Posts by Crypto Monad 597 publicly visible posts • joined 14 Dec 2017
Maybe. However, old cars which were manufactured without seatbelts, or even indicators, are still legal without modification.
There are some things on the road which almost certainly wouldn't be allowed if they were introduced today. Motorbike-sidecar combinations and horses are two that spring to mind.
Also, I don't think the ASX appreciated something fundamental: in the best possible outcome (i.e. it actually worked), it would have made the ASX redundant.
What were they actually trying to achieve? For example, did they want people to make faster local trades? Then set up little outreach branches of the ASX, have them broker local trades, and send them back to a central ledger for permanent notarisation.
If the Central bank finds 27 copies of the same digital note, how does it know which, if any, is the original?
??!
Nobody has ever said that digital cash will be in the form of "bearer tokens" that you just pass across. As you imply, any pattern of bits is trivially duplicated - there will be no "digital note".
Instead, the ownership of digital cash will be tracked in a central ledger, held in a central bank. All transactions go via, and are notarised by, the central bank.
This is pretty similar to blockchain, except that there's no need for an expensive consensus mechanism: the bank is the bank. The bank stamps the transaction, and the job is done.
Will the ledger be private or public? My guess is that although it will be technically private, it will be open for inspection to so many government bodies and their outsourced suppliers that it might as well be public.
.....but rather a random key for every message (calculated on each peer and thrown away after use)
And you think you're the first person to have thought of that?!
Have a read of the protocol called OTR ("Off The Record"): https://otr.cypherpunks.ca/otr-wpes.pdf
> if I want a mini laptop I'd rather shell out for something from Planet Computers which has a keyboard
Don't waste your money.
I was an early adopter of the Gemini in 2018, which was released with Android 7 and got its final update to Android 8.1 later that year. Then it was abandoned. No more updates for a phone less than a year old.
Incredibly, they're still selling it. A phone with Android 8.1. In 2023!
They were probably sold the idea that a migration from an Oracle on-prem system to an Oracle cloud system would be very easy, because obviously.
I expect they are totally different systems but with confusingly similar names - like Microsoft's Active Directory and Azure Active Directory, or Skype and Skype for Business (RIP), or Windows and Windows Phone (RIP).
Geolocation works exactly the same for IPv6 as it does for IPv4. The owner of the addresses registers the assignments of blocks.
For example, if an ISP's POP in Utah has a /20 pool of IPv4 addresses (enough for 2^12 users) and a /44 pool of IPv6 addresses (enough for 2^12 users with a /56 each), both blocks will (or should) be recorded as being used in that POP, and that will trickle through to geolocation databases.
It's not an issue for pornhub though:
$ dig +short pornhub.com aaaa
$
Computers (CPUs anyway) are 4-5 orders of magnitude faster than they were in the 1980's. A baseline of 1MHz / 8bit has become 4GHz / 64bit / 4 cores, even in a phone.
System RAM is perhaps 1-2 orders faster, if you include wider bus widths. Hence multiple cache tiers required inside the CPU.
And yet, somehow we seem to have squandered all this. I suppose much of it has gone in pretty graphics, animation and video playback, and thick layers of libraries to support all that. I would trade some of that for safety and security any day.
You don't *need* all that performance for a workable graphical UI. The Macintosh LC II had a 16/32-bit 16MHz processor (32 bit internally but 16 bit system bus), 4MB RAM (expandable to 10MB), and Color Quickdraw in 256KB of ROM. And I bet it booted up more quickly than a modern PC.
"Fresh kernels often include facepalm-grade fixes and this time around there's an update that ensures an Intel gigabit NIC can approach its promised throughput. Since 2020 the kernel has unwittingly throttled it to 60 percent of that speed."
And it's only been a problem for 8 years and 9 months.
> Why? Simple: What guarantee do I have, that there wont be another stunt like that in the future?
Yeah, but what's the alternative? What guarantee do you have that a.n.other language won't pull a stunt like that in the future?
Linux itself goes in for all this trademark stuff. Would that stop you recommending or using Linux?
