It's the end of the world as we know it, and we should feel fine

The death of Sir Clive Sinclair at the end of last week has seen an outburst of nostalgia. Understandably so. His glory days at the start of the 1980s marked the beginning of the digital revolution. He'd already put the first affordable calculators into pockets, and his computers were often the first in many households. They …

[Dan 55]

Does it work though?

There was a lot to make better in the early '80s. From RS232 to printer configuration, from moving files between computers to making the memory in your PC work, much of the experience of using computers was dire. Standards weren't. Software was late, buggy, and ugly. Computers wheezed if you asked them to go up stairs. You had to know what an IRQ number was, and why it mattered. We forget because it was all new and the next version would be faster, more colourful, less irksome.

The IT industry constantly promised that the next version would work. At first, it never did. Then it did a bit. Now it does.

With the early computers if they did something you didn't like or didn't do something you'd like them to do, you could write the software to bridge that gap. Nowadays there's no chance with Windows and MacOS. Fancy setting your own fonts and colour scheme? They stripped that out because marketing want a certain look, no hotdog stand for you, it's going to be low contrast greys to screw with your eyesight. Do you want to disable telemetry? No, you're having telemetry, it's good for you. Your old printer is going to stop working with our new OS because we say so and no you can't use the old device driver because we changed how they work and enforced device driver signing just to make sure. Your old computer is going to stop working with our new OS because we want you to have hardware with a TPM module, silicon shortage be damned. Or our new OS on our new architecture won't run unsigned apps that we haven't approved for our app store. And the constant patches because of the constant exploits. This is working, for small values of work.

Even Linux is looking a bit bloated these days, not everyone can build a kernel to strip out unnecessary stuff, and then there's systemd. You'd have to go to BSD to get something less bloated, more understandable, and something you're more in control of.

[Doctor Syntax]

Re: Does it work though?

"You'd have to go to BSD to get something less bloated, more understandable, and something you're more in control of."

Size, maybe but in terms of understandable and in control of, Devuan seems to fit the bill.

[J.G.Harston]

Re: Does it work though?

Over the weekend I started building a Wiki. My first thought on downloading the package was - wth??? Why so big? What can I strip out? How do I find what I can strip out? I've already got two Wikis, why can't the same single lump of code run all three with their separate datasets? Why do I have to upload an entire 'nother 80M? RT11 sorted one-code-multiple-data way back in 1974.

[Anonymous Coward]

Re: Does it work though?

my bugbear is foreign language packs. Nothing against Johnny Foreigner, but if I install an application and it asks me for English, please don't fill up my HDD with French*. German*, Italian, Korean, Russian, Spanish, Japanese, two versions of Chinese language files. I never going to have the urge to switch languages

* Especially these two - bad memories of a succession of hatchet faced women teachers at school with their irregular verbs and three words for "the" depending on what object you're talking about.

[Yet Another Anonymous coward]

Re: Does it work though?

And when you do try and remove the Korean font used only in 16 century court poetry, it says it's part of the desktop pack and removes your window manager

[Anonymous Coward]

Re: Does it work though?

Come to think of it - we had a HP LaserJet printer that a user kept fiddling with the settings on the control panel. Alas, we couldn't lock the control panel - next best thing was to use JetAdmin to change the printer language to Korean! That learn 'em!

[GrumpenKraut]

Re: Does it work though?

> ... irregular verbs and three words for "the" ...

Dass die der Teufel hole! (your German teacher).

[ICL1900-G3]

Re: Does it work though?

Ausgezeichnet..ich mag Deutsch sehr.

[TheFifth]

Re: Does it work though?

'three words for "the" depending on what object you're talking about'

I suggest you never try to learn Russian. There's no word for 'the', but instead the ending of almost every single word in the sentence changes depending on what object you're talking about.

And don't get me started on cases and the 16 different ways you can conjugate a verb...

[Mark 85]

Re: Does it work though? @Dan 55

Valid points you and the article make on things going obsolete or being "upgraded". As long as there's companies that need to make a profit we'll have these problems. Yes it's ridiculous, it's expensive, and for us in IT it's tedious as hell many times.

No answers here. Companies like Apple, et al, like the "here's a new shiney" even if there's no real improvement. There are those of us who some consider luddites because we don't run out every couple of years and buy new equipment, new software, etc. By "new" these things really aren't but have more stuff and bloat added.

I guess I'm one of those as still running a Gateway PC with Win 7. It works, it's no longer a prime attack surface for miscreants. For the things I do at home, it's all I need.

I'll await the downvotes from those who advocate for "new", "shiney", and insane corporate profits.

[adam 40]

Re: Does it work though? @Dan 55

Sometimes I think the OS vendors are in cahoots with the virus writers and malware peddlers, to maintain this conveyor belt of continual update/upgrades.

The bloatware needs bigger computers, which then allow more antivurus to run, more malware to hide in the cracks, and all the time nothing in user space runs any faster than before. And so it bloats even further until the next iteration.

The solution? Install something and never upgrade. Make OS stuff read only and keep your users out of it.

[Dave 126]

Re: Does it work though? @Dan 55

> Companies like Apple, et al, like the "here's a new shiney" even if there's no real improvement.

The iPhone 13 doesn't offer a great improvement over the 12, but it does over the iPhone 8 or 9. Most iPhone users won't buy a new phone each year. The same is true of Samsung's competing phones. How many years iPhone buyers go between new handset purchases varies of course. (The pictures of people queuing outside Apple stores likely do not fall near the centre of the bell curve, as I'm sure you good data-respecting people suspect)

So, it doesn't matter if a new model offers little over last *year's* model, only that it offers something over a user's last model.

[ChrisC]

Re: Does it work though? @Dan 55

That's one (optimistic) way of looking at it. The other (pessimistic) way is to consider that each time R&D release an incremental update, it gives sales & marketing another chance to turn the hype-o-meter up to 11, tell the world just how amazing the company is yet again, and drum up extra sales from the fatally infatuated who simply must have the latest version regardless. Come faithful, gather round, we've got an all NEW phone for you to desire. No, don't spend too much time comparing the specs against last years models, just LOOK at the shiny shiny, FEEL how nicely it sits in your hand, oh isn't it lovely, please buy it...

[DiViDeD]

Re: Does it work though? @Dan 55

TBH, that's truer than you suspect. I have several fanboi (and gurrl) friends and colleagues who have never missed a Jesus Phone model. Whether this is because, often they describe their new acquisition as 'fixing all the shitty things in the previous model', simply 'way better that the old model', or just regard it as an opportunity to sneer at those not sufficiently hip or wealthy to own the absolute latest shiney shiney, I leave as an exercise for the user.

That said, I just exchanged my Note 10+ 5G for a Z Fold 3, so glass houses, etc.

[KittenHuffer]

I just have to say ....

.... that nostalgia is not what it used to be!

[Doctor Syntax]

"But if we're not chasing the upgrade, what shall we do?"

An ever-lasting chasing of the industry's tail. Fixing the bits which were broken by the last fix of things which weren't broken followed by another bout of fixing things that aren't broken.

[DJV]

I suspect there's also a lot of "refusing to fix long standing bits because fixing them will break far too much other stuff that relies on them remaining unfixed".

As for Windows - I suspect there's now so much lipstick that the pig can no longer be found underneath it all.

[find users who cut cat tail]

If we stopped chasing the upgrade, I haven't noticed. Sure, the reason for upgrade used to be ‘it brings these nice things I need/want’, whereas now it is more ‘my old stuff is no longer working – again’.

But perhaps we are reaching some kind of equilibrium. I would not call it a good thing. It means about half of ‘improvements’ will make things worse. Probably more than half, knowing humanity and the second law of thermodynamics…

[Cuddles]

"the standards are settled"

Ha!

[Giles C]

Re: "the standards are settled"

Well if you are talking chat/conferencing software the standard is

There is no standard - everyone does it there own way

hence why we have zoom/teams/webex /slack other apps installed

[Henry Hallan]

Re: "the standards are settled"

...And Twits and Failbooks and Tics and all sorts.

We will not have arrived until there is an open standard for IM, filesharing, online meeting and blogging - and the SM companies are legally forced to use it.

[Greybearded old scrote]

Re: "the standards are settled"

There was one. Then the service providers realised that interoperability is bad for lock-in.

Guess which won.

[Anonymous Coward]

Re: "the standards are settled"

Obligatory XKCD:

https://xkcd.com/927/

[Version 1.0]

Re: "the standards are settled"

A few years ago, Friday, October 14 was World Standards Day. Or, at least, it was World Standards Day in *some* countries. However, in America, the celebrations were held on October 11th. In Finland, World Standards Day was marked on October 13th. Italy planned a separate conference on standards for October 18th. - an ASR sig back in the old days.

[Will Godfrey]

Thought Experiment

I sometimes wonder what sort of performance you could get from a single ARM2-like core, but on modern die sizes and with minimal single thread software.

[Yet Another Anonymous coward]

Re: Thought Experiment

Run RISC-OS on a modern Pi - it flies. And I don't remember it being a slouch on the original Archimedes

[werdsmith]

Re: Thought Experiment

RISC-OS flies on any pi, not just modern ones. It’s fine on a Zero.

