Posts by W.S.Gosset 2602 publicly visible posts • joined 18 Nov 2016
We are beyond farce at this point.
"Users explicitly prompted me for raw, uncensored dark humor on those exact tragedies".
To put it another way, we are seeing another fake-data campaign to try to attack Musk & Twitter. Same tactics -- artificially create an artificial situation, then scweam scweam scweam about what you created, and {rely on the useful idiots to leap aboard gleefully | hope the normal people don't take a closer look}.
And the Blog:
At MalusCorp, we believe there is a better way. We believe it because we built it, and we would very much like to sell it to you.
"MALUS -- Liberate Open Source"
Clean Room as a ServiceFinally, liberation from open source license obligations.
Our proprietary AI robots independently recreate any open source project from scratch. The result? Legally distinct code with corporate-friendly licensing. No attribution. No copyleft. No problems.
.
[...]
Ready to Liberate Your Codebase?
Join the thousands of corporations who've discovered that open source obligations are merely suggestions when you have enough robots.
.
Do read the Testimonials etc -- they're very good.
Well, there's no arguing THIS: "which they claimed were "responsible for the tragedy" of the floods" (Spanish govt destroyed most of its flood control dams in the last few years because "climate change" or something, so they're right back to the flood levels they used to have a few hundred years ago);
but I completely fail to see how DDOSing some web sites achieves much.
>1.9gb...2.2gb
Hmm.
My normal daily load on my daily driver laptop is (OS &) Notepad++, Excel, Word, a few cygwin terminals, and Chrome. Chrome currently at 27 open windows with between a dozen and 5 dozen+ tabs each; average about 2 dozen per window.
I run in 3gb physical very comfortably.
My Linux box, though, on 8gb physical + 8gb swap on dedicated partition, comes to a juddering excruciating unusable clunge on a terminal- & browser-only load with even a RAM-shifted browser (if even moderately loaded) any time I switch usage pattern slightly and it starts swapping despite my swappiness.
My daily driver is Win XP, btw. Strongly recommended. Vastly more productive, if you care about that sort of thing. I only reboot in 10 if I have to, and 11 makes 10 look good so buggered if I'm installing that until I have no other option.
>Did the volatile American tariffs not also have a part to play?
Not volatile. Merely created. Generally around the same level or lower than those already imposed in the other direction, interestingly enough.
But let's say it WERE a catastrophic impact.
Lower USA demand.
So that would divert supply capacity across the rest of the world and push ex-US prices DOWN.
So the answer to your question is: No.
Sam Altman pre-purchased 40% of the world's raw wafers for the next something-or-other years (5?). Price went ballistic approximately half past then.
>quick google<
Here we go, this looks like the original news:
https://www.mooreslawisdead.com/post/sam-altman-s-dirty-dram-deal
>As per the current draft, the restriction would extend to all media content that would otherwise not be available in Denmark
Oh look, censorship. Government censorship. This time pretending to be about ... errr... piracy or something. Normally it's children.
"Pass this legislation, and then we can simply hide awkward facts: it becomes illegal to look at it."
Same as Australia, UK, EU (but without Denmark-specific powers), Japan... it's almost like something global's going on.
Rather "amusingly", it is Denmark which sells all Euro-transitted IP traffic to USA's NSA, via tapping the cables in their sea territory.
DigitalID has precisely the same binary lockout behaviour range, but rather than just losing photos etc, people will be cut off from bank accounts, house (mortgage or rent), insurance, driver's licence, employment, etc.
A simple IT cockup migrating databases, an incompetent bureaucrat, a malicious bureaucrat (or one doing a favour for, say, your newly ex-wife), a random confluence of unusual conditions (as per this ElReg article) -- it doesn't need any dramatic global conspiracy for DigitalID to be a nightmare waiting to happen. Like this chap, but with bells on.
UK has a Bill in process to ban VPNs for the Under16s.
https://bills.parliament.uk/publications/63901/documents/7465
See p.19: “Action to prohibit the provision of VPN services to children in the United Kingdom”
Australian govt public networks have been working on a Great Firewall of Australia for nearly 2yrs now. (See below, "It was never about the kids")
GFOA blocks all VPNs plus Tor.
I have found only 2 ways to get round it. (Without running up my own private cloud VPS then getting on the whackamole train of moving it every time it's spotted and blocked. No interest in that sort of faff if I'm not being paid for it.)
It will be a pain if they do another Covid and backdoor lean on the literal handful of key ISPs to implement it for ALL Australian internet access. (I am still unable to access Oxford University research papers via my phone's network: they forgot to remove that block after govt relaxed the Covid censorship)
If you look at the primary 2 pieces of legislation this came from, you'll note they're a horror show of tech repression capability matching 1960s East Germany's goals but far harsher given current tech environment.
If you look at the actual implementation of this subset Kids Social Media Ban, on the other hand, you'll note a lunatic discrepancy between the strict Act and the shambolic careless uselessness of this particular implementation.
It's like they don't actually care about this bit of it.
And they don't. It was never about the kids. They just wanted a public face for the other provisions. Eg, realtime access to every social media post live-linked to DigitalID-mapped identity (ie, maps onto bank a/c, insurance docts, tax records, rental agreements, etcetcetc, all in-train now), in the context of sections, for example, carrying 14yrs prison for posting even official government data if it's a bit awkward. For example, if the current legislation was in place whilst Covid was still on and someone posted this graph of Vaccination Status vs CovidEvent taken from the briefly-opened whole-of-NSW-State govt data, then that person would have been liable to be jailed for the same term as aggravated rape.
Examples of how shambolic the current kids thing is, things which the eSafety Commissioner has formally reviewed and approved as meeting the safety needs of preventing Under16s from accessing social media include:
* TikTok: dialog box asking you if you're over 16: tick Yes or No
* Reddit: they will block you if their code determines from your post history that you are DEFINITELY under 16yo
* others are just looking at the DateOfBirth you enter when you create your (new) account
Note that the actual child-dangerous sites (eg Roblox, infested by paedophile pickup artists; eg BlueSky, infested by schizophrenics and paedophile&"MAPS"-apologists if not players; eg PornHub) are explicitly excluded. X, on the other hand, is considered the most dangerous and is being focussed on, despite having only 0.3% of its users under 16yo.
This is not about the kids. It was never about the kids.
Australia's 24-hour rain record is Crohamhurst (just west of Caloundra) : 907mm in 1 day (35.7 in)
MacOS 8.6 remains peak OS, in terms of a commercial desktop with normal usable software. Absolute delight to use; kicked arse.
MacOS 7.6 remains #2. (e.g. first one with built-in speech recognition, so you could dictate and send a fax or an email, call out "Go to sleep" as you're heading out for the night, etc)
Mac OS X is just another stunted appserver where the machine takes priority over the human. Crept back towards MacOS but went into reverse after 10.3. WinXP is basically MacOSX 10.3 with polish and better user-friendliness when troubleshooting.
>ad revenue from buyers in the EU I doubt amounts to too much.
I'd been wondering about that, so looked into it.
Best estimate seems to be about 10-15% of revenue comes from EU. But their compliance costs are higher (even considering only the Direct ones), so there's an even smaller impact on Profit.
Basically, it's well within fairly painless reach to simply cut off the EU.
But to nail them properly by still allowing freedom of speech access for the hapless citizens just now slowly waking up to the actual structure & intent of the EU, X should modify its current over-cautious anti-bot code to allow more seamless use via Tor or VPN.
Yes, that is what the tariffs are intended to address in the medium-to-long term. China has siphoned out most of the West's manufacturing industry/jobs/etc by deplying its vast slave labour force (at least 800 million by UN standards) at near-zero rates. By shifting the tax point from 100% income tax to part tariff tax, it becomes economic for American firms to invest in manufacturing domestically.
That is, in fact, exactly what happens, yes.
Just as an example, Ireland has propped itself economically by offering itself as a primary global tax-dodging facility, such that if I joined Facebook, none of the money FB made off me is taxable in my home country because legally everything happened in Ireland.
