I need suggestions on how I can publicly archive on the Internet a 30k
encrypted file that twenty years from now I will publicly release the
key to decrypt. It is important that the data be permanently archived
in a manner that it will be obvious that the data has never been
tempered with during that twenty years. I cannot at this time divulge
the nature of the data. In essense, the data must be "time-capsuled" so
that twenty years from now I must be able to prove the data has never
been manipulated.

At first I thought it would be an easy enough thing to just attach the
30k file to a Usenet posting and post it to an alt.binaries type of
forum. Google groups and other Usenet archiving robots would then scan
the file and archive it, effectly "time capsuling" it so people
throughout the world could reference it later. But I just toyed with
Google groups and it appears that Usenet binaries are actually not
stored along with their associated Usenet postings.

Do other Usenet-prowling robots archive the binary attachments?

Please post any suggestions you may have. Thank you.

Mike Vance wrote:
comp.databases is mainly about relational databases and not
cryptography. Please repost your question to a relevant group.

Regards,

Michael.
--
Which is more dangerous? TV guided missiles or TV guided families?
Visit my home page at http://michael.zedeler.dk/
Get my vcard at http://michael.zedeler.dk/vcard.vcf

Sorry about my having posted here. There is not a single newsgroup that
really suits my question that I can think of, but I will try taking it
to some newsgroup more suitable.

Michael Zedeler <michael (AT) zedeler (DOT) dk> wrote:

satisfied with his encryption. The alt.HowToCreateATimeCapsule group
does not exist so he asked in a group that is devoted to the storage
of data. It is probably the wrong group in that he won't get a
satisfactory reply but he does not deserve to be castigated for
trying.

Mike Vance <nospam (AT) nospam (DOT) com> wrote:

your lawyer. Put it on your, and your children's web site. In thirty
years it will either be easy to verify that there is only one version
or it will start a huge debate on who has the original version.

David Segall wrote:
Okay. I spent some time analyzing the question and it is not very easy
to answer, so lets discuss it since it seems that there are no other
places to get help...

I believe that the question should be split into two or more questions.

1: there is the problem of persistence - how do you make sure that the
data is still there in 20 years? My personal view is that just making a
wide distribution isn't sufficient. There is no guarantee that anyone
will keep the data, since it unencrypted and possibly considered
worthless. I spent some time searching the net for online perpetual
storage providers, but didn't find any. That is one thing that has to be
tracked down.

2: there is the question of availability. In 20 years, the adressing
scheme and processing of data may have changed so much, that the data
may be inacessible. First, there is the question of locating the data.
This can be done with one of the data handle schemes (see handle.net),
provided that the provider from (1) above supports this. Next there is
the question of data formats. Will the data be possible to decrypt?
(Ultimately yes - but how complicated will it be?)

3: Advances in cryptography may render the encryption used worthless. In
just 10 years from now we may have quantum computing that can break the
encryption in no time.

4: how to design the time capsule. It is obvious that some kind of
encryption is involved, but there are many details to work out. Should
the data be encrypted and the key be placed in escrow some where? Who
would provide such a service?

5: how to provide proof that the content is genuine when the data has
been decrypted some time in 2026? Two parts: that the payload itself
hasn't been tampered with and that the escrow provider has no way of
tampering with the data when decrypting.

I think that most of the questions can be posted to comp.security.

Regards,

Michael.
--
Which is more dangerous? TV guided missiles or TV guided families?
I am less likely to answer usenet postings by anonymous authors.
Visit my home page at http://michael.zedeler.dk/

David Segall wrote:
Okay. I spent some time analyzing the question and it is not very easy
to answer, so lets discuss it since it seems that there are no other
places to get help...

I believe that the question should be split into two or more questions.

1: there is the problem of persistence - how do you make sure that the
data is still there in 20 years? My personal view is that just making a
wide distribution isn't sufficient. There is no guarantee that anyone
will keep the data, since it is encrypted and thus possibly considered
worthless. I spent some time searching the net for online perpetual
storage providers, but didn't find any. That is one thing that has to be
tracked down.

2: there is the question of availability. In 20 years, the adressing
scheme and processing of data may have changed so much, that the data
may be inacessible. First, there is the question of locating the data.
This can be done with one of the data handle schemes (see handle.net),
provided that the provider from (1) above supports this. Next there is
the question of data formats. Will the data be possible to decrypt?
(Ultimately yes - but how complicated will it be?)

3: Advances in cryptography may render the encryption used worthless. In
just 10 years from now we may have quantum computing that can break the
encryption in no time.

4: how to design the time capsule. It is obvious that some kind of
encryption is involved, but there are many details to work out. Should
the data be encrypted and the key be placed in escrow some where? Who
would provide such a service?

5: how to provide proof that the content is genuine when the data has
been decrypted some time in 2026? Two parts: that the payload itself
hasn't been tampered with and that the escrow provider has no way of
tampering with the data when decrypting.

I think that most of the questions can be posted to comp.security.

Regards,

Michael.
--
Which is more dangerous? TV guided missiles or TV guided families?
I am less likely to answer usenet postings by anonymous authors.
Visit my home page at http://michael.zedeler.dk/

Mike Vance wrote:
Post the encrypted file and the checksum to a widely archived newsgroup
(and anywhere else you like). When you release the key in 20 years,
everyone knows you haven't altered it because the checksum is the same.
They can be double-sure by checking the file with the original post if
they want.

John