This is why I love Pi, it’s the cheap thing made for playing and learning and while it doesn’t have the mystery of the 1980~ price breakthrough micros, because we are all so used to tech now, it’s still addictively fun and broadly scoped in where it can be applied. For me anyway, nearest thing there is to the excitement of the early days.

[ICL1900-G3]

Raspberry Pi

And, though it might not be the computer most people want, it is probably the one that I large number would manage with very happily.

[_andrew]

Re: Thought Experiment

ARM2-like cores are still everywhere: the smaller Cortex-M series of microcontrollers are not too dissimilar. (Thumb2 instruction set there though, not the old ARM32 or ARM26 instructions, so you couldn't just run an old RISC-OS image.)

Those are now found in things so small that you would not even have thought of it, back in Clive Sinclair's heyday. Things like earbuds, passively-powered sensors, credit cards, ...

[Peter Prof Fox]

Some rotten foundations and missing bits

No. We don't have everything we need. Not by a long chalk. Pioneers and innovators are still needed. It's not a matter of polishing the edges and adding some shiny trim.

* How about date/times that understand 'Not yet' and 'September 2021' or '31st January 2020 plus one month'. (https://vulpeculox.net/day/index.htm)

* How about instant access to cells in spreadsheet files as if it was a database?

* How about single key accents and special characters. i.e. using the same key for all modifications? (http://vulpeculox.net/ax/index.htm)

* How about backup software that anyone can use reliably?

* How about a calculator that spots errors by 'understanding' the compute?

* How about database sanitising that management can understand and businesses can implement?

* How about weeding the wrong and obsolete technical 'documentation' and trouble-shooting pages sloshing around the web?

* How about web search engines that allow you to exclude aggregation and recommendation sites. For example look for a place only to be swamped by poorly curated junk or Trip Advisor twaddle or properties or 'news'. If I want say 'electrical supplies' then I'm perfectly capable of typing that in. Search Engine - he say you can have all the other junk because we think it's popular.

It's great we have stuff that mostly 'just works'. It's sad that the ace minds of yesteryear have been replaced by the 'anyone can do it' brigade. eg Track-n-trace. There's very little quality in IT projects. Planning? Experience? Knowledge? Problem-solving at the design stage? Bring back limited resources where you had to work really hard in your head to get the essentials into the system.

[Anonymous Coward]

Re: web search engines that allow you to exclude aggregation and recommendation sites

but this is just a side effect (or, arguably, 100% mainstream output effect) of 'monetization approach' that was taken, perhaps, 'naturally' as a role model. Internet is business, and nobody expects Spanish Inquisition to bring in free cushions. But those soft, FREE* cushions that fuck you 'pain-free'... well, look around.

[Anonymous Coward]

Re: How about backup software that anyone can use reliably?

not enough demand. No, I'm being serious. There's little demand, so there is only 'backup software'. If there was significant demand, there would be 'backup software that anyone can use reliably'. On the other hand... scrap that, there's plenty of demand for all kinds of software that can be used reliably, but the software that is there, is only reliable ENOUGH. Perhaps that's the key, than an 'x' doesn't need to be reliable, just 'good enough'?

[_andrew]

Re: How about backup software that anyone can use reliably?

The industry's response to the backup problem has been to keep the user's data safe, somewhere away from the users and their unreliable computers. In the cloud. (All of the other issues that go along with that move come down to differences of policy and contract terms.) Buy a new phone these days and getting your old stuff back doesn't involve restoring from a backup: it involves logging in.

[Denarius]

Re: How about backup software that anyone can use reliably?

Wrong answer IMHO. How about backups that happen without the user being aware ? I know, Google/M$/Faeces slurp. However, after trying to recover data from mangled disks because of HW failure, OS failure or plain malware, I see the problem of users (speaking of WFH and home users here) feeling the need for backups to a common portable hard drive that is normally offline. As for corporates, am only too aware of major Oz site I used to work where all the older staff (ie over 24) were made redundant and cheap newbies brought in. They watched the main Windows backup hardware controllers fail and ignored the warnings because no-one knew of their significance. So problem is manglement are not being made to be afraid, very afraid of not mandating and testing quick backups and restores to be achievable. Something for the stockholders to chant and company AGMs perhaps.

Lastly, OS that allow any user data anywhere near the OS disk has to be made illegal under consumer protection laws. At least unix derived OS try to keep user files away from system areas. Most of us have seen /tmp or /var/tmp fill up due to inadequate use of disk quotas, or worse, crappy software writing logs into wrong places like /etc or going feral. Looking at you big colour TLA. Thousands of empty files per day in software working allocation.

In short, still much decluttering to be done in software standards for config. binaries, working space, data storage and logs with automatic log control.