- So....in case you need to to be told....NO PERSISTENT OR PUBLISHED KEYS anywhere
[...] We can't see a problem. Perhaps someone out there can explain.
Well for one thing, Diffie-Hellman is subject to an active man-in-the-middle attack. Without persistent signing keys or some other means of authenticating your peer, you have no idea whether you're encrypting data to your buddy or to your adversary.
Oversec would most likely be categorised as a messaging app: its job is to encrypt data to be sent to other people, as its own website clearly describes, and there's no meaningful way to use it otherwise.
Banking websites don't need to break E2E encryption. The traffic is already decrypted at the server side, and if govt decided they need to monitor people sending messages through banking sites, they would do it there.
Hard drive backup is covered by existing powers that say you can be jailed for not revealing your encryption keys to plod on demand. Very recent example here.
I think the underlying point is: even though plod *can* get at encrypted data on your device today (without even grounds for suspicion, if they mention the word "terrorism"), they still need physical access to your device. The new powers are so that plod can do this remotely, and as a bulk trawl across *everyone's* devices.
But surely Oversec will be subject to the same rules as Whatsapp and Signal? Either they'll have to remove them from the (UK) app stores, or they'll have to add backdoors.
Sure, there will be people who compile their own encryption apps and side-load. So they'll be after them next. But I suspect most people won't bother: they'll quietly accept their loss of privacy and security.
Debian Bookworm will have LXD natively packaged, whereas Ubuntu only provides it as a snap.
As this is the last thing I use that needs snap, this will be a fine reason to switch from Ubuntu to Debian.
I am missing something very fundamental here.
When we're talking about optical losses along a fibre optic path, I imagine those are due to photons being absorbed or scattered, so the number of photons reduces over distance.
However, for this quantum entanglement stuff, we're talking about *individual* photons (one of an entangled pair), right?
Now, if a photon arrives at one of these repeaters, and is re-sent as a fresh photon, surely it's still just a photon? Is this new photon any more likely to get absorbed in the next 10km (say) than the original one was? Do photons carry a memory of how long they have been travelling, and therefore get "tired" and need to be replaced by a "fresh" photon every so often?
Alternatively, I can imagine that instead of single photons, bursts of photons with some sort of similar entanglement are sent, and over distance the numbers will go down as they are absorbed. But in that case, wouldn't the repeater need to replace one incoming photon with two or more identical outgoing photons, in order to make the numbers back up again?
CPU designers start to charge based on a slice of the total device value, not the CPU chip value.
4G/5G radio designers start to charge based on a slice of the total device value, not the radio chipset value.
Screen and battery designers start to charge based on a slice of the total device value, not the screen or battery value.
Soon there isn't going to be any value left in the phone at all.
My first use of ethernet was 10base2: it used coax cable similar to TV coax (but 50 ohm impedance rather than 75 ohm). Fortunately, we didn't have to drill into the cable: we had BNC T-pieces. And 50-ohm terminators to put on the T's at the far ends. Maximum length end-to-end was 185 metres.
The whole network would die, either if someone's foot got caught in the cable and pulled it out, or if a dodgy network card would start randomly chattering. Finding the computer with the dodgy NIC was a tricky game of divide-and-conquer, repeatedly dividing the network into halves and putting a terminator at each end.
It was a happy day when we got 10baseT hubs and Cat5/RJ45. Whilst it was still one big collision domain - only one device could transmit at once - the hub could detect a chattering device and block the port.
Or:
* create freshly imaged Linux server (or a fresh Linux VM with USB passthrough, if physical proximity isn't an issue)
* mount USB filesystem read-only inside Linux server
* fetch files using sftp/scp (or Samba if you value convenience more than security - but there are perfectly fine Windows sftp/scp clients)
For physical safety, I guess a USB extender with relays to allow remote activation.
Be careful what you wish for. Even if it were simply a useful tool meaning that "knowledge workers" are 3x more productive, it would still put 2 out of 3 "knowledge workers" out of work..
It's not a zero-sum game though, not in the long term anyway. Just because mechanised looms put weavers out of work, doesn't mean that there's more unemployment now than there was before then.
If it actually works as advertised (and that's a big "if"), workers will end up doing more and/or different things with their time, and probably under more pressure.
(Thinks: for example, handling the court cases from people who had their cases messed up by AI support agents and were unable to get through to a human. It's when the courts get AI-driven that we're really screwed)
Honda, Toyota, and Ford use approximately 3500 distinct key combinations. Therefore, the odds are reasonably good that someone else out there may have the same keys you do....
The birthday problem/paradox then says you only need to have sqrt(3500) = 59 of the same model of car, before it's likely that at least one pair of them will share the same key.
energy consumption due to roll resistance and wind pressure goes up exponentially
Are you sure of that? I thought it went up more or less cubically (i.e. speed raised to the power of three). Square of the speed for the force of wind resistance, multiplied by the speed to convert force to power. Something like that.
People are often inclined to say something has "exponential" growth when it isn't. Inflation is though - as is compound interest.
Yes.
In a fair world, they would charge a setup cost which matches their one-time costs (i.e. Openreach installation charge and supply of router), and then have a lower monthly cost for ever.
Doing it their way, if someone sticks with them, they recoup the initial setup costs many times over. However, the risk to them is that they'll get many customers who only want service for a month or two; they are banking on these being subsidised by the loyal long-term customers.
feasibly Evil ET could wipe out the entire UK online shopping industry with this bill. IF they push to make all secure connections an issue.
I doubt it, since they can always capture the communications at the server side (compelling the service provider to cough up). The same would be true of DMs sent via a website, where the messages are only encrypted via TLS in transit but are visible to the server in the clear.
It *could* be an issue with websites specifically set up for secure communication, e.g. Protonmail. However, services which are specifically set up to support criminal action can be taken down and already are. Clearly, no new powers are required for this.
The issue here is specifically around end-to-end, i.e. pleb-to-pleb, messaging - where there's no power equivalent to a telephone "wiretap". However, all the metadata is already visible.
That LeBron dunk NFT, for example, was purchased for six figures because its buyer believed it "was worth seven figures right away."
The question is, why on earth did the buyer believe that? How did they come to the objective assessment of its value being ten times higher than what the seller was asking? Or indeed, that it had any value at all?
Maybe they made a calculation that that there was a high probability of the existence of a Greater Fool who would buy it at a higher price. Maybe they had seen [claims] that similar items were selling for equally exorbitant prices.
But really, I suspect it was nothing more than a gut feeling about this "cool" thing they were buying. A thing that doesn't even exist - unlike baseball cards, which at least you can hold in your hand. And of course, the gambler's excitement that they were going to make a huge profit at the expense of the casino or bookie.
In the UK, the PSTN goes "stop sell" in September this year, so you won't be able to order a new analogue line (I guess not an E1 trunk either), but it is not going to be completely switched off until December 2025.
However, "all copper services" are not being turned off. Rather, remaining copper services will be digital only (FTTC and ADSL), without baseband dialtone. Any voice services will have to be provided as digital voice over-the-top.
There is a separate stop-sell process for copper services, but it only applies in areas which have reached a certain percentage of FTTP coverage, and it applies only to those properties that actually have FTTP available to order.
> $330/kWh = $330m/GWh x 137 = $45.2bn just for batteries. That won't even last a night, let alone a winter anticyclone.
They didn't say 137GWh: they said 137GW for 4 hours, which is 548GWh. So that's actually about $181bn in batteries. (Aside: the price of $330/kWh matches readily-available domestic batteries. There are inverter gubbins required on top, but I expect you'd get a bit of a discount buying that quantity of batteries)
Those batteries won't last forever; if they last 10 years then you'll need to budget $18bn per year for ongoing replacements, although they should get cheaper over time.
Four hours is also a reasonable time to fully charge or discharge a battery without killing it.
So how helpful is 137GW for four hours? For US peak usage, see https://www.eia.gov/todayinenergy/detail.php?id=42915
I believe the units of these graphs are "millions of kilowatthours per hour", which is a long-winded way of saying "gigawatts".
It's unclear how long the peak lasts after the sun has set. However it works in their favour that demand is highest during the summer months for aircon, and that there are large chunks of USA which are relatively close to the equator.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
