stating the obvious archives | about

Pyra's Killer App
May 01, 2000 :: Michael Sippey

I had lunch last week with Evan Williams and Matt Haughey, two of the folks at Pyra, one of the most interesting web app companies this side of the Mississippi. Pyra is best known for unleashing Blogger on the world, enabling thousands of webloggers to post snippets of content to their sites almost instantly.

While Evan, Matt and the rest of the Pyra team has been busy enhancing and stabilizing Blogger, I've been busy messing around with Pyra proper, their eponymous web-based project management and collaboration tool.

On its surface, Pyra seems to be a simple task management tool. A left-hand frame stores an outlined-based hierarchical view of project information. When an item's selected, information about it is displayed in the right-hand frame. It's incredibly easy to create a project, populate it with features, tasks, issues and URLs, and organize them into folders.

It's Pyra's teaming functionality that makes things interesting. Since Pyra's on the web, it's designed to be collaborative from moment one. Pyra will send out email invitations to new team members, and allow the project owner to designate permissions to each team member (e.g. "Can only propose new items" or "Cannot be assigned"). Once your team's in place, you can assign features, tasks and issues to them, and they can use Pyra's built in "add comment" feature to conduct inline discussions on very granular portions of the project. Pyra enables filtered views of project data ("Show me items assigned to me with a due date one week from now"), highlights new items since your last login, and automatically logs any edits to any item.

As a web-based app, Pyra also enables geographically disperse teams to collaborate on projects more effectively. The industry standard tool, Microsoft Project, has sported out-of-the-box web features for several years, but (obviously) favors an NT-centric workgroup environment, preferably with Exchange installed.

Pyra's just in beta, though, and if you look hard, it's beta-ness shows. It's lacking dependency support, a timeline (or even GANTT-based) view of project tasks, integration with one or more third-party web calendar apps (e.g. "Send your tasks to Yahoo! Calendar"), an import from Microsoft Project, sort by date, etc. It's also screaming for file management functionality, and even integration with source management tools like RCS. The scary thing is that when you mention obvious feature enhancements, or throw your wildest functionality or integration ideas at the Pyra folks ("Hey, how about generating an XML feed of project task data to enable, say, an XML-RPC-based feed for information display outside the app, like on a corporate intranet"), they usually smile, nod and say things like "Yeah, we've thought of that, and we have it running in our shop."

A collaborative, web-based project management tool like Pyra has the potential to change the way project management is practiced. PC-based PM software (like MS Project) is geared towards generating a document -- The Project Plan -- that without vigilant attention to detail (usually of the mind-numbing sort) quickly becomes out of date. This document-centric approach to managing tasks and schedules puts the people that know best about current status one step removed from the plan, and puts the project manager in the middle of what should be a simple transaction ("mark this task complete"). Furthermore, with its discussion functionality, Pyra enables team members to carry on asynchronous conversations about features, tasks and issues inside the project plan environment itself.

Of course, Pyra doesn't obviate the need for a project manager...just the need for the project manager to spend so much time maintaining The Project Plan. With Pyra, project managers could spend their time on true value-add project activities like setting goals, fostering inter-disciplinary communication, prioritizing features, resolving issues and communicating with project stakeholders that aren't part of the core team.

Contrary to popular opinion, Pyra's killer app isn't Blogger, it's Pyra. Of course, that's mostly semantics, since Blogger's an application built on top of the Pyra framework. Which means that Pyra could not only be your next project management app, but your next content publishing platform as well. An integrated content, template, task, issue, and discussion database? Sounds like a killer app to me.

Now they just need to figure out the business model...



Other pieces about server-side software: