[FoRK] PersonalWeb patents "unique" hashes of content

Stephen D. Williams sdw at lig.net
Tue Sep 18 19:39:41 PDT 2012


On 9/18/12 6:48 PM, Ken Meltsner wrote:
> On Tue, Sep 18, 2012 at 8:41 PM, Stephen Williams <sdw at lig.net> wrote:
>> Content addressable storage == hashtable.  Earliest use of a hash table?
>> 60's at the latest.
>>
>> Actual CAS granularity hash tables:
>> http://doc.cat-v.org/plan_9/4th_edition/papers/venti/
>>> The Stanford Archival Vault [2] is a prototype archival repository
>>> intended for digital libraries. The archive consists of a write-once log of
>>> digital objects (files) and several auxiliary indexes for locating objects
>>> within the log. Objects are identified by the hash of their contents using a
>>> cyclic redundancy check (CRC).
> Or 'rsync' -- version 1.0 released in 1996.  The first of the patents
> was applied for in 1997, so I suppose a "first to invent" rule might
> give them precedence.
>
> Heck, I had disclosures on a content hash-based access control system
> in 1996, so I suspect others had similar ideas even earlier.  There is
> a distant memory of a European company which used cryptographic hashes
> for identifiers of content to be distribution around that time.

Hash table + remote access of a file with a key == CAS.
I.e., a hash table with remote access to the table.

Relative to the Apple bounce scroll: I have been debating with myself whether an effect that is a natural consequence of touch-based 
UI development is patentable or not.  This function is something that is what simpler code does in this case.  You have to add code 
to prevent it from happening.  In other words, it is patenting a bug.

>
> Ken
>
> Ken Meltsner

sdw



More information about the FoRK mailing list