Show HN: Posterity Automations – Get things done from the afterlife | Hacker News Reader

Show HN: Posterity Automations – Get things done from the afterlife

posterity.life

120 pointsmohamedattahri88 comments

Hey there HN – I'm Mohamed, CEO here at Posterity.

I've seen a lot of threads lately discussing how stuff can be transferred safely to significant others and spouses. A key aspect of that transfer is often timing, and the ability to only disclose information if and when necessary.

This is basically what Automations are about; web-hooks triggered in the event something happens, and which can serve as a reliable signal because we verify every death manually.

Today, it's available via IFTTT (https://ifttt.com/posterity), and we'll be launching standard HTTP web-hooks very soon as a follow up.

Hope you find this useful.

Open articleView on HN

This is really interesting. The site is very beautiful! I'm curious: let's say I'm 30 years old. How can I be sure that 'posterity.life' will be around if I live a full and long life. A _lot_ can change in fifty years.

This seems like an obvious concern and I don't see it covered in your FAQ. This is not hypothetical. There are many similar (not identical) services that are no more. See SafeBeyond for one such example.

What do you say to folks who wonder about whether their lives will be longer than that of this site?

This is a good question.

The way to think about us is more like a home insurance. When you purchase a policy, your home is covered if something happens to it for one year. When it expires, you can renew it again. If the insurance goes out of business, you can choose another carrier or find another alternative.

It’s not so different with us. You purchase a subscription, and it covers you for one year if you pass away during that time, and you can renew it afterwards.

The services we offer never extend more than 6 months after one’s passing, so you don’t have to really be worried about us being in business for more than 18 months after you become a user, which is really not excessive and in the ballpark of most other services you use.

Edit — Typo.

I've been thinking about this topic after I got married, and realized that I wanted to back up my wedding photos for 100+ years so my great-great-grandkids have the originals.

I wonder if this is a good use of blockchain technology. The company sets things up in such a way where there are resources and mechanisms to keep things running after the company collapses. Theoretically, as long as the s3 bills are paid, things keep running - the money pool pays for that, getting annual refreshes from data owners. Then money also exists for maintenance - s3 will shut down some day, but with money in escrow, developers can create proposals for maintenance that the token owners vote on.

Just a lot of hand waving away problems on my part, but I think it's an interesting question: how do people store data for 100+ years?

> how do people store data for 100+ years?

Print it.

Don't do this. I somehow ended up with all the pictures of family going back to the 1800s. I have 5 suitcases full that I now have to scan.

Find the best representative photos, and upload them to Geni.com, Ancestry.com, MyHerritage.com, 23andme, etc. They are bound to survive from one of those sources.

Why don't you just pass on 6 suitcases instead of adding the risk of losing digital media. The current process has worked for your family for 200 years.

Color inkjet printing wasn't a thing 50 years ago

That's why you pick the best five from each year and print them in a small acid-free paper book.

You keep the gigantic digital repository as long as feasible (I burn to M-Discs), but for accessibility and long-term, pick your favorites in a way we have good reason to think can last for hundreds of years.

> print them in a small acid-free paper book

any good recommendations for printers?

Shutterfly's interface is (was? been a while) kind of painful, but we've gotten good results from them.

Chatbooks makes it fairly brainlessly easy to produce books regularly (and had a hilarious ad a few years back focusing on the ease of use compared to other options: https://www.youtube.com/watch?v=mF2eKaOc3wo).

I can't swear to the longevity of the books - I never dug into the details of how they're printed, because I'm not especially concerned about making sure they last a hundred years.

It's the right strategy in principle, though, if that is your goal.

On acid-free paper.

I disagree - digital is the only route. Printouts will get lost, bent, fires, floods, and lost.

The broader idea that I find interesting however is how to preserve digital content for > 100 years. It's okay if other people don't care about that - I do, and I think it's an interesting set of problem constraints, and was just wondering if anyone else has thought about it.

It's difficult to suggest an actual use case for blockchain, and I am far from an expert, but I think this could be an interesting innovative options for actually using distributed databases to control assets and make collective decisions in a decentralized fashion.

Looking at 200 year old photos is easy.

Trying to find hardware to access a 30 year old hard drive is not.

In a decent box.

With archival quality pigment inks

https://keithdotson.com/blogs/news/how-long-will-your-photog...

Do you have any insight on whether NeverTear would be good media for long-term archival? Waterproof and damage resistant is a big plus, but I wonder if they come at the cost of pigmant durability.

I can’t say for sure. I’ve never used the product you mention, but it is made out of polyester, which is supposed to be great for archival purposes. So…probably yes?

Arweave

I see a market opening: a service to track and manage the afterlife of afterlife services.

