JavaScript Devs: Is It Still Worth Learning jQuery?

Nerval's Lobster writes: If you're learning JavaScript and Web development, you might be wondering whether to learn jQuery. After nearly a decade of existence, jQuery has grown into a fundamental part of JavaScript coding in Web development. But now we're at a point where many of the missing pieces (and additional features) jQuery filled in are present in browsers. So do you need to learn jQuery anymore? Some developers don't think so. The official jQuery blog, meanwhile, is pushing a separate jQuery version for modern browsers, in an attempt to keep people involved. And there are still a few key reasons to keep learning jQuery: Legacy code. If you're going to go to work at a company that already has JavaScript browser code, there's a strong possibility it has jQuery throughout its code. There's also a matter of preference: People still like jQuery and its elegance, and they're going to continue using it, even though they might not have to.

Comments

VanillaJS Framework

[lgw]

I hear great things about that new-fangled VanillaJS framework. Very lightweight and fast, and already more popular than jQuery.

Re:VanillaJS Framework

[halivar]

I will admit: I looked at the page and was very interested and got super excited until I looked at the code examples harder. Well played, internet... well played.

Re:VanillaJS Framework

[Stewie241]

Not defending jQuery, but I think the VanillaJS page over simplifies things and it's examples are not quite equal, and seem to tout themselves as far better, when in fact, there is a lot of complexity that something like jQuery hides.

Examples:
1. AJAX - sure, you can memorize the special incantation of r.onreadystatechange and remember that you have to check if readyState is 4 (4? wtf?) and status isn't 200. Except in the little excerpt there there is no error handling, and you basically end up with an unresponsive page with anything except the happy path.
2. Fadeout - sure, you can do the same thing in approximately the same number of characters, but the vanilla example is far more difficult to read and interpret.
3. Selector speed - sure, it might be a lot faster to do getElementById or getElementByTagName, but again, you sacrifice a lot of readability and without good tools it is really cumbersome to write.

If performance is an issue, perhaps a different, Javascript compiler is the solution. But to suggest that everybody should hand code everything in native JS instead of using the more convenient syntax that a library provides is ridiculous.

Re:VanillaJS Framework

[_xeno_]

Basically this. jQuery is one of those things that's almost literally bloat: it adds nothing that your browser can't already do, it just wraps around it. You absolutely do not need to use it.

However it saves on development time. It's effectively a bunch of boilerplate code that you don't have to deal with. It's one of those things that if you were to decide not to use it, you're likely to end up rewriting a chunk of it by the time you're done anyway, so you might as well go ahead and use it from the get-go and save yourself some time.

(Which isn't to say you should always use it. I've written pages where the amount of dynamic code was small enough that using jQuery would make absolutely no sense. But the larger your project gets, the more sense it makes to use frameworks like jQuery.)

meh

[XO]

jquery makes an absolute mess out of javascript. Much of it involves DOM manipulation, which is something you generally want to avoid doing as much as possible. It's a pain in the ass to read, has a nasty learning curve, and it's slow as fk. Don't bother, unless you need to operate on existing jquery code, or have some other very specific reason to use it or interact with code that uses it.

Re:meh

[jandersen]

It's a pain in the ass to read, has a nasty learning curve, and it's slow as fk

Eh? It took me all of a few days to read through one of the many reasonable books about jQuery, and I found it makes it a whole lot easier to make sense of the DOM. In a browser, what else is there to JavaScript, other than messing around with the DOM? Of course, I only use JavaScript on the client side, for the server side I use J2EE and GlassFish. jQuery is perfect for my use and very, very easy to learn.

Re:meh

[meustrus]

Much of it involves DOM manipulation

Yeah? Much of jQuery is about DOM manipulation. If you don't want to make a hole, don't use a drill.

jQuery is for lazy, fat, "developers"

[Needs2BeSaid]

No. jQuery is bloated by their desire to make sure your pages are compatible with ancient Microsoft browsers.

Also, jQuery allows developers to be very lazy. Example: Many sites bring back JavaScript in their AJAX returns. Did you know that jQuery uses EVAL to process any JavaScript returned via AJAX?

jQuery UI is a horrendous memory and performance hog. There are billions of JavaScript code examples to perform the individual functions of jQueryUI without forcing your customers to download that monster.

Don't be lazy.

Don't forget legacy BROWSERS.

[Ungrounded Lightning]

A site may wish to continue using JQuery because some of its clients are using older browsers that don't support the new features that allegedly obsolete JQuery code.

Drop the JQuery code and you drop those customers. Develop future code without it and the pages with the new features won't perform with people using legacy browsers. And so on.

I've seen similar things happen over several generations of web technology. Use care, grasshopper!

Re:Don't forget legacy BROWSERS.

[PCM2]

This is tricky. It's tempting to support legacy browsers, but if you do too good a job of supporting them, you don't incentivize your users to ever get their sh*t sorted, and upgrade their browsers. It's a vicious cycle I am eager to avoid.

Yeah, but when your "users" are more properly called "customers" -- or even more important, "potential customers" -- then some web dev's desire to preach the gospel must take a back seat to doing the job the way it needs to be done, rightly or wrongly.

It's fine to push for strict browser standards when the only people who will ever see your web applications are within your own organization. Public-facing sites are a different matter.

JQuery is something to learn?

