What kind of projects are you working on so far? Been looking to get into Deno but haven't really gotten the motivation or need to. Theoretically, it's better than node and npm in my eyes and it's the direction of TypeScript development I want to get into. But yeah leaving node and npm behind just for the "ideal" is kind of impractical for me.
@@gadgetboyplaysmc APIs (both personal projects and for work out in production). Have also used Fresh and deployed with Deno Deploy for some personal websites. A lot of smaller hobby projects and scripts that I use locally on my machine. I believe Deno really excels at small to medium sized projects, as it gets you up and running really fast, while in node you would typically be doing more setup work. But the benefits can still be seen for larger projects. Being able to write code using Deno and then publishing to both Deno and NPM is also really cool. I have a library I made to improve fetching of data from APIs to clients, which I use in a NextJS project as well as on the API side, which is a Deno project.
Ryan revolutionized the world with Node and he'll do it again with Deno. Node helped me transition from Java/Android to a more fulfilling JS career. Thanks, man!
chalk requires env access. Looks like a previous version of deno (1.30.0) properly prompts to allow env: ✅ Granted env access to "FORCE_COLOR". ✅ Granted env access to "TF_BUILD". ✅ Granted env access to "TERM". ✅ Granted env access to "CI". ✅ Granted env access to "TEAMCITY_VERSION". ✅ Granted env access to "COLORTERM". But 1.31.0 and newer do not prompt.
there is not a single deno deploy server on the entire african continent .. how can you claim it is "truly global" when 1.2 Billion people (half which active internet users) will never benefit from this (low latency which is the whole point)?
It's super funny I was originally turned off of Deno due to the URL over NPM choice, and I didn't consider Node alternatives again until Bun came out a few weeks ago. But now I've tried Bun and it's buggy as heck, not stable at all or ready for v1, and so I'm wanting the fully integrated experience of Bun but with stability, and so Bun is actually turning me on to Deno again now that Deno supports npm. Funny how that played out.
excellent, would love to (at some point) see a) native modules like serialport compiling OR b) webusb/webserial/webbluetooth. it would be awesome for shipping software that talks to physical devices.
Have a bit of a love/hate relationship with Deno. I have a project in Node that generates a keypair, and the public key is exported with a particular padding. This is trivially easy with Node, but is currently impossible with Deno. The algorithm ("RSA/ECB/PKCS1Padding") doesn't exist on webcrypto, and its nearest equivalent doesn't allow exporting. And Deno's Node compatibility doesn't cover this area yet. And the there's NO workaround for this because you have no access to the internal key-data held in Rust. And this is ultimately my beef with Deno, or perhaps rather with the direction of web standards more generally, with this over-committal to deferring anything and everything even remotely taxing to Rust, and keeping all the data internally. It makes me feel like I'm metaprogramming. It makes me not want to code in JS, because not even Deno itself wants to do anything in JS. And the very same thing goes with Bun too.
Very good take. Originally JS/TS strength is async data IO. Anything data state really. When it comes to compute, encryption etc there is a good case to put on the polymorphic hat and use the tooling that excels at that and interface the two worlds. But you pointed out the problem: Web standards are inflating and getting into system territory, where stuff is weirdly or not at all supported. I have not used deno but it feels like, its going a bit to the roots of that IO async data strength (which covers a metric ton of business logic) but cuts overgrown end to end experiences frustratingly short. Like USB access, bluetooth, cryptography, network level stuff etc etc
There is no fundamental reason why JavaScript must be inherent to the web, it is today because it's part of the standard but the standard could evolve to support some other language (it doesn't even have to replace JavaScript, they could exist simultaneously, perhaps gaining market share over JavaScript gradually overtime or suddenly). The new language doesn't even have to be something humans code in directly, it could be something that programs get compiled into.
The slide mentioned that Deno supports NPM modules using common JS, but the latest versions of chalk use modules only, so it's no one's bug - it works as intended
![Cody Johnson - I'm Gonna Love You (with Carrie Underwood) [Official Music Video]](http://i.ytimg.com/vi/yy9PuYMU29g/mqdefault.jpg)
So much more consistant, reliable and focused, than Bun.
Deno is quickly becoming my go-to tool for most projects because it enables fast and easy development. It's been a blast to work with.
Great to hear that!
What kind of projects are you working on so far? Been looking to get into Deno but haven't really gotten the motivation or need to. Theoretically, it's better than node and npm in my eyes and it's the direction of TypeScript development I want to get into. But yeah leaving node and npm behind just for the "ideal" is kind of impractical for me.