Seeing this makes me also think... why not offer a service that will lock down all finance, identity and other related services upon the event of a perceived attack? Sure, it would be worthless if it was too sensitive. I key feature would be that it could be unlocked in 5mins or less by an account holder.

this would be vulnerable to DOSing. Imagine me, your competition, with a finger over a button that can make your money useless for the next five minutes.

This. Unless you are a mayfly, you are highly likely to outlive any new business. Simpler approach is to have an executor have a way to access a password manager to be able to identify and close out online services.

I wonder if the whole site could be put in a trust with legal stipulations to maintain hosting funds. The trustees would be the management teams, then potentially a corporate trustee who could keep the lights on in the event of a permanent code freeze.

I think partnering with a forward thinking legal firm that specializes in estate planning would be a cool avenue to explore that further.

or maybe structure the organization as two parts, an independent trust that acts as a custodian with set aside assets to cover the running costs over the expected lifetime of the covered population , and a technology company that builds markets and operates the service on behalf of the trust (or trusts)?

kind of like how a regulated brokerage, pension fund or insurance company operates. The difficult part is credibly ensuring that the companies will follow the rules when there is no legal big stick hanging over them. Maybe you could have the company buy and issue performance bonds to the policy holders , so if the company disappears, the policy holder or their family gets a lump sum? And have the (independent) company issuing the bonds have the contractual right to display the current pricing of the performance bonds, so if the bond company starts to suspects mismanagement, the price goes up, and that has an immediate effect on the company’s standing (and is therefore likely to deter bad behavior)

Cryonics service providers do that. Here's an example from Alcor, the largest: https://www.alcor.org/library/the-alcor-patient-care-trusts/

"At Alcor, patient storage costs are paid from two separate but interrelated Trusts: the Alcor Patient Care Trust, and the Alcor Care Trust Supporting Organization. … This conservative funding arrangement is designed to cover the cost of patient storage solely from the income from the Trusts, thereby assuring that such funding will continue indefinitely into the future."

The oldest organizations are churches and universities. Seems like they should be involved managing something like this.

Banks and trusts management firms generally have a good track record. Lately banks have been much more speculative but historically existed because they had peoples trust. Trust management firms that manage trusts have to last because it’s part of the purpose of a trust. I’d imagine a digital trust could have potential, where you add some sort of digital artifact and it’s expected to be accessible.

Churches and universities are not something I would bet heavily on in the next 100 years. Some will survive, but society may not continue to favor them like we’ve seen recently.

Churches surely have ahem a hand in the afterlife planning of many people, and a good track record for being their at time of death, but they’ve recently taken a massive beating in terms of attendance in much of American.

Universities were pretty rare a long time ago, and society has seen lots of new ones recently. I don’t know if I’d bet that any but the biggest names (endowments) eg ivy leagues, Oxford etc. can be counted on to last another hundred+ years.

They can ghost you

> Free 15-day trial, then $29.99/year

Time to start taking those big risks I always wanted to take. There are some real savings here if I manage to die in the next two weeks, plus I can still keep getting things done!

Will I need to keep paying to keep getting things done after I die? Can posterity automate that too? How can I cancel my service after death if it fails to deliver, or say, I, as a ghost become dissatisfied with it?

>How can I cancel my service after death if it fails to deliver, or say, I, as a ghost become dissatisfied with it?

I guess it should cancel subscription after a successful trigger? A self-killing business?

You don’t have to worry about that, we’re we’ll positioned to know if that happens, and the account is automatically memorialized with read only access to those you shared things with.

My mom was a hippie, and she always promised me that after she died she'd figure out a way to tell me what it's like. I was in prison when she passed. Leading up to that first holidays after she passed sucked. Thanksgiving, horrible. Four months after her death, and approaching Xmas, I get a package at mail call. WTF? All of my family/friends had fallen off, so by that point it was only my mom who had still sent me letters. I open it, and it was a book on 'what the afterlife might look like'. She had figured out how to keep her word, and let me know what it was like after her passing. I don't know how it got shipped it came as an Amazon gift but with no identifying info. Her last 6 months she wasn't in a state to arrange anything, so she would have had to plan this like 10 months earlier. But having this gift, that was also an inside running kind of joke between us, oh man, that was so friggen awesome.

I'm not doing great myself, and my boy just won't take it serious so we can communicate stuff so this will be super helpful. I have probably thousands of dollars just in re-sellable music software licenses that I need a way to communicate to him, and I can't even get him to bookmark a shared Google doc (funny how relationships go when you completely and utterly fail people and betray them by going to prison).

Thanks for sharing this.

This is probably a product of my current mindset, but I find the tagline "Get things done from the afterlife" hilariously grim - not only should I be obsessed with "getting things done" while I'm alive, but now I need to be productive after I die!

edit: I should add that I actually like this idea, and I'll check it out further.

Hilarious and grim are two words I never imagined side by side, so just for this, I think it was worth it :)

Jokes aside, it speaks to the need that many of our users have with regards to having agency in case of death, as opposed to productivity.

Glad it caught your interest. Hope you give it a try.

Mohammed,

Connect with me on LinkedIN.

We could be interested to include a posterity subscription for clients when we launch the US version of https://tontine.com in the coming months.

Cheers

Dean

Talk about testing in production. You have to die first to see if this triggers, and if not -- your automation becomes the ultimate in legacy code.

Seriously, this is a very interesting idea. However, after losing both of my parents in the last 4 years, there's a sad but legitimate intermediate state between alive & well, and deceased. People lose their ability to function and make decisions on their own, might not be able to renew your service.

I’m sorry for your loss.

Your raise a very good point. Two things.

(1) Incapacitation is something on our radar, and we really care about getting it right;

(2) We put a lot of effort into making sharing and keeping people up to date as simple as possible, so they can reach out to us if there’s a problem.

I used to really like IFTTT. They changed. They turned off stuff that used to work and added different licenses.

I would never rely on IFTTT to work in 5 years, much less 50.

This is a neat idea, but it doesn’t seem like it will work when I really need it to.

IFTTT is really the beginning. We’ll be rolling out webhooks, so hopefully, you can build something you can trust off of it.

Edit: Typo.

That’s a good direction. I think you’ll need to deliver a core set of functions that you can promise to run. And describe all the ways you’ll guarantee it to run and even what happens when your company ceases to exist.

Some services already have similar features that triggers on inactivity or death:

https://support.google.com/accounts/answer/3036546?hl=en

https://support.apple.com/en-us/HT212360

https://www.facebook.com/help/103897939701143

Manual action needed to begin the process

> We rely on a user’s circle of trusted contacts to let us know if something happened to them. It can be anyone they invited or nominated for a role on their account.

Absolutely, and by design. The only automated trigger we’re seriously considering is the social security master death records; if your SSN is there, means you passed away.

Are you really sure about that?

https://faq.ssa.gov/en-US/Topic/article/KA-02917

SSN theft is a real thing, but as far as automation goes, it's probably the safest in our opinion, and we'll obviously keep it optional.

We think the most reliable way is your loved ones signaling it, so they can access your plan and all the stuff you wanted them to know.

I'm fascinated by the pricing implication here. The older you are, and the more imminent your need for it, the cheaper it is!

Not really; it would be the same as saying that leasing a car is cheaper if nearer to death than not.

We're not a life insurance that you have to contribute to in exchange for a payout. We're a service that watches over you for one year to make sure your plan is revealed and shared with your loved ones.

Well if you plan to lease the car for the rest of your life, however long that ends up being, then yes the number of years left in your life determines how much it is going to cost you.

This comparison breaks down in a few ways: with the car, I realise the value of it immediately, as I get to drive it around, whereas for a service that only activates on my death, the value is only _potential_ until that actually happens.

Also, I can terminate my car lease and feel OK about it, I have already realised value, it did its job. If I pay for this service for a year, but I don’t die, have I realised any value? Maybe marginally in the peace of mind I will feel while alive, but there’s no arguing that the optimal way for me (or more realistically, my estate) to get the most value for money is to buy just one year of service, and then kick the bucket in that same year?

Edit: I’m not saying any of this necessarily wrong, it’s just a really interesting quirk of pricing incentives

The Law And Order episode "Rapture" is about a web site that automatically sends out its client's pre-written confessional and gloating emails after the Rapture takes place, which it cleverly detects when the two owners don't check in after a certain amount of time. Unfortunately they both die due to other less supernatural circumstances, the emails are triggered prematurely, and all hell breaks loose!

https://tvtropes.org/pmwiki/pmwiki.php/Recap/LawAndOrderS19E...

https://www.imdb.com/title/tt1343619/

Interesting idea, although I'm not sure it's more effective than drafting a will and including the same information about passwords etc. - which is admittedly more difficult for non-technical users.

How do you hedge against the risk of people using your service to commit posthumous crimes?

Interesting question for jurists; can you commit a crime if you technically and legally don’t exist?

Technically, I suppose the crime would be committed at the time the user configured the service to perform some action that would probably result in their arrest if alive. It just strikes me that you're sort-of automating the executorship process which might leave you holding the bag for a malicious customer.

I worked with an Estate Attorney on a similar product that was geared towards shutting down other online services. We started down the path of trying to figure out which services would let loved ones assume ownership vs which would only let you deactivate the account and there were a lot of headaches along the way.

