Hacker News Reader

Comment by ColinWright | Hacker News Reader

There are web crawlers that retrieve the data and store it, then making it available to everyone. Once it's here it's in the world, and removing it only from here is pretty much irrelevant.

As the web currently stands, if you put something on it you have already lost control, so framing the question as one of "owning your own data" is misleading.

I've up-voted the question because I believe it's important and should see a discussion of how future systems might be able to handle this, not because I agree with the implicit sentiment. Personally it would be fascinating to see a real world implementation of the "Oubliette" as given in Hannu Rajaniemi's novel "The Quantum Thief."

View on HN

I disagree with most of your answer, eg most crawlers discard old data in favor of the latest snapshot, but thanks for taking the time to provide your PoV and inviting to a discussion. Even if some piece of data remains around, one should be in control of his/her contributions and be allowed to delete, that’s far from irrelevant or misleading.

Suppose you have a comment at a given URL, say XXXX, an the wayback machine crawls it. Later you delete it so it's gone. It will still be at the wayback machine. So while web crawlers may discard data from an earlier crawl of a specific URL, they are likely to keep the data for a URL that subsequently vanishes.

This isn't to say that one shouldn't be able to delete one's comments, but in an open and distributed system you'd need to track down all the copies, and it seems there is no way to know you've got them all. And if you can find a copy, there's no guarantee that the host will respect a request to delete it.

These are difficult questions, and I have no solutions. Having said that, I don't think that being able to delete comments here is really that important or desirable. I respect that others will disagree with that.

For what it's worth, you can email the wayback machine admins with a request to remove archived versions of a specific domain.

They will also respect DMCA notices for your content when served.

This is sortof true for search engines, but HN is hit by a myriad of bots and scrapers non stop. There is no telling what they will choose to do with the data. I still get hits on my domains for URL's that ceased to exist 20 years ago and that have no internet references to them.

When you say that "one should be in control of his/her contributions"... you realize that you are "publishing" these contributions, right?

I understand where you're coming from in wanting that control, but I would encourage anyone who feels that they should be in control of what they put on the internet to meditate on the word "publish."

You would laugh at Stephen King if he suddenly demanded that everyone return their copies of the Dark Tower, right? At one point he made the decision to "pull the trigger" so to speak, and publish that work. Like any trigger with lasting consequences, once you pull it, they are there.

Generalizations aside though, you can have a comment removed from HN by the mods if the request is found to be reasonable (i.e. unintentionally disclosing medical history, etc).

Publishing something online doesn’t revoke your copyright rights of that content, especially if you weren’t asked to grant a non-exclusively license for posting it.

Sure, but let’s remember that copyright enforcement requires work. Are you going to be the one who goes from car to car, peeling off illegal images of Calvin and Hobbes? Out of the kindness of your heart? How’s that battle going to work out for you?

> There are web crawlers that retrieve the data and store it, then making it available to everyone.

I agree, but also people may quote your comment, repost it in another platform like facebook or tweeter, make a screenshot, print it in a newspaper, make a t-shirt, ... Once the comment is out there, deletion is only an illusion.

In practice it matters, I think, even if in principle you can't undelete something from the internet. If someone is not explicitly looking for it, or not looking very hard, then the fewer places it exists the less likely it is that they'll run into it.

Reply on news.ycombinator.com