> What you are proposing is akin to suggesting "any buildings should become public domain" or "any cars they own should go to a local charity".

I don't think this is accurate. I think they are proposing priority for the consumer in the hierarchy of creditors that is a little more specific about assets than the usual hierarchy.

All IP and licensing rights to IP could become specially encumbered during bankruptcy to enable a consumer priority on it, etc. I don't think it would be more difficult than some of the existing setup, but naturally much of the existing setup is more in theory than in practice.

> I don't think you're crazy, it's a good area to explore, but it's not practical to implement this in the business landscape we currently exist in.

It's relatively straightforward. You put the code or whatever else your customers need into escrow with a third party. The code gets released if you discontinue the product or it otherwise leaves the market.

You then market that you're doing this, so that everyone A) knows they can pursue getting access to the escrowed stuff, B) it's a selling point to remove some of the vendor risk, and C) it's not just a capricious action to undermine creditor rights.

Of course, it may be harder to borrow money if you do this. And customers are generally not savvy enough to require this (though if you do a large enterprise deal, requiring code escrow is not an unknown term).

The "simple" answer is to handle these issues long before bankruptcy, with right to repair laws. It is not actually simple of course.

I don't think it would be unreasonable to mandate software for hardware products be pluggable in such a way that you can modify the runtime with hooks.

During the commercial, supported life of the product the consumer doesn't know any different. But at EOL, repair shops can make a market for "modules" that can do things like change the remote server a IoT device talks to. Removes requirements for DRM, fixes bugs etc.

Threat to commercial interests is really low here in my view, an average consumer pays for a product and pays for the SaaS portion because its convenient and provides value. Then the product reaches EOL and the aftermarket takes over. No different to say the long, exhaustive use of used cars (at least in the decades prior to digital integration in cars, it remains to be seen how the next two decades go re:software services ridden cars.)

What would be the problem with making it a stipulation of the acquisition as a consumer protection law, regardless of the corporate structure? Either it goes to a buyer willing to take on the liability and continue operating it or wind it down cleanly in exchange for the IP (and bidders can price that liability in), or it goes to a government agency or NGO capable of open sourcing or extending its life.

I don’t know enough about how courts split up intangible assets like IP to suggest how to bundle it with this proposed liability, but I think it can be pushed up to insurance carriers like warranties are. Make the insurers liable so that they push the requirements downstream and make them first in line for the relevant IP as creditors. Allow them to argue in front of the bankruptcy court to define what bundle they are entitled to as creditors and them allow them to fire sale that bundle as part of the rest. If they can’t get enough to cover their calculations, they can dump it on the agency/NGO for free. (IANAL and I dont know what I’m talking about)

This is why the concept of "source code escrow" exists, as well as escrow in general.

This is why I really like the idea of establishing communism

Yeah so ultimately this means source code isn't considered an asset, but rather an integral part of a product that has been sold. This is how it should have been all along. It's the fundamental point of the free software movement. It's also how it was in the beginning, before companies like Microsoft realised they could use law and technical means to sell knowledge as a product. I hope one day we can go back to this and look back at this funny time when people made us think knowledge was a product.

Exactly. The problem here is that the company has set a policy not to support $100,000 medical equipment after 5 years. Mind boggling - imagine if an aircraft manufacturer said the same thing. Using aircraft as an example, they are still in service after 50 years and parts still available for many of them. A similar approach should be taken to equipment like this exoskeleton

^ this sounds like a long-term good plan to me. Submit all code to qualify, or not. That should be the standard for all authority-approved things globally.

There already exists some limitations from the FDA regarding software. For example if you receive clearance (NB this is different than approval) for a device which uses AI, it cannot modify its own source code and any changes need to go through the clearance process again. (Or something like that it's been a while since I looked this up.)

I don't know that this is how things are currently structured, but if a company was required to submit a device's source code as a condition of receiving a 510(k) clearance perhaps there should also be a condition that the FDA can release this code in instances where the public interest in doing so would outweigh the IP interests of some piece of abandonware.

Yeah who am I kidding that won't happen.

Have you completely misunderstood the concept here? This isn't some IP virus that infects anyone that touches it. J&J keeps the IP that it owns, in all cases, obviously. The bankrupt company must release the IP that the bankrupt company owns, unless someone like J&J wants to acquire it and keep things running. Obviously a company that fails wouldn't destroy the IP of another company. Obviously the failed company only impacts the failed company IP. I think you may be trying too hard to find flaws in this idea, instead of trying to make it work.

Exactly. Proprietary software and software patents need to end. In the case of medical devices, such restrictions violate the ADA by blocking access to reasonable accomodations.

