 |
VOOZH | about |
|
VOOZH | about |
We’re so glad you’re here. You can expect all the best TNS content to arrive Monday through Friday to keep you on top of the news and at the top of your game.
Check your inbox for a confirmation email where you can adjust your preferences and even join additional groups.
Follow TNS on your favorite social media networks.
Become a TNS follower on LinkedIn.
Check out the latest featured and trending stories while you wait for your first TNS newsletter.
Shawn Wang, aka @swyx, is a feature of the developer conference circuit and also has a knack for making thoughtful, forward-looking projections for web technologies. His latest theory is that we are about ten years away from developers dropping desktop files completely — it’ll be “the end of localhost,” according to Wang. To find out more, I sat down with him at the Infobip Shift developer conference last week in Zadar, Croatia.
Localhost is a computer networking term for a user’s machine, and in this sense, it refers to a developer storing their files on their device. By default, most developers save their work to their localhost and then upload to the application environment. Indeed, this is how Git works, as it gives a developer a local copy of the full development history and then syncs changes to a repository (which is often in the cloud). Yet as Wang pointed out in June, tech companies like GitHub, Etsy and Shopify have moved all or most of their dev environments to the cloud.
There’s a very real possibility local dev may be dead in 10 years.
– @isamlambert “Planetscale doesnt believe in localhost”
– @ericsimons40 Stackblitz runs Node fast in the browser
– @GitHub runs entirely on CodespacesThis would be the biggest shift in dev workflow since git.
— swyx 🌴 (@swyx) June 6, 2022
He pointed to the trend of Cloud IDEs as another sign of the (eventual) death of localhost, along with products that allow you to code on a local device while running on a remote server.
However, near the end of his Infobip Shift presentation, he was careful to put a “we’re not there yet” disclaimer on his theory. Latency is still an issue with online coding and he raised concerns about database branching, privacy, security, and more.
That aside though, Wang told me there is a pent-up demand for 100% cloud development environments.
“I think people want to code with beefy cloud machines, which bring back the mainframe,” he said. “Especially if they’re working with anything more complex. Like if they’re messing with Stable Diffusion [the latest generative AI program], or they’re running Kubernetes, for example.”
One of his main arguments is that “cloud has eaten the outer loop conclusively.” By that, he means everything which happens after the developer pushes their code to version control.
“We do everything in the cloud,” Wang continued, regarding the outer loop, “including committing stuff to GitHub together. But the inner loop has been on-device for so long. While at the same time, the inner loop for email has moved to the cloud, with Gmail. The inner loop for media entertainment — whether it’s streaming films, music, [or] games — has also moved to the cloud. It’s just code.”
His point is that we do almost everything online nowadays — from email to watching movies — and so why not shift coding online too?
One of the reasons for the relatively slow adoption of developers to 100% online is that cloud IDEs, in Wang’s view, are still not popular enough. One cloud service he now uses is Gitpod, which enables what the company calls “ephemeral cloud-based developer environments.” But Wang told me that it took him a while to realize the benefits of that approach.
“It took me a year of hearing about Gitpod… from, eh, why do I need it — I can always clone things down to my repo. It took me a year to realize, oh, every time a PR [pull request] comes in, I want to preview it. So I click a button and preview it live. It took me a year to try it out and then use it.”
He also name-checked Replit, an online coding platform targeted at youngsters learning to code. Replit describes itself as an “in-browser IDE to code in 50+ languages — without spending a second on setup.”
“Replit kind of started from the most pure version of this,” said Wang, “where if you are in the Replit world, you only do stuff in the cloud. You don’t know any other way, because you’re 13-14 years old. You learn to code on Replit and you’ve just always had this […] It’s just native in a way that most of us who grew up doing all this local dev, we’re not comfortable with.”
Despite the promise of fully online coding, Wang’s tweet in June received a lot of negative feedback from developers. Indeed, he himself still favors localhost coding for some parts of his workflow. “It takes a second to deploy a frontend preview with Netlify Drop and ~10 seconds with the Netlify CLI, but I still habitually use localhost for development because my iteration cycle is in milliseconds,” he wrote in his June blog post.
But he thinks it’s inevitable that issues like latency will get solved, giving developers less and less reason to continue with the localhost approach. He puts the timeframe for “the end of localhost” at ten years, although others have claimed it could happen in five.
If Wang is onto something, it means there is a big market opportunity for cloud-based developer tools. Companies like Gitpod and Replit are already doing innovative work in this area, but Wang also likes Brev, Coherence, and Railway.
Perhaps one of the above companies will be the Figma of web development — check back in five to ten years. In the meantime, I’d welcome feedback from developers about going fully cloud-based. Tweet me your thoughts on the cloud dev tools you currently use, or are testing.
Disclosure: Infobip paid for Richard MacManus to attend the Infobip Shift Conference in Croatia.
