Thanks for the article.

Honestly though, I'm not sure how useful this will be in practice. To begin with, as mentioned, this is not going to be a good way to deal with larger objects (larger than 5MB was mentioned at the end of the article). Also, moving a compress/decompress function into the database's process could be a nightmare for performance.

Overall, the idea of compressing data before pushing it into a BLOB is a great idea, but this should almost certainly be handled in application code rather than inside the database.

Still, it does get one thinking about it.