Re: AI for creating ASCII art diagrams

[Christian Huitema <[e-mail]>]

On 2/10/2026 1:47 PM, Donald Eastlake wrote:

> I usually use the text version. Furthermore, I might mention that as
> an author of 109 RFCs I have never, at least for these RFCs, run into
> a situation where I felt that ASCII art was inadequate to express what
> I wanted when I felt a diagram was needed.

I like the text version too. But I am no fan of specifying packet 
formats using
ASCII graphics.

Take for example the ADS-B 112-bit message for which Bob eventually got 
the ASCII diagram:

   0                   1                   2                3
  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|    DF   |  C  |                      24AN                     |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|    TC   |                         MSG                         |
+-+-+-+-+-+                                     +-+-+-+-+-+-+-+-+
|                                               |               |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|              CRC              |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Using the representation syntax defined for TLS 1.3, this would be:

ADS-B 112-bit Message {
      Downlink Format (5),
      Capability (3),
      Aircraft Address (24),
      ADS-B Data (56),
      Parity Check (24)
}

I think using a syntactic description beats the ASCII tables hands down. 
It is more compact, easier to read, and does not require any particular 
drawing skill. The only downside is that it does not look as much like 
RFC 791, but, well, time passes.

I think the IETF would be better off promoting a representation language 
than investing in ASCII formatting tools...

-- Christian Huitema