'We Need Software Updates Forever' (ieee.org) 264
The creator of the Virtual Reality Modeling Language (VRML) and founder of the first virtual reality startup, Mark Pesce, opines an IEEE Spectrum piece that we need software updates forever. Slashdot reader joshuark shares an excerpt from the article: Device makers are apt to drop support for old gadgets faster than the gadgets themselves wear out. Consumers have relied on the good graces of device makers to keep our gadget firmware and software secure and up-to-date. Doing so costs the manufacturer some of its profits. As a result, many of them are apt to drop support for old gadgets faster than the gadgets themselves wear out. This corporate stinginess consigns far too many of our devices to the trash heap before they have exhausted their usability. That's bad for consumers and bad for the planet. It needs to stop.
We have seen a global right-to-repair movement emerge from maker communities and start to influence public policy around such things as the availability of spare parts. I'd argue that there should be a parallel right-to-maintain movement. We should mandate that device manufacturers set aside a portion of the purchase price of a gadget to support ongoing software maintenance, forcing them to budget for a future they'd rather ignore. Or maybe they aren't ignoring the future so much as trying to manage it by speeding up product obsolescence, because it typically sparks another purchase.
Does this mean Sony and others should still be supporting products nearly two decades old, like my PSP? If that keeps them out of the landfill, I'd say yes: The benefits easily outweigh the costs. The devilish details come in decisions about who should bear those costs. But even if they fell wholly on the purchaser, consumers would, I suspect, be willing to pay a few dollars more for a gadget if that meant reliable access to software for it -- indefinitely. Yes, we all want shiny new toys -- and we'll have plenty of them -- but we shouldn't build that future atop the prematurely discarded remains of our electronic past.
We have seen a global right-to-repair movement emerge from maker communities and start to influence public policy around such things as the availability of spare parts. I'd argue that there should be a parallel right-to-maintain movement. We should mandate that device manufacturers set aside a portion of the purchase price of a gadget to support ongoing software maintenance, forcing them to budget for a future they'd rather ignore. Or maybe they aren't ignoring the future so much as trying to manage it by speeding up product obsolescence, because it typically sparks another purchase.
Does this mean Sony and others should still be supporting products nearly two decades old, like my PSP? If that keeps them out of the landfill, I'd say yes: The benefits easily outweigh the costs. The devilish details come in decisions about who should bear those costs. But even if they fell wholly on the purchaser, consumers would, I suspect, be willing to pay a few dollars more for a gadget if that meant reliable access to software for it -- indefinitely. Yes, we all want shiny new toys -- and we'll have plenty of them -- but we shouldn't build that future atop the prematurely discarded remains of our electronic past.
Open-source lifetime support. (Score:5, Interesting)
And the open-source movement will be the trend setters with support as long as it stays out of the landfill. I don't think they'll mind.
Re: (Score:2)
Open Source support is highly dependent on the size of people developing it. A software suite like, say, Apache Webserver will have continued support forever while projects that hinge on one or two people will very likely die with them.
Re: (Score:3)
When a manufacturer ceases to support a product that incorporates software of any kind, he shou
Re:Open-source lifetime support. (Score:4, Insightful)
Force open sourcing on dropped products (Score:5, Interesting)
There's a very simple solution :
Force open sourcing on dropped products.
You stop supporting a product ? Then open source it, by law.
There is no downside to this.
Re:Force open sourcing on dropped products (Score:5, Informative)
Even if you could get such a law passed, you run into the problem where a product sometimes licenses IP from other companies. When Be Inc. closed its doors, it couldn't open source BeOS because some fundamental parts of the OS used software licensed from other vendors. It's been a loooonng time, so I don't remember the details, but I remember this was a huge roadblock to open sourcing the OS.
Mind you, I don't think Be Inc. ever had any intention to do it anyway. They had debts to pay and had a duty to their investors to try to sell what they could to the highest bidder. And now that I mention it, the requirement to sell IP to pay back investors would be another roadblock to your idea.
Re: (Score:3)
When change the IP law then.
Either the people who licenced their IP to the deceased vendor get to maintain the zombie, or their IP gets to be public domain.
"Simples" as the Meerkats like to say.
Re: (Score:3)
So let me get this straight. If you license a library to a company, and that company goes under, YOU are willing to support the full software stack of that company? I really doubt that.
Re: (Score:3)
Nope.
