4 ms·
I think you're close, but missing something: The companies who wouldn't do this are run by developers or other people who don't interact much with the actual us
by chc 15y ago
I think you're close, but missing something: The companies who wouldn't do this are run by developers or other people who don't interact much with the actual users. Irreversible actions are horrible UX, and any irreversible action in a commonly used area of the app will make a large number of people steaming mad on a daily basis — guaranteed.
People who will blithely click "Delete" — and then click "Yes" even though they don't mean it on the confirmation dialog that comes up — vastly outnumber people who care that an invisible copy of their content might be buried in some Facebook database somewhere. Next to the unintentional deleters, the second group looks like a rounding error. And that's not even counting people who get their accounts hijacked.
So I would say people who know users are more likely to go with the delete flag, since that leaves you an avenue to help the user who emails support with "I got really drunk last night at the wake and thought it might be funny to delete everything on my Facebook and now all my photos of my dead Nanna are gone."
- pestaa 15y agoI highly doubt Facebook went with the delete flag because of the customer service perspective.
- sirclueless 15y agoI think that it is probably one of the major reasons, though certainly not the only one. In my experience, one of the most common customer service complaints/requests is "Oh no, I did this irreversible thing you warned me not to, fix it." I imagine facebook also gets its fair share of complaints like "Oh no, my ex-boyfriend deleted my account, fix it." This stuff is really common, and I can understand why a company would rather be able to say "Oh yeah, sure" than "Sorry, for privacy reasons your online life for the past two years has been lost irretrievably."
- pyre 15y agoI'd venture that a major reason has less to do with customer UI and law enforcement and more to do with data mining.
- CWuestefeld 15y agoI'm sure you're right, and not necessarily in any insidious manner. There's any number of entities that we won't actually purge when deleted from our system here. For example, when a sales rep wants to get rid of a Quote, we can't necessarily dump it, because we'll need a future record if there has already been a customer order placed against that quote. Sometimes the integrity of the system, and the ability to look at history in a (unfortunately necessary) CYA view, demands that some data be "deactivated" rather than deleted.
- code_duck 15y agoDo they even offer customer service of that sort?
- jfarmer 15y agoI don't know what it's like now, but back in 2007-2008, customer support was the biggest department in Facebook by headcount.
- FuzzyDunlop 15y agoI agree. I've already just replied to another reply on this but dealt more with the ethics of the matter. What I would actually implement is something akin to the recycle bin in an OS. Flag something as deleted, ensuring it's no longer published in whatever form on the website. Optionally, delete it properly after a set period of time, or otherwise allow the users to manually perform that action. In addition to that, log the delete actions along with the IDs of the deleted items. So if after all that the user regrets it and files a complaint, you can trawl through your backups to restore it. Irreversible actions in the UI are bad. Having no choice but to tell the customer tough luck is bad. Deceiving the user is also bad. Of course I'll concede my ideals are more compatible with the concept of deleting an entire user account, for example, as opposed to removing individual items associated with. But I don't think everyone else's intents are as pure as yours.