Thoughtworks Mingle vs. Pivotal Labs Tracker

by on January 8, 2009 · 12 comments

mingle_vs_pivotaltracker

As I’ve shown in my plea to Agile Tool Vendors: Please simplify my life, I currently see only Pivotal Tracker to be a tool which simplifies my life instead of managing complexity.

Enterprise Class vs Opinionated Software


Maybe the biggest differences between Mingle and Tracker are their core assumptions and target audience. While Mingle is targeted to the big enterprises in the world, Tracker appears to be there just to help agile teams.

Mingle wants to be able to adapt to every imagineable enterprise setup and already comes equipped with a set of project templates e.g. for XP or SCRUM based approaches. It is, at its core, a fully configureable card management database with all the bells and whistles you can imagine like its own query language for doing custom reports and graphs, or the possibility to define so called “card trees” – hierarchies of card types to e.g. manage story -> task relationships. Wow!

In contrast, about the only things you can configure in Tracker are the iteration length and the estimation scale (linear points, fibonacci points, etc). And this is not a shortcoming – quite the contrary. You can set up and understand Tracker in less than 5 minutes. All other things it simply does as the Pivotals think they should be done: Iteration planning – automated, Velocity tracking – automated, Status flow – predefined. Take it or leave it. But all the decisions they made magically fit quite well with what we want to do anyways. Ok, we had one additional status in our Mingle status flow called “ready for merge” for stories where coding is done, but could not yet be delivered. But we can happily live without that – it was a bit awkward anyways ;-)

Where Mingle is more like an assistant, whom you can tell to put a card on the wall or give you this or that report, Tracker is more like an agile coach, making sure all the administrative tasks like iteration tracking are “simply there”.

Project Backlog

In my eyes, one of the biggest shortcomings in Mingle is the missing support for a real backlog. Workarounds like priority meta data with “high, medium, low” values or position numbers simply don’t cut it. What I need is a plain list where I can re-order stories with drag-and-drop. The order of stories in that list is simply the order in which the stories will be done. No fancy priorities needed. You guessed it, Tracker offers exactly that: A backlog. Not a crutch.
tracker-backlog

Story Preview

mingle-story-preview

When working in a list or chart view of stories in Mingle, you can click on a story to get a pop-up div showing you the first few words of the story description plus the meta data attributes like Iteration, Type, Status, etc. In Tracker, a mouse-over is enough to see the complete description, all comments, and all attachments as previews including download links.

tracker-story-preview

This is a small but significant difference: As Tracker shows all the meta data already in the list, it can make use of the preview pop-up to show the real important stuff you need to know about a story (and this happens to be the description, comments and attachments for us). In Mingle, I have to open the whole story to get this. Which brings me to the next difference:

Story Details

mingle-story-details
In Mingle, there is a detail screen for a story. That screen works like expected: It shows you everything about the story: description, comments, attachments, history, metadata, etc.
tracker-story-details
Tracker doesn’t have a detail screen at all. Now you might think: That’s a shortcoming! But, you’d be wrong. Tracker, by default, shows all meta data in the list as icons. And for editing a story you expand the list item in place. This enables you to not only see other stories while editing one, but even edit multiple stories in parallel. This really makes breaking down large stories into smaller ones a snap as you can have the original story open for copying and the target stories open for pasting parts of the original description. Again, a tiny, but significant difference.

Why Tracker Might Not be For You

For me, Tracker is definitely the tool of choice for all my current projects. But your mileage might vary as some of the prerequisites or baked in assumptions Pivotals make might not cut it for you.

Pivotal Tracker

  • is solely available as a hosted tool. No local install at your company.
  • supports Features, Bugs and Chores. If you need more story types then you’re out of luck.
  • has neither project nor card hierarchies. If you need to manage your projects in a hierarchical way, you’re out of luck.
  • No custom reports or charts. If your boss demands a certain type of chart, you’ll have to put it together manually.

If any of these points is a show stopper for you, go for Mingle. Otherwise, Tracker is the tool for managing your agile projects.

Did you enjoy this article? Get new articles for free by email:

Comments

  1. Kevin Stewart says

    I agree completely. We’re moving away from ScrumWorks and Mingle was my primary choice until Pivotal released Tracker. The one potential sticking point is this language in the Privacy Policy:

    Access to your personal information will be limited to specific Pivotal employees. Specifically, Pivotal authorizes a limited set of people to act as “super admins” of the Service. While performing functions of customer support, site maintenance, research, debugging, or testing, these individuals may search and alter any information in the database, including but not limited to user information, project information, story information, and time records.

    Our Program Manager is worried about this more than I am, but one has to wonder why Pivotal would need to “search and alter” your information for anything but customer support without your explicit permission. I’m willing to give them the benefit of the doubt, but I’m sure our Legal department will not.

  2. says

    Kevin,

    thanks for highlighting this issue. From running our own web applications I know that we, as sysadmins, feel responsible for such an online application. But we do need to touch the data from time to time to fix data corruption introduced by bugs in the code. I trust Pivotal enough to assume that they will not simply drop all my stories ;-)
    But I can see your issues with your legal department. I’m curious to see Pivotal’s reply to your request on the Pivotal Get Satisfaction page.

  3. says

    This is not meant as a sales plug.

    We share the same idea of making developer’s life easier and not overwhelm people with two many options. Based on that we’ve developed a tool called Caimito One Team. It features the usual issue tracking things but is quite good at helping distributed teams communicate with each other. I’d love some feedback and would not shy away from a comparision with other solutions. Would love to hear back.

  4. Mark says

    I appreciate what you’re doing on this site, the information has definitely been useful to me, as I have been working through piecing together the bits that will help my operations team to work more efficiently.

    What we’ve ended up with so far works pretty well. However, we’re struggling with the physical board, and I’ve begun looking at digital solutions that might make this more workable.

    Sadly, it seems like most of these are still primarily focused around development, and not much work seems to be going into making them flexible enough to be useful for operations. Some of them do let you design your own workflow, but some of the really nice tools for small teams like mine (Pivotal for example) only let you use their predefined workflow, and there’s no option to switch even to a pre-defined flow that better suits operations.

    Kanbantool lets you do workflow to an extent, but it still simply doesn’t allow for the kind of crazy setup an operations team needs. For example, our team’s physical board actually has three “backlogs”. One for User Stories/Features, one for Pain/Technical Debt, and one for Interrupts. Our oncall person handles all interrupts, and in between works the pain projects. The rest of the team works the features/stories.

    On top of that, we not only have a workflow step for the customer to give feedback, we also have one for review. For changes/features/etc impacting production, it goes to a pending review state which then has to be confirmed as working post-change by another member of the team.

    We also have to have a pair of columns where we can put tags that are blocked waiting on circumstances beyond our control. Ideally, yes, they’d sit on their hands demonstrating the time lost to other teams’ inefficiencies, but that’s not practicable for us. Instead, anything we’re waiting more than 20 minutes for goes into the blocked column, and when the team it is delegated to is finished with their part, it goes into a “ready” column, which is basically another buffer we pull from.

    This probably seems *really* complex to some, but we’re only a few months in, and still working on ways we can simplify while still reflecting the reality of the way we have to work. For us, this has been a great step forward, and it’d be wonderful to have a digital tool that could support our team – which 3 days out of every week has at least one member working remotely. Preferably something inexpensive (read: free) for small teams, as we’re still too early to get any approval for spending $$$ on these ideas.

  5. James says

    Hi Mark,
    Have you tried KanbanTool? It’s perfect for teams like yours: you can split columns any time you need, sort tasks and assign tasks to team member. And it’s still for free :)

  6. says

  7. Stephen Smyth says

    Has anyone used Pivotal Tracker or Mingle beyond product development? For example for projects that touch legal, finance or marketing functions?

Trackbacks