The Long Road to EE2

Coming off last week’s Engine Summit we’re already hearing chatter about the next big release of ExpressionEngine. For me, and this site, however, it’s been a long road to ExpressionEngine 2.

I originally built EE Insider in late 2008 and up until a month ago it was still running a 1.x flavor of ExpressionEngine. It was the last remaining site that I regularly interacted with that was running the old friend we all knew so well.

So, why did it take me so long to upgrade? I mean, goodness, it’s been almost 3 years since EE2 was released.

Here’s why I stayed on EE1:

1. It was stable and worked perfectly for me at all times.
2. I had add-ons in use that were not updated yet (back when I originally looked at upgrading)
3. I had some functionality of the site that would require major re-working through an update. I didn’t see the investment as being worth it at the time.
4. Did I mention it was stable and worked?

ExpressionEngine 1.6 and 1.7 were (and are) excellent releases. I depended on them over and over again for all sorts of sites. EE2 has certainly matured into a great CMS (I use it to power my e-commerce store) but there was the risk of disturbing something that works fine and gets the job done. I could still post content from MarsEdit, my guest authors could still log in and post their content (although they did comment how weird it was to see EE1). We could do the business of the site without any problems.

It didn’t take long for all of my add-ons to be updated (the most impressive part of the EE1 to EE2 move was how responsive and quick the add-on developers were) so that reason didn’t last very long.

Up until EE Insider moved to EE2 I had the EE Insider Tips section of the site where people could submit their own EE tips. This section, while starting strong, never took off and the interest wasn’t there to sustain it as a great catalog of EE tips and tricks. As time went on I turned off the ability to add new tips. Now you can’t even access the tips unless you come from a search engine (99% of the traffic) or from another link on the site.

For EE Insider Tips I was using a handful of add-ons to make the posting functionality work properly. I would have to decide it was worth the time to rework that section of the site that was performing so poorly. In the end I decided it wasn’t worth even having it running and I turned off the ability to submit new tips. Another reason not to upgrade wiped from the board.

After those reasons were no longer, well, reasons, I decided that I did need to update but was working on other projects that needed my attention more than upgrading software on a site that wasn’t broken. I also saw the pain and struggle that Ryan Masuga and the Devto:ee team went through when they upgraded their site. I didn’t want that pain. I don’t like pain.

But one big frustration I had was that I couldn’t try the latest and greatest add-ons from our prolific developer community. Sure, I could use them on other sites but there were add-ons I wanted to use right here on EE Insider.

A month ago Chris Imrie and Eric Lamb announced the Entry Analytics add-on. Eric writes for EE Insider and told me he wished he could see the analytics for his articles on the site. I broke it to him that EE Insider was still running ExpressionEngine 1. He encouraged me to upgrade. I told him I don’t have time.

But then I thought, hey, why the hell not? Why not branch the repository and test the update again? Sure, I’d have to kill some stuff on the site to make it happen but it would be a good time to prune the dead wood.

I went through and made a list of everything I needed to have and then completely disabled all add-ons. After the EE2 upgrade I looked at what was broken and then fixed only the stuff I needed. It was a liberating and refreshing exercise.

There was only one issue with the upgrade, which involved some entry content that contained single quotes being truncated during the migration. This could’ve been disastrous (and, frankly, it’s a little concerning that this could even happen) but fortunately I had a good backup and the ability to whip together a quick bit of SQL to migrate over just the truncated content from my backup.

The upgrade took less than two hours, including all prep, planning and backups. The content migration to fix broken content after the upgrade only took about an hour to determine the problem, test a fix and do the final migration. For a project I didn’t want to do it only took me about 3 hours to actually get it done. That’ll teach me.

Do I regret not upgrading earlier? I only have a few regrets over the last 38 years and none of them have to do with software. The time was right when I did it and it worked.

What’s next? With the big update out of the way, it’s time to shake the dust off the design and code and make them more modern.