That can work for audio - Frank Zappa did this to archive 1" 12-track tapes - but in this case, it's probably a C-Format helical scan videotape. You would need an Ampex VPR or Sony BVH machine to play it back.

For 1982 I would assume C-Format, though you never can tell.

Star Wars through Return of the Jedi were available as CED discs, which is basically what you're describing.

There is an animated version of this, using the original soundtrack from the episodes. IMHO they did a really good job of it and I thoroughly enjoyed it.

Several reasons. The early devices (CE3 and earlier) did this thing where the whole filesystem was stored in RAM. So if the handheld was left to go flat, or the application crashed hard enough to bring down the kernel too, the device would be factory reset and you'd have to reinstall everything.

The UI was, for the most part, Windows stuck on a phone screen. Start menu and all. PocketPC/Windows Mobile made some concessions to being a mobile device, Windows CE without the PocketPC shell just looked like a slightly scrunched up version of Windows 95. The touchscreens were almost all resistive ones so that you used a pen (or fingernail) to push the screen. I don't think they ever did multitouch even when the capacitive screens started to appear.

For a long time, the platform had memory constraints, in that each application could access up to about 16MB before the allocations started to fail. You had to perform a lot of gymnastics with memory mapping to get past that and that memory was a lot, lot slower to access than what you get via calloc or something. (I cheekily referred to it as 'XMS memory' in our application). I believe CE6 removed that limit - but it wasn't backwards compatible IIRC and you'd have to rebuild the app specifically for that platform. I don't think I ever saw CE6 hardware.

CE and even the Windows Mobile/PocketPC did quite well in industrial settings, and I think they're still using those on my local supermarket for, e.g. printing barcodes for reduced items. I had a run of about a decade programming these things before they became largely obsolete and the company switched focus.

The iPhone was rather disruptive, but more in a dog-in-the-manger way than as a viable replacement. Around this period - 2000-2012 - the industrial handheld market in Europe was largely a cottage industry serviced by small mobile development firms of say, 20 people each, maybe less. So. What did Apple do? They released a device where you had to sell your application through their own marketplace - bespoke applications which were supposed to be solely used by a government authority or a supermarket. I don't know what they were thinking. In order to distribute the application yourself, you had to have 500 employees and a Bradstreet number or somesuch - a US-centric requirement that basically made the iPhone unusable for commercial purposes. To my mind, at least, it displayed a staggering lack of understanding of how mobile devices were used industrially.

The other interesting thing was that the apps on the iPhone were basically fart apps sold for 99p each - and so some of our customers started to expect that a RAD development platform which had taken 8 years to make, should also be sold for 99p, economics be damned. This was not helpful for the industrial market.

Finally, Windows Phone 7 came along. And while that was built around CE7 or something, it only exposed a javascript API. This was back in the days when javascript was still a toy language. Imagine if someone put a phone on the market and only allowed it to be programmed in BASH. That was the way this announcement went down. Microsoft honestly expected that everyone would rewrite their C++ application stacks - with 10 years of sunk costs - in a language that was basically useless. It was easier to port to Android - since you could salvage large chunks of the C++ code and call them via JNI. Most developers did so, and having stabbed their developer base in the back, Microsoft made the shocked pikachu face when nobody developed for WP7.

CE kept ticking along in the pure industrial market but around the time of WP7 we were exiting it and moving on to other things.

This problem has become a lot worse very recently. Basically, the UK was beholden to EU standards until late 2019.

There were concerns that leaving the EU would make it more difficult to obtain the chemistry needed to run the water treatment plants. True to their promise to slash red tape, one of the governments last year made it legal to dump untreated sewage in the rivers. I think this was supposed to be to absolve them if the necessary treatment chemicals were unavailable, but of course since it was cheaper not to treat the water properly, all the water companies started to skip the treatment process with happy abandon just to cut costs.

The previous Uno board was based around the AT328 chip, which does indeed have 2KB of RAM and 32KB of flash.

For things like displaying a clock or status info on an LED matrix, handling MIDI I/O for a synthesizer module or rotating a servo motor in 45 degree steps depending on what four I/O pins are doing, 2KB is plenty.

It's quite a laundry list, stuff like ensuring their competitor's software would fail on their OS (the whole "DOS ain't done 'till Lotus won't run" stuff, but they did that with Novell too), but here's a few off the top of my head if you want to know why some people might still be salty about Microsoft.

Embrace, Extend, Extinguish. Most of their misdeeds are centred around trying to ensure that Windows was the only viable system. Java and its potential to run anywhere as interpreted bytecode was a threat so they made their own incompatible fork, the whole Visual J++ thing (see wikipedia for details).

Likewise, Internet Explorer 6. Again, the possibility of being able to do everything in the browser and thereby allowing other OSes to the table was a threat, so they made that awful thing, tied it deep into Windows and ended up with a browser market share in the high 90 percent. At that point they had de-facto control over the web standards, but - and this is where you could argue they held back progress - as soon as they'd crushed all the competing browers, they disbanded the IE team and basically halted web standards there and then for a few years. PNG support for instance was held back for countless years because of IE.

The sheer inertia of IE6 made it quite difficult for other browsers like Firefox and even Chrome to get a foothold, even where I was working it was "Code it for IE because that's what ships with windows". Arguably it may never have been dislodged if it hadn't been for the smartphone market - and even Microsoft themselves found it a nightmare to end support for because they had got so many people using it's weird extensions (e.g. the entire South Korean banking sector).

The Halloween Documents.

One of the things which struck me as pretty underhanded - to the point of fraud - was the deal they had with OEMs selling PCs during the 90s. If you wanted to sell 1000 computers, you paid Microsoft for 1000 licenses, whether or not they had Microsoft software on them. Want to sell 200 with OS/2? That'll be 200 MSDOS licenses! Want to sell a line of SCO Unix machines? Linux? DR-DOS? Better pay up for that software you aren't even selling, or you won't get any Windows licenses at all, ever! This led to litigation and they were eventually forced to stop, but it had decimated the competition by that point.

More recently we had the business with the Open Office ODF format being made a government standard, to which Microsoft's response was to create their own rival "open" format which they called Office Open XML to confuse matters. This standard was something of a mess since they basically took binary blobs from Word etc and embedded them in the XML for certain object types and these were not documented.

However, where things took a dark turn is when they subverted the ISO organisation to try and get them to ratify it as a standard. They did this by stuffing it with new members, a flood of obscure countries which had never had ISO membership before, suddenly turned up, voted on OOXML, and then never voted on anything again. Because there were so many new "single-issue" members who lit up briefly to support Microsoft and then went dark, ISO was left paralysed and unable to ratify anything because they couldn't obtain a quorum until the one-shot members had expired.

Nearly all of these are things which were discussed on Slashdot at the time.

GoTek disk drives are quite popular for that kind of thing. Very handy for vintage samplers, workstations and similar keyboards from the 90s. Korg in particular store most of the preset sounds in RAM on machines like the Triton and Trinity, which means you have to reload them from disk when the internal battery is replaced.

There is a certain appeal to that, and yes, some people do still do this (Slugbug, for instance) but it tends to be the domain of audio nerds these days. However there's an interesting quirk about cutting vinyl. It has a crude data compression scheme where you can vary gaps between the grooves depending on how loud the audio is. Louder and bassier audio requires a wider groove wall, quieter audio does not. So you can pack more audio onto one side by varying the width depending on the content of the signal.

However, for this to work, you need to know what the audio is going to do one rotation ahead of the cutting stylus so it can be used to alter the groove wall spacing. The old way of doing this was to play back the master tape on a special deck with an extra playback head placed so that it delays the recording by 1.8 seconds or whatever the duration of a single rotation is. The more modern approach is to use a digital delay - which means that it's the digitised signal being cut on the record. (With the Slugbug album they disabled the variable spacing system and cut the album direct from tape)

Yeah, looks like it's one of those annoying differences in terminology. 'Rental' is sometimes used, but generally if you need some expensive equipment for a specific job here, it's called 'hiring' it.

"Plant hire" is specifically the rental of construction machinery - referred to for some reason as "plant machinery", which is why you occasionally get road signs saying "Heavy plant crossing". (What it's really trying to say is more like "Danger, bulldozers")

We may be seeing some difference between US and UK English. I don't think you could hire a shovel because it's cheap enough that you might as well buy it outright. But if you need a mini JCB or portable generator for a few days, you absolutely can hire one for the duration. In fact, I think the term you're looking for it "Plant Hire".

If this is not a valid expression in US English that's something I'll have to bear in mind for the future.

I always assumed it was cost-cutting. I bought one when they first came out and it worked so well that my Dad bought one some time afterwards. His drive, and all the subsequent drives we tried, kept getting the Click of Death, while my earlier unit just carried on working happily. Maybe I was just lucky, but my gut feeling was that they'd changed the design after the first run, and broken it somehow.

In some cases this is because they've secretly swapped out the TLC flash for QLC flash, in which case reliability is going to go down the pan as well as the data rate. This is definitely the case for the Crucial drive mentioned, and other articles point to ADATA doing this. If WD are playing that game as well, then yeah, I'd be worried.

It's nothing so petty. The layering system gets in the way of what ZFS and BTRFS are doing, which is why BTRFS also has to reinvent its own wheels as well, e.g. it has its own RAID implementation. The bitrot protection scheme on both systems works by checksumming the data and metadata, which requires it do understand the filesystem disk format. If the checksum mismatches the data on one disk, it then has to pull another copy from the mirror, which it can determine is correct because the checksum will match. That requires interacting with the RAID layer in ways which a generalised RAID system won't have.

Basically, more advanced filesystems have more advanced requirements that Linux' layering system can't provide at this point.