A grim proposal

I’ve been a customer of Dreamhost for 11 years now. As far as I’m concerned, I’ll be a member until the day I die. Which leads me to my suggestion.

I’d like some kind of plan add-on that would lock in a website for someone that has died.
I’d like to be able to set this up for 1 or more domains I own.
How it would work is you’d designate a contact to serve as a trustee solely for this reason. Another acceptable way of being informed would be an autonomous message from a system like Google’s Inactive Account Manager.

Once it’s been determined that yes, so and so is dead, their website is converted from dynamic (If it is) to a static site and moved to a lower-priority silo where it will then sit for however long it’s domain lasts for. Once the domain is within 30 days of expiring, an email can be sent to the designated trustee informing them of the impending expiration. They can then choose to pay for domain registration to continue or let it expire.

If domain becomes expired, then DH can move the website to something along the lines of a memorial. With a format similar to something like www.dreamhost.com/tombstones/www.originaldomain.ext

By converting it to a static page and moving it to a farm that isn’t Tier 1 stuff there’s very minimal use of resources. You could even have stipulations on top of the fee, such as the domain using DH as the registrar, you have to prepurchase your domain name for XX years in the future, etc.
Potentially, you could include rewards for customer loyalty - such as a discount in price for being a customer based on time. And for the really long time customers - lets say 30 years - the memorial service is free, and DH will maintain the websites domain name registration for life*.

I don’t know, just thought it would be kind of a cool service to have as a way of ensuring a legacy stays behind.


*Life is define as the year $customer becomes deceased plus $number of years

This is an interesting idea. One variation would be to use the Internet Archive’s Wayback Machine for the static and long-term “tombstone” storage. Many sites are already in the Archive, so it would just be a matter of URL forwarding the domain. For example, example.com could be forwarded to web.archive.org/web/*/example.com (or similar Archive link).

Although DH doesn’t, some DNS services provide simple URL Forwarding as a feature of their DNS registrations. A new Dead-Man’s URL Forwarding feature could forward a domain to the Archive if/when the website’s hosting goes offline.

Once the DNS registration expires (max 10 years), all that will be left are the snapshots of the site in the Archive. But by then conscientious webmasters will have update their links to the archived site.

This is certainly a job for spidering archivers like internetarchive.org

An immediate issue would be that if an expired domain is purchased by another party

  • the archive can be flushed programmatically (or upon request) at any time without notification.

  • /tombstone/domain – having someone’s (newly acquired) domain name in a website link not controlled by them might be a potential legal concern.

I agree with @habilis that this is an interesting thing, but I think retaining the domain name in some form is going to be the tricky part and needs some extra consideration.

That’s a good point @sXi. I gather they’ve modified their robots.txt policy in recent years to avoid some forms of programmatic deletion (like large-scale domain parking), but it can still an issue:


Nevertheless, on the question, “Where can I find a 2020 site in 2050?”, I’d still bet on the Internet Archive being around in some form.

1 Like

Interesting stuff. I was unaware of WayBack’s more aggressive approach with concern to no longer honoring a domain owner’s robots.txt file on what I’d consider extremely thin reasoning (some domains get parked, therefore we will encroach upon everyone’s Intellectual Property by default).

Off Topic: I chuckled at this little Wikipedia article nugget…

Furthermore, the site is used heavily for verification, providing access to references and content creation by Wikipedia editors [citation needed]

I’ve thought about this many times, even blogged on the topic. Grim, I guess…
I think this would be suited to a third-party where you provide encrypted credentials which can only be decrypted by another entity if the owner has not updated a “ping” in some amount of time or based on some other events, web calls, etc.
I think it’s a challenge worthy of development, just not something that I think should be done by a provider like DreamHost.

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.