[Denarius]

Re: How about backup software that anyone can use reliably?

sometimes, if one keep the same brand. Why cant I just use phone backup to microSD card on new Android 10 phone. It might be hidden somewhere but all I can see under backup is a Slurp. On Android 7 I could do that.

[Anonymous Coward]

Re: How about weeding the wrong and obsolete technical 'documentation'

sorry, too much shit already. Humans just love to leave their junk around, be it the internets, neighbourhood, or outer space. They keep shitting, and the best they can muster is stay on top and pretend they're not on top of a pile of shit, but in a sparkling castle.

Another argument re. weeding: no need. Clearly, if there was a business 'need', or if a business saw this as a 'need' from their perspective (say, if too many people say: you know MS, your castle sits on a pile of shit, we'd rather go to one that stinks less), then a business might consider serious weeding. As it stands, people don't say that, they don't mind that shit, or look away, so clearly, there is no business incentive to clear it. Sure, they clear the biggest, stinkiest, steaming piles, but essentially, most people don't care about 'sparking clean', as long as they, themselves, don't step on shit too often.

[Doctor Syntax]

Re: How about weeding the wrong and obsolete technical 'documentation'

At best "weeding" isn't going to bring a RoI in sufficient time. In reality we all know what happens if you chuck out old documentation. Murphy arrives bearing a slightly broken piece of kit you thought had been chucked out but is (a) indispensable and (b) irreplaceable.

[Doctor Syntax]

Re: Some rotten foundations and missing bits

For dates, add in the likes of "Friday in the octave of Easter 1312".

But "instant access to cells in spreadsheet files as if it was a database?" - isn't use of spreadsheets as databases part of the problem?

[Anonymous Coward]

Re: Some rotten foundations and missing bits

> * How about backup software that anyone can use reliably?

Only people who have the requisite hardware can perform reliable backups, even if they are geniuses and the software is faultless. To be thorough, the user would also need several remote secure premises on which to keep their hardware - though a garden shed would at least be a good start, offering some mitigation against a house fire.

My point being: software is only a part of a backup *system*. What we mean by 'system' can be very broad.

I used to wish it was the law that all new PCs came with a second (or better yet, array of) HDD that, whilst invisible to the average user, performed backups in the background. When the hapless user inevitably calls on you help fix their computer, you restore files or images as appropriate. I know it's an unrealistic idea on a few levels, but hey!

[Dagg]

Re: Some rotten foundations and missing bits

The one problem with all of this are humans. Humans are stupid, most of the time they don't even know what they want!

Just look at how many times humans can stuff themselves up with no help from anything else.

Your comment about the "ace minds of yesteryear" doesn't cut it. I started in this game in the 70's and even then we had "jokes" about the miracle developer that do anything and get it all working. And the punchline was "You gave me what I asked for, but not what I wanted"

Your comment How about a calculator that spots errors by 'understanding' the compute? well we have something similar with text auto correct for spelling mistakes. And how many times does the auto correct still stuff up.

If you really want to see what you are up against work on a help desk...

[Crypto Monad]

You used to be able to get interesting applications that were 10KB in code size. Now try finding any useful application which is less than 10MB.

GUIs? The original Macintosh had a well-designed GUI with most of the OS (QuickDraw) in 128KB of ROM, and 128KB of RAM to run applications, and a 400KB floppy drive. Multiply those figures up by a factor of 1,000 or more today.

In short: the *complexity* burden we carry today is enormous. The chances of any individual understanding the whole system end-to-end is virtually zero. But for those people prepared to put in the effort (i.e. hackers), the holes waiting to be found are everywhere.

On top of that - performance gains have been frittered away. Computers used to have clock speeds measured in MHz, now it's GHz. However, whereas computers used to boot in 5 seconds, it's now common for a server in a data centre to take 5 minutes. (Spoken by someone who has suffered the pain of repeated rebooting to find which BIOS setting needs to be tweaked to make the OS recognise its system disk)

Those who think we live in a golden age of compatibility are probably working with Virtual Machines. The actual hardware underneath is as terrible as ever.

[Electronics'R'Us]

Hardware

The actual hardware underneath is as terrible as ever.

I would argue that the underlying hardware is extremely capable but every advance in hardware capability has resulted in ever more code to do the same (and often, in defence of it, improved) task but the increase in computational resources used never seemed to be matched by an equivalent increase in functionality.

I started out on 8 bit machines with very limited resources and I remember well adding a test for a particular product (it tested a 32K flash memory if fitted - they were very expensive at the time).

The test added a grand total of 33 bytes and ran very quickly (walking 1s and 0s on both address and data bus).

[Anonymous Coward]

Re: Hardware

Any evolved software has re-inventions of functions because people didn't realise they existed already. Even I am sometimes surprised that something I have built actually caters for a condition which has only now become a requirement.

[J.G.Harston]

Indeed, I've been looking at an integer-to-decimal routine I've written, and am annoyed it wastes 40 bytes on a table of power of ten on top of the 40 bytes of code, and have been twisting my mind to try and get rid of the table. I think I've managed to dump the table and replace the divide-by-tens with multiply-by-fives-and-shift and push the digits backwards onto the stack.

[ICL1900-G3]

The joy of making an efficient assembler routine...I do so miss it. I bought an editor for the Atari ST a long time ago. The author said it was fast and efficient because he'd written it in assembler; he loved writing in assembler, but it was not commercially viable, but he'd produced this just for fun.

[Lotaresco]

Sinclair and the SSD

It's one thing I haven't seen any journalist mention is that back in 1989 Sinclair pioneered solid state storage with the Anamartic SSD. Initially 40MB but soon 160MB of storage. Which may sound piffly today but at the time my computer was a Mac SE/30 8/80 - 8MB of RAM and 80MB HD. Friends with PC/ATs and IBM PS/2s looked at the Mac and declared that no one would ever need that much storage! So Anamartic's product was cutting edge.

Sinclair had that rare thing, vision. He didn't necessarily know how to get to where he wanted to be and his ambition exceeded his talent more than once, but if he'd been in Silicon Valley rather than Cambridge, I suspect he'd have been a multi billionaire and Sinclair would be a name alongside Microsoft, Apple, Tesla.

We also shouldn't forget that the reason we had Acorn, the BBC micro, and ultimately ARM is that Sinclair ticked Chris Curry off enough that he left Science of Cambridge (the precursor of Sinclair Radionics) and created Cambridge Processor Unit Ltd which became Acorn.

[Yet Another Anonymous coward]

Re: Sinclair and the SSD

We wouldn't have any IT industry if it wasn't for management pissing off techies who then formed much more successful competition.

[Anonymous Coward]

Re: Sinclair and the SSD

Eleven grand for 40MB! Maybe that's why it didn't take off?

[Admiral Grace Hopper]

Re: Sinclair and the SSD

Not long after that ICL used us a testbed for a 1 GB solid state mainframe storage device. That thing flew! We had a dual-processor 3900 processor bound, which I'd never seen before. When we asked how much it would be if we wanted to but rather than just play around with it for a while the number quoted was Very Large Indeed.

[ICL1900-G3]

Re: Sinclair and the SSD

A truly lovely mainframe.

[Gene Cash]

Re: Sinclair and the SSD

The same thing happened in Silicon Valley: Shockley left Bell to start Shockley Semiconductor Labs in a disagreement over the handling of the invention of the bipolar transistor then As a result of Shockley's abusive management style, eight engineers left the company to form Fairchild Semiconductor; Shockley referred to them as the "traitorous eight". Two of the original employees of Fairchild Semiconductor, Robert Noyce and Gordon Moore, would go on to found Intel. according to Wikipedia.

[werdsmith]

Re: Sinclair and the SSD

but if he'd been in Silicon Valley rather than Cambridge, I suspect he'd have been a multi billionaire and Sinclair would be a name alongside Microsoft, Apple, Tesla.

We also shouldn't forget that the reason we had Acorn, the BBC micro, and ultimately ARM is that Sinclair ticked Chris Curry off enough that he left Science of Cambridge (the precursor of Sinclair Radionics) and created Cambridge Processor Unit Ltd which became Acorn.

And if Acorn had been in Silicon Valley instead of Cambridge? The Acorn Market Hill barely noticeable door entrance was a few strides away from Clive’s Kings Parade rooms above a shop. They went off the same start line at the Baron of Beef.

To look at Fulbourn Road campus now, and compare legacies.

[Anonymous Coward]

"Software was late, buggy, and ugly"

Some things haven't changed then.

If anything, its worse now. Back then, updates were hard. You had to find a way to distribute them.

Now its just "release it broken and patch it later."

[Shak]

Self-sabotage?

I've convinced that "most of the problems have been solved" is why software folk create new programming languages and frameworks to stave away boredom.

[werdsmith]

Re: Self-sabotage?

I might not choose C++ to manipulate a dataset, but the language and libraries I do choose might well be written in C++ and C++ might be written in C.

[Electronics'R'Us]

Most of the problems have been solved

Much the same was said of physics just before 1900 until Max Planck uncovered quantum mechanics.

Some of the problems uncovered then are still mysteries to us and I suspect the same will be true in computing.

As for standards, many exist (some might say too many) but they are always getting updated with the added complexity (in a hardware world at least) with the requirement of backward compatibility.

A PCI express gen 4 switch for example has to be able to connect to any of gen 1, 2, 3 and 4 (and the differences between 1/2 and 3/4 are significant).

