A week ago, we posted a proposal to use Jekyll to build our lessons rather than Pandoc. The immediate reaction was almost uniformly positive, but in the days since, people have pushed back on two fronts:
We have been changing the way we manage lessons far too often: for example, one instructor has taught four times since early 2013, and the procedure has been different every single time.
We should be putting our energy into writing better lessons, not into tweaking our formatting rules and build system.
Both criticisms are valid, so after some discussion with our lesson maintainers and the Steering Committee, here's what we're going to do:
We will push back the release date for Version 5.4 of our lessons from mid-August to the end of November. (The mid-August date was probably always unrealistic...)
We will stick with our existing build system for Version 5.4, i.e., there will be no changes in tooling until the end of November.
We will focus our energy between now and then on revising our lessons. In particular, we will convert our Python lesson to Python 3 and figure out how to act on this discussion about its future.
This decision is another reflection of the fact that Software Carpentry is a community project: the volunteers who create its lessons, teach them, and sit on the Steering Committee are the ones who decide what happens next. The process might have some rough edges, but I think the result is something we can be proud of.