We use a Reader distribution model exclusively (at this point). While it is quite easy to 'crack' a twbx file (after all it's only a zip file with a different extension) I am not aware the cracking at tde is easy. Please advise me on how to do this, as there are times this skill would be quite helpful.
I'll take your word on the whole 'underlying data' issue as I've not experimented with it in Reader much.
There is one considerable advantage to using Reader versus Server, and that is how Reader uses the Page shelf. In Reader you can play/loop through pages (creating animations), while this is not true of Server (which requires you to click through the animation one page at a time).
Unfortunately, I don't believe there's a way to get to the .tde data in a clean format that would be easy to manipulate in another program. That said, you can still open it up as a messy .csv file and see the sensitive data jumbled up a bit. I find this to be super frustrating because the data is formatted legibly enough for me to consider it insecure (especially since I'm currently working with very sensitive patient medical records), but not legibly enough to use it in a more meaningful way. I'd certainly appreciate hearing other users' views on this.
Thanks for the pages tip. That is good to know.
I too am looking for best practices or checklists when creating extracts. Perhaps a checklist of items to be sure to consider, when it makes sense to use an extract, when it makes sense to use one data extract as data source for several workbooks, scheduling vs republishing etc.