Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> I concentrated on asking lawyers -- legislators, prosecutors, judges, and litigators -- to quit hurting harmless people.

Every person generally thinks they're doing good in the world. They certainly don't think they're hurting harmless people for fun, they honestly think they're protecting other people. You need to understand this if you want to have any hope of getting to root causes.

> If an owner of a Web site has some data they don't want spread around, then they should not put it on their Web site. Just don't do it.

See, this is the exact opposite of the truth - it's quite possible to publish data online and employ access methods to control who can access it. By arguing your case this way, you are undermining yourself - the reason we have a legal system with laws etc because things like murder/robbery/etc are impossible to prevent, so the only thing we can do is punish transgressors. Computers give us the ability to implement absolute restrictions formally. The push to implement ambient-authority retributive laws comes from management types who don't devote want to spend the effort to implement formalized restrictions, and want to cry foul after the fact.



> Every person generally thinks they're doing good in the world.

I think we mostly know who's harmless. We know for kitty cats and puppy dogs: In my neighborhood, cats are free to wander but dogs are not.

> See, this is the exact opposite of the truth - it's quite possible to publish data online and employ access methods to control who can access it.

Wrong, fundamentally technically and practically. I told you it can't be done, and you just didn't believe me.

Once that toothpaste is out of the tube, it can't be put back in.

Once a secret is out, can't pull it back.

Once some data is sent over the Internet, the person who receives it has it, all of it, and can technically can do essentially anything with it. E.g., when a Web site sends a Web page, it's gone, the whole thing, HTTP header lines, HTML mark-up 'elements', CSS 'properties' and values, software in JavaScript, text, files in JPG, PNG, GIF, MP3, WAV, etc. The computer receiving this data is free to store it on hard disk, manipulate it many ways, back it up, send it, etc. Net, that data's gone, out'a there, out in the public, beyond control, with no self destruct mechanism, no time out clock, no string to pull it back. Can't track it; mostly can't trace it; in practical terms can't say where it came from, can't claim ownership of it, in no practical terms can enforce copyright for it. Etc.

Sure, can embed a secret 'watermark' in a PNG file, but that doesn't do much good, e.g., as the file passes from person to person. Besides, such a watermark might get lost if the image is resized.

Yes, can put some carefully constructed errors in text and numerical data, but maybe only parts of the data get used or copied, and, again, after the data passed through many hands tough to say who originally 'stole' it, and for the rest they had no knowledge that the data was stolen.

Yes, can set up strong authentication for users and use strong encryption when sending the data, but eventually some user gets the data as 'plain text', that is, not encrypted, and now can redistribute it to friends, family, etc. It's just bits and can be stored, copied, transmitted, modified, etc. So, one 'authorized' user leaks the data and it's gone and essentially out to the public.

For a Web site trying to block a user, it is essentially impossible to know if the user returned -- with a different IP address, MAC address, ISP, Web browser string HTTP_USER_AGENT, etc. Searching the user's house or office also is unpromising since the data could be on DVD below the insulation on the floor of the attic, stored in the cloud, etc.

Again, yet again, as a practical matter, now and over the horizon, if a Web site doesn't want their data out in the public and usable by every Tom, Dick, and Harry for whatever, then they should just never have their Web site send that data. And, really, there's nothing laws or lawyers can do about this except cause a lot of trouble.

Again, more generally, lawyers should just f'get about the Internet.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: