I evaluate below a bunch of backlog managers. I picked them based on what looked interesting. Not an ideal methodology but there are so many of these I had to narrow it down. The one that did everything I wanted was YouTRACK by IntelliJ, even the pricing was outstanding. But I rejected that option (for now anyway) because their UX is just too confusing for me. I actually had settled on Flying Donut and started to use them but I quickly realized that they were too simplistic. They didn’t do a good job of allowing me to manage iterations, epics and releases separately. So Tim Park had mentioned he had used Pivotal Tracker at his previous company and I tried them out. They aren’t perfect and their beta has some bugs but they had a really great balance between simplicity and flexibility. So hop on over to our new tracker and see how we are using them!
1 Requirements
We really need a backlog manager for Thali. Right now our stories, features and issues are spread out across two issuer trackers and some email. This is obviously a mess. To make this sane we need a backlog manager that can:
Handle epics, stories and features Or, put another way, I need some kind of tree like structure for stories so I can start with big stories and break them down.
Hosted online I really have enough to do without having to run another server. And yes, I look forward to the day when we can run the backlog manager on Thali but I got to get Thali going first.
World readable We are an open source project, so we need everyone to see what we are doing.
Exit story I need some kind of story for how we take our data if we decide we want to move to another solution.
GitHub Integration This is really just a nice to have but it would be cool if we could close issues in our backlog manager using commits from GitHub. And if they can coordinate issues in GitHub with stories in the backlog manager, even better!
Can we afford it? I know everyone thinks a project backed by Microsoft would have a ton of money and everyone would be really wrong. If something is going to cost us a ton of money then we can’t afford it.
Our basic requirements aren’t terribly fancy. Yes, we would like various kind of charts, burn downs, etc. But honestly the team is tiny. We don’t need anything that fancy.
2 Contenders
Here are the folks I’ve looked it. All the options I evaluated are hosted online so I skipped that row in the table below.
| Agilefant | easyBacklog | tinypm | flying-donut | Yodiz | YouTRACK | Pivotal Tracker | |
| Nested stories | They explicitly support epics, stories & features. | They support themes which can then have stories. But this is not nearly as flexible as it might sound. Themes are really little more than a field that just marks a story. You don’t have the ability to do any kind of roll up or hierarchy management. | More or less. You can create projects which contain users stories which contain tasks. So it’s workable. | You can create buckets which can then contain cards (stories basically) which can then contain specific tasks. | Yes | Yes, in fact they seem to support many more levels of nesting. | Yes |
| World Readable | Not as far as I can tell. You get a certain number of users and that is it. | Not that I can figure. Only provisioned users can see things. | Not really. They do support unlimited viewers but each and everyone one of those viewers has to be manually configured. There is no way to say ’let everyone see’. | You can create a public project that seems to do what we want. | Not that I can tell :( | Yes | Yes |
| Exit Story | They have an open source version and I believe there is a way to export your data. | You can manually go through and download all of your data in their local XML format. It’s not much, but it’s something. | They have an API so in theory one could write a custom program to try and retrieve all the data. That’s more work than I’m every likely to do and I don’t know if the API can actually see everything (haven’t checked). | None that I found | They mention export but that is about all I saw | They do have an API with an explicit export call that returns a bunch of XML. But I don’t believe it really captures all of the available state. | They have an API that might allow exporting the data but I wouldn’t be on it. They also support exporting to CSV. |
| GitHub Integration | None that I could find. | Note that I could find. | Yes. You can put in comments in GitHub checkins that will then match to items in tinypm. | Not that I found | Yes, they support using comments in commits to link them to tasks and issues in Yodiz. | Yes, like everyone else, they hook comments in commits to issues in YouTRACK. | Yes, comments in commits can affect stories in Pivotal Tracker. |
| Cost | $5/user/month with 5 free users, we can make that work. | They haven’t announced pricing yet. | At $9/month for 5 users we can probably figure it out. | Free for public projects | At $10/user/month they are pricey given the competition but their feature set is pretty amazing. | Their free tier would meet all of our needs but their 15 user tier is $20/month. | They are completely free for public projects so we are good. |
2.1 Notes on my use of flying-donut
I started trying to actually use flying-donut. On the plus side it’s very simple. On the negative side, it’s way too simple. The lack of GitHub integration is annoying but not a show stopper. But I found the way I had to do things in Flying Donut to be annoying. For example, you can’t just have a card be moved as a whole from to do/in progress/done. Instead it has to have tasks and those tasks are what has to be moved. There is a default card you can put random tasks into but they generally get collapsed in most views so you can’t see them anymore! Again, the data is there and if you click around you can find it. But the process felt very unnatural to me.
There also doesn’t seem to be any support for release tracking. The best I could do was create a bucket for the release and then drag cards there. But that was unbelievably annoying because I was also using buckets to collect major feature areas together. So now I had a bucket (my release) with cards from various feature areas so it just made things a mess. There is an unimplemented feature called ’labels’ that probably could have helped here but only a little because I doubt labels would let me see things like progress to completing a release.
When I switched to Pivotal they had solutions for all of these problems that worked really well.
I talked with flying-donut support (on 10/16/2014) and they said that they will be introducing the ability to track milestones and they will be adding GitHub support and a way to export data soon.