@@gadgetboyplaysmc APIs (both personal projects and for work out in production). Have also used Fresh and deployed with Deno Deploy for some personal websites. A lot of smaller hobby projects and scripts that I use locally on my machine.
I believe Deno really excels at small to medium sized projects, as it gets you up and running really fast, while in node you would typically be doing more setup work. But the benefits can still be seen for larger projects.
Being able to write code using Deno and then publishing to both Deno and NPM is also really cool. I have a library I made to improve fetching of data from APIs to clients, which I use in a NextJS project as well as on the API side, which is a Deno project.
Love it, but waiting on Storybook tooling for Deno Fresh@@deno_land
Ryan revolutionized the world with Node and he'll do it again with Deno.
Node helped me transition from Java/Android to a more fulfilling JS career.
Thanks, man!
We appreciate your comment!
chalk requires env access. Looks like a previous version of deno (1.30.0) properly prompts to allow env:
✅ Granted env access to "FORCE_COLOR".
✅ Granted env access to "TF_BUILD".
✅ Granted env access to "TERM".
✅ Granted env access to "CI".
✅ Granted env access to "TEAMCITY_VERSION".
✅ Granted env access to "COLORTERM".
But 1.31.0 and newer do not prompt.
there is not a single deno deploy server on the entire african continent .. how can you claim it is "truly global" when 1.2 Billion people (half which active internet users) will never benefit from this (low latency which is the whole point)?
Thanks for your comment. We're working on it.
@@deno_land That said, I use deno heavily and am a big fan of its principles. Keep up the good work!
It's super funny I was originally turned off of Deno due to the URL over NPM choice, and I didn't consider Node alternatives again until Bun came out a few weeks ago. But now I've tried Bun and it's buggy as heck, not stable at all or ready for v1, and so I'm wanting the fully integrated experience of Bun but with stability, and so Bun is actually turning me on to Deno again now that Deno supports npm. Funny how that played out.
That's awesome. How's Deno working for you? We love feedback: github.com/denoland/deno
I am a simple man. I see the hero speaks, I hit like.
😂
excellent, would love to (at some point) see a) native modules like serialport compiling OR b) webusb/webserial/webbluetooth. it would be awesome for shipping software that talks to physical devices.
22:47 "DNT takes care of all of that garbage for you"
Love it!
Have a bit of a love/hate relationship with Deno. I have a project in Node that generates a keypair, and the public key is exported with a particular padding. This is trivially easy with Node, but is currently impossible with Deno. The algorithm ("RSA/ECB/PKCS1Padding") doesn't exist on webcrypto, and its nearest equivalent doesn't allow exporting. And Deno's Node compatibility doesn't cover this area yet. And the there's NO workaround for this because you have no access to the internal key-data held in Rust.
And this is ultimately my beef with Deno, or perhaps rather with the direction of web standards more generally, with this over-committal to deferring anything and everything even remotely taxing to Rust, and keeping all the data internally. It makes me feel like I'm metaprogramming. It makes me not want to code in JS, because not even Deno itself wants to do anything in JS. And the very same thing goes with Bun too.
Very good take. Originally JS/TS strength is async data IO. Anything data state really. When it comes to compute, encryption etc there is a good case to put on the polymorphic hat and use the tooling that excels at that and interface the two worlds. But you pointed out the problem: Web standards are inflating and getting into system territory, where stuff is weirdly or not at all supported. I have not used deno but it feels like, its going a bit to the roots of that IO async data strength (which covers a metric ton of business logic) but cuts overgrown end to end experiences frustratingly short. Like USB access, bluetooth, cryptography, network level stuff etc etc
Great talk, thanks Ryan!
Amazing platform, 🎩 off.
🎉
There is no fundamental reason why JavaScript must be inherent to the web, it is today because it's part of the standard but the standard could evolve to support some other language (it doesn't even have to replace JavaScript, they could exist simultaneously, perhaps gaining market share over JavaScript gradually overtime or suddenly). The new language doesn't even have to be something humans code in directly, it could be something that programs get compiled into.
The slide mentioned that Deno supports NPM modules using common JS, but the latest versions of chalk use modules only, so it's no one's bug - it works as intended
Ryan my model... wat a simple but complex guy. amazing work mahhhn 🔥want to be like you someday
Use hono, stop using express.
Ryan is literally talking to me 😂
Love what you guys are doing! Thanks a lot!
Amazing talk! Thank you Ryan and all Deno team.
Nice compression demo, but can we have higher res, video?
Let me see if I can get a higher res copy. Thanks for your patience!
Unfortunately this was the highest res we could get.