The Svbtle Promise | Hacker News Reader

The Svbtle Promise

dcurt.is

102 pointssp33288 comments

This feels a bit like a marketing gimmick, especially when you read the fine print. If forever means "till Svbtle becomes financially insolvent", that is no different from any other company that shuts down because they don't have money. The only thing that this promise promises is that if Svbtle gets acquired then as part of the purchase agreement the buyer will be required to keep Svbtle "alive". But what guarantees the buyer from selling again and not including that in that agreement and the new purchaser from shutting it down? And don't make a promise where your fine print says these policies are subject to change. That's not a promise that's a promise with your fingers crossed behind your back.

It's not a marketing gimmick--it's a promise. It's how I want my own blog to be treated.

We announced this publicly so that you now know: if you write on Svbtle, your content is vastly more likely to survive for a long time than if you used your own server (probably) or if you used another publishing startup.

Ok I understand your motivation, I really do. However announcing something publicly with fine print that says you can't be held liable for that announcement basically negates that announcement. Either you promise and stick to it (no fine print) or you announce something that catches someone's eye because they wish there was something out there that worked this way and you get them as a paying member but has a way for you to go back on your promise (thus making it a marketing gimmick).

I think that you mean a pledge.

Anyways, I think that it would have been more effective if there was an actual plan associated to it. For example explaining that all the static content is stored on S3 and there is money set aside to fund the static serving for 5 years. And that users can get a dump of their website for after that period.

your content is vastly more likely to survive for a long time than if you used your own server (probably)

I had to chuckle.

What if you published monthly dumps of (public) Svbtle content and distributed them by bittorrent? Then in the event of an acquisition, someone else could host the content. Much like how Geocities was mirrored.

You could even set up a separate legal entity that would have a trust set up to keep these backups alive, or partner with e.g archive.org.

Edit: looks like you're already doing something like that :)

I could see it as substantial if it was attached to a public strategy of longevity, but sans a sweeping organizational shift designed for sustainability over the long term, this amounts to nothing more than words.

I'm thinking something on the level of changing into a non-profit organization with a backing trust to support the continued existence and maintenance of the product. Similar models have worked throughout history, and it would be ground-breaking to do something similar for the concept of digital content.

(Edit: apparently this is what Posthaven is trying to do, and has taken steps to at least attempt to set up. Sounds difficult!)

A promise the buyer will be required to keep a service alive is certainly worth something when services are frequently shut down by acquirers despite being possible to operate profitably simply because they're not that interested in it.

But unless there is something in place that prevents ALL future buyers ad infinitum to keep the service alive is going to essentially make the first buyer not buy this in the first place. So if you don't put in the clause that ensures this promise is kept for all subsequent buyers there is nothing stopping the first buyer from essentially selling it to a shell company controlled by the buyer then shutting it down and then buying the shell company.

I think you're greatly overestimating the level of determination of the average corporate acquisitions team to ensure the legacy product gets killed rather than left in "maintenance mode" or merged into a new service with grandfathered rates.

Agreed it might be a rare occurrence but still a possibility that "the promise" does not address.

I'd be shocked if that's not open to negotiation during an acquisition process.

As someone who built a content platform to address most of the same issues, I tend to agree. I'm definitely starting out with the mindset of "I want to grow this into a modest and long-term sustainable business, but...".

It turns out making sure my product isn't on the critical path for you always having a copy of your data, should I go out of business with zero warning, is actually a lot of work.

I should add: if you're using such a tool, it's valuable, and you want your data to stay around then pay Dustin for all of his effort! The biggest challenge I've seen here is people wanting things to last forever for zero cost.

Startup idea: a service that prevents your startup from fucking your users over when you go out of business with zero warning.

Our Incredible Journey as a Service (OIJaaS.. kinda sounds like ouija...)

It almost sounds like a type of insurance company. While the actuarial tables would be horrible on startups, the cost of keeping relatively small amounts of data alive could be easily spread across lots of startups.

Besides, you wouldn't have to buy the servers until the startup died.

I wonder if this would work as an add on for a backup company? Calling @backblaze!

Or possibly The Internet Archive (http://www.archive.org/) and/or Archive Team (http://www.archiveteam.org/).

It's been done: http://www.shutdownify.com/

... and they've shut down. How predictable.

Honestly that was a very very bad idea for a startup.

If you've read some of the rationale and things Dustin is trying to put in place "when you go out of business with zero warning" can mean get hit by a car. Being all hand-wavy as though not fucking people over in that instance isn't particularly helpful or realistic. Especially when such a high percentage of people expect to the services they use to be completely free in perpetuity.

Business continuity and succession plans aren't free.

Nobody starts a company going "and if we fail, fuck the users." The problem is that maintaining service takes resources that cease to be available when you fail. This "Promise" does very little to allay those concerns.

