[Standards-JIG] Avatar Image Hashes

Daniel Henninger jadestorm at nc.rr.com
Tue Dec 6 00:38:26 UTC 2005

That's exactly what I wanted to hear.  =)  Still, the question did  
come up and it sent some of us into a flurry of "hrm, are we reading  
this right?"  I mean just because an image is encoded with base64  
doesn't mean it's not still data.  =)  I agree that if it were a hash  
of the encoded form that that would be horrible.  ;)


On Dec 5, 2005, at 6:33 PM, Sebastiaan Deckers wrote:

> If it says "hash of the image data" then you take the hash of the  
> image data.  Simple enough.
> Otherwise the spec would say "hash of the base64 encoded image  
> data." :-)
> Besides, hashing the base64 representation would introduce more  
> uncertainty: should that include the newlines?
> -- 
> Sebastiaan
> Daniel Henninger wrote:
>> Hi folk!
>> In all three of the avatar implementation JEPs, you are told to   
>> generate a hash of the "image data".  Until now, we have operated   
>> under the assumption that this is a hash of the actual image  
>> data,  not of the base64-encoded image data.  However, after it  
>> was brought  up on our list, it seems that this is not explicitly  
>> specified  anywhere.  So:
>> 1. Are we indeed correct in thinking that it's a hash of the real   
>> image data, not the base64 encoded data?
>> 2. May I request that JEP-0008, JEP-0084, and JEP-0153 all  
>> receive  minor updates to clarify this?
>> Thanks!
>> Daniel

More information about the Standards mailing list