Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Parcel: Fast, zero configuration web application bundler

github.com

Post reply on HN
Print view View original

51–60 of 98 posts

Re: Parcel: Fast, zero configuration web application bundler

#51

Post by dmitriid »

yzmtf2008 wrote: post #24
joshwa wrote: post #2

Ah, the neverending cycle. 1. The existing tool is way too complicated to configure and slow. Let me introduce a new, zero-configuration, blazing fast tool 2. OK so because it's strongly opinionated there are some things it can't do. So we'll allow you to extend it via plugins. You'll have to configure this. 3. Due to the increasing popularity, we are making a meta-version that lets you re-plug any element in the ent…

I think this is a great exaggeration of the "javascript build tools are hard to config" meme, and it honestly is obsolete. Maybe the meme was relevant two years ago, but every since (1) create-react-app and (2) webpack 3, the configuration story has greatly improved. With Webpack 4 and Parcel especially, these tool are both zero config and configurable . Your point 1 and 3 are actually not mutually exclusive, and rep…

You are so wrong on so many levels.

Have you ever looked at what CRA does? Just do an `eject`. It layers and layers and layers of configuration. The moment you need something not covered by CRA you'll be stuck with the insanity that is webpack config. [1]

webpack 3 did nothing to solve the configuration problems. Heck, just half a year ago one of the core devs of webpack argued that there could be no defaults in webpack and that webpack config is horrendous "because it's javascript" [2] [3]

webpack 4 addresses just some of the long-standing problems with webpack configuration leaving the actual config as insane and inscrutable as ever. And the only reason it exists is probably due to the public outcry of the larger community and tools like Parcel somewhat stealing the thunder.

Parcel is probably the best thing to happen to JS bundlers in recent years, but it's still not without its problems.

[1] I did: https://twitter.com/dmitriid/status/908677755046449154

[2] Webpack team arguing there can't be defaults in October 2017: https://twitter.com/dmitriid/status/920542261020160000

[3] And continues pretending config issues are because of Javascript: https://twitter.com/dmitriid/status/920714918625775618

Re: Parcel: Fast, zero configuration web application bundler

#53

Post by oweiler »

spankalee wrote: post #29

I love that Parcel is out, even if yes it is another bundle, because it helps highlight the major problem with bundlers right now: they (but especially WebPack) are a meta-platform over the web unto themselves and they encourage use of features that are not supported by the web platform, in a way that's not easily transformed into what the web platform does currently, or will realistically, support, creating lock-in.…

Except we don't have great modules support in the browser.

Re: Parcel: Fast, zero configuration web application bundler

#54

Post by oweiler »

jacobsimon wrote: post #13

I've used used Parcel for a few projects this year and really enjoyed it for the most part. I've found it's basically effortless for building static projects, but its lack of documentation can be frustrating when errors pop up (e.g. I was stuck on an older version for a while due to some issue with a newer release) and I still haven't quite figured out the best pattern for extracting the hashed bundle names for use i…

Webpack's zero config is a joke.

Re: Parcel: Fast, zero configuration web application bundler

#55

Post by nostalgeek »

crooked-v wrote: post #33
bradknowles wrote: post #12

IMO, the fact that 1726 modules is a “reasonable” number for a web app is a clear indication that there is something seriously wrong with this process.

That's the result of JS lacking anything resembling a standard library. Just look at how long it took just for Object.entries to make its way into the language.

This is both an education issue and a std lib issue. There is no need for packages such as "is-number" or "is-odd" when the language already provides the facilities it check if a value is a number or is even/odd, yet these packages are downloaded a thousand times a a day because they are dependencies in this or that popular package.

For some the first reflex is to add a dependency instead of spending 5 minutes thinking of simple solutions for their use case. But the latter isn't restricted to the node ecosystem.

The avalanche of useless modules profits mainly to NPM, a for profit organisation operating a closed source package repository which imposed itself on the node project in dubious ways. Even nodejs creator regrets it.

Re: Parcel: Fast, zero configuration web application bundler

#56

Post by spankalee »

oweiler wrote: post #53
spankalee wrote: post #29

I love that Parcel is out, even if yes it is another bundle, because it helps highlight the major problem with bundlers right now: they (but especially WebPack) are a meta-platform over the web unto themselves and they encourage use of features that are not supported by the web platform, in a way that's not easily transformed into what the web platform does currently, or will realistically, support, creating lock-in.…

Except we don't have great modules support in the browser.

Chrome, Safari, Firefox and Edge support modules. Firefox is landing import.meta.url and working on dynamic import() now.

