> On November 15, 2024, the Fable product will be shut down. All users will no longer be able to sign up, log in, or access files. All customer data and project files will be permanently deleted [...]
> We recommend you look through your Fable files and export your projects off the platform (e.g. to mp4, gif, lottie… etc).
I genuinely am at a loss for words here. Somebody sat down and decided that giving you your project sources in some form is too much work, just a GIF export will do.
(For those looking to save their data: go for Lottie, it’s a bit more versatile and you can preserve vector data at least.)
EDIT: the ideal way to go here would of course be open sourcing the whole thing!
This shouldn't be surprising anymore. This has happened every week since 2010. This is the VC cycle, and it happens over and over and over and over and over again. In fact, I'm sure there's some flashy new thing on the front page right now that will have its own corresponding "goodbye" thread in 2-10 years, and everyone will get screwed by that, too.
We can lament the current state of affairs, but let's not pretend it's abnormal.
I run a solo bootstrapped SaaS. When people sometimes ask me whether I think this is a stable business, I tell them that with a VC-funded company, it's almost certain that there will be an outcome in several years that will not be good for customers. VC-funded companies go bankrupt, get acqui-hired, or get strategically acquired most of the time, and all these outcomes are bad for customers. The only outcome which is slightly less negative is going public, but very few companies reach that stage, and even then you get endless feature creep and progressive ensh*ttification (like with Dropbox, for example).
Self-funded sustainable businesses can be much more stable in the long term.
Most people (as in 99.9%) don't give a shit about venture capital and don't follow it at all. I would guess that most people (as in >50%) couldn't even give you a half-way accurate definition of it.
If you're buying enterprise software you're more likely to know that VC exists but you're not any more likely to care about it. None of the people writing 5-7 figure enterprise software checks are tracking which VCs are at which firm when, and who invested in what, and what company had a negative client impact when, and all that. And even if you were tracking it what are the odds that you'll need the same software 5 or 10 years later when it might be relevant to the decision?
Things evolve based on evolutionary pressures only and the VC world is too disconnected from its negative impacts to feel any pressure from them.
Put another way, credibility (from the standpoint of the customer) does not positively or negatively impact a VC in any way.
My whole point is the association is very, very far removed from anyone signing checks. Nobody knows who the VC investors of their enterprise vendors are.
VCs care about their returns. They get returns by selling the company to the next entity in line — a strategic acquirer, an acqui-hire acquirer, or through an IPO. To maximize their returns, they insist on maximizing growth, specifically user base growth. And when they sell, the acquiring entity usually ruins the experience for users, shuts down the business, or both.
I think the biggest issue would be the most "successful" VCs aka the ones that make the most money, have the most money to invest in other companies.
If they can make the most return on their investment by fucking over customers than they would have a greater impact.
I suppose if we had the ability to "pierce the VC veil" we could see who the main VCs are and choose weather or not to use a product based on their track record.
Was Fable using a well-known format, or is it a proprietary format specific to the app ? If it's proprietary, would it make any sense to export it directly? Is it even a file format, or is it possible they were simply storing various pieces of data related to your project in various places in backend DBs?
I agree. I deciphered a few weird formats, for example the raw format of an old camera, the save files of drawings in WinWord 1. It's easier if you can make a few custom small examples, for example in the drawings use only one line, then two, then tree. They were binary formats, so the trick was to use an hex editor and look at them for a long time and make a few guess. I never deciphered them 100%, but I had enough info to read the files I care about and create new files.
Sometimes you get a riff format and a big chunk of the guesswork is solved.
And most modern formats are just a bunch of xml files inside a tar.gz , so now the first step is to change the extension to .zip and cross your fingers.
I think it says more about us that we are at a loss for words or shocked that a startup goes under and its unwinding is messy. It is neither a big deal, nor is it particularly unethical that this should happen. Caveat emptor. This type of reaction should be reserved for real catastrophes like a cancer diagnosis, a war, the death of a loved one, etc.
Have we all become so fragile that something as minor as a startup with a small number of users shutting down causes us shock and disgust?
It always amazes me how many of these "we're winding down X" stories hit the front page here and I've never heard of X. We really do work in silos, even if it doesn't seem that way.
Or maybe it's that you're well informed and X wasn't well known...
When a product shuts down it's usually because it wasn't profitable (or not profitable enough). Either they didn't get enough paying customers or they set their pricing too low. For whatever reason- but there's probably some correlation between a product shutting down and many people not being aware of its existence.
Even without being well known, these stories can hit the front page if a few people upvote a post quickly, and then the post gets a few sympathy votes as it rises up the page.
When a VC-backed company shuts down, it's usually because they weren't growing. You can only get to the next round of funding if you're still growing—otherwise you run out of runway.
Profit plays the biggest role, growth is secondary. Another round of funding is only necessary if they aren't profitable enough to sustain themselves on their own.
Although, as I think about it, maybe as an AI company nobody ever expected them to ever get to profitability so the only way to continue their existence was through continued funding.
IMO the biggest detriment of VC-backed companies is that they can still get killed when they're profitable because they're just not profitable enough.
You found a company and own 100%. You get investment and bring on partners, you set aside 20% for employee shares, etc. All of a sudden there's a dozen owners and nobody owns 50%. You're profitable in the sense that everyone's payroll clears and you're making rent, with money left over, but it's not enough to invest in anything big, and the investors won't put up anything more because your growth has stalled.
Every investor who was previously your "partner" will be trying to shut the company down and sell it for parts.
VCs don't make their money by having a portfolio of companies throwing off $200k/yr in profit to a dozen different investors.
This is what happens when massive amounts of free money slosh around courtesy of ZIRP. Question is: will everybody rush to join whatever new darling graces the front page? Or will they finally learn their lesson?
