I am glad to see the growth of momentum on the development side of lwIP;
people stepping up to implement new features, wanting to clean the APIs
and people concerned over performance regression, etc. Several
suggestions on the list would call for breakage of the lwIP for some
users, because of API change. My personal preference would be to ditch
the current API's and go for a sole BSD sockets compatible API, stripped
for features as appropriate.
As I am no longer using lwIP in current projects (and probably will not
spend time on it for the next year), it makes no sense for me to be a
project co-leader anymore. I simply am no longer able to oversee the
As Adam Dunkels (the original author) and Kieran Mansley (co-leader) are
in similar positions (no resp. limited time to spend on lwIP) I hope
other people (which are actively involved with lwIP) can step up for
this position, which mainly involves proper progress of the future
development of the lwIP stack and proper release engineering.