I remember the quote that ISDN was a standard in search of itself and many newer ones seem to meet that description.

The problems still to be solved might not be recognised as problems yet, just as happened with quantum mechanics.

[Yet Another Anonymous coward]

Re: Most of the problems have been solved

So the 'solution' to today's IT woes is going to be the equivalent of what QM did to physics - oh joy

[Doctor Syntax]

Re: Most of the problems have been solved

"the differences between 1/2 and 3/4 are significant"

That sort of thing is the problem. if Xv3 is so different to Xv2 it really shouldn't be Xv3 at all. It should be Yv1. There may still be a market for X to be supplied for some time to come but without the demand that Y be back compatible with it.

[Anonymous Coward]

Where are the Tee-shirts ???

"All the drivers for new, exciting experiences that the Children of Clive have come to accept as intrinsic to their IT lives are running out of steam."

We need to have these tee-shirts now !!! :)

I would buy one, if available !!!

RIP Sir Clive Sinclair.

[Yet Another Anonymous coward]

Re: Where are the Tee-shirts ???

Never mind the cult of Mac, Children of Clive / Hordes of Specy sounds cooler

Although a Commodore Barbarian myself

[theDeathOfRats]

Re: Where are the Tee-shirts ???

"Never mind the cult of Mac, Children of Clive / Hordes of Specy sounds cooler"

Totally agree, Mr Commodore Barbarian.

Dragon Rider here.

[Yet Another Anonymous coward]

Re: Where are the Tee-shirts ???

What are we calling Oric owners ?

[PerlyKing]

Re: Oric owners

Owners? ;-)

I once saw a Jupiter Ace (owned by an Afficianado Ace?) but I don't think I ever saw an Oric in the wild.

RIP Sir Clive.

[werdsmith]

Re: Where are the Tee-shirts ???

Many of us Children of Clive had a 6502 dalliance on the way up, but ZX was always the springboard.

[Jedit]

We shouldn't feel fine

And the reasons why are spelled out in the article. Apple are producing new phones that do nothing the existing phones can't. As the executive says in Tron Legacy when Alan Bradley asks what the differences are between the new version of ENCOM's OS and the old one: "We added 1 to the number". And they don't last, by design so you have to buy the new phone. 80s computers may be obsolete, but many of them still work. How many iPhone 13s will still work in 20, 30 years, even if it's just as an iPod because the phone functionality has moved on (for which Apple can't be blamed)? We already know: the answer is "none", because due to planned obsolescence none of them are likely to work in ten years. And it all uses resources that can't be replaced.

We're spinning our wheels in pursuit of maintaining an unsustainable status quo because the status quo makes money. No, I can't feel fine about that.

[doublelayer]

Re: We shouldn't feel fine

You're broadly correct, but the specifics aren't:

"80s computers may be obsolete, but many of them still work."

In the sense that they turn on, maybe. But that's if you kept them in relatively good condition and kept the boot disks as well. If you didn't have all the parts in storage, they wouldn't work very well, and there are lots of pieces that wouldn't work. Try getting a computer to display anything if you didn't also keep a compatible analog TV, with either the right RF or analog input. Or to produce a disk if you didn't already have one.

"How many iPhone 13s will still work in 20, 30 years, even if it's just as an iPod because the phone functionality has moved on (for which Apple can't be blamed)?"

Basically the same amount. If you put it in storage, you can take it out in decades and turn it on and it will be sort of fine. The battery is the only part that is likely to die, and in twenty years it won't be in great condition, but it will still boot up. You will certainly have lots of problems with it, but they parallel the problems you would have with an 80s computer. You probably couldn't access Apple's app store, so you couldn't install things that weren't already installed, similar to how you couldn't easily load new software on an 80s computer without rebuilding the hardware to load it onto media. You could get the old XCode to run so you could build from source, but only if you're willing to virtualize the old Mac OS, just like you still can write in Z80 assembly and have that work. You probably couldn't buy replacement parts for your old phone, just like how you couldn't replace a chip on your 80s computer should it decide to release the magic smoke. In fact, if there's any difference that jumps out at me, it is that you have a greater chance of the magnetic storage from the 80s computer degrading without external damage than the flash in the relatively sealed phone, making the phone more likely to have the software it started with.

Phones don't last very long, and they're not designed to, but they would keep working if you preserved one. For the same reason, a lot of old computers didn't get preserved and ran their last halt instruction back in the decade when they were bought. Your preservation of 80s equipment doesn't make it particularly resilient.

[Dan 55]

Re: We shouldn't feel fine

Try getting a computer to display anything if you didn't also keep a compatible analog TV, with either the right RF or analog input. Or to produce a disk if you didn't already have one.

