I've been a CodePen user since the very early days. Have always loved the platform. I hate to be negative, but I'm honestly not a fan of the 2.0 release. What I loved about CodePen was just being able to try out something really quick and simple. But the new interface makes me feel more like I'm building a website inside of a website.
I have absolutely no idea if CodePen is feeling pressure from their shareholders or want to cook before the IPO, but I often wonder why Spotify needs 3,000 [1] developers working full time on a streaming app.
4 weeks ago there was a thread about Spotify making over 4,000 production releases a day (or 10 in 10 minutes in another comment in that thread).
WinAmp has been dead for centuries but it's still a perfectly valid player, foobar is being developed by a single dev. Of course I do understand that these people have a lot of auxilary tasks at hand, and "music player" is just a small component of the whole system, but still, it feels bizarre to think you need such an army to maintain a working app.
I love cool features as much as the next guy, but I'm also a fan of coherent products that knows their identity.
[1] the number might be inaccurate but that's beside the point
That's the hard part about tools like this. They're too simple to be used for anything important, and as soon as the creator follows that path you end up in a situation where it's too simple for anything complex and too complex for anything simple.
Amazing platforms like Glitch ran into this problem.
(Some did pull it off! Like how Vercel built a kingdom out of `now`.)
The more complex the more fragile. The simpler the solution the more robust. This, I feel, is an overall truth to software development. And simpler solutions, as far as I know, are usually the more useful ones.
Congratulations, Chris! I’ve always been competing with you, and many users came to Playcode from Codepen. In fact, I even offered to buy playcode.io about 9-10 years ago, but you didn’t respond and 1-2 months later you released a copy product named Codepen Projects.
This is awesome. Been a CodePen user/fan since it launched (and a fan of Chris/CSS-Tricks for even longer). Really nice to see CodePen evolving.
I don't use it as much as I used to but every so often I hack something up in it for a quick PoC. I'll say the new ability to deploy got my attention. Really handy to be able to deploy a prototype or demo of something and toss it over the fence.
It will be interesting seeing how this evolves. Any free hosting solution tends to spiral towards abuse at some point. Still trying to solve it for my own stuff.
That said, I do wonder how valuable something like codepen is in the current meta. I really no longer look at code. I'm not referencing examples, I'm just prompting for what I want / inputting images as context.
Any free hosting solution tends to spiral towards abuse at some point.
Does it work? I'm a paid user, but set up a new free account to see if the deployment stuff worked there too (as I was linking to it from elsewhere) and it said I had to upgrade to a plan that supported it.
I've been wondering how CodePen is surviving the AI onslaught - the core thing CodePen has always represented to me is to showcase the amazing hand craftsmanship of CSS HTML JS developers. Obviously that has changed, so I wonder what it means for CodePen - presumably they need to change their business model in some way.
I was wondering that too. Google Trends shows the following story:
* COVID basically doubled the traffic (everybody stayed home and crafted something)
* AI hit and the interest was cut in half.
I think that makes sense. You are less fiddling around and only need CodePen when you are really stuck but AI solves many problems now.
The shift from sharing little HTML/CSS/JS components to a website builder is probably their answer to this problem. They want to position themselves to compete with a Loveable, Replit and others. It's probably not the same, a little more technical, but the direction definitely shifted.
A CodePen api/protocol that supports a local ide opening a CodePen, making changes in the user's developer tooling and pushing back to the pen would be a game changer.
VCS extensions for jetbrains and vscode could hook into an api and smoothly pull and push pens. This feature was always my hangup for using CodePen as a UI playground.
I made https://jsmess.com as one of my first vibe code projects and have been incrementally improving it over time when I have leftover tokens...
It's closest to JSFiddle but completely static, hosted on GitHub pages. No server side processing. Projects are private, saved on localstorage and can be shared via encoded URLs.
This has been a really fun way to ramp up with LLM based coding and I learned a ton, and made a tool I actually use quite often.
https://jsbin.com is what I've used since forever, it's fast and easy, used to use JSFiddle before that. I think you used to be able to create "bins" without having to login way back though. Honestly don't find much use for them anymore nowadays, if I want to share any web/web-like I do it via my own servers + domain and send links to people.
Always felt like the worst aspect of code pen was the exploratory side. Looking up codepens is extremely poor, you often only look up one "tag" and after the first page quality gets worse.
Wish it had an advanced search that I can filter codepens for something useful like this: animejs, svg, forms; with the result being codepens that use anime.js around a form.
I'd probably use codepen as the defacto web ui reference if it had better search capabilities. Right now it's mostly github + blogs, but only because it's so hard to find anything worthwhile on codepen. Gold is there but it's deep under the strata with no willing to actually get it.
This should be the top comment, especially since organic search has been lost. Not to be overly critical because Codepen was great in its time, but it's now on individual websites to upgrade their search game if they want to stay relevant and useful.
Edit: Just noticed that logging in is now required for search. I'm sure this is an attempt to prevent AI scraping, etc., but it's a big mistake IMHO. While I have an account, I don't log in much and tend to abandon sites that require log in. Among nerds, I'm probably not the only one like this.
I've been a CodePen user since the very early days. Have always loved the platform. I hate to be negative, but I'm honestly not a fan of the 2.0 release. What I loved about CodePen was just being able to try out something really quick and simple. But the new interface makes me feel more like I'm building a website inside of a website.
A textbook definition of "feature creep".
I have absolutely no idea if CodePen is feeling pressure from their shareholders or want to cook before the IPO, but I often wonder why Spotify needs 3,000 [1] developers working full time on a streaming app.
4 weeks ago there was a thread about Spotify making over 4,000 production releases a day (or 10 in 10 minutes in another comment in that thread).
WinAmp has been dead for centuries but it's still a perfectly valid player, foobar is being developed by a single dev. Of course I do understand that these people have a lot of auxilary tasks at hand, and "music player" is just a small component of the whole system, but still, it feels bizarre to think you need such an army to maintain a working app.
I love cool features as much as the next guy, but I'm also a fan of coherent products that knows their identity.
[1] the number might be inaccurate but that's beside the point
> or want to cook before the IPO
What IPO is that?
That's the hard part about tools like this. They're too simple to be used for anything important, and as soon as the creator follows that path you end up in a situation where it's too simple for anything complex and too complex for anything simple.
Amazing platforms like Glitch ran into this problem.
(Some did pull it off! Like how Vercel built a kingdom out of `now`.)
The more complex the more fragile. The simpler the solution the more robust. This, I feel, is an overall truth to software development. And simpler solutions, as far as I know, are usually the more useful ones.
Not necessarily! Linux is incredibly complex. AWS is incredibly complex. Next is incredibly complex. GitHub is incredibly complex.
We need simple tools and we need complex tools. It's the middle ground that's hard to navigate.
Solvable with reasonable defaults and putting most things behind an advanced section?
no, it's just not solvable... you can only mitigate
anything hosted free will inevitably be abused by scammers, you need constant vigilance just to stay a step ahead
> it's just not solvable
Watch out, good hackers find those words incredibly motivating.
This is something they're certainly aware of as they kept the old style available (https://codepen.io/pen) and created a tutorial on how to make 2.0 work like the old version: https://www.youtube.com/watch?v=MMxXwupwURQ
I suspect they're doing this because of AI.
Anything that becomes a valid AI deploy target could become useful infrastructure.
The rate of humans testing things in a CodePen will slowly fall off, similar to Stack Overflow.
The rate of AI models deploying to CodePen 2.0 will pick up significantly.
In some number of years, the amount of code written per month by machines will dwarf all code written by humans from 1957 - 2027.
In that world, CodePen 2.0 is valuable. CodePen not so much.
Congratulations, Chris! I’ve always been competing with you, and many users came to Playcode from Codepen. In fact, I even offered to buy playcode.io about 9-10 years ago, but you didn’t respond and 1-2 months later you released a copy product named Codepen Projects.
This is awesome. Been a CodePen user/fan since it launched (and a fan of Chris/CSS-Tricks for even longer). Really nice to see CodePen evolving.
I don't use it as much as I used to but every so often I hack something up in it for a quick PoC. I'll say the new ability to deploy got my attention. Really handy to be able to deploy a prototype or demo of something and toss it over the fence.
> every pen is deployable
It will be interesting seeing how this evolves. Any free hosting solution tends to spiral towards abuse at some point. Still trying to solve it for my own stuff.
