Talk:JavaScript Object Notation: Difference between revisions
imported>Tom Morris No edit summary |
imported>Chris Day |
||
Line 1: | Line 1: | ||
{{subpages}} | |||
== Suggestions for development == | == Suggestions for development == | ||
Latest revision as of 14:53, 7 September 2008
|
Metadata here |
Suggestions for development
I'm delighted to see this, under a couple of hats. Personally, I happen to need to do my first JavaScript programming. As a Computers Workgroup editor, I think there is valuable information here.
The title is a little confusing. As a style note, and for other reasons, I'd do a MOVE early, changing it to JavaSecript Object Notation, with a redirect from Javascript object notation. I myself remain confused about some suggested editing conventions that don't capitalize a proper name. In the first sentence of the introductory text, a simple statement of what it is, put JavaScript Object Notation (JSON) in bold (as shown here); that bolded material in a sentence otherwise of normal text
More importantly, this appears to be the first material we have on JSON. Now, good or bad, people here have called me a "top-down" rather than "bottom-up" editor and author. When I coauthored my first book, my colleague and editor beat me with assorted editorial whips until I got into the habit of both defining a term when it was first used, and then providing a top-down context. In this case, JSON is linked to JavaScript, but not the other way around. Somewhere in the JavaScript article, there should be a few sentences mentioning JSON, and then linkig to it.
In this article, there should be more context, such as a brief introduction to the JSON notation, and perhaps generic comment on data interchange formats. You've mentioned XML. I discover Domain Object Model isn't defined anywhere, so a new reader would have trouble, I suspect, making the transition from XML, where there is an article, to this one without anything on DOM. Very shortly, I'm going to edit some unpublished work of mine on DOM into an article. Please feel free to edit it so it fits in the flow; I am a networking person that understands packets, but XML more from the perspective "if this were like ASN.1"...
It's worth double-checking if there are any conflicting abbreviations/acronyms. Since I'm on my first cup of coffee, it took me a few minutes to realize that I was thinking of DOI (domain of interpretation) in IPSec; IPSec doesn't use DOM. If anything else does, there needs to be disambiguation (e.g., DOM (foo) vs. DOM (JavaScript)).
So, in this article, you need some subheads. The introduction defines the problem JSON is trying to solve. Next, who saw the need to develop it, what was the need, and how was it developed? Is there an open group for it?
Following that, give some examples of its notation, and whether there are tools that help write JSON.
Comparing it to other exchange formats deserves its own section, if not an independent article. Certainly, explain how it differs from XML, XML with DOM, and, if relevant, HTML, XHTML, and ASN.1--or any other competitive methods.
You're off to a good start by bringing in a new subject. The challenge now is to integrate it more with other CZ articles, and put more structure into this article.
Howard C. Berkowitz 11:22, 12 August 2008 (CDT)
- I agree with Howard. Thanks for the article, but I think it should definitely be moved to "JavaScript Object Notation" now. I'm going to be bold and do it right this very second. --Tom Morris 18:41, 12 August 2008 (CDT)
- Also, one generally doesn't "write" JSON by hand nor use a tool (as one might for XML), but rather JSON serialization is a library or built-in feature of one's programming language so that one can simply output JSON strings by, for instance calling a "to_json()" method on an object or associative array (etc.). In some languages, there are multiple JSON implementations which have a variety of looser parsing rules and speeds - see, for instance, this blog post comparing Python JSON implementations (no surprise: the C library runs rather quickly). --Tom Morris 18:47, 12 August 2008 (CDT)