Head is swimming. Almost seemingly on cue, a lot more strategic discussions have been coming up lately. Hiring, product road map, etc. It's a pretty clear reminder how I've been failing at communicating back to the teams about the high-level goals. Everybody's been in "heads-down" mode with clearing out low-level work items for the past couple of weeks, but things are starting to clear up.

A process has been defined for getting acceptable updating messaging onto our corporate website that has been very successful over the last month. In the absence of a marketing department (how small we are!), I've been helping Aaron with the copy on the site. I feel like I've been doing a satisfactory job on the pages that are expected of me ... so I'm feeling better about my writing. Now that I don't need to micromanage the details of deploying those changes, I've been slowly removing myself from that process, which has allowed me to immerse myself back more onto the engineering side of things.

With the Minneopa release looming, I've been engaging in some pretty trivial bug fixes (31, though!), which has been a nice break from everything. I also managed to hack together a early prototype of <form> support inside MindTouch (blog post forthcoming on the dev blog!). As much as I like DekiScript for what it does, it is just misusing the technology if you try to accomplish any type of form control with it. And I do NOT like Javascript-heavy versions (accessibility, people!)

Along with the Minneopa release, we've got a pretty big feature delivery to a rather large customer, that is a bit stressful. It is unfortunate that one of the primary devs scheduled a vacation during July, and that OSCON was scheduled during this week as well. Their team is flying out to SD next week, so that'll be fun to have them in the office and work with them closely on getting this feature polished and ready for them. We've got our work cut out for this.

One of the major areas where I've been unable to really push forward with all this happening is professional services. While we have two other large projects outstanding, I don't feel like I've been able to make forward traction on them. It's really painful to be running these large deals basically through two people: me and Damien. Pete's been incredibly helpful in assisting on any deployment issues, but it's incredibly unfortunate that I need to keep tapping engineering resources to complete these projects. This is something that needs to be resolved in some way - as it's not sustainable.

I managed to get back into wearing more of the product management hat today, and it occured to me as I was doing some sketches and mock-ups of our solutions pipeline that we won't be able to execute on this without beefing up the front-end resources significantly.

Back around when Lyons was wrapping up, it was clear that marketing & professional services was our huge weakness. To that end, I siphoned off engineering resources - and the two resources which were most convenient were the UI devs (talk about the flexibility of web devs!). This essentially meant that any UI development ground to a halt, which was most inconvenient. I'm going to amend this by switching some of them back over to UI tasks, because it's clear to me that if MindTouch expects to deliver the high-quality turnkey solutions, there's no way we can avoid that without significant UI overhauls.

So I was sketching out mockups for one of our solutions today, and I kept going back to a navpane-less design. No left navigation! When I looked at our existing customer base who fit this solution, none of them used it, and I found myself reaching the same conclusion! I also found myself suppressing editorial tools completely from the UI when they weren't necessary, instead of showing them as a disabled state.

This leads me to a discussion of the Deuce design, which is my baby:

This design has been around for about 3 years now, and I still love it. (That says a lot). It accomplished exactly what it was designed to be: a simple wiki design that sits in the background. A ton of focus was given to the content designs to make sure that the design would never overtake the content. I love authoring content in it, and I think content (for the most part) looks fantastic in it.

But it was designed for content, not for dashboard views that are now becoming the primary focus of MindTouch. The more I look at our existing default skins, the less I see it possible to use these skins to present anything besides a wiki tool. The skins are too generic in their approach (they do a great job of the 80% case, but now we want to actually succeed in the 20% case!)

My original concepts were to take Deuce, and plug-in the solution templates into the content, but it's clear that is not working. The left navigation and additional editorial elements are just confusing. I'm still trying out a few more concepts, but I'm not sure if it'll work.

So I'm in a situation where I'm pretty sure that if we are to deliver anything sellable, we will need to come up with new skins that are targeted for each solution. Thematically, I would ideally like to keep the same simple aesthetics of Deuce in any future design for solutions, but one has to really wonder how maintainable that approach is.

Posted by roy on July 23, 2009 at 01:58 AM in MindTouch | Add a comment

Related Entries

Want to comment with Tabulas?. Please login.