That said, I do wonder how valuable something like codepen is in the current meta. I really no longer look at code. I'm not referencing examples, I'm just prompting for what I want / inputting images as context.
> Any free hosting solution tends to spiral towards abuse at some point.
Have github pages done this? Has cloudflare pages?
I would imagine that both companies have extensive internal tooling to address this concern; likely at multiple levels.
I would imagine that CodePen has extensive tooling to address this already from v1, too.
Any free hosting solution tends to spiral towards abuse at some point.
Does it work? I'm a paid user, but set up a new free account to see if the deployment stuff worked there too (as I was linking to it from elsewhere) and it said I had to upgrade to a plan that supported it.
Where do you think the AI is getting the code from your prompts from.
I've been wondering how CodePen is surviving the AI onslaught - the core thing CodePen has always represented to me is to showcase the amazing hand craftsmanship of CSS HTML JS developers. Obviously that has changed, so I wonder what it means for CodePen - presumably they need to change their business model in some way.
I was wondering that too. Google Trends shows the following story: * COVID basically doubled the traffic (everybody stayed home and crafted something) * AI hit and the interest was cut in half.
I think that makes sense. You are less fiddling around and only need CodePen when you are really stuck but AI solves many problems now.
The shift from sharing little HTML/CSS/JS components to a website builder is probably their answer to this problem. They want to position themselves to compete with a Loveable, Replit and others. It's probably not the same, a little more technical, but the direction definitely shifted.
Source: https://trends.google.com/explore?q=codepen&date=all&geo=Wor...
Used to love this site back when you could see what other craftsmen were actually creating by hand.
What you can do with it: https://blog.codepen.io/2026/07/23/two-point-oh/#what-can-it...
A CodePen api/protocol that supports a local ide opening a CodePen, making changes in the user's developer tooling and pushing back to the pen would be a game changer.
VCS extensions for jetbrains and vscode could hook into an api and smoothly pull and push pens. This feature was always my hangup for using CodePen as a UI playground.
If the owner reads this, I'd consider getting codepen.app registered as a public suffix [1].
[1] https://publicsuffix.org/
CodePen is like an HTML/CSS/JS temple. Greate place to explore the limits of what native web offers.
https://jsfiddle.net/ used to be also very popular, I see that it is still available, but haven't used it in a while.
Which other alternatives do people use now days?
Used JSFiddle years ago too. I barely touch these anymore. Google Trends shows JSFiddle lost the fight: https://trends.google.com/explore?q=codepen%2CJSFiddle&date=...
I made https://jsmess.com as one of my first vibe code projects and have been incrementally improving it over time when I have leftover tokens...
It's closest to JSFiddle but completely static, hosted on GitHub pages. No server side processing. Projects are private, saved on localstorage and can be shared via encoded URLs.
This has been a really fun way to ramp up with LLM based coding and I learned a ton, and made a tool I actually use quite often.
https://jsbin.com is what I've used since forever, it's fast and easy, used to use JSFiddle before that. I think you used to be able to create "bins" without having to login way back though. Honestly don't find much use for them anymore nowadays, if I want to share any web/web-like I do it via my own servers + domain and send links to people.
Always felt like the worst aspect of code pen was the exploratory side. Looking up codepens is extremely poor, you often only look up one "tag" and after the first page quality gets worse.
Wish it had an advanced search that I can filter codepens for something useful like this: animejs, svg, forms; with the result being codepens that use anime.js around a form.
I'd probably use codepen as the defacto web ui reference if it had better search capabilities. Right now it's mostly github + blogs, but only because it's so hard to find anything worthwhile on codepen. Gold is there but it's deep under the strata with no willing to actually get it.
This should be the top comment, especially since organic search has been lost. Not to be overly critical because Codepen was great in its time, but it's now on individual websites to upgrade their search game if they want to stay relevant and useful.
Edit: Just noticed that logging in is now required for search. I'm sure this is an attempt to prevent AI scraping, etc., but it's a big mistake IMHO. While I have an account, I don't log in much and tend to abandon sites that require log in. Among nerds, I'm probably not the only one like this.
this is neat. Thank you.
any intersection with val.town use cases?
Whoah… CodePen… didn’t remember that… since the overflow of ai tools, this feels a bit like ancient tech. Used to love CodePen.