I just can't get over the fact that this starts with
> This is a big change, but it’s not an “Our Incredible Journey” post
and then never, nowhere, at all, makes clear that the post means "We're shutting down Glitch". At least the "incredible journey" posts are clear about that, somehow this here is even worse. It suggests that they're just shutting some small part of it down but actually if I read it right (and the comments here) it actually means they're shutting the whole platform down but might someday want to try and do something else with the name and the userbase.
I simply don't understand why people write like this. What's the benefit of trying to fool your userbase into thinking that maybe not that much is changing when in fact everything is? Who wins when, the day after they pull the last plug, lots of people email them in panic, because they hadn't realized that "important changes" means "we'll delete everything"?
It's a bit ironic coming from Anil, who has quite a following on social media due to "calling it like it is" when big tech or VC culture does something rotten or self-serving.
The doublespeak and evasiveness in this "incredible journey" post is exactly the sort of thing he'd typically criticize!
Yes, they tried to avoid the cliche by pointing at the cliche, but they weren't able to refute Our Incredible Journey.
Though, this was a real product, with almost 10 years of operation; very few products last this long, and the product wasn't immediately shuttered after the acquisition, so it is unfair to categorize them into the same OIJ bucket.
$30M Series A by Tiger in 2018, acq by Fastly in 2022 for undisclosed price, shutdown in 2025. Fastly is a CDN and edge network. Glitch was used by Fastly for “internal training” and “internal sales tools” (kinda?), a far cry from “fullstack application platform in the edge”.
is it me or does it feel like the american dream is quickly becoming "work super hard developing a new product that meets people's needs, then get bought out and shut down by one of the supergiant orgs that owns everything"?
You’d have to convince someone to say no to a multimillion dollar VC check, bootstrap themselves financially (and often on “bread and water” levels of available money for their project), and be willing to potentially run or be at least somewhat responsible for whatever they build for potentially decades.
An entity would have to really want to do all that for the “American Dream” to happen.
This is why civil design software has not meaningfully improved in 20 years and is indistinguishable from its state 10 years ago. We’re living in a Dark Age. We just don’t realize it.
Anyway, all I can say is thank you Anil and team, Glitch helped me learn web dev 7 years ago, and also teach others. It’s just very sad that it’s ending like this, sugarcoating much.
> We’re thinking through what’s next. I’m really interested in how we can look at all the other amazing creation and app experiences out there (I really love stuff like Val Town and Fly.io and Deno and Netlify, etc.) and bring all those together for easily making and remixing new apps. Will take a bit to figure that out.
They're shutting Gitch down. Anil's just trying to be high-minded and optimistic without saying anything specific. I'll be surprised if they do anything new with Glitch.
This absolutely is a "Our Incredible Journey" story, but just with a good migration off-ramp.
Well, it has a better migration off-ramp than most “Our Incredible Journey” stories, but they don’t even promise to keep redirects working beyond the end of next year. Fastly can absolutely afford to do better than that.
I'm similarly lost as GP, and even reading this, I feel like I'm missing something.
If HN said they're disabling new submissions and comments, that'd effectively mean they're closing the site, and would probably say that outright.
But Glitch aren't saying they are shutting down, but they're removing the feature that I thought made Glitch a thing. So nothing is left, yet they continue? Is this a 180 pivot or? Are there other features that will still be usable?
It could go the Piczo route—a website builder aimed at kids and teens, not for polished blogs or ecommerce, but more for one-off, personal and playful friend-to-friend sites.
They're still just hosting websites in that scenario, which seems to be exactly what they don't want to do. A bit like if GitHub decided they didn't want to host source code anymore.
Glitch has been a key piece of the A-Frame community (open source project I maintain) for almost a decade. I'm super thankful to the team! So many people started programming and had first steps in game and 3d graphics development with Glitch.
It's sad to see it go. I was always somehow worried. They had an awesome and super generous free tier. You don’t even need to create an account! Unfortunately, it looks they couldn't make the numbers work.
As someone who has been using glitch.com because of A-Frame and was teaching students to also use them together I am really sad by the news. It was a great thing to see students who had no experience with programming to create a virtual spaces in under an hour and after few more lessons using A-Frame in a creative way that I wouldn't even thought off.
I guess I would also like to thank you @dmarcos for keeping A-Frame alive. My especially fond memory of it was during Covid lockdowns when my friend reached out to me that he had students of Art University feeling pretty bummed because they weren't able to do a end of a year physical exhibit and if I had any ideas to do a simple site to show the works. What we went with instead was to create a virtual gallery that evolved also in to place where people could stream live performances through the internet and even had a premiere of a CD of a band done there (https://vimeo.com/428814586).