Re: Parcel: Fast, zero configuration web application bundler

#57

Post by Hurtak »

modernerd wrote: post #50
cjhanks wrote: post #4

The code looks nice, I hope it has helped you in your life. But frankly, if a JavaScript developer ever really wants to solve the problem... Learn enough C++ to write the equivalent code that compiles 6.5M in 0.1 seconds (assuming a nice HD).

Others feel the same. Fastpack is a binary JS bundler written in OCaml: http://fastpack.io/ It’s new and missing some features ( http://fastpack.io/docs/get-started.html ), but already faster than Webpack / Parcel. It also supports Webpack loaders.

So it is 3 times faster than Webpack, ATM it does not seem worth to be switching to this. Also the Parcel comparison seems wrong (maybe they are using too old version of Parcel?), see Parcels own benchmarks https://github.com/parcel-bundler/parcel#benchmarks

Re: Parcel: Fast, zero configuration web application bundler

#58

Post by yzmtf2008 »

dmitriid wrote: post #51
yzmtf2008 wrote: post #24

Earlier quoted context omitted.

I think this is a great exaggeration of the "javascript build tools are hard to config" meme, and it honestly is obsolete. Maybe the meme was relevant two years ago, but every since (1) create-react-app and (2) webpack 3, the configuration story has greatly improved. With Webpack 4 and Parcel especially, these tool are both zero config and configurable . Your point 1 and 3 are actually not mutually exclusive, and rep…

You are so wrong on so many levels. Have you ever looked at what CRA does? Just do an `eject`. It layers and layers and layers of configuration. The moment you need something not covered by CRA you'll be stuck with the insanity that is webpack config. [1] webpack 3 did nothing to solve the configuration problems. Heck, just half a year ago one of the core devs of webpack argued that there could be no defaults in webp…

Hey, I looked at ejecting as early as September 2016, but why let that stop you from saying “you’re wrong on so many levels” :)

Looking at JavaScript build tools today and ridiculing it makes as much sense as ridiculing the width of car lanes (from width of wagons) or ridiculing the width of rockets (width of railroads).

Create-react-app works just fine with the 80% case you wanted so badly for webpack team to implement. (Not to mention there are easy ways to make incremental changes without ejecting.) But why let that stop you from writing a rant about webpck’s lack of default either :)

Re: Parcel: Fast, zero configuration web application bundler

#59

Post by tomnipotent »

some_account wrote: post #49
tomnipotent wrote: post #21

Earlier quoted context omitted.

You mean iterating on something, sorta the entire conceptual model behind the scientific method of experimentation?

Iterating without improvement is just making a different wheel, if you know what I mean. It gives the impression of progress without actually making progress. This is also why I got tired being a programmer after doing it for many years. First version is always super nice. Then you add features, and refactor. Doing this is slower and slower the bigger your software gets. At the end, your software is filled with compl…

> No. At the end, his product is going to be replaced by other products that does the job better - until they also meet the same fate.

Yes, this is the very definition of innovation. Just because what you're working on is going to be replaced, doesn't mean it's not innovative. Of course it's going to be replaced, that's the whole purpose. Hopefully we get better and smarter along the way. Sometimes something that seems trivial to you as an engineer is mind blowing to a layman with domain expertise that goes on to leverage your software to help make the company more money in ways no one was expecting (or maybe save money). In other words, innovating.

Re: Parcel: Fast, zero configuration web application bundler

#60

Post by dmitriid »

yzmtf2008 wrote: post #58
dmitriid wrote: post #51

Earlier quoted context omitted.

You are so wrong on so many levels. Have you ever looked at what CRA does? Just do an `eject`. It layers and layers and layers of configuration. The moment you need something not covered by CRA you'll be stuck with the insanity that is webpack config. [1] webpack 3 did nothing to solve the configuration problems. Heck, just half a year ago one of the core devs of webpack argued that there could be no defaults in webp…

Hey, I looked at ejecting as early as September 2016, but why let that stop you from saying “you’re wrong on so many levels” :) Looking at JavaScript build tools today and ridiculing it makes as much sense as ridiculing the width of car lanes (from width of wagons) or ridiculing the width of rockets (width of railroads). Create-react-app works just fine with the 80% case you wanted so badly for webpack team to implem…

You can look at it today, it will be mostly the same: hacks, inscrutable configuration, custom overrides for stupid things etc. etc. etc.

I just looked at it a minute ago.

As for the rest you are equally wrong in anything from assuming that webpack 3 made configuration easier to assuming that configuration became easier just because CRA hides all this complexity from us.

Post reply on HN