Some reasons for building Mingle
The trouble is that the tools available - from Excel to the enterprise agile project management tools - focus on tracking and on making agile software development legitimate, rather than supporting agile development. In this focus they forget one very tangible benefit of agile methods and one not-so tangible benefit.
Firstly, they overlook the key practice of agile development: ‘reflection and learning’. Feedback loops are built in to provide the ability to alter paths and change what is being done, so the team can constantly improve and perform better from week to week. Once you start using a tool to help your software delivery process, you have introduced a barrier to change. If the software tool cannot change and adapt as fast as the team needs to, it quickly becomes a burden or, worse, stops the team improving. Mingle attempts what the other tools do not try, or fail at doing. Mingle is a tool for agile development which is itself agile. By reflecting on the hundreds of projects ThoughtWorks has delivered, we are able to distill what works well, as well as provide flexibility in areas we know need to be adapted to the different contexts in which software is developed.
The second reason for building Mingle is personal and harder to measure. There is something powerful about moving physical cards. The ceremony of moving a card from 'in-progress' to 'done' is a token that becomes more rewarding than changing a value in a spreadsheet. Planning and estimating go faster with real cards to visualize and then move. At a glance, a group of cards all sized at '3 story points' by their association is easy to validate. When looking at all the cards in group '3', one may jump out as being bigger - and can quickly be moved to 5.
Wouldn't it be great if some of the benefits of using cards could be kept and extended in a tracking tool, rather than compromised, or worse, lost altogether? This feature of cards can be enabled in agile project management systems - and is a key feature of Mingle.
Check out 10 reasons why we think you should consider using Mingle on your project.
Comments > (HTML is allowed)
-
Mike WApril 2nd, 2007 @ 09:17 PM
"If the software tool cannot change and adapt as fast as the team needs to, it quickly becomes a burden or, worse, stops the team improving" Hi, can you describe how your product will "adapt" to our process? This seems rather magical.
-
Andrei NadinApril 6th, 2007 @ 06:52 AM
Unbelievable, myself and a collegue had long conversations this week about a tool that uses cards as it's basis for manging projects - a more visual project managment system - and here it seems you are indeed doing that. One one hand I am sad that I miss out on making it and on the other glad that in fact there may be a tool that I could use to manage my projects. I very much look forward to seeing the screenshots (PS please be cross platform...)
-
Julian BootApril 10th, 2007 @ 04:59 PM
Mike, not so much magical as simply flexible. We are not going down the path of learning machines or anything - just trying to build a tool based on the premise that its the stuff that is different about a project that's critical - not the stuff that's the same as every other project. That's not to say that the things we repeat from project to project are not important. Just that we are trying to allow for every project to have its own set of nuances. How this plays out in practice is stuff we've already seen on the web. Flexible data models, tags, loose dependencies, flexible linking and light-weight customisation. Nothing magical - just applying standard techniques to be supportive of teams, not govern them.
-
Julian BootApril 10th, 2007 @ 05:13 PM
Hi Andrei, I'm a huge fan of interesting visualisations. Presenting a data set in the right way can allow for fast human analysis that's often more effective than trying to do excessive analysis in the form of computer algorithms. For Mingle we've been cautious not just to replicate blindly what we do with cards on the desktop or on a whiteboard - the metaphor does have its limitations. A database and a browser presents lots of opportunities for interactions that is cannot be done efficiently with paper cards.
-
Andrei NadinApril 11th, 2007 @ 12:43 PM
That's really interesting Julian, as you seem to be on a similar wavelength to us, as you say, once you move the concept into the computer, there are indeed many things that can be done, I imagine it more like a node structure where the nodes can represent many items, Product Backlog Items, Tasks, Actions, Documents, Graphs etc., even saved queries giving different snapshots on the data in the system. I am really looking forward to seeing what you have come up with.
-
Sean WoodhouseMay 17th, 2007 @ 06:55 AM
Guys, Just make sure you're able to easily print out stories as 'cards' that can be used on a physical story board. We currently use JIRA for iteration planning, and have a simple custom report that allows us to print all the stories (4 per- A4 sheet) so they can be cut up and used on a story board. For extra points, allow me to take a picture of the story board (with a mobile phone or something) and have Mingle automatically update the status of the stories in the system. Cheers Sean.
-
Yves HanoulleMay 24th, 2007 @ 07:14 PM
What about giving point to a story? It would be nice to implement some kind of plannign poker inside the application.
Sorry, comments are closed for this article.
