Comments on: Call for Feedback on OpenAjax Conformance and OpenAjax Registry

By: Tribulus

Tribulus — Mon, 22 Sep 2008 17:53:11 +0000

Yeah it is right
I can agree.

By: polterguy

polterguy — Sun, 01 Jun 2008 09:35:47 +0000

Hi uize, add me up as an “open challenger”…
I couldn’t agree more with you :)

By: uize

uize — Sat, 31 May 2008 21:23:32 +0000

“Repairing” the JavaScript language by adding extensions to its native objects is kind of like being a vigilante and taking the law into your own hands. This reminds me of the wonderful and fun days of different browser vendors “taking the law into their own hands” and “repairing” HTML by adding their own extensions to HTML and to CSS. Standards bodies to the rescue, and finally it is reasonably livable to do cross-browser coding without wanting to kill yourself.

I don’t want to turn the JavaScript language into a kind of Wild West where you can’t do cross framework coding and mixing code that uses different frameworks, because everyone is trying to extend and repair JavaScript to suit their own specific sensibilities. Someone’s gonna want to make it more like Python, Ruby, or Java, or whatever. Before you know it, JavaScript no longer can maintain a single core identity.

Introducing new features to the language should be channeled through a standards process. I feel adamant about this, and so – in this respect – I openly challenge Douglas Crockford’s viewpoint on the matter. In other respects, I deeply appreciate his efforts in championing the utilization of JSON.

By: PeterMichaux

PeterMichaux — Sat, 31 May 2008 20:19:20 +0000

I agree that augmenting built in objects is a mistake. Unfortunately some influential people disagree and use the technique even when it is completely unnecessary. For example:

http://tech.groups.yahoo.com/group/json/message/1028

By: uize

uize — Sat, 31 May 2008 18:05:35 +0000

Oh, and let me add… I do think it’s a good idea to have the registry and I applaud the people behind it.

By: uize

uize — Sat, 31 May 2008 18:03:39 +0000

It’s amusing to see how many globals Microsoft declares. On the opposite extreme, The UIZE Framework tucks all its goodies under the single global of the Uize object. Being as short and as sufficiently unique as it is, this minimizes chances of namespace collisions. I don’t feel bad about staking claim to just this one global.

Also, it’s sad to see discussion around managing conflicts in extensions to native objects, such as new methods for String, Object, Number, etc. I realized a long time ago in my JavaScript career just how bad of an idea it is to utilize this tantalizing capability in JavaScript, in terms of interoperability of code. So, UIZE takes the approach of providing utility packages of helper static methods. I think the reason extending the native objects is such a teaze is because it gives you the sense of the power to shape the evolution of the language – but it’s a potentially disastrous idea for real world application interoperability.

On the subject of globals in general, if you find you’re having to rely on too many of them, then it suggest that something may be going wrong in the architecture design – insufficient encapsulation in the design. I would see this is a red flag.

http://www.uize.com