You could support your binary file of your library.
Users would have to take that and compile it in the source of the product.
Re: (Score:3)
What's the difference between open-sourced abandonware and closed-source abandonware?
That's a serious question: If it's been abandoned it doesn't matter if it's open source or closed source, you're up shit creek either way. Sure, in theory you could spend six months figuring out what some piece of custom embedded firmware does, try and set up a build environment where you can compile it, try and figure out how to load it into the target device, and try and figure out why it's not working as you expect it t
Re:Force open sourcing on dropped products (Score:5, Insightful)
What's the difference between open-sourced abandonware and closed-source abandonware?
Source code availability. Now ask a hard one.
That's a serious question: If it's been abandoned it doesn't matter if it's open source or closed source, you're up shit creek either way.
It matters very much, and while you may be on the same creek either way, at least one way you have a paddle.
Sure, in theory you could spend six months figuring out what some piece of custom embedded firmware does, try and set up a build environment where you can compile it, try and figure out how to load it into the target device, and try and figure out why it's not working as you expect it to, but in practice it doesn't matter if the abandonware is open-source or not, once the original creator loses interest, you're screwed.
People have figured out solutions with even less to go on before. Your pessimism is unwarranted.
Commercial stuff actually has an advantage there because the creator is being paid money to continue taking an interest in it. For non-commercial stuff, you're at the whim of the creator, once they get bored with it, you're on your own. Of all the abandoned devices I've got, I'd say about three quarters were open source, and in no cases is the availability of the code useful.
Can you provide any examples?
Not just that... (Score:5, Informative)
...we also need a government-appointed advocate who will argue on our behalf that something really is a bug and needs to be fixed. Otherwise it's the consumer's word against the device manufacturer's, and who do you think usually wins that argument?
Open source by law. (Score:2)
There's a very simple solution :
Force open sourcing on dropped products.
You stop supporting a product ? Then open source it, by law.
Re: (Score:2)
Make it both civil and criminal. If the feds don't force the issue, the plaintiffs will.
Re: (Score:2)
Re: (Score:3)
Serious? I have products from China and Europe that are discontinued and are no longer receiving support.
The idea that if you sell something you are required to support it FOREVER is insane. And the fact that ya'll want the government to force that is about as anti-freedom as one can get.
Re: (Score:3, Funny)
I think that's just called a democratic government.
The USA should get one sometime. ;)
Hold on! A "democratic government" is bought and paid for every 2-6 years (depending on the particular component). Too bad it comes as-is with no warranty or SLA.
End of support must come with end of copyright (Score:5, Insightful)
All unsupported software and hardware must be put into the public domain so it can be supported by other means.
Re:End of support must come with end of copyright (Score:5, Insightful)
This is a much better solution than the one in the story... literally supporting something forever would mean supporting DOS 5.0 and Windows 3.5 Does anyone really think that is a possibility? The problem with this is making windows 3.5 open source would reveal much of the codebase of the software that came after it... so companies would need to be certain they were 100% confident in the security of their living software before releasing code for the dead versions. (this is desirable in my opinion, but also impossible, so it will never be done)
Re:End of support must come with end of copyright (Score:5, Interesting)
Who said anything about open sourcing it? Just making the binary DOS 5.0 and Windows 3.1 public domain would enable third parties to distribute binary patches without fear of getting sued to death by Microsoft. Sure, it would be nice to have open source code, but public domain binaries would still be a lot better than nothing.
Re:End of support must come with end of copyright (Score:5, Interesting)
Who said anything about open sourcing it? Just making the binary DOS 5.0 and Windows 3.1 public domain would enable third parties to distribute binary patches without fear of getting sued to death by Microsoft. Sure, it would be nice to have open source code, but public domain binaries would still be a lot better than nothing.
And then we get into an escalating war of checksumming and obfuscation and ever increasing bugs. The correct medium for altering software is source code. Binary patching will just mean substituting one small elite set of owners for another set that happen to be good enough to do the disassembly. Sure, that could be done, but since it's mostly for the benefit of a small elite then why should the whole of society support it with legal changes?
As an alternative, how about hardware specs and secret keys allowing updates and changes have to be held in escrow and are released to device owners at the moment that the company stops providing free guarantee support? That way companies can keep their secrets for future versions but all hardware can be repurposed. A requirement to include the source code for the software at the moment the customer bought it doesn't seem wrong to me, but how about just a requirement to provide some software that provides similar features to the original? That way a PC manufacturer could provide Linux source code instead of Windows, for example.