The very best example of that is the Kagi search engine which, if you spend on time of HN, looks very widely used, but actually has less than 10,000 customers.
Sad, I’ve been waiting for the “figma of after effects”—one of the few positive (albeit unlikely) potentials around the failed adobe purchase.
“However, by the time we started to take Fable to market, AI was beginning to challenge the very nature of software itself, and our multi-modal bet with Prism wasn’t enough to cut through. And while we hold a strong pov on how the next generation of creative workflows should evolve, unfortunately, we don’t have the time to get there.”
It's a great reason why webapps suck. A licensed local app would be great here, as the customer wouldn't lose value when Fable disappears. Yet now customers lose all value. Great.
Yeah bummer that Safari and Firefox are blocking any origin file access for security reasons.
> when the company owning the DNS records for that origin goes out of business.
Likely not a problem. Local data in OPFS is synced with a remote server for collaboration reasons anyway.
> There's also the issue with actually having an executable that can just run without any external dependencies
Assume you target the browser as OS, then it's already possible to have an executable: Run the JS file that renders a UI. Related: We will likely have an app launcher demo soon for https://lix.opral.com/
Yep. I lament the loss of native, local apps all the time (and the prevalence of "always online" local apps - I've seen LoB apps that were promising, native apps only to find out they won't run offline because the app does a license check online at startup and will refuse to launch without a connection).
Everything as SaaS web apps is just yet another chapter in the enshittification of tech.
> I've seen LoB apps that were promising, native apps only to find out they won't run offline because the app does a license check online at startup and will refuse to launch without a connection
Isn't this fine as long as this is the only thing it does?
"Making sure this has been paid for" seems completely legitimate.
Totally fine and legitimate, the problem comes in situations where you might need to launch without an active internet connection. Arguably this wasn't the best vendor in the world, but IMO there should be a grace period, i.e., must connect at least x times per months instead of at every start up.
Keeping an app running requires personnel, personnel is expensive, and good personnel is unlikely to cool their heels on an EOL'd app at an EOL'd company.
It really doesn’t though. You could scale down to a 1-3 person team. Stop building new features. Keep a full stack dev on for 1year to fix bugs and the remaining team does customer support.
Hiring a contracting company for 3 mo per year to perform updates.
My guess is the decision maker doesn’t think it’s worth their time to run a business like this. Instead of rapidly shifting focus into a higher growth area, they would continue to be attached to a shrinking product
A 1-3 person team in the US is a half million per annum proposition, and, again, even if you could make the payroll, you have to retain qualified engineers for a role that has zero growth potential.
cool, so you hire one for 2x the the price of a full time one for 1 week per month. 12 weeks * 2x price < 52 weeks * 1x the price.
I've done this job for a 3-4 startups before. they can't afford a full time engineer, so they hire someone to work part-time to keep the lights on. its not a hard job and many engineers appreciate the consistent work and extra cash.
> However, by the time we started to take Fable to market, AI was beginning to challenge the very nature of software itself
No it didn't. ChatGPT spitting out snippets of broken shit where you have to tell it exactly what you want and think through the entire architecture yourself is not a challenge to the nature of software. Especially older models that were even worse than the current 2024 ones.
Also, AI-generated code is still code. The underlying hardware still works the same.
Open Sourcing is a lot of work. Plus it might contain (even huge) parts that can not be open sourced.
And even unknown parts where real ownership and patents might be unknown.
Where did this bizarre myth start? You apache-license the code you wrote, write a paragraph about code isn't yours or you're not sure about and is therefore missing, and move on with your life.
Patent indemnification is not your problem, it's the problem of whoever wants to use the code.
Erlang, FoundationDB, there's plenty of success stories of products who were EOL at some point and open sourcing was the thing that made them explode in popularity.
I think Fable is a very-high quality product, it's unfortunate it didn't materialize as a business, but the software has enough quality and popularity to maybe become something like the Blender of that niche; a thing that would be extremely beneficial to any of the founders' future endeavors.
Huh, it's interesting the supposition is that AI tools are the future, therefore they're unable to compete in time especially when they seemed perfectly poised to adopt that strategy.
Why wouldn't they use that to sell another round to bridge?
I was just looking at motion graphics (I guess it's called motion design now) tools yesterday. I have a small YouTube channel and I was noodling around with the idea of making a little intro screen that just does some kind of swishy flooshy thing. I am very good at operating technical software, but I have no graphic design skills, so as much as I am concerned about and skeptical about AI "art", I thought I'd see what's out there. Maybe instead of starting from a template, it could generate an idea that I could finish my own way.
To get to the point: I didn't find anything that looked usable. Maybe it's telling that I didn't find Fable/Prism at all. But from what I can make of their site, it seems to be about adding (to my eye, unpleasant and unnecessary) textures, rather than generating new animations.
This looked like a nice piece of software with a lot of functionality and a meaningful customer base. It sucks that everything is grow, grow, grow, grow, exit, or die. I've been contemplating starting something, but I don't know if there's any oxygen left for the idea of simply making something people want and selling/operating it at a reasonable profit.
I was wondering the same, was it a decision or a realization? I suppose the decision part is to do it now rather than later when it can get messier. I've gone down with a startup ship and it was rather pointless.
> We recommend you look through your Fable files and export your projects off the platform (e.g. to mp4, gif, lottie… etc).
I genuinely am at a loss for words here. Somebody sat down and decided that giving you your project sources in some form is too much work, just a GIF export will do.
(For those looking to save their data: go for Lottie, it’s a bit more versatile and you can preserve vector data at least.)
EDIT: the ideal way to go here would of course be open sourcing the whole thing!