RF/composite/SCART as an output still works if the screen has the equivalent input, and most usually have the first two. RGB component can be converted to YPbPr with a dongle.

Your preservation of 80s equipment doesn't make it particularly resilient.

If you turned on an iPhone a decade from now, it wouldn't be able to talk with the different services at Apple's end and would probably be pretty useless as anything other than a simple phone... that's if it could still verify with the mothership if it's stolen or not. If your iPhone decided to release magic smoke, there's little you could do to fix it.

[doublelayer]

Re: We shouldn't feel fine

"If you turned on an iPhone a decade from now, it wouldn't be able to talk with the different services at Apple's end and would probably be pretty useless as anything other than a simple phone... that's if it could still verify with the mothership if it's stolen or not. If your iPhone decided to release magic smoke, there's little you could do to fix it."

I said much the same myself, but in fact, you're not entirely correct. I have an 11-year-old iPhone here. It doesn't even have the benefit of having been preserved as it's one that a friend discarded five years ago and I couldn't find anyone who wanted it. It's not very useful, but it does actually connect with Apple's services. I can get apps still. I can use their messaging service on it. I can sync with it. It all still works as well as things can when you only have 256 MB of RAM trying to run an app written for much more.

[werdsmith]

Re: We shouldn't feel fine

There are available replacements for RF tuners and expansion packs like ZXpand that do the job of replacing tape loaders with SD cards. I run a ZX81 version 1, with a ZXpand and replacement tuner.

Replacement parts are out there too.

Old phones will see limited use when 2G spectrum goes, but 80s computers go on working because they don’t depend on external services.

[Fading]

Re: We shouldn't feel fine

All my Amstrad CPC 6128 needed was a rubber band to replace the perished one in the 3" disc drive (note: 3 inch floppy) and a bit of a clean up (nothing Isopropyl couldn't handle). Even the multiface II was still functional. Cleaned it up over Christmas and had some retro fun. No batteries to worry about in those old 8-bits. Amstrad's may have been all about cutting corners and "all on one plug" but the all bits still work.

Note - it wasn't in "storage" just a cardboard box in the wardrobe. Apologies for mentioning the hated Alan Sugar Trading's name on a Sir Clive Thread.

[Mike 137]

"now we've caught up and we're puzzled about what to do next"

How about working on producing code that isn't still littered with bugs after all the years we've had to practice coding?

As an engineering product, software is the worst. No other branch of engineering would allow constant fixing of flaws in design and implementation for the entire life of a product, and in some cases you'd be prosecuted if you tried it on.

[Yet Another Anonymous coward]

Re: "now we've caught up and we're puzzled about what to do next"

Yes but we would also still be on Analytical Engines and arguing about which brass alloys IEEE approved for use in computers.

You can have reliable computing, stick to BSD and only use Awk + Tex

[Boris the Cockroach]

Re: "now we've caught up and we're puzzled about what to do next"

Come on down to the land of real engineering

You pump out buggy untested code you'll be the one getting a 10lb cutter lobbed at you at 8000 rpm...

Which is very expensive

Especially when you consider that everyone nearby will need clean underwear.

Yes I do put bugs in the code, no way I can avoid it, however we test, and then we test and just to be sure we test again, all in slow motion.(my common fault is failure to retract tooling clear of the clamping before moving onto the next clamp station.... thats very loud

But as you said, you pump out a buggy pile of crap in an OS or a game or phone app and an adult can send an update out the next day. this is not engineering as it is known... its lazy and rushed coding with no engineering behind it. take the money and run springs to mind and stuff the consumer rights acts by a 47 page EULA that claims you have no claim if it causes the computer to explode with the force of 20 tons of TNT.

[Prst. V.Jeltz]

We've fixed the problems. Not all of them, but the standards are settled, the memory and compute are more than adequate, the graphics as good as our eyes, and the software does what it's told. Like the dog chasing the car, now we've caught up and we're puzzled about what to do next.

Well , looks like we've finished!

hurrah , everyone can have the afternoons off now

[Big_Boomer]

Machine Learning

Thank you for using the term Machine Learning and not the term AI that the marketeers are still incorrectly spouting.

For the obsolescence problem, that always came from the crazy upgrade cycle. Now that a level of sanity has returned to the phone markets I would expect a "quality" phone to be good for at least 5 years, but don't expect too much if you buy a budget phone. The same holds for PCs and other electronics goods. I currently have a 4 year old Honor 9 phone that does everything I need, my desktop PC is a 6 year old i7 Haswell with 16GB, GTX960 4GB, and a 2TB SSD + 2TB HDD so is quite capable for my needs. My TV is a 10 year old 32" Samsung from before the SmartTV era and works perfectly. All of this will get replaced when I feel the need, not when some manufacturer tells me to. For the phone & TV it will probably be when they die. For the PC I could be persuaded if the specs for GTA6 or Fallout5 are above what the current PC is capable of, but even then, I might just upgrade components.

