why thank you! :)
I just got to this line on Riot: , and is the *best pattern for unit tests.* I got into TDD in 2008 and have been hooked ever since. Back to Riot reading...
I couldn't agree more with you! It all seems like big conspiracy, people using frameworks like crazy. it feels so wrong.
Should have used ExtJS (combined with sencha touch) in google trends comparisons also... and with the massive decline in jquery mobile...
I'm currently scouring the web for a decent framework, but I noticed you said something and wanted to pass some advice. If you want the two-way binding from Angular without all the extra stuff, knockout is a great framework. It's only 13K and crazy fast in my opinion: http://knockoutjs.com/
If you read trough all the coments you are definitely not the first one to bring out Knockout :) I totally agree with you. If you want two-way binding without extra stuff Knockout is clearly a better option than Angular. However I personally prefer to build a programming API and send (documented) events for JavaScript presenters. Two-way data binding is another way to build dynamic views and Knockout seems to be a good / minimal choice for that. I just belong to the event camp (aka. backbone.js).
Oh, but I thought jQuery was large too! Is it worth it to you? I'd love to hear how you ended up using it. I do agree that Vanilla JS is the best toolkit :) And the recent fast.js revelation that the side effects of many JS Array methods makes VJS MUCH faster than native (and also works with Typed Arrays, important for my project. It would be nice if Object.observe was in most browsers, would make "binding" that much easier.
jQuery is large. No doubt about it. But it is used by [94.1% of all the websites](http://w3techs.com/technologies/details/js-jquery/all/all) whose JavaScript library we know. It's a safe dependency. I ended up using it roughly 10 years ago when it was in version 1.2. I mainly use it to cover all cross browser issues and it makes DOM manipulation easier.
Unfortunately, software with overheads/delays/bad-performance always wins in the software industry (that's history: Assembly->C-> C++ ->MFC-> .NET -> Javascript). That's because every company wants the best results in the shortest time and usually that kind of software has the ability to provide it. Now, I think Angular is great, even with its large API. And I also think that its API should expand to replace Javascript's broken core functions (because I see Javascript core functions as a pile of obsolete and buggy trash). I am terribly sorry that browsers run Javascript. That's a fail on its own. Fortunately, JQuery makes JS a little bearable but the trash is still there. I am also sorry for the mess of HTML. This is a recipe for failure too (the people that make the standards must be serious alcoholics to produce this trash). There, Angular comes in to save the day. It certainly can't change the way our crappy browsers work, but at least it tries to protect our intellect from this garbage...
changed my mind...
Changed your mind? About what?
"If you look at the history of frameworks in any programming language, it’s a history of failures." How do you define "failure"? There are popular, long-lasting frameworks actively used to this day in several languages, eg, Struts 2, Spring, Wicket (Java); Django, web2py (Python); CakePHP, Symfony (PHP), just to name 3 languages. These frameworks don't seem to be going away any time soon.
I'm sure you know what I mean with that. There are far more failures than success stories. And your list also looks a bit suscpicious; slightly outdated tools that will eventually fade away since there are already "modern" alternatives available.
[Riot 2.0 is out!](/riotjs/) This is large rewrite brings custom tags for all browsers starting from IE8. Think React + Polymer, but squeezed into 2.5K. It's a full stack with virtual DOM, "reactive views", router and observable.
'3.5 years to build Moot' - r u serious?
> @resolution1441 > 3.5 years to build Moot' – r u serious? Absolutely. There are multiple reasons for why it took that long, speed, scalability and (apparent) simplicity of the UX being some of them. There's a blog entry about this called [You spent 3.5 years on this??](https://muut.com/blog/technology/you-spent-3.5-years-on-this.html)
Are you recommending then using YOUR framework then? That's not exactly framework less, is it? =)
This is absolutely the correct way to handle this. If you're embedding JavaScript into someone else's site, you can't go chucking Ember in there, or Angular, or even Backbone. You've built some simple MVC, it's enough. That's not to say frameworks are bad, just not here.
David Walsh is Mozilla’s senior web developer, and the core developer for the MooTools Javascript Framework. David’s blog reflects his skills in HTML/5, JS and CSS, and offers a ton of engaging advice and insight into front-end technologies. Even more obvious is his passion for open source contribution and trial-and-error development, making his blog one of the most honest and engaging around. Website: davidwalsh.name
David Walsh is Mozilla’s senior web developer, and the core developer for the MooTools Javascript Framework. David’s blog reflects his skills in HTML/5, JS and CSS, and offers a ton of engaging advice and insight into front-end technologies. Even more obvious is his passion for open source contribution and trial-and-error development, making his blog one of the most honest and engaging around. Website: davidwalsh.name