I wasted able to get the CMS preview to work. Using the debugger, cmsPreview() nor PreviewLink were getting called. I followed the instructions inthe document I'll paste my code below.

Seems like these docs are duplication of what's already on framework - https://docs.silverstripe.org/en/4/developer_guides/customising_the_admin_interface/preview/ / https://github.com/silverstripe/silverstripe-framework/blob/4/docs/en/02_Developer_Guides/15_Customising_the_Admin_Interface/04_Preview.md

We should aim to not have duplicated docs if possible - would it be possible to just remove the duplicated bits and link to the framework docs?

Seems a lot better now that the Previewable examples have been removed.

I've followed the example code locally, the BlogPost's aren't publishable, which is unexpected given the content blocks are inline publishable

I haven't actually tried running the example locally yet, but I figured I may as well submit this much of the review while I do that.

There are some questions here that you may have already gone through with Steve - if so, sorry about the duplication. Just want to make sure I fully understand what is happening here.

Nice work!
There are a few minor tweaks to make on the documentation, and some code quality tweaks to make, but I think it's just about ready.
I haven't done so in my review, but once you've made these changes can you please make sure to run phpcs? (You probably already do this but just in case)

At some stage we'll want behat coverage to ensure elemental blocks continue to work in arbitrary DataObjects in the future... I'll let you decide whether you want to do that with this PR or create an issue for it to be added later, given the time you've already spent on this one.

After thinking about this I'm taking this decision for you :p just creat an issue, it should be a separate piece of work. Adding behat tests for this will be a big piece of work.

Tests are failing. Ignore the code coverage one for now - there's not really anything you can do about that. I'm hoping it will resolve itself by the time you push a fix for the other failing tests.

Sorry, should have noticed this in the last review.
Looks good now, I'll check that the behat tests run with the fixtures in frameworktest when I have a moment. If that passes fine, I'll approve the fixtures and then get travis to run the tests again here before merging.

Tests are failing that aren't failing on the 4.8 branch

Tested locally, works as expected.
The only remaining failures are also observed on the 4.8 branch, and are unrelated to this PR.