On the leandevelopment list, there has been a bit of a discussion on how long the “backlog” should be, and why. This was triggered in part by a posting from Mary Poppendieck, in which she said:

It doesn't matter who owns the backlog, the responsiveness of the development team to real needs of customers will be in direct proportion to the length of the backlog. This is a supply chain issue, and in general, the shorter the supply chain, the faster its response.

I’m not clear whether the manufacturing analogy applies well to a software backlog, which is, after all, just a list of things one might do, not a bunch of partly-built products tying up capital and storage space. I have suggested previously that the usual list of priorities that software people have should be eliminated, as in my article Petition the King.

My view there was that it is better to have a hard line that says “These things will be done and no others,” than a huge soft list that makes people think they are, well, “on the list”. It seems very clear to me that there should be such a line, so that managers who want things can be made vividly aware that they aren’t going to get them any time soon. That’s the sort of thing managers should manage.

But if someone wants to keep a giant list elsewhere of every idea they’ve ever thought, and their reasons for not doing anything about those thoughts, is that inherently bad? Mary’s note suggests that and I can make up arguments as to why it is, if not bad, at least useless:

  • Few people make any use of such lists anyway;
  • If it's a good idea you'll think of it again;
  • Sorting the list takes time that could better be spent;

I could go on forever. I have vast quantities of writings, notes, cards, stored up and only rarely do I manage to glean anything out of them. The most common result is to find a few cards that I thought I’d like to write about. I’m reminded what cool ideas they are, and I move them to the front of my stack. Then I don’t write about them because I have better things to do. They slide back in the list at a rate of one day per day.

Some people expect to find value in recording why they did not choose to do some good idea. Bless me (I was going to use another verb there), if I started listing why I don’t do things it would be a really long list. Why didn’t I ask her out? Oh, yeah, right, I’m married. I forgot that.

Some people claim to have discovered really good ideas in the list and to have brought them forward. Well, as I said above, I do that too, and then they still don’t happen. I suspect most everyone has a really great story of a fine treasure they discovered in the attic. I also suspect that the odds are that most everything up there really is trash, and that of the five wonderful items you brought down to clean up and use, four of them are now waiting to be taken back up.

Some people even claim that their organization has such a list and that they make really good use of it. Well, fine. Have a ball. I think that’s rare, and I think it should be out of the face of the people who are actually doing something.

I still don’t know the whole answer, as if I ever would. I do sense some bad things about a long work backlog, and I have a feeling that a list of Might Have Been or even Do Someday is mostly a waste of time. But is it really harmful to progress in some way, as Mary suggests? If so, why?