Deno is Coming

preview_player
Показать описание


----
Checkout my side projects:




----

----

----
Follow Me Online Here:

#benawad

----
#benawad
Рекомендации по теме
Комментарии
Автор

Already seeing the job posts coming: Atleast 3 years professional experience with Deno

katywings
Автор

I opened my mouth and tried to sell a Deno project to my boss. I'm now a carrot farmer.
Good video either way. Imma go farm some carrots.

alanhegewisch
Автор

Deno has a few more advantages that just the ones you listed:
* While short import is nice in Node.js, it requires `npm install`, which would slow down your development iteration. Deno solves this by auto-fetching needed libraries.
* Duplicated `node_modules` in Node.js is a waste of storage space. Deno solves this by having a shared cache (~/.cache/deno).
* Setting up common tooling such as TypeScript and webpack is a waste of time. Deno has them built-in.

khaix
Автор

Next in the YouTube recommendations:

How to farm carrots

josephkim
Автор

I remember when Php developers werent exited about node here we are . Web always evolves. You create a project with npm and you have 600MB node modules

Ozuqam
Автор

When you don't have enough problems to solve, you create your own

bhupeshimself
Автор

I think the point you made about deno not having anything super flashy to show off to others might be its biggest downfall. As you said, it will be very hard to convince bosses to use it for future projects if on the face of it, there is not much that is getting added to the current state of affairs. Couple that with the incompatibility with current NPM packages, and it feels like deno really has an uphill climb ahead.

CodingWithChaim
Автор

Recruiters tomorrow: "Looking for a new Deno developer, 5 years experience"

SimonCoulton
Автор

Seems like a natural evolution more than a complete revamp. Keep the best, clear the rest.

willmakk
Автор

what an incredibly reasonable take. love it Ben

swyxTV
Автор

"legacy angular codebase" -> "spiraling depression" -> "carrot farmer"

jamesweber
Автор

I think there was an important thing left out which is that each project doesn't need its own node modules folder. Once you import something, it will update by itself and it will be globally accessible, so if different projects use the same import, it won't create a separate folder for each and every project, it will only download it once to the machine. I think impacts development quite a bit, it's one less thing to worry about or wait for the install and have redundant files and folders.

theSUBVERSIVE
Автор

Thank you for covering this in such a thorough fashion. This reminds me of the promise of Pacel JS instead of Webpack. At the end of the day, it needs to work and make my life easier, or else I'll just use the technology I'm already comfortable troubleshooting. The part about removing the package JSON doesn't sound like a positive because it limits some of the control over things that may cause errors due to incompatible versions of dependencies that can be addressed inside the package json file.

busyrand
Автор

Ben Awad: "Deno js community is about to get flipped"
Me: What the auditory hallucination was that?

philippefutureboy
Автор

I have used Deno for a few projects already both terminal-oriented and web-based, and some things that I like so far about Deno is:

- Dependencies are files, not "modules". They are easier to inspect and manually audit for security.
- Shared dependency cache, shared across projects. No massive node_modules with thousands deps for a single module.
- Permissions... Node really needs this.
- TypeScript is first-class citizen – Deno compiles your source code.
- You can mostly write your applications that can be consumed both by web browsers and Deno (given that you don't use Deno-specific APIs)
- Built in test runner (although it could be improved; there's no mocking and async tests can fail in some edge cases)

I personally don't think Deno should care about Node. Deno is its own thing. It adds too much surface area. I believe this kind of tooling should be third party for mainly three reasons;

1) there's going to be third party solutions to this anyway, like how we have hundreds of different package managers for Node.
2) It adds a lot of friction to Deno's development cycle. It needs to be maintained and it needs to be compatible across Node versions.
3) We likely will end up writing modules with Node-like APIs, which goes against the point of Deno which is to be compatible with web standards, not Node standards.

dealloc
Автор

I think the “secure by default” part of Deno is being undervalued here. For me this is the killer feature. Imagine not having to think about whether a library used by a library used by a library that you installed and use is mining bitcoin on your computer or stealing your ssh keys. What other programming language / runtime can make the guarantees that Deno makes? I don’t know of any.

MarkVolkmann
Автор

Gotta agree with you if a new technology doesn’t give us 200% more efficiency or performance its not worth moving too. But the creative engineer in me loved the clean and unambiguous design in dino

hnasr
Автор

deno is literally an anagram for node 😂

danylfernandes
Автор

OMG, it's been TWO years already

ЭнрикеЧурин
Автор

Mom, can we have Deno?


No. There is Deno at home.




Deno at home: Node

chanlito_