[YoungManKlaus]

I always used it like "fuck it, I'm lazy, lets go"

Re:JQuery is something to learn?

[Tailhook]

Same here. $('select something').do_stuff(function (e) {...});

That's about it. I don't feel as though I have some big investment there.

Uh... yes

[ZipprHead]

That is, If you like frameworks that are bug free, extensively tested, universally known and hide a lot of the weirdness between browsers and browser versions providing a consistent interface, then yes, definitely learn it, it takes a lot of pain away.

If you like pain, or always want to use the shinny new fancy thing, or want to learn all about IE9 and Opera edge cases. Then don't use it.

SharePoint....

[oh_my_080980980]

So yes you need it to do the stuff that SharePoint doesn't do.

Yes...

[hyperar]

As long as the web is used to publish extremely complex apps that should not be on the web, yes. Web is for video playback, reading news and blogs, Business app?, desktop, using web services.

Re:Yes...

[xxxJonBoyxxx]

>> Web is for video playback, reading news and blogs, Business app?, desktop

And how long have you been out of work?

Re:Yes...

[Literaphile]

Web is for video playback, reading news and blogs, Business app?, desktop, using web services.

2001 called, it wants its antiquated attitude back. The web evolves: deal with it. You bitching about what the web is now isn't going to make us all suddenly go back to "video playback, reading news and blogs".

Re:Yes...

[dissy]

Web is for video playback, reading news and blogs, Business app?, desktop, using web services.

I thought the web was to add inline images to your gopher menus :P

Re:Yes...

[SoCalChris]

Yeah, those are legitimate concerns with a web app. But pretending that developing a standalone app doesn't have its own set of equally painful problems is ridiculous.

With a standalone app, you've got to worry about things like .Net versions, DLL hell, installation packages, keeping users up to date, windows versions, etc... Yes, modern languages will typically take care of most of this for you, but to pretend like they don't exist is just being naive.

Yes

[Anonymous Coward]

JQuery as a framework is, frankly an utter joke. However, as a utility library it's invaluable. Learn Angular, React, or some other modern framework, and lean on JQuery for things those frameworks don't do. Simple example: Ever try to submit a from from an AngularJS controller? Yeah, not doable. Add 2 lines of jQuery? Done! So yes, learn it, and use it where it fits.

I don't get it

[quietwalker]

JQuery is just encapsulating some primarily dom-related javascript mainpulation routines with the added bonus that they try to eliminate browser differences. So, when you're saying that the browser provides features that jquery was needed for, you're really saying that the browser does things that javascript is no longer needed for.

I'm just not seeing it though. With pure HTML & CSS and a fancy new browser, can I:

Write ajax requests and parse and conditionally apply the results to various page elements?
Dynamically add and remove elements?
Perform liquid resizing based on a layout approach with glue elements and fixed-but-scalable areas - that is dependent upon the size of other elements rather than explicit browser viewport height/width?
How about perform custom input box validation?
Maybe set the value of a text box only when a value in a linked select box is changed?
Pop up a dialog when a button is clicked?
Start an image upload when you drag an image over a browser region?

In the age of ever-closer-to-desktop-application websites, I'm only seeing more and more use of javascript frameworks - of which jquery is one - and frankly I don't see how anyone could do without it. Maybe if you're making static brochure sites, I suppose, but then you weren't using javascript for that anyway.

Maybe the original poster meant to say "is it worth learning jquery instead of another framework or library" ? Otherwise I can't see anyone who actually creates web applications for a living even asking this.

5 Minutes

[wisnoskij]

It takes like 5 minutes to learn to use Jquery, it is used in 90% of all current websites, and is still one of the best if not the best library for dynamic DOM manipulation and has incredibly easy to use ajax requests. I cannot imagine creating a dynamic website without it.

Re:IE 6

[Lodlaiden]

$20 and I'll turn off my regression suite that checks if your site is still IE6 compatible.

You forget memory usage...

[edxwelch]

When you load a javascript library the browser has to allocate memory to every function in the library even if they are never used and most web sites are using dozens of javascript libraries. While this is ok on a desktop, on a tablet - which has much less memory - it means you only have enough memory to have one web page open at a time. Some web pages are so infested with javascript libraries they cause the tablet browser to crash. And they are just displaying static text and images, something that doesn't require javascript.

jQuery is a crutch.

[extranatural]

I saw a very insightful & funny talk on this subject last year. The very clever Josh Broton lays out exactly why jQuery has become an excuse not to do it right the first time. Basically it comes down to this:

A few facts about latency and user behavior: "...250 milliseconds can be the difference between a return customer and an abandoned checkout cart." "...every 100 milliseconds of latency resulted in a 1% loss of sales." "...lose 20% of their traffic for each additional 100 milliseconds it takes a page to load."

The average overhead jQuery adds to a website: "... add roughly 150ms to 1 full second of load time..."

He goes into many other good reasons too, it's well worth a read.

Slide here: https://github.com/joshbroton/...

Expand details of part of the document

[tepples]

Instead of throwing many small fragments at the browser and stealing user cycles to cobble it all together, just serve up the content already.

I have served the document. Now the user has activated a control to expand details of a particular part of the document. How should this click be processed?

Or I have served the document. Now the user has opted into real-time updates of part of the document. How should these updates be served?