Backplane is shutting down
backplane.ioI've been bitten by service shut downs like this and over the years, here is the "rulebook" I've made for myself to reduce the risk:
- Don't use products from startups with unknown, dubious business models that are clearly subsidised by VC money until they "figure out how to monetize". (recently I saw a company saying "our business model is still firming up, whatever that means).
- Open source doesn't mean it's safe. Very few open source companies have solid business models and if the vast majority of the contrinbutors to the project come from the same company, then if they go down or get aquihired reviving the project and its community is not always guaranteed or simple. OSS projects need major adoption to be safe from this sort of damage.
- I sometimes even research the VCs backing the founders as well. I have seen companies founded by associates in VC firms leaving the firm to start a company because they know 1. they can get funding from their old pals in the firm, 2. they are going to be selling the company quickly to X because of some insider information about X's need or an internal project to find a company to buy in the space. Some founders are serial "build and flip"ers and I avoid using their products.
I feel much safer buying crtical services from a bootstrapped and profitable startup than a well funded one that doesn't have a clear business model.
2. Open Source is relatively safe when you are able to self-host it. It is good to setup an environment and test the features on your own machines and also check how easy is to extract the data.
1-3 is a roulette. Unless they have a technology that you cannot substitute with an open source solution then it is difficult to justify using it. There must be a solid extraction plan available so that you can move your data to an alternative solution in reasonable time. Sometimes it is better to hire a team and expand existing open source solution with necessary features. Sometimes this could lead to a side product and another stream of revenue.
Having source and license to run something, and being willing and able to self-support the software through (eg) security issues are two very different prospects.
This is network infrastructure stuff. Not the kind of thing you leave to fester.
At least, it gives you a much longer runway to migrate off the no-more-supported solution.
SaaS, if it closes shop, becomes instantly unavailable.
You can always outsource the support but being in the control as well.
> Don't use products from startups with unknown, dubious business models that are clearly subsidised by VC money until they "figure out how to monetize".
This is why I tell every startup I talk to (not that many, to be fair) to charge as soon as they possibly can, preferably in the first months after launch (b2b of course, consumer is a different beast).
Wow. If I were a user, I'd hate to have this dropped on me today. Christmas eve is in 10 days and your traffic is about to stop routing in 15 days.
Yeah I think in certain circles this is known as a "dick move."
maybe they're doing it before the new year for tax reasons. still done in poor taste though.
Cases like this make me always think very hard before admitting a closed-source / SaaS solution into the critical path of my stack.
In this regard, large established players have the benefit of the doubt when using a proprietary SaaS service: they are unlikely to fold, and if they sunset a product, they will likely give ample warning well ahead of time. (But not always even so: I see any new Google consumer product as a "while supplies last" sale.)
As a SaaS author: I would also suggest taking into account the business model of the SaaS you are looking at.
The old way of thinking was that a Serious Company is safer than a one- or two-founder operation. But Serious VC-funded Companies are unprofitable most of the time and burn through VC money, subsidizing their business. Even if they don't crash, a "successful outcome" is an acquisition, which most of the time results in shutting down the product and a post about what a "wonderful journey" this was. And don't forget all the products that Google just shut down over the years.
A self-funded slow-growth profitable startup can be a much more stable bet, even though it seems counter-intuitive.
> A self-funded slow-growth profitable startup can be a much more stable bet, even though it seems counter-intuitive.
It's not counter intuitive, it's just that it's really hard to guess which company is doing OK, getting slow steady and profitable growth and which does not. Additionally, the slow and steady company can still be acquired / decide to "bet big" / ... . So we are back to Google and co.
>A self-funded slow-growth profitable startup can be a much more stable bet...
I think his point is that Microsoft, Amazon, and Google are even safer.
I may be mad, but I'm at a point where I would trust a self-funded slow-growth profitable startup waaaay more than Google to run a service in the long run.
Decades of "quickly started, even more quickly deprecated" services have taken their toll.
Sure Google can and does shut down products but they tend to give more than two weeks notice during the holidays.
A company with a known working profit-making model is best, big or small.
GitHub used to be like that. They started very humbly, rejected most investments, grew slowly, but eventually owned most of the market.
I'm happy to hear that as a self-funded SaaS founder (https://partsbox.io/). This is exactly what I'm trying to tell my customers (some of which would rather see a large VC-funded operation): if you consider the probable exit paths for a VC-funded startup, very few are good for customers.
Agreed
Two week notice during the holiday season seems pretty sudden and painful for users. I feel bad for those who now have to scramble and probably miss time with their families as a result.
This seems like an all-in-one solution and quite complex, I highly doubt that this can be exchanged in a short time or maybe even at all...