What's your plan to maintain those links and user data if you get acquired/shut down?

As I mentioned elsewhere, there are certain things that can be done to keep the service running for at least your lifetime--I'm working on a trust that will be created in the event of my death (or the company's death) that can keep the servers running for a very long time to come, even with no paying customers. We will make that public when it's finished.

Great idea! Could the trust somehow have an arrangement with the Internet Archive to ensure all content is archived? It'd act almost as a last will and testament for a company.

Wow that is some incredible commitment. Mad respect.

  We will make that public when it's finished.

Well, why did you announce now, rather than when you actually had the legal agreements lined up that could lend real weight to a promise to keep the site up forever?

Because it's going to take months to work out the specifics and I wanted to make the promise now.

I feel sorry for you whenever you post. I was an asshole about your forks post, (and I'm sorry for that, (EDIT: I should have apologised before now)), and people are kind of being assholes now.

I mean, here you're saying "I'm thinking of the users in case I FREAKING DIE" and even that isn't good enough for some people.

Are you kidding us? We all need to feel a bit better by criticizing everyone as long as they're actually doing something. Next time they'll think twice before making us feel bad for not really being able to be proactive and tackle even the smaller projects.

Love this way of thinking. Why not charge something, anything, like even $1 a year to cover server costs? While it may not seem worthwhile perhaps it would protect this promise by ensuring that the service at least pays for itself.

I would have found it more meaningful if you had waited until you had those things in place to make the announcement of the promise.

Or at least mention you're working on them in the promise.

Cool idea! Looking forward to your post.

I would have to say that a lot of companies (including startups) act that way. After being up for years they give users just a couple of months or weeks or sometimes no notice at all to export the data. Many don't have export functions at all until they announce that the data is going to be deleted soon, so users can't easily make regular backups and there's a high chance they won't see the notice in time.

While true, I don't attribute any of that to malice. If the founders' backs are against the wall, there may no longer be the resources left to plan for a graceful migration.

I applaud Dustin's consideration of the "What If," but this would have been a much stronger announcement with a concrete plan, e.g.:

- what tools will be provided to backup a user's data (e.g. Google Takeout),

- how do you guarantee access in perpetuity in a scenario where you and/or the company are no longer around

I presume most people have good intentions in this area. The difference between a noteworthy step forward and the status quo is having a concrete plan in case Bad Things happen.

>Nobody starts a company going "and if we fail, fuck the users."*

with one notable exception...

In addition to keeping the permanent URLs live, why not pledge to release the source code? Seems to me that this would be the easiest way for users to take control of both the data and the platform, in the event the company becomes financially insolvent.

Good idea--we'll add that.

I'd also add to that - build in a simple way to export your data (maybe it already exists). That kind of trust and confidence in your product being so good that you give people the keys to leave at any time would remove any hesitation at all.

This would actually be a brilliant solution. Pledge to release the source code in an installable form (code is useless unless you can run it) if the company becomes insolvent or otherwise stops trading. Ensure data can be migrated. Then, in the meantime, provide the best service possible.

I am a paying Svbtle user. I like the interface, and I like that platform.

I sent Svbtle some comments about their platform last year. I wanted links on the blogposts to use target="_blank", and a way to tweak which image gets shown when you share a post on social media. There was also a few issues with images you couldn't get back to thumbnail-mode after having click-zoomed in on it. I didn't hear back, but a "No we don't want to do that" or "We read the mail" would have been appreciated.

But I guess my priorities aren't aligned with Svbtle, which is sad. I care less about how long my posts live in their infrastructure (and beyond), and more about how my data is consumed and shared today.

If you want to customize tiny details like that, you should have made a blog yourself. It's really not that hard. For me the point of paying for a service like Svbtle is not having to care about these details and only on the content.

This looks similar to Posthaven's pledge[1] to "last forever". There was a similar discussion on establishing a non-profit trust or foundation[2] to carry out this objective. What happened?

[1]: https://posthaven.com/our_pledge

[2]: https://news.ycombinator.com/item?id=5230137

Garry and I have the same goals and motivations (and we are good friends).

Great!

I've literally written Posthaven into my estate plan to make sure there is funding for it, if that helps. There's pretty considerable paperwork involved in a nonprofit and it's been trickier than I had hoped to get it there.

Awesome!

I was a little bit active on Posterous. I was a broke student when Posterous closed and so I never made the transition to Posthaven. So my little blog isn't on the internet anymore, but I hope I have a backup somewhere.

You know, there are a lot of blogging and micro-blogging sites popping up in the last 10 years (Wordpress, Blogger, Medium, Svbtle, Octopress, etc), but I can't help but feel that nobody cares what most people have to say. Twitter has devolved into this weird sort of micro-marketing medium, Medium and Svbtle have devolved into Wordpress, Facebook has devolved into Yahoo's old front page. Any given person's software blog is probably only used as knowledge-portfolios on resumes, and are almost always abandoned for anything else more worthy of that person's time. The whole social aspect of blogs and micro-blogs seems completely useless compared to more transient services like snapchat or whatever. I just keep wondering when all these tons of mass-communication services are going to go the way of our old geocities pages.

I tend to agree, but with the following qualification: I don't really read any particular person's blog regularly, but I do end up reading a range of blog entries when they are linked from some sort of aggregator (HN, Twitter etc).

Additionally, as a software developer I often find myself on the blog entry of another developer who has blogged a solution to just the exact problem I am facing, which is extraordinarily helpful.

For those reasons I feel that these blogging sites should and will continue to exist in some shape or form.

There's a lot of hate in these comments, and I think a bit unnecessary.

The fineprint says of financial insolvency: "In the event that one of these extreme cases comes to fruition, the founder of Svbtle currently intends to take additional actions in order to continue honoring the above promises". Surely the meat behind Svbtle is the user accounts, the kudos, the database. That's expensive. If Svbtle were to shut down couldn't the owners just create a snapshot as a bunch of static files, wack it on a $10p/m VPS behind nginx and a free cloudflare plan? They could even release it for anyone to mirror.

The idea is noble and seems great. Stop jumping to conclusions, this guarantee is still better than any other shaky blogging startup without one.

While I like both Svbtle and Medium, I'd still prefer to have the blog fully controlled by myself. I'm thinking about git(github pages)+jekyll because I don't want to pay for a server.

Having said that, Svbtle's kudos are a nice thing. Is there any simple, trusted upvote-as-a-service thing out there so that I can embed it in my blog and don't have to maintain a server+DB myself?

Everyone and their dog offers free advanced analytics, but I want a simple upvote button API :) Edit: nope, FB is not what I want.

Well, there's this startup called http://thefacebook.com. Not sure if it'll go anywhere, though.

Sadly, Facebook's Likes are probably the only version of this that has enough of a network to meaningfully make comparisons. It's essentially an upvote-AAS.

Since you are using "git" & "Jekyll" I would say you are a technical person, so you should just do it.

A couple of days scripting, designing and configuring I managed to have a decent blog [1], without paying a dollar.

I'm using a combination of metalsmith, Amazon s3 & gulp.

1: http://www.drinchev.com/blog

Is your sidebar suppose to have a vertical and horizontal scroll bar on it?

  While I like both Svbtle and Medium, I'd still prefer to 
  have the blog fully controlled by myself. I'm thinking 
  about git(github pages)+jekyll because I don't want to 
  pay for a server.

If you're not paying for a server, you don't "control it" in any meaningful way.

Someone needs to make a better static site generator, with an interface that a content team can use.

> I'd still prefer to have the blog fully controlled by myself. I'm thinking about git(github pages)+jekyll because I don't want to pay for a server.

And that's the fundamental contradiction. If you're not paying for it, who is? He who pays the piper and all that. You can maybe set up an organization intended to serve your goals forever (charitable foundations are the current tactic), but society deliberately opposes perpetuities, and for good reason.

> Everyone and their dog offers free advanced analytics, but I want a simple upvote button API :) Edit: nope, FB is not what I want.

If you want a trustable service then that service needs a way of determining identity (to avoid vote spamming) which is a very big task. Only the large social networks (facebook/google/twitter/etc.) really have the level of resources needed to handle that; anyone else pretty much has to delegate to them.

Man, this is such a breath of fresh air after seeing the startup scene in general treating user data as trash when the service folds. That's not your data; your users just let you hold it for them.

Note that not all user data (e.g settings, drafts, etc) is being saved; only the published content.

Actually, we do save all user data right now and don't plan to change that--but it's hard to promise to keep that stuff forever.

That's not immediately clear from the verbiage (both places about the promise only mention "content you publish"), so that might be something worth adding.

He actually addressed that. He said its hard to promise to keep that stuff forever, implying that limiting the promise only to published content was an intentional decision in order to make it feasible to make the promise at all. Notice also that he did mention that they save that stuff right now, despite not promising. Better to promise to save what you can than to not save anything at all. I thought his response was quite adequate.

What I read into this is that a lot of people have moved from Svbtle to medium.com so hence they are looking at giving reasons for people to stay.

Guess a complement to how successful medium.com is becoming.

Is it legally binding in some way? Is there a funded trust set aside with a charter and instructions on how to handle caretaking services once svbtle itself has shut down? No? Well then these are just words, which will be meaningless when the shit hits the fan.

You can have all the best intentions in the world, when the walls close in and you need to think about your mortgage, paying the light bill, and keeping food in your stomach I can guarantee you that keeping someone else's blog running will be so far down the list of priorities it'll get trampled underfoot.

