The CMS Experiment at the LHC has released many large datasets of proton-proton collision data
as well as simulation to the public as part of its commitment to data preservation and open access.
The collision data released totals around 2 $\mathrm{fb^{-1}}$ at 7 TeV and nearly 12 $\mathrm{fb^{-1}}$ at 8 TeV.

Data preservation describes the efforts to not only preserve the data itself but the conditions
in which it can be analyzed. This requires archiving and documenting information such as run
environment and conditions and tools such as analysis software and workflows. Open access describes
the efforts to share these tools and information with the public, including educators and researchers.

These efforts present many challenges, partly due to the complexity and size of the CMS datasets
and also to the knowledge required to make meaningful use of them. We describe here how we dealt
with these challenges and the usage of CMS open data not only in public education and engagement
but also in fundamental research. Furthermore we describe plans for future releases of data.