Not that I'm aware of. I like glitch because the dev experience is the same as working on a local machine: a list of files, text editor, viewer. Anything you learn on glitch transfers directly to local development.
Codepen, jsfiddle abstract away too much with the UI and different panels. You're coding their way.
Yes! It's exactly what it says it's not, an Incredible Journey post, from Anil, who is a laser-sharp writer.
I kept reading because I assumed they were going to - idk open source it or allow you to host Glitch apps elsewhere - but nope, it's just a straight shutdown announcement. What have I got wrong?
I'd assume there is some obfuscation required by owners or whatever pending a change of ownership, something like that?
Anyhow, RIP, if the internet has taught me anything it's that we can't have nice things for more than about 10 years.
This is one of those classic examples of awful communication dressed up as some kind of “good news everyone!” treatise on a founders personal feelings while being clear as molasses about reality.
Sad, one of the first of its kind. Created a bunch of one-off tools for friends and colleagues on glitch, but could see why it didn't really take off.
For me I switched to other platforms like codesandbox, replit because the editor UX wasn't great for a long time. I get wanting the simplicity angle but having poor hints/autocomplete/etc is a hard sell for writing code.
I really enjoyed using Glitch as it allowed me to quickly publish and iterate on various experiments or try out new libraries like Datasette or HTMX.
I am curious what Glitch will look like after July. If they aren’t hosting apps, will they still be hosting code and letting it deploy elsewhere? It says it’s not a full shutdown, but it doesn’t appear to say what will be left to do on Glitch after that date.
Trello sold to Atlassian. Stack Exchange and Stack Overflow sold to Prosus FogBugz sort of lives on at https://ignitetech.ai/softwarelibrary/fogbugz but it looks like one of those companies that buys software solutions and retains the minimum staffing to keep lights on.
I've been a long-time Glitch user and am now looking for good alternative platforms. My primary use case involves online coding with a Node.js backend using Express and some React apps. If anyone has recommendations, I'd greatly appreciate it.
Me and a buddy built https://pico.sh to make it easy for developers to prototype and share their projects. In particular we have a static hosting service (https://pgs.sh) and a tunnel service (https://tuns.sh) that should cover most apps in the prototype phase.
Thank you for building the pico services. I use pgs for hosting my simple static website, and prose to host my blog, and it’s just such a wonderful experience.
Also it’s been a while since I mailed a check to pay for anything, let alone an online service!
It's not nearly the same experience, but I'd argue a bit nicer of one, I'd recommend giving coolify a shot https://coolify.io/
You have to bring your own server to selfhost but it's dead dead easy.
If you have a nodejs app you can basically just click "new project from github", select the repo, and click deploy. Then it'll be there on your domain (or a free one) and auto redeploy any time you push to master.
I pair program JS-based games with my nephew remotely using the glitch editor. In the end, it’s just static files. Can anyone recommend another place to do that? What we needed from glitch was the collaborative editing and instant update.
I could stand up a server if there’s some open-source glitch-like for just static HTML and JS.
I've been thinking about offering some kind of independent free hosting for the last few months or so. It seems like there's some need for more free, unenshittified hosting services aimed at non-commercial projects (like Miraheze is for wikis), but I'm very concerned about the legal risks, being outside of the USA (not that it's any better in the USA right now but traditionally it was). Servers are plenty cheap now, but lawyers aren't.
I'd recommend, unfortunately, buying a cheap $200 netbook and running cloudflare tunnel on it. As long as you're relying on other people's computers for hosting you'll keep looking forever
My vote for doing this is to get a second hand Lenovo/HP/Dell mini pc.
They're cheap (thanks to corporate upgrade cycles and the sheer number of "obsolete" models that are out there on eBay et al.), quiet, reliable, low power consumption, and generally pretty capable for the money.
The Cloudflare dependency is for networking, not hosting. It would be very impressive to see a self-hosted service available over the internet without introducing a 3rd-party dependency or two.
Yes, but my point is that there are many web hosting services, but only very few such proxy services to my knowledge. (Tailscale is another great option if you don't need public reachability.)
I don't think it's a great tradeoff, when optimizing for independence of specialized solutions at least.
I don't work for Racknerd, but my business uses them for our clients. Most of them have low-end requirements. I mean that's less than $1/month right there.
