I write so little about something which consumes so much of my life: MindTouch.

First things first ... we moved offices. What was once a 0.1 mile walk to work has now turned into a 1 mile work (boo!). The new office is really swank compared to the old office, so I guess it's much better.

We launched a new homepage, which was the work of the new designer who joined our company not only but a few weeks ago - we ditched the neutral browns and went with a more aggressive red/grey branding scheme.

We launched MindTouch Deki, which allows IT administrator's to install and run a wiki on any OS - you can install Deki on its native Debian, or you can install one of VM's many products and get it to work on other operating systems (like Windows). We've pretty much abandoned focus on selling the DekiBox (a real shame, cause that was pretty cool) to focus on selling the Deki software.

We took over the hosting of wikis from the now-defunct Wiki.com to wik.is - we'll be relaunching that site in a few weeks for public registration. Unlike wiki.com, I really want to take the time to make it NOT suck - DekiWiki (the name of the wiki software MT makes and I work on) is NOT geared to the general public, which was a huge stumbling block for wiki.com adoption. It says something when people think Wikia is better than your product...

Some of the stuff happening internally is the most exciting. We've taken a HUGE risk forward and gutted the whole technology powering DekiWiki - we're dropping all PHP backend code for a C# (powered by Mono) backend. I'm not enlightened enough to talk about the benefits of one over the other, but the one cool thing is that this PHP to C# backend transition is forcing us to write a powerful API that will power DekiWiki. We're writing an API for everything - user creation, file management, page management ...

Designing and watching an API being borne is an amazing thing, although it's been incredibly frustrating due to the technical walls you keep running into.

We've consolidated most of the dev team to San Diego (3 of us in SD, 1 still in MN), so I've been able to directly benefit from the daily interaction I have with my coworkers ... being able to quickly debug a problem that spans across many stacks ("Is this problem in PHP or C#?") with your coworkers is really helpful. We also have developers in Russia who have been working the PHP side of LDAP implementation (the bulk of that functionality is still being handled by C#), but coordinating with them has been difficult.

In an effort to make development easier, we have an internal version of our Deki project for deving - in theory it should mean everybody develops on one environment (crucial for a product like us, which has so many dependencies), but it's been rather buggy lately, and I've spent most of my time tracking down problems in the Dev VM and getting them fixed.

Anyways, I sometimes feel with our dev process, we've sort of thrown up all these balls, and now we're scrambling to catch them all in time. We've gutted all backend functionality - an active rewrite of all services is being done. Because of this, we're also getting the flexibility to implement things 'as they should be' and not 'as they are because that's the way MediaWiki did things.'

So far, everything has been manageable (albeit a bit frustrating), and I'm feeling very hopeful about the technical underpinnings of our new product. *If* we get our current release done in time, on a technical level, NO other wiki product could even be close to ours.

What's pretty cool is that even before these huge updates, I had gotten a lot of work done on improving things on the templating/UI side - the currently released open source version doesn't support multiple templating or skinning - I had implemented features to support both of these features. I also worked on making the image gallery not suck by using a ton of Ajax and simplifying the view (managing images is a BREEZE now) ...

I admit, we still have a lot of UI stuff to deal with (how do you push 1 meg worth of javascript down the pipe and still make your product responsive?!), but so far, the hectic pace we've been working at has made me ignore those concerns :)