zip

Kevin> So we should store the images using a sub-atomic scan of the Kevin> disks, so that it can be reproduced back to an identical copy. Kevin> Did people 500 years ago archive information using easily Kevin> recognizable pictographs because people 500 years from then Kevin> (today) will have lost all capability to read? Kevin> Do you believe future people will be that much stupider than Kevin> people are today, and that all information about computers Kevin> will have been totally lost? Kevin> If they lose and then rediscover computers, then they may have Kevin> to go through the same process that we have had to go through Kevin> to read Egyption writings. Kevin> Or, they may think it is just too primitive to care about. Probably not, if they are at all like us. The question is: do you want to assume a continuous record and knowledge, or do you not want to assume that? Examples (from language studies) of the former: Greek, Hebrew -- and of the latter: Sumerian, Etruscan, Easter Island writing. Egyptian was read given knowledge of Greek, it was not deciphered on its own. Sometimes you can decipher things from internal data, if you are lucky to have enough material and enough hints pointing to other stuff. The decipherment of Linear B was an example (when it was done it turned out to be Greek, but that wasn't the starting assumption). Then again, sometimes you're not that lucky, which is why the Easter Island writing remains undeciphered to this day -- and may well remain so forever. You can rely on the skill and the luck of those who come after us -- which is what happened in the past (because it wasn't considered). Or we can do explicit planning to improve the odds. Again, I like the work of the Long Now foundation because it seems like a fascinating example of what you get when you really dig hard into these questions and assume as little as possible. paul