Re:End of support must come with end of copyright (Score:5, Informative)
This is actually already part of a copyright exemption in our law over here. Disassembling, reverse engineering and changing copyrighted software is allowed provided it is done to (re)establish functionality and compatibility.
Re: (Score:2)
it's not enough. We need source.
Re: (Score:2)
It's actually part of the DMCA over here, sort of. Reverse engineering for purposes of interoperability is explicitly protected.
Re: (Score:2)
would enable third parties to distribute binary patches without fear of getting sued to death by Microsoft.
I think you get your view of how software works largely from Hackers and Johnny Mnemonic rather than an actual programming course. Binary patching is incredibly difficult. Binary patching a binary patched version even more so.
Re: (Score:2)
I'd pay a few bucks a year for DOS 5.0 support or even MSX-DOS, find another 100,000 people and there is a viable market for one or two developers working full time. Something more obscure would be less practical. Getting firmware and OS updates for the AT&T UNIX PC (PC 7300) is significantly more work and there are far less public interest and thus less financial incentive.
I think if our operating systems and so-called modern computer hardware were designed better that bugs and security problems in a p
Re: (Score:3)
If it were viable there would be a lot more money flowing into the FreeDOS project.
Re: (Score:2)
There probably would be if companies knew it existed. You have to understand that most company leaders, and even their IT department heads, usually don't even think laterally that there may be a cheap and workable alternative, so they could keep their old piece of software running instead of having it recreated for a lot more money.
Re: (Score:3)
Good software well written to standards doesn't even need support, I like to use Paint Shop Pro 5.0 copyright 1996, it runs perfectly on windows 7.
Re: (Score:2)
Re: (Score:2)
Has the exposure of DOS caused the sky to fall? I ask because pretty much every byte of it has been reverse engineered for a long time. Most of it's internals were published while it was still in common use.
Re: (Score:2)
yes Haven't you noticed?
Re: (Score:2)
It's far more simple than that. We merely need a fair-use exemption for non-commercial derivatives in the form of backups and patches.
re: Backups; If the distributed original version is unmodified and provided free of charge and unbundled, this is a fair use. Such an exemption would apply to any version of firmware or essential operating software for a device. The difficult part is defining the exemption in a way that does not apply to anything except software essential to the core function of a device such
Re: (Score:3)
All unsupported software and hardware must be put into the public domain so it can be supported by other means.
I support open source and all but at some point you need to be practical.
Writing software and producing gadgets should not be heavily legislated. Adding red tape just makes things more difficult for start-ups and small/medium sized businesses. Open source has proven to be very successful without the government forcing the issue. If the government wants to promote open source, what they should do is only utilize open source technologies so they never get locked into a vendor and lose their ability to seek co
Re: (Score:2)
The issue is forced by government granted copyright/patent privileges. Ending copyright/patent protection for unsupported devices actually reduces legislation. There has to be a price on this stuff. If they want the protections maintained, the support has to be maintained too. I am totally unconcerned about their entangled licensing issues. Let them sort it out. In the meantime, cough it up.
I'm also for compulsory licensing of everything that does remain under copyright/patent protection, so we don't have s
Re: (Score:3)
Sounds simple, but it is full of loopholes that would be difficult to patch.
For example some video games have never had their full software stack released. As an example many multiplayer games now do not release the server portion, so the publisher/developer can run servers that cheaters have no insight into how they function exactly. Once these servers shut down, the game is unplayable, and open sourcing the released software would not change this. Perhaps a new server could be built from scratch, but that
Re: (Score:2)
This is an excellent idea.
Re: (Score:3)
With the problem that trying to prove intent is one of the most difficult things to do. Mens rea, while a quite simple concept in theory, is an insanely complicated thing in reality.
Re: (Score:3)
Take Nintendo as an example. They supported the original NES until they ran out of spare parts around 2007.
By that point the hardware was completely reverse engineered and understood, and third party support was available.
The NES is relatively simple though, so these days some kind of mandatory publication of design data might be required.
Re: (Score:2)
Gimme a rebuilt Starfox.
Re: (Score:2)
The NES is relatively simple though, so these days some kind of mandatory publication of design data might be required.
