<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: Web apps need scriptable interfaces</title>
	<atom:link href="http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/</link>
	<description>experiments in contemporary web development</description>
	<pubDate>Wed, 07 Jan 2009 12:18:27 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.1</generator>
		<item>
		<title>By: mauve</title>
		<link>http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/#comment-88</link>
		<dc:creator>mauve</dc:creator>
		<pubDate>Tue, 07 Nov 2006 14:18:39 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/#comment-88</guid>
		<description>Yes, that's true, I suppose, but not really what I'm getting at. I'm focusing on web applications installed on Linux because that's the problem domain I've been facing. The big Web 2.0 sites realise that people want APIs, but they are not open-source and their databases are not accessible to their users in any other way. Those huge databases are the one asset that differentiates them from locally-installed software that does much the same thing so freedom for user to work with them really builds their strength.

Besides, if they don't offer them as APIs, users will write flaky screen-scraping APIs to do the same thing but using more bandwidth.

Open-source LAMP web apps have evolved a very formulaic structure. A design pattern. Or maybe an antipattern? There's no justification that these patterns solve anything other than constraints imposed by LAMP (mainly MySQL and PHP).</description>
		<content:encoded><![CDATA[<p>Yes, that's true, I suppose, but not really what I'm getting at. I'm focusing on web applications installed on Linux because that's the problem domain I've been facing. The big Web 2.0 sites realise that people want <acronym title="	Application Programming Interface">APIs</acronym>, but they are not open-source and their databases are not accessible to their users in any other way. Those huge databases are the one asset that differentiates them from locally-installed software that does much the same thing so freedom for user to work with them really builds their strength.</p>
<p>Besides, if they don't offer them as <acronym title="	Application Programming Interface">APIs</acronym>, users will write flaky screen-scraping <acronym title="	Application Programming Interface">APIs</acronym> to do the same thing but using more bandwidth.</p>
<p>Open-source <acronym title="Linux, Apache, MySQL and PHP (sometimes Perl or Python)">LAMP</acronym> web apps have evolved a very formulaic structure. A design pattern. Or maybe an antipattern? There's no justification that these patterns solve anything other than constraints imposed by <acronym title="Linux, Apache, MySQL and PHP (sometimes Perl or Python)">LAMP</acronym> (mainly MySQL and <acronym title="PHP: Hypertext Preprocessor">PHP</acronym>).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sil</title>
		<link>http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/#comment-87</link>
		<dc:creator>sil</dc:creator>
		<pubDate>Tue, 07 Nov 2006 12:55:35 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mauveweb.co.uk/2006/11/01/web-apps-need-scriptable-interfaces/#comment-87</guid>
		<description>There are some apps that have APIs, and most web APIs have Python bindings:

Python Flickr API: http://beej.us/flickr/flickrapi/
Python Google API: http://pygoogle.sourceforge.net/

and so on...</description>
		<content:encoded><![CDATA[<p>There are some apps that have <acronym title="	Application Programming Interface">APIs</acronym>, and most web <acronym title="	Application Programming Interface">APIs</acronym> have Python bindings:</p>
<p>Python Flickr API: <a href="http://beej.us/flickr/flickrapi/" rel="nofollow">http://beej.us/flickr/flickrapi/</a><br />
Python Google API: <a href="http://pygoogle.sourceforge.net/" rel="nofollow">http://pygoogle.sourceforge.net/</a></p>
<p>and so on&#8230;</p>
]]></content:encoded>
	</item>
</channel>
</rss>