The service eventually launched, but it never got much transaction. I believe the Attorney tried to get investors at one point to fund marketing and advertising and everyone felt like it was a good idea, but was worried about adoption. The funding never came and the service was shuttered.

I still think it's a good idea if executed properly. Will have to give this a look.

I would like to pass along my master password and safe combination, but those would allow an attacker to help themselves to most of my estate. Is there a way to do that with your service that you feel includes sufficient security?

Check out https://posterity.life/planning/crypto-vault/

It's E2E encrypted, and can only be opened by whoever you made a recovery key for after your death.

I'd pay for the service that would post to my Facebook/Vkontakte/Telegram accounts afterlife.

IFTTT is not very reliable IMO, since you can't control the tech once you dead. But if you have a script that human needs to follow whatever what, then it can convince me.

Moreover, if the service periodically reminds me to write something to be posted later, it would probably work very well.

In other words, my digital life is not around IFTTT, until IFTTT can guarantee on their side that everything is going to work as expected

We started with IFTTT to tap into their ecosystem, but webhooks with your own code is probably what you’re looking for.

The “bring your own code” part is interesting, but what happens if there is a mistake in my code that causes one or more of my automations to fail to run? How do you ensure that user created code is in a functional state so that the purpose of the service is fully executed? If my code breaks and I’m dead, then some potentially very important (to me) stuff doesn’t happen?

It’s not different from a smart contract. Testing, low is the only remedy, and that can be simplified with a great toolkit, which we’ll be offering.

This is a brilliant idea, and terrifying experiment, in finding out what people were harbouring to their deaths, and how they wanted to project back into the future some material change. Most of it will be to cause harm, and some of it will be so beautiful as to make tolerating the human smallness of the rest of it worthwhile, or so I'd venture.

Excuse the harsh question, but how many people using this have actually died yet? As far as we know, you could just collect money and never trigger anything. It's also very hard to collect reviews for this service. I love the idea and I would like to have it, but I find it a hard sell to actually rely on it.

"Never trigger anything" is actually not the worst outcome, compared to prematurely releasing your secrets, accidentally or intentionally.

Yes, great point!

Very legitimate on the contrary. Luckily enough, none.

You can see how incentivized we are to really go above and beyond if something happens to one of our users; the loved ones we help are our best advocates.

This kind of product/service is launched every few years, but unless users hurry they find they typically outlive them. It would be very interesting to see a post-mortem of the various "schedule stuff to happen after you die" projects, at least one of which I believe was YC funded.

most of those I've seen have been shut down your other services after you die, which is really a subset of schedule things after you die.

The ones I'm thinking of were variations on "send these emails after my death is somehow confirmed"

actually it made me imagine scenarios of lifelong enemies sending a message every anniversary after their death - keeping seat for you in hell, don't dawdle!

Love the idea ( played with similar concept, but could not push it past the design stage ) and the approach ( manual verification as opposed to say.. automatic email check ). I think you did the right way. Good luck!

Thanks!

I couldn't help dark-laughing at "if dead, then that"

Love it! haha

Is this a one time trigger, or can you, for instance, periodically continue to annoy some folks by sending them a message or a photo? If, periodically, how does billing work then?

The trigger is a one-time event, and you can use it to start some other process than does things periodically.

Billing stops if you pass away.

An answer I didn't see in the FAQ: why should someone rely on a tech startup to exist and function correctly after their death, for something important, usually decades later?

I tried to answer a similar question above.

Death may happen at anytime, and we provide coverage one year at a time. If we go out of business before, you have the time to make other arrangements.

PS: Will definitely add to the FAQ.

Tech has changed _a lot_ over the past few decades. Auth tokens expire. I'd be interested to hear how you're planning to future proof your solution.

You are using a dead mans switch or another method to confirm death?

Death is manually verified, you can read about the process here:

https://posterity.life/how/death-verification/

This is for USA only?

For the time being yes. It’s the only place where we know how to reliably verify someone’s death.

What other country are you interested in?

The UK has a public register so should be easy if you want to expand.

Sweden too IIRC - and there the population register is public too so you can even confirm with the identity number.

In Germany the registrar of the birth place (or last address if born outside Germany) will be notified in case of death and can be asked. Your client would register his address, birth date and birth place with you anyway.

>It’s the only place where we know how to reliably verify someone’s death.

No need to verify. You can modify the software to trigger if it doesn't receive periodic signals from someone.

That is quite the opposite of verification/its intent.

Yeah, dead-man switches are risky when there’s a lot at stake and you forget your password.

An alternative we’re considering is the death files published by the SSA if you provide us with your SSN.

Much needed service! Keep us posted please!

Reply on news.ycombinator.com