Disqus - Latest Comments for zaach

Re: Mozilla launches Persona Identity Bridge for Gmail, lets users sign in with their existing account credentials

zaach — Thu, 08 Aug 2013 23:44:29 -0000

I don't see where there is any loss of control... this would be no different than Google deciding to support the protocol themselves. Users still decide what identity to use, and where.

Re: Mozilla launches Persona Identity Bridge for Gmail, lets users sign in with their existing account credentials

zaach — Thu, 08 Aug 2013 22:21:43 -0000

Persona's protocol has been open and open source since day one. Any email provider can implement the protocol and become a "verifier of identification". This "bridge" is to allow a major email provider to support the protocol without waiting for them to implement it, which is a huge win for users, today. Users can still use any email they chose to sign in, so it's no less open.

Re: Guest Post: Finland – Freezing Cold, But So Hot Right Now

zaach — Wed, 19 Jan 2011 01:37:55 -0000

I think you undervalue the affects of roadway congestion and public transportation. Quicker, more efficient transportation benefits all businesses within the vicinity. You're right in that it probably wouldn't be profitable for one company to take on, which is why it's better suited as a "group effort", in effect, taxation. Privatization is a hammer, but not everything is a nail, as the cliché goes.

Re: Slides and code from "Build Your Own Programming Language with JavaScript"

zaach — Wed, 21 Jul 2010 16:39:26 -0000

Hand written parser can be useful, but parser generators do help manage the complexity a bit, especially for complex langauges. Glad I could help. :)

Re: DailyJS: Jsawk and JSONLint

zaach — Tue, 23 Mar 2010 17:17:33 -0000

I threw together a pure JS version (no server requests) of JSONLint a little while ago http://zaach.github.com/jso...

Re: DailyJS: Models with js-model

zaach — Wed, 10 Mar 2010 20:29:35 -0000

Underscore is useful and all but shouldn't be a dependency of another library, IMHO.

Re: MetaLab Goes Open Source?

zaach — Tue, 09 Mar 2010 22:02:43 -0000

Well, at least you guys are semi-famous now! I'm sure thousands of people just discovered Metalabs existed. Still a fairly big oversight by Mozilla, I'm guessing.

Re: Jison, now with more Bison flavor.

zaach — Fri, 12 Feb 2010 09:35:55 -0000

There are a couple others already out there:
* http://boshi.inimino.org/3b...
* http://code.google.com/p/es...

But no, it should not be too difficult since you can reference the Bison grammar JavaScript Core uses. I was planning on building a parser based on that, but if you want something already implemented I'd take a look at those other two.

Re: Suit My Mind | devdiari.es : a sample Cappuccino/Rails application open sourced for your pleasure

zaach — Thu, 14 Jan 2010 13:39:44 -0000

Currently not working in Firefox 3.5+

Re: EtherPad Blog: Google Acquires AppJet

zaach — Fri, 04 Dec 2009 14:55:33 -0000

Congrats! But this centralization of everything is worrisome. I'd rather Google keep it separate and perhaps integrate the two.

Re: Affiliate Click Tracking With Rack And Redis Because I Care

zaach — Thu, 29 Oct 2009 08:23:50 -0000

Oh, yeah, that makes more sense.

Re: Affiliate Click Tracking With Rack And Redis Because I Care

zaach — Thu, 29 Oct 2009 00:56:23 -0000

Sinatra runs on Rack, so I'm not sure how Rack could be slower. Rails also runs on Rack, so you could conceivably use this middleware with either framework, or a minimal Rack app as I believe you suggest.

Re: Stop Using arguments.callee

zaach — Fri, 24 Jul 2009 20:12:42 -0000

I'm aware of what the link says and the drawbacks of JScript's implementation. What that article does not mention, and indeed what caused me to write this article, is the recommendation by the ECMAScript committee, and the next implementation of JavaScript, ECMAScript 5. In ES5 strict mode, arguments.callee will throw an exception. Implementers are hedging for strict mode to become the defacto default mode for JavaScript developers, so use of arguments.callee should be discouraged on those grounds. I think the advantages of strict mode might be worth it, but we'll have to wait and see.

I wouldn't expect this to be noted on the MDC page yet, as the ES5 spec is not final yet.

Re: bitquabit - The One in Which I Say That Open-Source Software Sucks

zaach — Wed, 08 Jul 2009 23:01:28 -0000

I'm not sure how you're measuring popularity or how relevant that actually is, but Android is very usable.

Re: Twitter Launches Verified Accounts

zaach — Fri, 12 Jun 2009 00:54:54 -0000

I'm pretty sure those are bots, but yeah.

Re: A Naive, JScript Compatible Getter Pattern

zaach — Mon, 01 Jun 2009 17:52:51 -0000

You're right, I wouldn't rely on this in production.

But you bring up an interesting point! myArea1 would indeed store a *reference* to the area function and, when coerced, would continue providing the current area of that object. It's like a detached getter, but is still bound to the object thanks to a closure. Useful? Probably not. But cool!

Re: jsUnity

zaach — Sun, 01 Feb 2009 00:13:01 -0000

I know with(){...} is considered passé, but I saw a neat use of it within the Firebug source code. jsUnity could come with a default set of assertions, say in jsUnity.Assertions, and you could wrap the contents of your test suite with a with(jsUnity.Assertions){...} like so:
function myTestSuite(){ with(jsUnity.Assertions){ // use assertion functions as normal // i.e. assertEquals(jsUnity.Assertions.assertEquals, assertEquals) }}
The underlying test code would remain the same. It's a pretty neat trick.

You should make a Google group so I can post to a more appropriate venue. :)