5 ms·
The other day 8 people did a stand-up for 27minutes. That's a crazy loss of time. It's not necessarily a loss of time though. The mindset "any time spent doin
by onion2k 10y ago
The other day 8 people did a stand-up for 27minutes. That's a crazy loss of time.
It's not necessarily a loss of time though.
The mindset "any time spent doing things other than writing code is a waste" rarely produces great products - you need to talk about things. No one knows everything about a project once it gets to a even moderately large size. Working in isolation from the rest of your team means you're not influencing, or being influenced by, the knowledge and discoveries that other people are making as they work. That means you're very likely to be working with an out-of-date model of the problem you're trying to solve.
Clearly a meeting where nothing happens is a waste of time, but the solution to that problem shouldn't be "no more meetings", but "better meetings with real outcomes".
- adrianN 10y agoUnless those eight people work on very closely related problems, there is little value in having all of them in the same room talking for half an hour. Communication doesn't have to happen with the whole team if different parts of the team work on different parts of the software.
- sidlls 10y agoIf their work is that independent then the fact they're all on the same team seems problematic.
- Scriptor 10y agoIt's entirely possible to be on the same team as someone and not need to know the details of what they are working on.
- sidlls 10y agoWhat do you mean by "details"? The level of detail is highly context dependent. I'd argue that if there isn't a level of detail that all members of the team agree they should be interested in, then either the team is not composed properly or there is a misconception about the level of detail that is appropriate.
- dasmoth 10y agoIt's problematic because current business organisation fads seem to mean that anybody doing hands on work has to be "on a team". Recognising the existence of one-person teams might help a bit.
- pjbster 10y agoIt's more than a loss of time - there's also the context switch and attendant energy expenditure to restore the "head space" after escaping the meeting. If the discussion can add more value than those costs, then go for it.
- criddell 10y agoWe did stand-ups at my previous employer and for some reason they happened early in the morning. That's my most productive time. I always thought we should move them to after lunch or maybe later in the day. In that office, productivity and energy definitely dropped as the day wore on and it seemed foolish to spend some of the best time in discussions.
- mixmastamyk 10y agoI had the opposite problem, but it was a morning problem also. First thing in the morning, I am very foggy, and remember very little from yesterday. It is not until I've worked for an hour or two that everything comes back. Paged in from disk, so to speak. So I often had to make things up on the spot to sound busy when I had no idea what happened yesterday at the crack of dawn. What a waste of energy.
- onion2k 10y agoThat can be mitigated somewhat by having the stand up meeting first thing before people have started other tasks, but really that rarely works if people are interested in their work enough to get to the office early. You just have to take the hit, and readily accept that people will skip the meeting if they're deep in to something else.
- curun1r 10y agoThe other time to hold it is just before lunchtime. This has the dual benefit of encouraging people to be brief (hunger is a powerful motivator ;-) and breaking at a time when people would have to break anyways. I also like holding it midday rather than the very start or end of the day because those times encourage staying late or coming in early to resolve an issue before standup. Lunchtime is also usually a time when everyone (apart from remote team members) is in the office no matter what schedule they keep.
- fsloth 10y ago" That means you're very likely to be working with an out-of-date model of the problem you're trying to solve." This is highly context sensitive. The models I deal with when I develop software change very seldom, if ever. Different projects have different cadences for inputs and outputs, even within this wide domain we call "software engineering".