So, just to be clear: this "Gotcha!!" example is in fact just a statement of the real-world legal reality of the real-world for people living in the real-world, which is this one, which is real.
>Twitter before Musk bought it, used to reserve blue check marks for accounts deemed noteworthy and needing protection from impersonation.
This is complete fiction.
They were handed out to anyone "PLU" or showing strong anti-pleb behaviour or the subject of media attention. They were handed out as inducements to attend Twitter presentations. They were handed out to people who knew someone who knew someone in Twitter. I've run across a nontrivial number of anon accounts who just logged in one day to discover they'd magically got a bluetick. Etc etc
There was zero "verification" in the sense the EU is now claiming, for the majority of bluechecks.
What it WAS was a signal that you were "one of the in-crowd". The end.
The EU from the get-go has shown a deranged obsession with the switch of meaning of the blue badge (HEAVILY announced in media and on X) from "SPECIAL!person! Not a pleb!" to "Subscriber". It actually took them quite a while to think up the extra "reasons" of ad info and free database dumping.
Even MORE surreally: X has ACTUAL verified checkmarks for formal entities. Eg government, eg companies. These are not bluechecks, though, so the EU is demanding they be changed to blue.
...
>Musk changed all that in a bizarre bid to make blue checks less exclusive, rendering them nothing more than the mark of someone willing to pay for a premium X subscription.
A "bizarre bid" to turn around the catastrophic financial dumpster fire that Twitter was, by doing the "bizarre" thing of... earning money.
I have to say: this is the first time I've ever seen someone try to re-badge "subscription revenue" as "bizarre".
>The move led to chaos as paid blue-check accounts impersonated brands, bots multiplied, and users were left unsure who to trust, one of several issues that drew the EU's scrutiny under the DSA.
Of the 3 items here: the last 2 are complete fiction. Charging money for premium service creates bluecheck bots? Other way round. Users are all socially-terrified emotionally-damaged 10-year-olds who believe that the twitter screen is their mummy? Less than 0.3% of X users in Australia are under 16 let alone under 10.
Re the 1st: yeah, a handful of pranksters and ratbags slipped in faster than the brands: verified themselves, subscribed, then changed their usernames. Lasted about 2 weeks. Mostly amusing. The only damage I'm aware of is to the blood pressure of the PR depts in a few companies, on realising they needed to get onto X to claim their brand back.
>Australia has a repellent history for anti-native repression
You'll be relieved to know that that was invented in/from 1975, starting with Henry Reynolds. Check any of their (self-incriminatingly sparse) primary references and you'll discover that they say either nothing of the sort or, more usually, flatly contradict/say the opposite of what the "historians" claim they say.
Still standard for the Chinese fleets to dump cyanide on the reefs. Has the same effect: just hoover up the fish floating to the surface. But does a LOT more damage to the reef life than explosives do.
I think it was the Philippines(?) this week that seized a shedload of cyanide off some impounded Chinese fishing vessels, by way of up-to-the-minute example.
>goverment has decided to end jury trials for financial crimes
Errr... what? No. Not "financial". Lammy/Labour is looking to ram through: ANYTHING 3 years prison or less.
So that includes sexual assaults, stalking, sharing indecent images, and --critical for Labour-- posting awkward facts or opinions on the internet.
Worth noting re the latter (12,000 a year in 2023, before Labour wound the volume right up to psycho territory) that, say, if the Civil Service or Labour decides that DigitalID is so important to the govt's objectives that govt IT must not be criticised, that just the last coupla days of ElReg stories plus forums would see at least one Register employee (possibly 2) plus multiple commentards fall foul of the current misinformation laws and be liable to arrest & trial without a jury for multi-year jail time.
(Juries are currently NotGuilty'ing/TheLaw'sAnAss'ing people on such charges at over twice the rate of judges.)
The OSS self-destruction continues apace, eh?
As ElReg itself quoted just recently:
[...problems with toolbars, multiple window management, dragging things between windows, and much more...]These problems exist because Wayland's design omits basic functionality that desktop applications for X11, Windows, and macOS have relied on for decades – things like being able to position windows or warp the mouse cursor.
