You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Kai brought this up in MATSim advanced. One of the main reasons, why scientific (open source) software is able to compete with commercial software is that it doesn't have the business constraints like having a big support infrastructure (like high documentation coverage -- "code is documentation" -- or long backwards compatibility).
There are some questions regarding this topic (tbc):
Which parts of the code are so important that we maintain backwards compatibility?
e.g. external (input files like config, plans, network) vs. internal API (public/private guice modules)
Isn't backwards compatibility also for us a nice and important (?) feature? Code from older projects might break much earlier.
Are there explicit examples where backwards compatibility make the code worse and our life harder while maintaining?
Kai brought this up in MATSim advanced. One of the main reasons, why scientific (open source) software is able to compete with commercial software is that it doesn't have the business constraints like having a big support infrastructure (like high documentation coverage -- "code is documentation" -- or long backwards compatibility).
There are some questions regarding this topic (tbc):
accessEgressModeToLink
the default AccessEgressType for Routing and deleteNetworkRoutingModule
#3354The text was updated successfully, but these errors were encountered: