I used NROFF and TROFF in the mid-1980s in university, so I was familiar with markup early on. I then used TeX and LaTeX in the late 1980s, and ever since, for academic papers. I even persuaded Computer Services (with the help of Paul Doyle) to use a LaTeX workflow for user documentation. This did not last long, as the learning curve was too steep for non techies to engage as authors. Anyway, I was a big fan of structured markup to manage content and produce various formats of output.

In the early 1990s I got into the web after seeing a presentation at NSC92 conference (Network Services Conference) November 1992 in Pisa, Italy. That’s a separate thread, but of course HTML was originally much easier than LaTeX, and conceptually similar. Thus I got engaged early on in deploying websites, and loved it. This is when I learned Perl, at the suggestion of Joe Desbonnet in Galway, who thought it would be useful for the UCG website. I was fascinated by CGI, and the power of the web paradigm. I generally avoided higher level authoring tools and did raw editing (or generation from LaTeX).

I moved from Galway to Waterford in the mid-1990s, where I helped setup the TSSG with Willie Donnelly and Eamonn de Leastar, later joined by Barry Downes. I was interested in RSS as it emerged in the late 1990s with Netscape support, and I followed Dave Winer’s blog. I was able to incorporate feed aggregation into an early TSSG project, NITOURA, in 2000 as a way of providing customised information. Eventually I created the space to setup my own blog in early 2002. The initial system I used never really took off, it was called GreyMatter.

As a Perl guy, I was then attracted to MovableType (MT) and migrated the original blog to this. I persuaded the TSSG to adopt this for their own blogs, and some content management. I migrated from own hosted solution to one shared with the TSSG. This exposed me to the dangers of Berkeley DB version incompatibilities, ho hum. I then moved to a full SQL DB backend instead.

As I became more interested in W3C standards, not that I was ever uninterested, I wanted to ensure my stuff complied. I found that with MT it was always generating invalid pages and invalid feeds. This was not the platform’s fault per se, but due to my habit of pasting in clipped URIs and text that broke the rules. However, I could not find a smooth workflow that gave me confidence that I wasn’t breaking compliance, and I let it slip. Browsers are very forgiving, even if validators are not, though it always irked me.

Now I know that the original markdown was implemented by John Gruber and others in Perl, and available for MovableType. I was interested in the concept, perhaps more for its use in automated program documentation workflow than for blogs. I didn’t implement it on my own blog, but did use it elsewhere.

So now, I’m experimented with Octopress and Jekyll to manage a statically generated blog. I have since embraced git and github, and I can see that for some types of blog this low-tech approach may suit well. It is much less tooled than the MovableType web browser embedded editing, but is very powerful. It feels right for me just now.

Ironically, in FeedHenry I’ve been re-engaging with my scripting roots, now JavaScript rather than Perl. The server-side JavaScript in Rhino was conceptually similar to the Perl CGI paradigm. The server-side JavaScript with Node.js is streets ahead, and very powerful. I plan to use this blog to promote code snippets exploring the power of hybrid apps client-side and Node.js server-side. The Github gist integration is the feature I plan to exploit to make this easier to format, and easier for readers to access the code examples.