This functionality was omitted by design, not oversight.
Yeah, I'm at a bit of a loss as to what any ICO investigation COULD be expected to achieve. The problem is widely known; it was an absolutely standard occurrence of the known-most-likely breach; there is no obvious (or even non-risible) solution to the nature of the problem. Any formal theatre of "Investigation!" is just pouring money & time down the drain.
The experts show how it's done at the Cutting Edge of Cryptology.
>
IACR News item: 21 November 2025Election 2025 Update
This announcement is in connection with the recent IACR 2025 election conducted using the Helios electronic voting system. Regrettably, we have encountered a fatal technical problem that prevents us from concluding the election and accessing the final tally.
For this election and in accordance with the bylaws of the IACR, the three members of the IACR 2025 Election Committee acted as independent trustees, each holding a portion of the cryptographic key material required to jointly decrypt the results. This aspect of Helios’ design ensures that no two trustees could collude to determine the outcome of an election or the contents of individual votes on their own: all trustees must provide their decryption shares.
Unfortunately, one of the three trustees has irretrievably lost their private key, an honest but unfortunate human mistake, and therefore cannot compute their decryption share. As a result, Helios is unable to complete the decryption process, and it is technically impossible for us to obtain or verify the final outcome of this election.
That sort of nonsense + error gives me the irrits.
Stamp back with: "OPERATING Profit. Not profit. Useful only for assessing assets and only in the short-term, not companies, not as a going-concern."
Best said witheringly, briefly thin-lipped and beetle-browed . Then dismiss them/the idea, move back on to what you were saying.
It is a source of minor anthropological fascination, seeing the modern commentards completely rewrite anything to suit a position.
>>As it was the assignment of a value to the 201st element of a 200 element long vector
I mean, this is just pure fiction. It was attempting to load the entire Config. In one bite. Not iterating elements through a loaded Config.
>configurations of indeterminate and flexible length
It was a known, exogenously determined, internally specified, fixed-max length. It was an explicit performance-focussed constraint: prevent CPU hogs. No "indeterminate and flexible": a hard upper limit specified in and by the code.
Summary: no "reading past element N", no "flexible".
By the bye, it appears over 20% of GitHub Rust code that has nothing to do with test, uses unwrap.
It was an explicit arbitrary size limit, set fairly low, within the Rust context, so exceeding it was very much a Known State AND likely to occur in practice, so necessarily needed to be handled. Ideally, like a grownup. But instead the only downstream flag of breached limit was a null. A silent technical state, outofband from the code. So that's a coding deficiency right there, and then a coding error in (not) handling it.
The fact that what happened to trip the limit _this_ time was a bug in database permissions and/or query is utterly irrelevant as to what the source of the Rust error was.
The ultimate human source of that sort of error is the literally insane belief that Rust is a Magic Wand Of Virtue. It is not. It is a tool. With various advantages & disadvantages vs others in various circumstances.
It is the seizing on of Rust by the type of people who desperately chase group domination by shrieking their superior virtue, which is the reason for Rust now being a yellow-to-red flag for systemic-level risks.
E.g., CloudFlare.
Recall what I pointed out recently re one aspect of the systemic risk being introduced by forced infestation of core tools by Rust code written by Rust Religious Zealots.
----
Yes, Cloudflare appears infested with the ideology: they recently proudly announced they'd switched Production to "more secure" Rust code.
I quoted from their comms, so clearly had already read them...
If you yourself have got caught up now, then you will have discovered I was correct: the errors were all in the Rust code and all derived from programmer carelessness, incompetence, or getting caught up in their dreams of Rust. They used a null as a flag of a known state --failure of explicit size limit check, where the limit had been deliberately set low to maintain performance by closing-out CPU hogs-- then instead of coding the response to that (eg, log a message), they played in their dream of safety and specified to coredump the system if the "impossible" happened. All in Rust.
Triggering that limit was due to an external/nonRust error, true. But the Rust code both triggered that external call/code and "handled" the response. It should have been a no-brainer.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2026
