Re: [Drip] drip-rid-18 - CTA 2063-A encoding changes

[Robert Moskowitz <[e-mail]>]

On 3/28/22 08:33, Robert Moskowitz wrote:
> I am going through the changes I missed needed since decreasing the 
> HID to 28 bits.
>
> Just caught that sec 4.2, CTA 2063-A needs to be reworked.  Need 4 
> more bits, it MAY be easy, I have not done the work yet, but heads up 
> on this one...

I have time to think more on this.

If I stay with the Base32 encoding, I will have to use 2 bits from the 
1st character in the Serial Number field.  This would work if we reserve 
values of W,X,Y, and Z (first 3 bits 1s) for that encoding.

This approach would allow manufactures to still have 7 possible 14 
character encodings under their manufacturing code. (0 - V)

I do note, now one 'error' in this section.  I say to avoid using 'Z' as 
it is too similar to '2':

"Thus, here a Base32 encoding will be used by also excluding the letters 
Z and S (too similar to the digits 2 and 5)."

Yet then I go and use it to identify HHIT encoded in 2063!

"The first character after the Length Code MUST be 'Z', followed by the 
14 characters of the encoded HIT Suite ID and ORCHID hash."

Argh.

Suggestions?  Should I thus use V, W, X, and Y to encode 11100, 11101, 
11110, and 11111?

Or I can use the Base34 encoding and then there is plenty of space in 14 
characters for 72 bits.

Comments please.

I would like to get this done by end of week, and Friday I have 3 
grandchildren (and their parents :) ) descending on the house! Won't get 
too much writing done then as the youngest (5 mo) uses my office for her 
bedroom.

Bob