[Doctor Syntax]

Re: Machine Learning

Now that a level of sanity has returned to the phone markets I would expect a "quality" phone to be good for at least 5 years

It depends on what you mean by a quality phone but I suspect Jedit's post is closer to the mark. A quality phone that lasts 5 years isn't going to bring continuing profits.

[Yet Another Anonymous coward]

Re: Machine Learning

> A quality phone that lasts 5 years isn't going to bring continuing profits.

Apple profit margin on iPhone 5-10% ?

Apple profit margin on 20% app store commision = 20%

Apple profit margin on you streaming content from Apple = 99.99999%

[veti]

Re: Machine Learning

I don't believe I've ever had a phone I used for less than five years. Typing this on my second-ever Android. And both of them have cost less than one third of the price of a contemporary iPhone.

[adam 40]

Re: Machine Learning

My Sony Z1 is still going strong at 8 years old. Just replaced the screen though as the touch surface was wearing out a bit.

Replaceable battery, LTE, and waterproof, what more do you need?

As the OS is getting a bit slow now (I suspect app bloatage) I might go over to LineageOS.

[Filippo]

Some things work

It's true that some things that did not work in the 80s now work. Generally speaking, even Windows now does not crash unless there's a hardware problem, or some exceptionally poor third party device driver. You also can just plug in most peripherals and they'll either work correctly all by themselves, or work after you perform a reasonably straightforward driver install. This state of affairs is a major achievement, which is not to be underestimated.

That's a far, far, far cry from thinking that IT is a solved problem.

Networking has gotten better, but it's still far from "just works", except for very simple configurations.

WiFi and Bluetooth are still horribly finnicky, even when the signal should be loud and clear.

Every single aspect of security is an unholy mess.

The fundamental problems of GUI development remain unresolved; frameworks proliferate as everyone tries again and again to solve the same problems, finding that this only raises other problems, in an endless game of whack-a-mole.

Modern hardware gave us orders of magnitude more resources, which we've invested in creating more and/or less efficient layers of abstraction. This has made development easier, but not very much.

Targeting multiple platforms is still a major problem. If you're doing web development, it's less of a problem - except that web development itself is as messy as any three different platforms put together.

And that's just the technical issues. I don't even want to get into the social and economical issues.

[Anonymous Coward]

Re: Some things work

"Generally speaking, even Windows now does not crash unless there's a hardware problem, or some exceptionally poor third party device driver. "

..or a forced update that hasn't been comprehensively tested.

[steviebuk]

Also forgot

You no longer own anything. Sir Sinclair never asked you to pay for a yearly subscription for software and he gave you the option to buy it in kit form and make it yourself.

Now, you're locked into subscriptions and you no longer own the kit you've purchased. Open and iPhone and expect and knock on the door from CrApple as they moan "Don't open our kit. You've now voided the warranty. And pointless as you can't fix it as we don't supply parts or schematics cause we're cunts"

[Version 1.0]

Excellent story

While I can be pissed off with things today, I can remember my dad telling me that calculators sucked and when numbers were thrown at us he always beat my calculator results with his foot long slide rule. So Rupert Goodwins - Cheers for describing the world we all live in, whether we like it or not.

[Anonymous Coward]

We have gone past the plateau stage - and are now on a downward slope. Increasing "simplification" just makes it harder to sort out any underlying problems. A bit like cars - so much embedded complexity that you can't do much yourself apart from fill the washer bottle.

[Webmannie]

Running a Nuclear Plant

As one of the first to get a 16k Spectrum my delight was short lived as it had to be returned due to a fault, but boy did that shape my future when i got it back. Leaving school with 1 O grade in Arithmetic, little did i know how that computer would shape my future career, more than any schooling. Set up a computer club, got published by Sinclair User (Shopping List program, wish i'd patented the concept) set up a web design company whilst working on shifts on a Nuclear Site. On-site, got involved in Y2k project 'because i was into computers', trained to be a SCADA system engineer, became a member of Safety Critical Systems club, travelled on special projects for IAEA, presented at IAEA..

But the best bit? whilst clearing out stuff, i came across the original ZX81 that Sir Clive claimed to be 'running a nuclear plant', backed up by a purchase order to Clive from DNPDE (Dounreay Nuclear Power Development Establishment) and this was checked with Sir Clive by a mutual friend..i kept it safe but unfortunately it got 'skipped' when i left UKAEA to work for the company (Navertech) i formed back while working on shifts..best move thing i've ever done.

RIP Sir Clive, forever grateful to you for proving my teachers wrong!!