A couple of Alex’s points in particular stuck out for me:
Santiago: There’s always a learning curve. Every platform is different, specially when you want to fine tune your layout and deviate from the provided templates.
Alex: This one strikes me as a bit silly. There is a learning curve when building your own system too – especially if you haven’t written your layout/templating system yet.
Santiago: You’ll never get to experience the satisfaction of engaging in a conversation about how you developed your own platform from scratch.
Alex: … if what you want is engagement then joining a bountiful and vibrant community of developers is a much bigger opportunity than the potential for a conversation with another NIH hacker.
Santiago finished his post with:
It takes a few evenings of work to get it done. It’s that simple.
Honestly I doubt it. Although I’m an experienced web developer, if I were to attempt to roll my own solution from scratch, it would be a huge undertaking, fraught with many potentially fatal problems:
- First I’d have to choose a programming language and platform, with very little in the way of criteria with which to make the right decision—at least not without doing a lot of research first.
- I’d need to decide what features I really need and what I could do without.
- I’d have to write (and debug) the code—probably a lot of code.
- If I wanted to be able to use a native app to post to my blog, I’d have to implement a well-known API, with a dialect that the app understands. (Mo code, mo problems.)
- I wouldn’t be able to take advantage of the vast universe of WordPress plugins: I wanted a feature a plugin implemented, I’d have to write it myself. (Mo code, mo problems)
- And so on…
And after all that, I’d still have to find a way to export the content from my current site, and import it into the new one, which was something was going to have to do anyway. 🙁
Plus, as Alex hints at by pointing out the vibrancy of the WordPress community, I wouldn’t be able to leverage the experience to actually learn WordPress (and some PHP, and some optimization, and some Apache config, and…).
Update: Santiago has a follow-up post:
“I’d never ask someone to do this. Rolling your own engine means a lot of work, and unless you are really on the nerd side (like I am and Brent Simmons is), it will be a waste of your time.”
Update: More dialog on Twitter
Ps. In the end Brent decided to stick with the self-built engine he’s been using for years, and write an iOS app for himself to post to it remotely. Moral of this story: Stick with what you know?