Run iostat -c 1 and keep an eye on %steal column. IME these cheap VPS cannot be used for anything where response time is a factor. It's fine for static content if you put cloudflare CDN in front of it.
Low volume, low resource usage is definitely the best use case for these things. Yeah, you can get a bit more out of them with CDNs or maybe even just some caches.
The problem is, having a server on the internet is painful because you have to be constantly on guard for patches - if not you'll get hacked sooner than later.
You are getting some pushback but you are not wrong. I thought I was being pretty careful with my DO droplet but just last weekend discovered that it had been hacked and was consorting with mysterious IP addresses in Russia and Brazil.
This was on a box that was firewalled and ssh was locked down. It was running an older kernel - that was probably my downfall.
I immediately shut down and rebuilt the droplet with a more modern kernel. It wasn't too hard because my site is (mostly) static with a simple custom service but is very discouraging to find that somebody has damaged your home project just to (I assume) make a small amount of money.
Unrelated: but the whole glitch labour union that lasted a year before it dissolved itself is the first thing that comes to mind when I read about glitch.
> Glitch’s 14-person staff will all make the move over to Fastly. However the Glitch union, which made history last year when it signed the white collar tech world’s first collective bargaining agreement, won’t be coming with it.
Just coming to say this is a great example of how communication should be done for such changes. So many companies get this wrong, but this is thoughtful and to the point. Glitch was a springboard for myself as well, so very bittersweet (and I guess I need to migrate really old stuff... but to their point, this is super easy now).
glitch.com brings back memories of the pre-slack era Glitch MMO. What an amazing time that was, really beautiful game made by talented people, pivoted the communication side into what we now know as Slack
I thought of that game as well, but I never played it or even knew about it at the time. I also forgot about the connection to Slack. What I DO remember is that they released the game art with a CC0 license after shutting down the game. The site seems to just redirect to a 404 on slack.com now.
I love the brass of putting "Our Incredible Journey" posts on blast right up top and then writing a post that even less clearly explains if they are shutting down entirely, why they are doing so, or what is going on.
I’m not a fan of those overly sweet, corporate-style messages that try to sugarcoat the truth: Glitch was simply too good to last, and you’re losing money...
That said, my only real complaint is this: if you knew the situation wasn’t financially sustainable, you could have at least announced it a year in advance—not overnight.
It was a platform to create / edit / host frontend and node.js backend. It was popular because it was an easy and free way to deploy something on a playground.
When it went out I was super impressed, but they failed to monetize it and got badly abused by bad actors
> As explained in NSD’s Data Security Program Implementation and Enforcement Policy Through July 8, 2025, NSD will not prioritize civil enforcement actions against any person for violations of the Data Security Program that occur from April 8 through July 8, 2025, so long as the person is engaging in good faith efforts to comply with or come into compliance with the Data Security Program during that time. These efforts include engaging in compliance activities described in that policy, such as amending or renegotiating existing contracts, conducting internal reviews of data flows, deploying the CISA security requirements, and so on.
> At the end of this 90-day period, individuals, and entities should be in full compliance with the DSP. This policy does not limit NSD’s lawful authority and discretion to pursue civil enforcement if entities and individuals did not engage in good faith efforts to comply with, or come into compliance with, the Data Security Program.
The same July 8th deadline here doesn't feel like a coincidence. But the provisions of that executive order wouldn't apply here unless they have been quietly selling user data. Do we have any evidence of this?
The provisions will apply to Glitch parent company Fastly, to my reading, and so would apply to subsidiary Glitch. Same logic applies to Pocket and parent company Mozilla.
The implication that Mozilla, a privacy focused company, would have been selling user data in any form is deeply concerning. I'm inclined to withhold judgment until I've seen evidence of this, because that would be a significant scandal given their mission.
Well, I'm not implying that Mozilla did it, but rather that Pocket might have done it, or been unable to prove that they didn't if they were audited. The fact that Pocket was owned by Mozilla during the relevant period is not really up for debate, but whether they were aware of anything that did or didn't happen there is unknown, but they have a reasonable expectation for knowing such things as the corporate owner of Pocket.
I mean, it's kind of curious that the Pocket server source code never got fully released, even though Pocket was promised to be open sourced by Mozilla when they bought it. Now, Mozilla is ending Pocket, without ever delivering on their obligation to their users and donors. I don't mean to cast aspersions here, but this is not a good look for Mozilla, as their hand was forced, if our reading is correct, and this upcoming enforcement deadline is the reason behind Mozilla ending Pocket. Would they have closed Pocket on their own if they didn't have a reason? What was their stated reason?