I really feel bad for those who committed their company to this company.
Yeah, definitely not fun for those impacted.
In my eyes it's evidence that you definitely shouldn't bet your company on a startup unless you have a backup plan that you can implement quickly or the benefits are massive (ie: the reward is worth potentially losing your business). Larger companies also kill products but they tend to give a lot more heads up.
The company I work for was heavily invested in Adobe BusinessCatalyst, which Adobe shut down.
The difference is that Adobe have money, so it's not a two weeks notice, but two years notice - which is reasonable!
Unfortunately, new customers where still annoyed and chose alternatives.
Customers have options, scramble or not. My sympathy is with the staff, sorry "teammates" (per the webpage). It's a lousy time of year to be out of work.
I'm willing to bet the team worked with customers ahead of the announcement in good faith. Hacker News threads aren't always looking behind the curtain at what's really going on.
At KickoffLabs we used them to handle a bunch of routing on behalf of our hosted customers.
The timing was less than ideal for sure.
However they arranged a good support system with the folks at Fly.io. We’ve already transitioned new customers to it and they are helping to migrate existing sites and certs next week.
I think it’s a risk you just have to take, but we learned a lesson that for things that are dependancies for your business it’s good to have a couple of alternatives lined up or working side by side. :)
I like the basic technical idea behind Backplane, which is that your backend servers "dial-out" and connect to the edge load balancers. Does anyone know why this technique is not used more? Or is it actually common but I just haven't heard of it? If so who/where is it actually done this way?
(shameless plug) you might find this interesting: https://github.com/bitnami-labs/udig
It focuses on Zero configuration by skipping the account creation and hostname selection (and long term state) by using the hash of the tunnel destination public key as hostname.
> "your backend servers 'dial-out' and connect to the edge load balancers"
I believe Cloudflare offers this as an option: https://www.cloudflare.com/products/argo-tunnel/
I love the idea of app servers tunneling out (we have an OSS project that does similar: https://github.com/superfly/wormhole). I really hope we get to where that's the default.
I like it as well. It seems like it would provide significant improvements in latency and failing-faster.
Why not just use nginx and zerotier?
2 year old seed round company with <10 employees. I don't imagine they have a whole lot of customers, and hopefully no customers that rely solely on them.
I understand not wanting to appear on "our incredible journey" blog or whatever with a bunch of sap, but surely more information can be made public without everyone emailing? Why shutting down? Why only a couple of weeks? Will the code be open sourced?
Customers got emails a few days before this, and they're getting a lot of help. It's easy to criticize but I think they're handling it pretty well.
"A few days before this" really isn't "handling this pretty well".
It's great that they (you?) are offering some assistance in transitioning to an alternative, but it's still a very crappy situation for customers during the holiday season with many people taking extra time off.
Considering the fact that they're going out of business, I wonder how many customers that is? Still a shitty time of year to do it, and a horribly small time window.
you're the kind person helping me move from backplane to fly.io!
That's me!
Not trying to criticize as I don't know the reasoning behind the decisions, hence my questions.
Open source would be very interesting to me (at least). Looking at their GitHub, it appears only some minor components are currently open : https://github.com/backplane
What is Backplane? Why is it shutting down?
> Backplane is a service that unifies discovery, routing, and load balancing for web servers written in any language, running in any cloud or datacenter. Traffic shaping, request enrichment, blue/green deploys, and other difficult to implement networking operations become trivial when using Backplane's combination of hosted and on-premise software.
From their docs page here: https://www.backplane.io/docs
In the last ten years I have also seen 'backplane' be:
* A protocol for helping web widgets on a page communicate and share data https://openid.net/wg/bp/ . We used to support it at Livefyre to interop with certain Janrain products, but the ecosystem wasn't so big at the time. It wasn't under OpenID at the time.
* This roller coaster of a social network startup https://techcrunch.com/2015/03/06/the-backplane-black-box/
I remember being at a hackathon of theirs (the Lady Gaga backplane) years back and seeing dozens of what looked like Herman-Miller Aeron chairs all over the office. Maybe they got them for cheap, maybe they weren't actually Aerons, but also maybe this pretty small (I don't actually know how many employees they had..) startup had a couple tens of thousands of dollars worth of office chairs
The cost of a good chair is nothing compared to other expenses like salaries. Aerons aren't my favorite chair but I sure as hell don't want to work for a company that makes me sit in a terrible $50 Office Depot chair.
Besides, you can get Aerons for much cheaper than retail if you look.
Even if you give zero shits about your employees, the $50 office max is going to fall apart fast with daily use. HM sells somewhat stripped-down cheaper versions of certain models specifically for the bulk office drone employee market and in bulk the pricing is competitive with other makers.
I don't disagree, I just somehow got the impression they had way more chairs than employees
Backplane was also an internet services company founded by FreeBSD and DragonFly BSD developer Matt Dillon:
Backplane the social roller coaster was a different company.
So basically what ZeroMQ does at far better speeds with better latencies and reliability with a fraction of cost and it doesn't crash or stops operation out of nowhere.
Why is everything SaaS, when it is fairly easy to use existing proven solutions?
Is it just money?
Wow. What a nightmare to shut down just before Christmas.
So cause of death is "Eaten by Amzn"?
I tried looking for their landing page on the wayback machine, but every snapshot I tried just shows a 404 page for some reason.
Not to be confused with Backblaze, in case anybody else made that mistake.
“Backplane” refers to the rear interior surface of an equipment cabinet, where there’s often a maze of electrical conductors for routing power/data around the various rack units or cards it can hold. Examples include blade servers, very serious routers, and theatrical lighting dimmer racks.
The system can be expanded and serviced (without interruption) by installing and removing modules, but the backplane is forever.
A hot SaaS startup purporting to be your backplane is about as backwards as it gets. Smart companies are using cloud services in exactly the opposite way: as temporary, interchangeable capacity slotted into a backplane they own.
> rear interior surface of an equipment cabinet.
it refers to a component interconnect backplane, an electrical / signal bus. not the rear interior surface of a cabinet.
They couldn't wait till after new years to do this?
They're probably doing it now for tax reasons.
If, like me, you asked "what is backplane?" Here you go: https://www.crunchbase.com/organization/backplane
so i'm curious then...
considering that this is displayed on the CNCF interactive landscape page, does it mean that backplane will be quickly removed from the service proxy group, or does anyone think that CNCF is already aware? i reaalize it's not backed project or anything, but it clearly made it there somehow. considering how hot CNCF is right now, made even more apparent by the 8K+ sized crowd this week, i would imagine that the CNCF website brings a nice bit of traffic to these small companies.
it's too bad...backplane's vision seemed promising albeit lofty, and against some heavy odds (with some big players in the space).
Thanks for the mention. I updated https://landscape.cncf.io/format=landscape to remove them.
If I miss the initial announcement, we would remove them when they don't tweet for 3 months: https://github.com/cncf/landscape#non-updated-items
(Disclosure: I maintain the landscape.)
Is this the same company that at one point was making social networks for celebrities?
No, different people entirely.
I'm neither an employee nor a customer, just someone who was following the project on twitter because it looked very intriguing. I just want to say that the comments on this thread are absolutely ridiculous and I expected better. Does anyone actually think the customers would find out at the last minute? That the company would leave its users without any support? It's baseless speculation and my guess is it's totally wrong. - The company is founded by Blake Mizerany https://twitter.com/bmizerany?lang=en an engineer known for Sinatra and a bunch of other well respected projects. - The users adopting an early stage startup's product are likely friends/former colleagues who are putting personal trust into the team. Does anyone really think nobody got a heads up, or possible support deals while they migrate?
Second, Backplane really looked like great tech https://www.youtube.com/watch?v=43wFJBRTHG0
The customers did find out at the last minute. Weeks notice is last minute.
I think zalmoxes' point was that they may have found out through direct contact from the founders, or a customer-only email 2 weeks ago - rather than just by loading up the homepage today.
> Weeks notice is last minute.
Well, no, it's literally not. It's actually a reasonably long lead time for “going out of business”, which basically no one ever announces before essentially all hope of finding a way to keep the business running is exhausted, which intrinsically means there is little runway left.
Yep, it literally is last minute for a platform product used by others during a time of the year when the developers responsible for finding/building a replacement would otherwise have gone on vacation.
If this were a tangible product, a week notice would be fine since it gives customers time to stock up.
> Does anyone actually think the customers would find out at the last minute?
Two weeks notice at a time where people frequently take vacation time is very much "last minute".
> That the company would leave its users without any support?
"It's all gonna stop working on December 29th" does seem to do exactly that.
> I just want to say that the comments on this thread are absolutely ridiculous and I expected better. Does anyone actually think the customers would find out at the last minute? That the company would leave its users without any support? It's baseless speculation and my guess is it's totally wrong.
Well, it's their baseless speculation and your guess, so a level playing field. You can make the same point without baiting other commenters—in fact you just make your point without baiting other commenters, that's how.
It sounds exactly like customers got last minute notice right before the holidays when everything shuts down.
Of all the times they could have chosen to shut down, they chose the absolute worst.
If I still programmed, I would never use this guy's products ever again.