rmurphey Adventures in JavaScript

EnterpriseDojo.com uses jQuery!

21 Aug 2010 edit

And you should too ... for your WordPress blog that came with jQuery already installed and probably barely even needs JavaScript in the first place. 

Here's an idea: Let's have an honest discussion about when a given solution makes sense and when it doesn't, rather than mocking a contributor to that discussion for choosing a perfectly appropriate tool for a perfectly mundane task. EnterpriseDojo.com is a blog about using Dojo in the enterprise, not a blog about using Dojo on blogs. Snickering at its use of jQuery is approximately as useful as pointing to this graph and saying case closed, leaving more reasonable, thoughtful people to explain all the ways that graph doesn't tell us a damn thing

I want to be very clear, as the dust settles around my several recent rants: jQuery has its place, and it is a very, very big place. It is a lovely DOM, Ajax, and events library, and a great way to get certain things done quickly, especially for people who may not have the luxury of learning the inner workings of JavaScript (noble a goal as that is). There was a time when DOM and Ajax and events questions felt like the questions of the day, and jQuery showed us how to answer those questions using a simple, easy-to-learn API that suddenly made JavaScript accessible to the masses. Other libraries solved the same questions but those solutions felt ugly and clunky and goshdarnit, hard, and I give jQuery the utmost credit for making JavaScript suck so much less when all I wanted to do was show or hide or slide a thing or load some content onto my page. I'll even grant that it can be used as a piece of a large application solution, though I question the wisdom of doing so for reasons outlined in another post. Heck, I'll even grant that "enterprises" aren't off their rockers to use it -- for the things it was meant to be used for. 

What upsets me is when smart people seem to say that jQuery's victory in an internet-wide popularity contest suggests, well, anything at all when it comes to more complex needs. It upsets me because my clients hear those suggestions, look at that same well-marketed graph, and I am left explaining to them that, yes, jQuery is popular, but it's popular because it answers a small set of questions easily and well -- even for people who don't even know JavaScript! -- and you, dear client, have vastly larger questions than that.

At the end of the day, these toolkit decisions ought to be about more than a popularity contest; jQuery may be the right answer, or part of the right answer, but it's imperative, to me, that my clients understand the scope of the question first. It's imperative that they make their decisions based on a full understanding of pros and cons, risk and reward, cost and benefit -- not based on a graph, not based on a popularity contest.

So again I say: we, as a community, and especially the influential ones among us, do well when we elevate the conversation beyond that contest and acknowledge that choosing the right toolkit depends on first understanding what you're choosing the toolkit for. We do well when we educate teams and decisionmakers on the lessons we've learned in the time since DOM, Ajax, and events were the big question of the day, on the best practices that have emerged, on the situations where we've, gasp, had to look beyond jQuery -- either to other tools or other toolkits -- for the answers. And we do well when we start showing them how they can do the same.

An aside: Some people I like a lot have pointed out that in the midst of all of my complaining, I have not come out and offered a solution. This is fair. Two things: One, I have not wanted this conversation to collapse into Dojo vs. jQuery vs. YUI vs. Ext vs. MooTools vs. whatever, because if it does, I think we've missed the crux of the matter: that different tools do different things, that some set out to answer complex questions and some do not. If anyone has been unclear, which I rather doubt, my personal preference of late has been to use Dojo. I do not recommend using Dojo for everything under the sun, but I find it offers a lot of utility when writing non-trivial applications. I cannot make a compelling argument for using it vs. YUI, simply because I don't know YUI well enough. I can't even make a compelling argument for using it vs. Ext, except for licensing issues that may or may not be relevant in a given scenario. Two, if you're disappointed that I haven't offered a solution, especially a jQuery-based one, I apologize. However, I feel there are too many viable existing solutions out there already, and I haven't come up with a good reason to promote a jQuery-based solution besides jQuery's popularity. And, well, see above for my thoughts on that.

Comments