No, I'm not asking that go out of their way to help out. I would just have any lawsuits against publication by a third party be dismissed with extreme prejudice.
This is bad because it creates a perverse incentive to build more complex / difficult to fix and improve hardware. Any law needs to require a very basic level of access to the hardware at the very least. That will encourage simple easy to fix hardware because otherwise some of the cost of complexity will be borne by the original manufacturer.
Re:End of support must come with end of copyright (Score:4, Insightful)
Re: (Score:2)
Re: (Score:2)
Sure it needs a change of the law, but copyright law is obsolete anyway.
This. (Score:2)
This.
Time to change a few laws.
Re: (Score:2)
Absolutely agree. The "right to repair" doesn't guarantee that the dealer will fix your stuff - it gives you the right to fix it yourself (or pay someone else to do it). If a company stops supporting a product, they should be required to provide all necessary info for someone else to support it. Specifically including all source code, with an OSS license.
For many products, though, it goes farther. How many IoT devices are utterly dependent on some server. A server that the company may decide to kill after
Re: (Score:2)
The phone companies lock the bootloaders to placate Verizon so we can't have security patches. This is insane.
Open Source it (Score:2)
Does this mean Sony and others should still be supporting products nearly two decades old, like my PSP?
They should either support it, or open source it so other people can support their own products.
Re: (Score:2)
Re: (Score:3)
Escrow service is a good idea. It should also include the ability to install modified code onto the device, so any keys or signing tools. Code is not much good if you can't install it on the device.
Re: (Score:2)
Re: (Score:2)
For many of these devices, the software is also complete and utter rubbish, so the ability to install custom software should be a possibility from the beginning.
With some of these devices, the manufacturer actually relies on that...
Re: (Score:2)
Re: (Score:2)
And this is one of the reasons why IoT devices are the laughing stock of the security industry.
Re: (Score:2)
You mean IoSD -> Internet of Shitty Devices :)
Re: (Score:2)
Exactly this
Consumers aren't willing to shoulder the price (Score:3, Interesting)
If you put two otherwise-equivalent doodads in front of a consumer, and tell them one will get driver updates for 10 more years than the other but also costs $1 more, the overwhelming majority will buy the cheaper one.
Re:Consumers aren't willing to shoulder the price (Score:5, Insightful)
Just call it a "consumer protection plan" and somehow tons of people will buy it.
Re: (Score:2)
That's why it should be obligated to keep supporting it, so the consumer doesn't get the choice to shoot himself in the foot.
Re: (Score:3)
Re: (Score:2)
Did you think a "lifetime support" meant your lifetimes?
Re: (Score:2)
Don't even have to go that far. Just put two devices in front of them with two cards with checkboxes of feature lists that they have not idea what these features can be and they will buy the one that has more features they don't know, don't understand and very likely will never use, no matter which one has the longer support.
Hell, they'll even take the one that costs more, isn't supported anymore 2 years down the road but has one checkbox more ticked.
Re: (Score:2)
And with good reason. The doodad will likely end up in the landfill in 5 years anyway without any concern of the state of the software, so it makes perfect sense to not pay for something you don't use.
Re: Consumers aren't willing to shoulder the price (Score:3)
And allow a competitor to use the code and screw them over as the competitor didn't have to invest in the R&D or developers hence making the competition cheaper.
There is a reason corporations do not want to open source software as unpopular as that is on slashdot, but software development is not cheap
Re: (Score:2)
And allow a competitor to use the code and screw them over as the competitor didn't have to invest in the R&D or developers hence making the competition cheaper.
First of all, it doesn't, at least not for GPLv3. If the competitor uses the code and distributes their product, they have to release their code. BSD licenses on the other hand does allow this kind of screw over.
software development is not cheap
But it is cheap. You really think these device manufacturers have armies of developers? They just take open source code that they didn't write and then glue it all together.
Re: (Score:2)
There's more to R&D than gluing together free libraries. All the non-coding research, testing, marketing, etc. costs fortunes just to figure out what niche you can fulfil and how to fill it. Most of the time the only protection you have against someone else eating your lunch after you've put all that effort into making it is that there's enough proprietary gubbins to slow them down and make it not commercially viable to clone, especially if it will get pulled for violation of copyright when you point ou
Re: (Score:2)
>> And allow a competitor to use the code
Giving the code to community on a reasonable licence will forbid competitors to rip it off, mostly.
and this will spark innovation, and much better support.
Consumer is not willing to pay (Score:3)
If you have a right to maintain law that forces reputable manufacturers to budget for ongoing maintenance, prices will go up. Less reputable manufacturers will continue selling through ebay, aliexpress and other channels where they can fly under the radar of such regulations, and consumers will buy from them because they are cheaper. If you think otherwise, then you haven't been paying any attention to the downward spiral of consumer electronics for the past 30 odd years.
Re: (Score:2)
The downward spiral is pretty much headed by Apple and John deere and other corrupt big corporation deciding you have to throw away your devices after 2-5 years.
If Sony thought that you would pay for updates (Score:2)
GPLv3 (Score:4, Interesting)
Most of those companies make no money from the software, so there's no reason why they have to prevent people from being able to maintain their own devices.
Re: GPLv3 (Score:5, Interesting)
People who released BSD licensed code know what they were doing.
I've got a bunch of code out there mostly under the MIT license (some GPLv2 or BSD, but most MIT). If someone else finds my solutions for certain problems useful and wants to incorporate it into their product, then great, I'm happy for them. I'm not going to bother myself with worrying about what they use it for or if they want to build a business model around using it.
If I didn't want anyone to make money off my code without my permission, then I would have picked a license that enforces that.
Re: (Score:2)
That's legitimate, but it's good to understand the implications of that decision to the users of those products that wouldn't have existed without your code. If you don't bother about how those users will get their freedom to use the product limited by the product creator, at least be conscious that you're allowing this to happen.
Also your last comment about others making money off free code without permission also applies to GPL and other share-alike licenses, it's not exclusive for BSD-like licensing.
Re: (Score:2)
If I didn't want anyone to make money off my code without my permission
That's got nothing to do with GPLv3, and that just points to you mindlessly accepting what you're told without independent thought.
worrying about what they use it for or if they want to build a business model around using it.
Again, more nonsense.
The GPLv3 explicitly forbids limitations on use, or business model. It only requires that code is distributed along with the binaries. You can use it for whatever you want, and make money off it however you want. If you're the copyright holder you don't even need to abide by the GPLv3.
Re: GPLv3 (Score:3)
I view the BSD as more free. Who are you to tell others what to do with their money and IP they paid for?
The GPL3 may benefit a user but not the developer. Or may harm the user as the gplv3 can be viral if it's source code has a single to it from another program. GPL restricts developers unlike LGPL and BSD.
BSD is great. You can do what you like and they can contribute if they wish or not.
Re: (Score:2)
Restricting rights has been "freedom" on slashdot for quite a while now.
It isnt just the copyright mafia that has destroyed the public domain.
Re: (Score:2)
"Freedom" is a meaningless term when applied to a single person. All freedom is defined in terms of how the actions of one person affects the others.
You recognize that BSD restricts users of the software more than the GPL, which restricts developers more. That doesn't make one "more free" than the other, they simply balance freedom at different points with respect to the different roles of people accessing the code.
Re: (Score:3)
I view the BSD as more free.
Yeah, but you'd be wrong.
Who are you to tell others what to do with their money and IP they paid for?
What does that have to do with open source licenses? Those companies don't pay for open source. Irrelevant argument.
The GPL3 may benefit a user but not the developer.
But it does benefit developers.
Or may harm the user as the gplv3 can be viral
No, it can't harm the user. It's not viral. You can't get it unwillingly. And GPLv3 only covers distribution, not use. GPLv3 explicitly forbids any limitations on use.
GPL restricts developers unlike LGPL and BSD.
It restricts developers as to what rights they can take away from the user. The developer is restricted from barring the user from using the program in anyway they want. The
What a waste (Score:2)
There are extremely few technology products that I would be even willing to consider paying $2-3 more for 10 years of support instead of the the couple of years I get now. Technology moves fast and stuff becomes obsolete.
Really expensive and important stuff (like mainframes from IBM running COBOL) get support contracts indefinitely - those contracts are expensive, but IBM still sells them 40+ years later.
Re: (Score:2)
Cell phone?
Re: (Score:2)
Shove your 5G up your ass
I can't. Earlier this year it was jabbed into my left upper arm.
Re: (Score:3)
Technology moves fast and stuff becomes obsolete.
