The Second Way To Skin A Cat
You know the season of a t.v. show when you realize that they’ve run out of plots? One of the most common signs is when the characters start reversing roles. Suddenly Sylar is a good guy, or Dharma stops being a hippie and Greg stops being a Republican. You know the gig is up when that happens.
This week I’m left wondering if a software company is “done” when it starts releasing products that compete with itself. Google has had me a bit confused with Chrome OS and Android for a while. But this week things hit closer to home for me when the makers of FogBugz, a project management tool, released Trello, a project “organization” tool.
Project Management Pegs
I’m a big FogBugz fan. Heck, I even wrote a blog about it for a while. I’ve promoted FogBugz at work for years and I’ve used it for personal projects many times. I’m in the software dev world, so a tool like FogBugz is essential for almost every project. However, I’ll be the first to admit that FogBugz is not for everybody or every situation.
I’ve tried lots of other project management tools. I’ve used some “agile” tools like Pivotal Tracker and Rally. Recently, I’ve pushed BaseCamp strongly for lots of teams at work that needed something different from FogBugz. And I’ve tried various kanban tools like KanbanPad.
So, when I load up Trello for the first time I’m not trying to squeeze a square peg into a round hole, but I also recognize a square peg when I see it. Trello is clearly a project management tool in the spirit of kanban (kanbanpad, leanitkanban).
Why Trello?
Okay, I’m cool with that. Maybe they’ve made an electronic kanban system that works. But wait, why is Fog Creek producing two apps in the same market? Why is one of them free and the other one is charging $25 a user a month (for the hosted version)? And why isn’t FogBugz getting some of these features?
So, I go back to the source to see what Joel Spolsky has to say about the release of KanBugz, err, I mean Trello.
“After ten years in management I still never knew what anyone was supposed to be working on. Once in a while I would walk around asking everyone what they were doing, and half the time, my reaction was “why the hell are you working on THAT?” So one of the teams started working on finding better ways to keep track of who was working on what. It had to be super simple and friction-free so that everyone would use it, but it had to be powerful, too.”
Joel – JoelOnSoftware – emphasis mine.
So, Trello (man, I like KanBugz, do you mind if I call it that?) lets you see “who was working on what.” Really? I think like most project management software, what it really lets you see is who is SUPPOSED to be working on what. What feature shows you what folks are really working on at the moment?
Okay, it actually does have that feature! It’s the “Activity” feed, or board, or panel, or window, or thingy on the side of the screen. Whatever they call it. And that feature seems to work okay, based upon my initial experiences. The allocated area is a bit narrow, but it works.
But all the other stuff, the cards flipping and dragging, the lists, the boards, etc., all of that stuff doesn’t tell you what anybody is actually doing.
And I have to wonder, doesn’t the FogCreek team live in FogBugz already? What are they using internally now? Trello or FogBugz? Because FogBugz already has the features that you need to run a project and to know what everybody is working on.
Working On
The “Working On” feature in FogBugz literally let’s one identify exactly what they are working on at that moment. If that’s not enough, Working On provides time tracking and provides the raw data that powers Evidence Based Scheduling.
There is a drop-down labeled “working on” in FogBugz that lets the user, very easily, identify what they are currently working on and how long they think it will take. This feature is very smart. It adjusts for your lunch schedule and your normal beginning and ending of the day. So you don’t have to micromanage the app to accurately record the key activities of the day.
FNN – FogBugz News Network
So, maybe your users refuse to enter estimates or indicate what they are working on during the day. That’s okay, if they are doing anything with the project management software, like opening, re-assigning, or resolving cases, then everything gets recorded and can be reported.
FNN is a nifty little FogBugz plugin that, wait for it, tells you what people have been doing. Here is a quick screenshot of the plugin in action:
Kanban Plugin
And just to be clear, you can kanban in FogBugz. (say that 3 times fast). There is a plugin for that too! Kanban boards are useful in software projects and they are available for FogBugz. The Kanban plugin provides a nice little board for your card sorting pleasure.
So, About This Trello
I’ll be the first to admit that FogBugz is not perfect. But Joel’s claim that he didn’t know what his employees were working on doesn’t pass the smell test. He already has one of the best project management tools on the market and it includes specific features for the problems that Trello claims to address.
So, what is good about Trello?
It’s a huge step forward in interface design for Fog Creek. The initial interface of a “board” is simple and well implemented. Drag-n-drop features work well, and like most kanban boards, it provides an understandable high-level view of a project.
Fog Creek doesn’t have the best reputation for building fancy user interfaces. FogBugz is powerful, flexible, and full featured, but it is not pretty. Trello shows that they can embrace design at Fog Creek.
And I’ll admit, we are giving Trello a shot at our office. The R&D team is using Trello to describe high-level projects and tasks. Essentially, we are using it for release planning. But we are still managing all of the details and sprints within FogBugz.
Why? The problem I’ve always had with Kanban boards, both physical and electronic, is that when they get too big they start to lose their effectiveness. When everything fits on one screen, or one whiteboard, and when each card can hold all of the relevant information at that level of the project, then they can convey a lot of information very quickly. But when the card gets too full and the board starts overflowing, then the metaphor starts getting in the way of actual project management.
Take a look at the Trello Development Board for example. It’s an impressive example of eating your own dog food. But at the moment, there is something like 36 cards in the ideas column and I can only see about 5 cards at a time on my laptop screen.
And then you flip a card over and things really get messy. For example, this is part of the back of a card about sorting cards:
Buried somewhere in that list of over 100 votes that scrolls forever, you will find almost a dozen comments. And that’s on a card that can be described in 6 words. I don’t have many cases in FogBugz that are only 6 words long.
And if you were actually a member of a board, you would see that the right hand side of the card has all kinds of other doohickeys that I haven’t taken the time to figure out. Let’s just say that the first time I turned a card over it reminded me of the story cards from Rally and I don’t mean that as a compliment.
Project Management for Real People
The frustrating thing about Trello is that I sort of get why they created it. Back in 2009 I said this about FogBugz:
Perhaps the greatest flaw in FogBugz is built right into the product’s name. Bug. FogBugz was born as a software defect tracking application. It is hard to escape one’s heritage and yet there is a much larger market available that has thus far been out of reach.
http://fogfanz.wordpress.com/2009/03/17/7-and-beyond-glass-half-empty/
Trello seems to be Fog Creek admitting that FogBugz will never be adapted to appeal to a larger market. And that makes me a little bit sad.
Of all of the project management tools I’ve used, to manage software projects, house projects, blogs, customer support, clients, and co-workers, none of them have made me as happy as FogBugz over the years. And yet, I know not to use it for certain projects because I know my client or co-worker will be overwhelmed by the density of information. What works great for programmers doesn’t work great for everybody else.
Clearly, since our own R&D team is trying to use Trello to manage releases at a high-level, even while they live and breathe FogBugz daily, there is a need for this type of view for software teams.
Yet I shudder to think of managing a large-scale software product via Trello. One look at the Development Board for Trello convinced me that is not how I want to work when we get down to the details of a large project.
Conclusion
I’d rather see some of Trello’s approach integrated into FogBugz than to see more of FogBugz’s features integrated into Trello. And I’d like to hear Joel and company speak the truth about the purpose of Trello and where it fits within their product lineup. I just don’t buy the “what is everybody working on” story and I’m not sure I like what it says about their long-term belief in FogBugz.
10 Responses to “The Second Way To Skin A Cat”
Trackbacks / Pingbacks
- - September 18, 2011





Hear, hear! I do agree!
The Kanban plugin would be sufficient given multiple instances and better filters, like the current drop down of case filters
Bradley,
Insightful as always.
FogBugz is great at what it does, and we are absolutely not abandoning it. In fact, a major new release is coming out in the fall with some (I think) massive UI improvements. The team is very excited about it!
But FogBugz just doesn’t get much traction outside of the dev community, which is understandable since it’s built for developers. Sure, people, including us, press it into service for all kinds of other things, but the marriage is always a slight mismatch, even though it often works pretty well. But everything, the names of fields, the name of the product, source control hooks, etc. tend to scare off more general users, as you point out.
We wanted to make an app for managing projects that was almost instantly understandable. The fact that it looks like a kanban board is really a case of evolutionary fitness–it just an implementation of how people organize things in space with index cards or post-it notes. But notice that we deliberately left out the software-centric verbiage. No story points, or sprints. That’s the stuff that scares non-dev people off.
We also wanted to make it extremely lightweight. It was not designed to carry the weight of a large complex software dev team. It can assist with that by giving a high-level feature overview and product map, but it would not be good at holding the dozens or hundreds of bugs in a project of that size. In fact, we built the interface to subtly discourage that kind of “hoarding.”
We also wanted to play with some technologies that let us make the app fast and super-responsive, and provide some of the eye candy, which at first might seem pointless, but which actually helps people understand how to use it.
There might be some tighter FogBugz integration in the future, but none of that is settled yet. Yes, in some small ways it competes with FogBugz, but mainly it doesn’t. So far it is appealing to a far broader audience who want a low-barrier way of jotting down what they need to get done with their team. It takes seconds to sign up, it’s free and when you open the welcome board it takes a couple of minutes at most to get what’s going on. We thought there was a need for something that fit those criteria, and so far that’s turned out to be true.
And lastly, yes, the voting system is broken. People start to use your stuff and you have a lot of “oh, yeah, that doesn’t work so well” moments.
We’re already working on a way to make it so voting doesn’t clutter up the cards so much.
So how do FogBugz and Trello fit on the roadmap? FogBugz makes us money and is used by thousands of software teams. It’s not going anywhere, and is still actively being developed. Trello can be a complement to FogBugz, but it breaks out of the software market. It’s a tool for designers, and writers, people moving house, or planning vacations, feature planning, weekend house projects, PR efforts, sys admin teams, general small business projects, content creation, and classroom projects. Those are just a few of the things I’ve seen people mention on Twitter. Trello is much more general purpose and lightweight than FogBugz could ever be, so in the end, I think there is much less overlap than suggested in your post.
Great post. I totally agree with your conclusion: FogBugz is for the detail-level, Trello is for the big-picture. You might have a single Trello card that says “Fix FogBugz bugs.” That’s how most of the teams work at Fog Creek — they’ve got Trello to track things at a high level and FogBugz to track things at a detailed level.
Also keep in mind that Trello was very intentionally designed for a general purpose audience, not for software teams. Everything about FogBugz is tweaked and optimized for software development projects. I used to preach that FogBugz was the ideal tool for my dad to track his apartment renovation, but in reality, it wasn’t, and Trello is.
So… while there IS overlap between FogBugz and Trello, there’s not as much as it might seem!
P.S. I agree that it’s non-optimal for every single vote to have an entry in the activity feed… we never noticed that problem until thousands of people started using the public Trello roadmap to give us feedback on features, and you can be sure it’ll change.
I see Trello more as a way of organising across multiple projects and the FogBugz Kanban boards as visualising tasks within a single project. In my case I’m using Trello for a mix of work and home projects (a tenuous distinction at the best of times) and it’s really helping me to manage the queue – essential as I work in a small team and often have to run off and do maintenance or emergency bug fix tasks – and provides me with a solid way of capturing those otherwise fleeting “hey, wouldn’t it be cool if we built a….” ideas
Sure it still needs a few bugs working out but it’s brand new so that’s ok. And as a non-paying user I get the chance to drop stuff into the team’s board if I find something that’s really holding me back, which is pretty cool too
Adding a Kanban module to FogBugz is understandable – JIRA (main bug tracking app) introduced Grasshopper. And I bet Joel watches them closely.
Then Kanban just got more popular with developers. But why built something so lightweight, when there are apps around now like smartQ (www.getsmartQ.com) – that supports far greater customization (even custom fields). Plus lighter ones like KanbanPad… And who know how many more…
I like the voting approach to it though… But then it should not be promoted as a project management app, but more like an idea poll management app… Or may be something will come up with a better term.
.
Why all of them are promoted as project management apps anyways? What’s wrong with staying a bug tracking app, and then build a voting up for ideas, etc.
I couldn’t agree with your POV more. In fact when Trello was first revealed I slapped my forehead and said, I should have filed a patent on my invention.
I wrote an extension to FogBugz (we have our own private installation) where I can pull up a page that shows me who is working on what, how much time has elapsed and whether they are in danger of exceeding their estimates.
Maybe I should turn this into a plug-in.
Bill,
sounds like you have a pretty useful plugin for FB there, it is hard to get the big picture on elapsed and estimated time.
As a FogBugz customer for a long, long time, I think your conclusion is perfect. I really don’t see where Trello fits in FogCreek strategy. It competes too much with FogBugz and makes us feel like using an obsolete product.
Hmm, I get that in general that Trello does not compete with FogBugz. It has a larger potential user base, will be used for a different role, and it is free. I guess I’m more concerned as a FogBugz user about how Trello will affect the continuing evolution of FogBugz.
On one hand, it is great that Fog Creek is experimenting with new ways of doing things. Best case scenario, projects like Trello inspire new and better things for FogBugz. On the other hand, the concern is that Trello like features never make it to FogBugz because it is assumed that those features don’t fit within the FogBugz paradigm.