Mozilla Firefox collects user data and shares it automatically:
> Looking at about:networking I can see connections to pocket (despite me disabling pocket in about:config) as well as connections to "firefox.settings.services.mozilla.com".
> And after research, it appears some of these are hard-coded into the source code on purpose for "security reasons" which is ridiculous.
> Mind you, my browser is hardened to it's best.. just felt like sharing this for anyone unaware that even if you harden Firefox, even if you go the extra 10 miles and edit about:config, it will still spy on you!
> As explained in NSD’s Data Security Program Implementation and Enforcement Policy Through July 8, 2025, NSD will not prioritize civil enforcement actions against any person for violations of the Data Security Program that occur from April 8 through July 8, 2025, so long as the person is engaging in good faith efforts to comply with or come into compliance with the Data Security Program during that time. These efforts include engaging in compliance activities described in that policy, such as amending or renegotiating existing contracts, conducting internal reviews of data flows, deploying the CISA security requirements, and so on.
> At the end of this 90-day period, individuals, and entities should be in full compliance with the DSP. This policy does not limit NSD’s lawful authority and discretion to pursue civil enforcement if entities and individuals did not engage in good faith efforts to comply with, or come into compliance with, the Data Security Program.
> the Data Security Program establishes what are effectively export controls that prevent foreign adversaries, and those subject to their control, jurisdiction, ownership, and direction, from accessing U.S. government-related data and bulk genomic, geolocation, biometric, health, financial, and other sensitive personal data
>> the Data Security Program establishes what are effectively export controls that prevent foreign adversaries, and those subject to their control, jurisdiction, ownership, and direction, from accessing U.S. government-related data and bulk genomic, geolocation, biometric, health, financial, and other sensitive personal data
> How do you think this could be related?
I didn't quote that when I said it may be related. I have some ideas of my own, but I just want to be clear so that I'm not being held to account for things I never said. How do you think it may be related?
People in other threads on this post were saying that Glitch was being used by bad actors, and if Glitch was aware of this or reasonably should be, then they would likely have issues with complying with audits for compliance with the new order, I would think.
You linked to that announcement. I quoted the most substantive section explaining what the DSP is. Reading that description, it’s pretty clear that it has absolutely nothing to do with Glitch (or Pocket). Glitch isn’t a data broker.
> You linked to that announcement. I quoted the most substantive section explaining what the DSP is. Reading that description, it’s pretty clear that it has absolutely nothing to do with Glitch (or Pocket). Glitch isn’t a data broker.
If anonymous people are abusing free tiers on Glitch, it's hard to say if some of those bad actors are also foreign adversaries. People will make C2C infra out of anything these days, and even if they were making a net profit on the whole enterprise, which remains to be seen, perhaps the costs of compliance for Glitch (or Pocket) were not worth paying.
I went out of my way to do original research for HN, and made a case for why I think the results are respondent to the original query as asked. I think that it's likely that the new parent company of Glitch, Fastly, is subject to the DSP, and that obligation probably extends to their subsidiaries. Same for Pocket and Mozilla.
It's clear to me that this situation deserves further study.
I tried to read up on this DSP thing and was immediately lost. How do I know if I need to comply? What does compliance look like? The FAQ is 100+ questions long!
It feels likely that it could apply to any company that draws the current administration’s ire.
I agree it’s very reasonable to imagine that legal departments across tech are worrying, and might consider shutting down — especially if they are already losing money.
> How do I know if I need to comply? What does compliance look like?
These are questions for your lawyer, and if they don't have good answers, questions for your new lawyer. If you don't have a lawyer yet, there's your problem.
Or, to another reading, they are making every and all good faith effort to comply by ceasing business operations entirely. Why else is this announcement so strangely worded and timed?
It isn't unheard of for 2 companies to announce a shutdown on the same day. What is less common is for them to both announce it on the same day and last day isn't 4 weeks/30 days.
When I saw this announcement after seeing pocket earlier, I immediately started to wonder if there was a regulatory change on 8 July.
This is excellent communication and respect for their users. Applause for Glitch. Especially allowing users to retrieve resources for a full 6 months after the closure.
If you appreciate this level of communication and respect, avoid Digital Ocean at all cost. They will fail to send you emails for a few weeks and then delete your resources permanently with no recourse. They are the literal opposite of Glitch. Avoid Digital Ocean.
I would recommend Glitch remove Digital Ocean from their list of alternatives.