I'm writing this from my 11 year old W510 laptop, which is perfectly serviceable. I use it for my home day-to-day tasks for everything except the odd bit of video editing and deep learning for which a laptop wouldn't really cut it anyway (I have a Ryzen dekstop).
stinginess is generally the consumers not makers (Score:2)
No, you don't. (Score:2)
If you "need" something, you're willing to pay for it. Practically nobody in almost no field on Earth is willing to pay for "forever updates".
Re: (Score:3)
But it's ok to pay forever royalties for IP?
Re: (Score:2)
1) You haven't paid for "forever updates"
2) Odds are you aren't ACTUALLY willing to pay for "forever updates" when you see the price tag.
Better to recycle (Score:2)
consumers would, I suspect, be willing to pay a few dollars more
But how many dollars more?
If a company made a successful product and (as in the exampe given) then has to support it for 20 years, what is the cost of that?
For a start there is the cost of maintaining a development environment, of employing programmers and testers for all that time. Then there is the cost of making updates available - which would require freezing in time the update delivery system and of maintaining that, too. Should we expect makers of old stuff to still be sending out CDs with software
"If you wish to make an apple pie from scratch..." (Score:5, Insightful)
"If you wish to make an apple pie from scratch, first you must invent the universe."
There are some excellent suggestions kicking around in this thread, but reading through them, it occurs to me that before any of them could come to fruition and realize their potential for improvement, we first need to 'invent the universe' and create the environment where these ideas have a chance of working as today's comment authors intend.
Taking software as our first example... we need to take a long, hard, look at copyright law. Although it was originally intended to protect the rights of the original author of a work - to safeguard there interests and prevent another person from simply copying that work and then reaping financial and other rewards without having to invest the original effort in creating the work - it is now being used (see e.g. Oracle vs. Google) by large corporations to stifle innovation and create "barriers to entry" in markets and market sectors.
Ditto software patents - in fact, these are much worse, given USPTO's tendency to grant a patent for anything that is submitted with an application fee - and the problems inherent in expecting someone picking up the duties of maintenance on "end of life" code to be aware of all the patents that it potentially implements, who the holders of those patents are and so on.
In short, I think that as we brainstorm solutions for extending the service life of modern hardware and software, we must necessarily include consideration of the intellectual property minefield that has been created by technology companies in the last 20 years. We have to remember that they went to the trouble of filing patent claims precisely to prevent this sort of "third party development" from being permitted.
And that's even before we start thinking about commercial company "A" releasing hardware and software at end-of-life and then commercial company "B" stepping in and providing on-going maintenance and support for older products in an attempt to deprive company "A" of "new product sales"... Let's be honest here - how many times have we seen appliance hardware deteriorate after an initial version launches, as a manufacturer re-calibrates to ensure that their product "wears out" sufficiently quickly to allow the manufacturer to make a "replacement" sale?
But perhaps the biggest problem of all that we face is the economy. No, not the local economic question of "how much will it cost to maintain an ageing product"? Rather, the macro-economic question of, "If we make this pivot towards enabling for the extended maintenance of hardware and software, we are going to have an implicit impact on "replacement sales" across entire families of product - from the humble kitchen fridge and washing machine through televisions and DVD players and consumer electronics to computing hardware and software.
I think it's also important to be honest about some of the broader challenges we face. Most of us will remember "BatteryGate [bbc.co.uk]" - the scandal involved in Apple using software updates to degrade the performance of older iPhones. But in the small print of Apple's settlement agreement with the Court on this matter was the following observation, as reported on the above BBC article:-
"The tech titan also agreed for the next three years to provide "truthful information" about iPhone power management across its website, software update notes and iPhone settings. "
For three years. Three years. So Apple have only agreed to be honest about iPhone power management until 2023. What then? More importantly, how are Apple going to cope with continued support for older iPhones when so much of their trillion-dollar valuation is driven by a business model which maintains the average age of an iPhone at 4.3 years [voucherix.co.uk]? I'm using Apple as an example not to
Nice fantasy (Score:2)
It's a nice fantasy. But anyone who has been on the inside of these companies knows it will never work, especially if there are economic downturns and layoffs. Some companies don't even manage to stay in business, let alone support products with software updates indefinitely. But enjoy your fantasy. I mean, sure there are huge companies like Sony, Apple, Microsoft, Cisco etc and it probably seems like they would have the resources to do this. And they probably DO have the resources (if they make that decisi
Security; Repair or Recall (Score:2)
There should be a law that obligates companies to ensure firmware can operate in a fit-for-purpose manner for the lifetime of the physical device.
