Tom Morris

If you don't know who Dave Winer is, let me tell you. He was blogging before you were.

He still is. Today is the tenth anniversary of Scripting News. "Praise Murphy!" as Dave would put it.

Here are a few things from the Scripting News and DaveNet archives:

Programmers (7 May 1997): "When a programmer catches fire it's because he or she groks the system, its underlying truth has been revealed. I've seen this happen many times, a programmer languishes for months, chipping at the edges of a problem. Then all of a sudden, a breakthrough happens, the pieces start fitting together. A few months later the software works, and you go forward... I think, in some ways programmers, who live the scientific method, are better prepared for life than non-programmers, but the opposite is often assumed. We all have a visual image of the programmer, but this is just the outer package. A great programmer is a seeker of truth and beauty. Successful programmers know how to ask questions, and they know how to ask the *right* question."

Perils of publishing (13 Feb 2004): "If you scan an email quickly and there's a bunch of nasty words with someone's name in the middle, the mind has a tendency to connect the words and the person. Unfortunately this is how people read on the Web. Charles noted that I'm a friendly person, in person. I like to think that's true. Sometimes it's hard to see that in the writing. I know."

Dave has some probably true things to say about certain standards types too (Molly will hate this) - 17 Mar 2005: "The WaSP guys aren't interested, because as designers, their livelihood depends on it being hard for users to create content. Kind of like a book publisher working to keep a format complicated so there will be demand for their book." Heh!

Another nice standards post from 28 Apr 2001: "I'd like to see something like the Hero Machine for new web technologies. Create a customized super-hero out of any of the following: XML, HTTP, UDDI, WSDL, RDF, SOAP, Jabber, Jxta, XML-RPC, RSS. What logo would your super-hero wear? What kind of a shield?" Mine would be a giant XML and RDF robot fighting back against badly-formed markup. And he'd be home in time to put his feet up and enjoy a comedy film marathon in the evening. Speaking of which...

Seriously though. Dave. Congrats. Thanks for the continuing inspiration and having the courage to speak your mind. Don't let the fact that the fashionable people keep trying to write you out of history get you down.

If I hadn't started reading Scripting News, I could have ended up as a lawyer. Or worse.

Comments | TrackBack

The blog comments service Haloscan is great if you are using a hosted service like Blogger or OPML.org as it allows you to add comments to your blog using a fairly simple bit of JavaScript. But if you are transitioning and moving your blog over to a web host you control, it's quite neat to be able to deliver your content without relying on document.write().

To that end, I've written a scraper which reads in the JavaScript for the Haloscan service and outputs either XML or JSON containing all the data for the data set delivered in the JavaScript array.

Why is this useful? Well, if you are doing server-side scripting like I am, it means you can add comment and trackback links statically - meaning that the links apear for people with JavaScript turned off. And the JSON part makes it easy to call your comments data in JavaScript and Ajax applications for extra client side scripting. I'm thinking what could be quite neat would be to semantically mark-up your comments page, pull the JSON in of your Haloscan data and then dynamically insert your comments in to the page. That's a bit too fancy pants for me though.

You can access a beta version of the Haloscan script using the following URL:

tools.opiumfield.com/haloscan/[format]/[username]

format should be either 'xml' or 'json'. If not provided, it will default to XML.

username should be the username used for your Haloscan account.

Take a look at mine - XML and JSON. I'd be quite grateful if someone who knows more about JSON than I do could take a look at the JSON one and tell me if it's useful. I'm just taking the PHP array I've created and running it through json_encode.

Once I finish the last part off - dynamically replacing the comments links using this data (probably using PHP5 and SimpleXML), I can start serving up valid, MIME-type compliant XHTML from my new blog.

Comments | TrackBack