This is what we're building with Tent (https://tent.io). The idea is since you can always host your own Tent server, it doesn't matter if your hosting provider shuts down.
You can migrate your data to another server and change your entity (global user identifier) as part of the protocol, so there's no messy "please update your address books" (that part happens automatically.
Data is stored as posts and post types are developer extensible (posts are JSON and binary attachments), so if your favorite app is discontinued, another developer can make a great app that interacts with the same post types.
We're working on a Tent app that imports Posterous posts right now.
Like Brad said -- this one is not to be sold. We'll charge money and keep the lights on no matter what. We want to provide this service like a basic utility like water and power, because that's what having a voice on the Internet should be.
I'll believe that once you actually have an escrow policy that e.g. ensures both source and domain names go into the trust of <public domain, non-profit org> if you sell or fold.
Until then, that pledge is just words which can be rescinded any given time.
To be very clear: I am not insinuating you would break your pledge, but as a user, I'd be unwise to trust a non-binding statement on a web page.
Why should I trust that this will last longer, given the same people are involved? The fact that you charge $5 / month? What do I get for $5 / month versus, say, my own VPS running whatever blog software I choose for the same money (or less)?
I have run many personal servers. They all come and go. They get hacked. I lose interest. The content goes away forever, except if I'm lucky enough for it to get picked up in http://archive.org. Posthaven is like that, but you don't have to hope. It will get archived.
Self-hosting is great and I don't think we'll ever replace it. But for the vast majority of personal content out there, you want a caretaker. I hope you'll think of us when you do.
Instead, though, you have to deal with moving all your content, repeatedly, as your chosen "forever" or "lifetime" company goes under, folds up, gets sold and shuttered, changes terms of use and claims rights to your content, or any of the other bad things we've all seen over the years.
Vigilance is going to be required whichever solution you choose, but having said that, a properly set up, locked down server based on a stable server-oriented OS distro, is going to require very little maintenance, if you subscribe to the old ways. Don't fix it if it isn't broken, and keep installed software to the absolute bare minimum required. Analyze upgrades to see if they're actually necessary (the answer is usually "no"). Pick your blog software carefully, from projects with proven track records for stability and security, and apply the same practices of keeping modules and add-ons down to the bare minimum, and you'll find there really is not much you have to keep up with.
Unfortunately no business regardless of the desire can guarantee it will run forever. It's not a matter of will alone. Any agreement to never sell your shares would imply that a entity structure without shares would better fit your stated motives.
Sorry, unless you either open-source all your code so anyone can pick up the pieces once you are unwilling or unable to continue or you show me the part of the business that details your contingency plan, this is just sweet talk.
>And they charge $5 per month, so
>the sustainability is more clear.
In 1996, Geocities began offering a deal: pay $4.95 a month and no ads will be shown on your site, and you get to use your own domain for your website. I signed up immediately. 3 year later, Geocities was acquired by Yahoo. 10 years later, Geocities was closed:
There is nothing really permanent about any of these services. Ma Bell may have offered a reliable phone service for almost 100 years, but in the era of the Internet the half-life of consumer sites tends to be much shorter.
Both 2 and 3 are certainly in the plans. How could we not have an API? And custom domains?
Brett and I have discussed #1 and it's an option we want to keep on the table. First priority is to build and create a great hosted experience, however.
That's good to hear. You would be surprised how many services (even new) don't have public APis. I didn't see it mentioned on the site which is why I assumed it didn't exist.
By custom domains I mean user supplied top level domains. Definitely have a read on how we view URL permanence [1]. I wish more sites thought about it and glad you already are.
I wish there were some open source resource sharing protocol a bit more high-level than HTTP.