If firmware because faulty, the obligation should be on the manufacturer to provide repairs, or the means to repair the device.
Similar to car air-bag recalls in recent years, or other physical goods that are assumed safe.
Zombie IoT devices can become a threat to the internet, and everything the depends on it.
My Commodore Amiga doesn't need updates (Score:5, Interesting)
My SNES and PS2 don't need updates. And I'm not sure why the author of the article thinks his PSP still needs updates. Does it run his PSP games? I'm guessing that it does. He might want to install some homebrew software, or repurpose his gadget, but I'm guessing that 99% of PSP owners just want the games.
On a different tangent, I've got a Windows 7 laptop that I use for some gaming and internet use. The copy of Windows 7 dates back to 2010. I deliberately switched off all Windows updates just to see what would happen. Years later ... all my games still work. My laptop hasn't been infected by viruses, ransomware etc. I've taken no special precautions to shield it from the internet. Dumb? Probably. But my non-essential laptop seems to work fine on an EOL OS with no updates for 11 years.
Pay for updates (Score:2)
Maybe we should legislate that consumers should always have the right to install open-source software after the product's commercial lifetime has expired.
Dying products have no vendor support, though. (Score:3)
For most of these products, the vendor has destaffed, and often enough the vendor itself is dying. So having a mandate about releasing software at that point is useless, the people left probably can't accomplish that task. In many of these cases, the remaining people probably can't even build and push new releases successfully.
Better would be to require escrow of the software before sale. If you want FCC approval or copyright protection in the US, you have to file your software with a central authority, and it will be released after support is terminated, or perhaps N years after to accommodate business kerfluffles. If your later versions rely on much of the same code and that is too scary ... keep supporting the older versions.
There's a lot of comments in here about how this will make things more expensive, etc - that is the point! Right now, companies are often making things less expensive by releasing something they don't intend to allow you to use for the long term. So instead of selling you one thing you can use for 10 years, they're selling you three things that barely last 3 years apiece.
Of course, the core problem with any of these proposals is that most companies can barely build their software as it is. Publicly releasing source code sounds like a great solution, but in most cases it would just be a bunch of dead code with no ability to build new versions. Only the most popular products would draw enough core obsessives to get the software stack buildable.
Someone needs to pay for it (Score:3)
Even many open source projects go dormant, and are no longer possible to run on modern OS versions. Used to play a lot of Koules on Linux console, I don't think I can run it anymore. It will require a dedicated player to bring it back to life.
For closed source, it costs at least an engineer just to keep things running. Some do it well, like Microsoft. Yes, they might not be the nicest company around, but at least if you give them money, they will support even Windows XP for you: https://www.techradar.com/news... [techradar.com]. I am sure you can even get MS-DOS for a fee as well.
For devices like an IoT camera from a no-name brand in Alibaba, I don't think there is any hope. No amount of legislation can fix that.
Re: (Score:2)
Re: (Score:3)
Re: Some manufacturers have already figured it out (Score:3)
Yes, there are hardware-limiting reasons you cannot install the new MacOS as Apple has dropped libraries those old chipsets require in the interest of improving performance and compatibility for their newer chipsets.
Don't like it and want an OS still getting security patches? You are free to move your old Macs to Linux.
The original Macintosh isn't getting security fixes anymore either. Is that really a problem?
Re: (Score:2)
Yes, there are hardware-limiting reasons you cannot install the new MacOS
Yeah bullshit.
Don't like it and want an OS still getting security patches? You are free to move your old Macs to Linux.
Then it's not a mac in any meaningful manner any more.
People will defend Apple for bloody everything including planned obsolescence it seems.
Re: (Score:2)
Don't like it and want an OS still getting security patches? You are free to move your old Macs to Linux.
Then it's not a mac in any meaningful manner any more.
No, but it's a useful computing platform for someone that doesn't have to go into landfill and doesn't have to go through some horrible recovery processes in order to be recycled. That's the major step forward we need. The government has no duty to provide you with access to Apple's proprietary software and services - they just shouldn't have to deal with the environmental impact and costs of getting rid of hardware that could perfectly well be used for something different.
Re: (Score:2)
No problem. Third parties will support it if the source is available.