Yes, thats the current law in most places. The op is proposing that not all agreements made by a company should be void on bankruptcy.

Really, there are cases where this is obviously a change that needs to happen. Eg, if a company holds personal data, the agreements that allowed a company to obtain it should live across bankruptcy with the data, not be terminated if it is transferred.

This sort of arrangement is common in real property. For example, if you buy a house where you have to access it via some other private land, the agreement that you can do so is arranged in such a way that it can't be voided when that land is sold, even during bankruptcy. You can't do that for data AFAIK (would love for a lawyer to comment). It needs to be enabled and made mandatory in cases like this.

There is some limited ability to do this: creditors can get a "charge" registered against intellectual property, so theoretically in the OP's case one might be able to design a legal construct that would cause a bankruptcy court to give rights to some special purpose vehicle that is set up (in advance) to represent the patients. But I don't know if that would work in all cases

:) that’s a sad but excellent point

they didn't appear to do that either though - they just uncritically report that it is an FDA rule preventing this company from doing repairs.

> I mean, I guess it does make me curious how you should go about regulating a device like this.

That's actually my point. There's an interesting policy discussion to be had here, and instead of starting it, they just decided to smear a company.

> Like, if the metal fails and you have a horrible fall and break your hip and shoulder, that's pretty different from an iPhone that won't turn on.

> If this is only approved for 5 years, shouldn't the guy be replacing it rather than repairing it? And shouldn't health insurance be covering that, at least beyond the deductible or whatever?

Probably, but (to throw out a hypothetical alternative) you can also imagine a situation where e.g. inspection can tell him if the whole device actually needs replacement.

I don't have the answer here - I just know I'd like the discussion to revolve around the merits of the situation.

What other $100K products do you throw out after five years? I suppose if the insurance is on the hook, they might require repairability.

It's almost like there's no money left in the profession, so there's no more expertise or time available to do proper investigating or writing.

How long is long term?

Forcing companies to support cutting-edge prototypes (like exoskeleton) for decades is the best way to ensure there won't be any company making them.

> For decades, and for most devices, it was trivial to find replacement parts and/or swap out identical parts…

This seems like a stretch for anything other than cars, pcs, and large appliances. I doubt this has been true for a long time for anything under $100 and more electrically complicated than a switch and a motor.

So just no exoskeletons ever then.

> "Straight’s path to paralysis started in the 1990s at the Saratoga Race Course".

why is that ridiculous? seems tragic to me; not to mention the horse broke his neck as well.

It doesn't sound like a hard and fast rule. It seems like it's whatever the manufacturer asks for in terms of "intended working life." This _seems_ to be the original FDA certification for the device:

https://www.accessdata.fda.gov/cdrh_docs/reviews/den130034.p...

The problem here, of course, is it's not the device itself, but the simple remote control peripheral that is designed to switch device operating modes.

The FDA should force manufacturers of "two piece" systems like this to have backup controls on the device itself and to exclude these non-medical components designed for control from any regulation covering "intended design life."

It seems like a daffy middle ground that the FDA lets exist and manufacturers take advantage of when they can.

Your point is valid, and is improved when we recognize it's not an either/or blame game when acknowledging the parts of the system that need improvement for failures like this person experienced.

Or, you know, the company works with both the FDA and the customer to iron out this legal wrinkle instead of taking a zero-sum approach.

Not sure about the US but in some countries this is regulated for cars. Companies are required to provide parts for a certain amount of years after discontinuation of the product.

It seems abundantly reasonable that a similar requirement be imposed for prosthesis. And it would also be very reasonable for the required period to be longer than that of cars.

So yeah, not forever, but definitely not a a short period either.

They shouldn't be making such unnecessary proprietary components. For cases where there really is no way around using custom hardware, at end of life the specifications should be made public so that a third party can manufacture them.

In this specific case, the real issue is just the incredibly short service lifetime. While different medical devices are going to have different lifetimes, manufacturers need to continue to provide support for at least 36 months after reporting that they plan to discontinue support, which is 60% of the lifetime of this product. Typically medical devices are supported for much longer.

But is a compelling reason to have solid right to repair laws.

>He just needs a part.

Right to repair, in a broad sense, also covers access to parts. This is definitely an edge case and we might want to just consider that if we're going to do experiments on disabled with the aim of helping them, and they want to continue using the tools, we might have to subsidize access to the parts until they die.

Yes it does.

https://www.francetvinfo.fr/sante/politique-de-sante/sante-l...

Oh no. You get what is owed to you very easily and very quickly, whatever how much it cost.

I've got relatives who got cancer medication just like that, appointment to the doctor, go to pharmacy and pay nothing at all. In a day he got it.