This isn't rocket science, lots of human beings have figured out how to set up organizations and funds to continue doing work after they are long dead. Alfred Nobel died over a hundred years ago, for example. But it requires setting aside money now to make that guarantee, not just some nice sounding words on the internet.

You mention design customization in this post, and the Svbtle homepage claims "the best reading experience".

Have you considered offering a toggleable dark mode to readers? The glaring white background of every Svbtle post, combined with the large amounts of blank space, agitates my eyes.

this skips the entire point - it says within their power. Servers cost money. How is it going to be "within their power" if they have no money?

I'm working on a trust that will be created in the event of my death (or the company's death) that can keep the servers running for a very long time to come, even with no paying customers. Will make that public when it's finished.

Perhaps it would have been better not to make the promise until you had the backing for it. I think that's what's causing most of the negative reactions here.

Having just seen Mandrill's "free forever!" tier turn into a $240/year minimum spend, color me a bit skeptical.

If you stop paying (e.g. if they increase prices on you), you lose the ability to publish new content. This is different than "free forever." Their "promise" is that existing content will, to the best of their ability, never disappear from the web.

I'm not sure how this is a response to my comment.

Mandrill's free tier was "12k emails, free, forever". It didn't last forever. The use of "everything in our power" in this announcement makes it essentially meaningless.

Which is why I'm moving to SendGrid.

Same here.

Pardon my language, but fuck free. I don't want to rely on anything that's free. It's a perversion of market principles. I don't take any startup offering seriously that starts with free. Charge money and make the value proposition clear. Yes that means you can't compete with Amazon or the big boys on price. Differentiate with better service or find a better idea.

Free is for entertainment.

Fuck free, but fuck forever too. Neither one is realistic.

And it's not even $240/year, you need a paid MailChimp plan as well, so the total costs is at least $360/year.

https://news.ycombinator.com/item?id=11170713

"free forever" -> $360/year in a matter of days. Not sure what they were thinking how this reflects on their brand.

BitPay is an even more brazen example, using the "unlimited" and "forever" combo together in the same sentence.

September 2015: https://web.archive.org/web/20150709093806/https://bitpay.co...

    A free plan with unlimited processing, forever

February 2016: https://bitpay.com/pricing

    FREE: 30 Transactions / Month, Up to $1,000 Daily

Why not just open source the code and make the data easy to import/export? My thought is that Svbtle is two things: 1) a website 2) a company/brand. This promise seems to be mostly about the website. As long as I can export the data, keep the domain name and use the same code/design/theme options then it doesn't matter to me if my Svtble branded blog is hosted on their website or $2.99/mo shared hosting. It'll look the same to my readers and function the same to me as a writer. I don't need a promise, just let me export/import everything or at least pay someone to do that for me.

What I think is more interesting is the question of what will happen to the brand and culture of Svbtle, and Posthaven as well. Would it even be the same company if it was run by the executor of the founder's will? And if so, then why not just hire an executor now? I've always been intrigued by the idea of Autonomous Corporations, and it seems like something kind of similar would be an interesting way to solve this problem. It really comes down to, can a brand and style be maintained into perpetuity without daily human effort and upkeep, or will it lose its haecceity over time? I guess we'll have to wait a long time to find out.

I know it's usually frowned up to reply with media references here, but this is such a preposterous claim that I couldn't help but think of this: https://www.youtube.com/watch?v=8yaTCXcvTGY

Any plans on using IPFS and Archive.org to accomplish this?

No, we want to do it ourselves.

Only in the case of something horribly catastrophic would we resort to using an external service like Archive.org--but unlike other startups, we would work very closely with them to archive all data. And perhaps even give them the domain name.

Does a feature such as making account content a git clonable repo in the background ? Then all changes and revisions are easily maintained where ever a user wants.

Thats essentially what IPFS is (content addressable by its hash).

http://ipfs.io/

That is a bold promise that they will have to break at some point. Everything is ephemeral in this world, especially applications on internet.

There is one thing they could have done: commit an amount on money in an official bank account, or with a contract, etc... that will serve as the back up plan to keep the content online if only they were to shut down.

Keeping static text online is actually pretty cheap.

But then if you want your content down one day you would have to have some sort of dynamic back end, maybe even a support if a bug comes around, etc... Another problem.

Anyone wondering how this is going to end up should read the story of TextDrive's "lifetime hosting": https://en.wikipedia.org/wiki/TextDrive

When I hear "forever," my ears perk up because... that's a long time.

Will my content be on Svbtle for, say, 200 years? I'm very curious about how you can honestly promise that.

There's a lot of negativity and people poking holes in the promise in this thread. I'm glad that Svbtle is taking steps and thinking about the (really) long term. So kudos for that.

Reply on news.ycombinator.com