For example, Conway’s Law: Let’s assume you have a web app. You also have enough developers to split them into three teams. You can either split according to the parts (Frontend team, Backend team, Batch team) or you could use some other method, like split by function (Chat team, Video team, Account Management). You could also split them by skills (Dev team, Ops team, Designer team).
You can now pick your poison, where you want the app to be more consistent according to Conway’s Law. If you split by parts, then there will be a well defined API between front- and backend and changing it takes some discussion. If you split by function, the API will be fluid but the different functions will not feel as consistent to the users. If you split by skill, new features will take longer but the servers are more reliable and the logos are very consistent.
The effects will be weak if the three teams are in the same open floor. They will be extreme if each team is on a different continent and timezone.
I think that Conway’s law and most of the others are not necessary meant as a recommendation to do something different but are more an observation. It does not mean you need to change something if you are happy with how things are but you should be aware of the effect and make sure you are still happy in one year. For that these laws or effects can help.
In the other hand, following all these simple rules will not save you either. 😥
how would you follow any of these?
For example, Conway’s Law: Let’s assume you have a web app. You also have enough developers to split them into three teams. You can either split according to the parts (Frontend team, Backend team, Batch team) or you could use some other method, like split by function (Chat team, Video team, Account Management). You could also split them by skills (Dev team, Ops team, Designer team).
You can now pick your poison, where you want the app to be more consistent according to Conway’s Law. If you split by parts, then there will be a well defined API between front- and backend and changing it takes some discussion. If you split by function, the API will be fluid but the different functions will not feel as consistent to the users. If you split by skill, new features will take longer but the servers are more reliable and the logos are very consistent.
The effects will be weak if the three teams are in the same open floor. They will be extreme if each team is on a different continent and timezone.
I think that Conway’s law and most of the others are not necessary meant as a recommendation to do something different but are more an observation. It does not mean you need to change something if you are happy with how things are but you should be aware of the effect and make sure you are still happy in one year. For that these laws or effects can help.