Re: Donatas Smailys said:
Hmm, internet connected allegedly 'smart' washing machines from the likes of Samsung, LG, Kenmore and no doubt others rather suggest that it's already changed...
7666 publicly visible posts • joined 18 Apr 2007
Aliexpress search is perhaps already doing it: certainly, their search for (e.g. EOL memory chips) returns very few EOL memory chips, an awful lot of generic electrotrash, the occasional toaster, shoes, and ladies' underwear. Also tyres, for some reason. One day I'll understand the logic, but I may have to meditate on a mountaintop first.
Isn't it funny how almost the entire internet is predicated on a method invented a hundred years ago to sell soap?
I'm curious to find out how Kicad works with XWayland, given the dire warnings here: https://www.kicad.org/blog/2025/06/KiCad-and-Wayland-Support/
As it happens, I have a spare laptop with nothing to do (it's sound is broken but I don't think that's an issue for Kicad) so I may play and find out.
The version of Abbyy I was using worked really well for this: my application needed to extract the original text (obviously) but also metadata relating to font size, weight, and style, page layout, white space etc. If converting from a printed page to a text stream this is essential (curiously, one very important thing is correctly identifying paragraphs and scene break white space between them - you're trying to locate something that essentially isn't there).
Abbyy produced an output consisting of individual lines of text with the necessary data - and specifically, the location of the line on the page - in an XML file which could easily be parsed to reconstruct the original text and (streamable) layout. Tesseract, for example, was very good at reading text but doesn't track 'empty' lines at scene breaks, and nor did it identify e.g. italics or relative font size. It handled multiple columns fine, though.
In a previous life I made much use of Abbyy running under linux, with a paid-for lifetime license, 12,000 pages per year from memory.
Some years later they turned off the license server and provided neither an alternative way to activate the product if you updated the distribution or computer, nor a replacement product (indeed, this looks like a replacement, but it's only available for a year outside enterprise?).
I really liked the product - it provided essential (for me) metadata in a convenient form that e.g. Tesseract doesn't - but once bitten and all that, I'm not inclined to return to the fold.
Hell, I don't even buy laptops for myself less than two or three years old. End of lease Dells, usually, at a tenth of the new prices. Then replace the OS with Mint and they're happy for years.
Though my wife's fifteen-month old Acer is about to be replaced by something else: it has developed an odd fault - I think in the USB hardware - that started with noisy audio, then no audio at all, no audio devices listed, external usb audio devices not recognised, and if plugged into my normal desktop USB expander, it passes video fine, but still no visible audio device, and also no external wireless keyboard or mouse. Annoying.
But really, I think all the important points have already been mentioned.
Chromebooks were a good idea when they were launched, right up until you realised you could do little offline. They were a better idea when people realised you could replace the OS with Linux. But now they're just locked down extensions of Google. No thanks.
I had kinda the same questions. From their experimental layout photo, it looked as if they were expecting an audio frequency (i.e. under 20kHz) trace of their injected signal, while the implication is that megahertz signals were used (it'd take a bloody big antenna to couple audio frequencies with anything other than near-field). I really wouldn't expect headphones and a generic microphone to do _anything_ much above 20kHz, and nor would I expect the codecs used to compress the sampled audio for distribution to allow it to pass, so that tells us little.
I do note though that many (older) phones used an earphone/headphone *cable* as a receiving aerial in the 100MHz band...
Injecting a heterodyne carrier into a class D amplifier to disturb its operating frequency would seem to be reasonable way to do this, and that looks like what they're demonstrating in the 'can you hear it' picture (no headphone cord). But if that's how they're doing it, rather than interfering with the source, then a cabled headphone is probably immune. Unless, of course, sufficient injected signal finds its way back to the headphone amplifier at the other end.
one of the few aspects of computing that haven't got ridiculously expensive.
Perhaps not, but three thousand bucks for a screen is not something I'm likely to splash out on in the immediate future: it's over six times what I just paid for a 55" Panasonic TV, and probably ten times what I paid for the (pre-loved) laptop that's driving the 29" 4k monitor in front of me.
and discovers, to my amazement, that eggs don't pack noticeably closely. Even in a hexagonal grid, there's a lot of space with a variable but decreasing amount of egg and very little where the ends are touching a nominal top and bottom.
Is the improvement in impact strength quoted related to the position at which the impactor strikes?
Yes and no. If your definition of 'works properly' is 'shows a picture presented to the HDMI input' (as I do) then the Panasonic I set up this morning, after a bit of head-scratching, makes me happy(ish).
I don't use streaming, just local sources one of which is a Magenta TV (German TV) set top box. I route the sources through an HDMI switcher what extracts the audio and lets me hear it through a decent sound system synchronously with the pictures.
The Panasonic selected none of the 'would you like' options, though it did position the selector over the 'accept all' - you know, personalised advertising, audio spying etc - and to its credit it did not require a network at any time (unless you want to use the supplied applications, which I don't). The network is set (now) to do not remind me.
One minor issue which took me a while to sort out: the Panasonic was turning off the Magenta TV box as soon as it connected; it recognised the box, advised me that keypresses would be passed through, and then turned it off. Amazing. Turns out that the HDMI ports are by default 'enhanced' and when that's deselected, things work as they should. Though I'm amused that on/off and channel selection is passed through, the volume control isn't. So you might still want two remotes...
Anyway, we'll see how things go.
Interesting 'lookalikes', thanks.
I was extremely ill for several days after eating - in an Austrian restaurant - chantarelles. My wife who had the same dish was unaffected... who knows? But I've avoided them ever since. I hunt mushrooms only in the supermarket.
You can indeed. One day I shall tell the tale of formatting (somebody else's) two thousand page document, each paragraph of which had a different style, even though they were all theoretically the same.
I refer to projects using LaTeX like LyX, which does not allow, without jumping through a lot of hoops, random styling. More significantly, it offers by default a selection of styling libraries such that your document will come out matched to a house style.