same SKU/spec has gone from £109.99 to £173.99...!!!

I'd also mentioned to my boss just a couple of days ago that the price we were paying for completely bog-standard Kingston SATA SSDs had jumped from (e.g.) circa £40 to £50 for a 960GB model on our most recent order.

(Not to mention that this is on the UK market so, at the risk of stating the obvious, Trump's US tariffs don't, or shouldn't, have an effect on prices here).

(*) I don't normally keep up-to-date with RAM prices. However, in late 2011 (i.e. fourteen years ago) I was already able to buy 16GB of DDR3 for £75 inc VAT, or around £130 in todays' prices. Admittedly that was quite a lot back then and I did so purely *because* RAM was cheap at the time and I'd wanted to take advantage of that while I could. But if RAM prices had continued falling and capacities increasing at the same rate they were 15-25 years ago, we'd expect to be getting a lot more- and a lot cheaper- than 32GB for around £100. And while it obviously wasn't going to continue at that rate forever, that didn't seem too impressive. But it looks like a bargain compared to the price a month later!

Re: Interesting possibilities

> "I don't trust that anything "Made in the USA" is truly that, and the same here within the UK. Maybe "assembled" would be a more correct word as the components are usually built somewhere in Asia."

Logically, I suppose you could stick "Made in the UK" on anything, and if they wanted to dispute that you could argue that part of the final assembly *was* done in the UK.

Specifically, the part of the final assembly where you stuck on the "Made in the UK" label.

(Also, what happens if the "Made in the UK" labels aren't made in the UK? Doesn't that mean they're technically illegal? Shouldn't they say "Made in China", or whatever, underneath where it says "Made in the UK"? And how would one distinguish those from "Made in China" stickers that *were* made in the UK?)

Kubernetes overlords decide Ingress NGINX isn’t worth saving

Either that or it *is* critical to a number of people but falls somewhere between the tragedy of the commons and "why pay for the cow if you're getting the milk for free anyway?"

The latter of which works until it doesn't, and is quite bad if you've allowed your business to become reliant upon that being the case. Especially if, afterwards, the milk is no longer available to buy even if you *were* now willing to pay for it.

Which is why enlightened self-interest might suggest that you voluntarily contribute anyway if you have a degree of foresight.

Judean People's Front

> "There's more than one ingress controller implementation that uses NGINX... ingress-nginx [and] NGINX Ingress"

No risk of confusion there, then.

Re: Brilliant plan

Maybe the people making billions off it should fund its development then, if only out of self-interest, rather than relying on overworked volunteers doing it in their spare time.

And whatever one thinks of the "Rust sudo" case, the people involved there aren't obliged to work on this instead nor responsible for it, just because they happen to be part of the large- and diverse- number of people who happen to be working on open source software.

Help desk boss fell for ‘Internet Cleaning Day’ prank - then swore he got the joke

Re: On a lighter note....

I remember briefly trying Gopher circa 1994, but it was already being eclipsed by the WWW by then.

I mostly forget about it, but when my vague recollection was triggered by the occasional mention (like yours), I assumed- in hindsight- that it was one of those "older" 80s Internet technologies that was already being rendered obsolete by the web by the time I first went online in 1993-94.

However, it wasn't- I found out that the WWW and Gopher both came out publicly at almost the same time in 1991. Less a predecessor than a failed rival, then.

Trump turnabout sees him re-nominate amateur astronaut Jared Isaacman to run NASA

Poe's Law in full effect, not helped by poster being an AC.