1) It requires implementation. Implementation is often costly and sometimes infeasible (e.g. Vimeo having to build feature which lets you download all XXXgb of

1) It requires implementation. Implementation is often costly and sometimes infeasible (e.g. Vimeo having to build feature which lets you download all XXXgb of your videos at once). 2) It opens the door to litigation. "I told gave <web service> <piece of data> and now I can't get it back easily. Give me money". Don't laugh now, because we all know that laws designed to protect IP tend to be abused (patents, copyright, etc.). I'm sure that Europeans will set the precedent for this one with their tracking opt-out laws. 3) What format should it be delivered in? Can Facebook just give you a 100Mb HTML file with all of the things you've "liked" or does it have to be more accessible? Is a purely machine readable format alright or must it be human readable too? Of course, it's possible to create a law which gives just the right amount of data freedom to users WITHOUT turning every social website into a horrible Dropbox clone, but there's a very fine line between making legislation toothless and making it pedantic.