On 3/23/19 2:43 PM, Christian Grothoff wrote:
> On 3/22/19 7:40 PM, amirouche wrote:
>>
>>
>> It would rely on GNS and a TXT record to publish the gnunet fs URI of a
>> map of the
>> database of meme each identity or ego published. Think of this as some
>> kind web
>> sitemap.xml.
> 
> Please do NOT use a TXT record. This part of what you describe fits what
> I had in mind for GNS/FS integration, but I would define a new record
> type, say GNUNET-FSURI or something like that.

Clarification/correction: we should probably call it GNUNET_FS_CHK or
something like that and NOT make it a URI but the binary encoding of the
information encoded in a FS URI. That's more efficient, and we can still
define the human-readable FS-URI encoding for the text representation.
That way we gain ~2x efficiency on the network, and the usability stays
pretty much the same.

>  It's pretty easy to add
> new record types to GNS, we don't need IETF approval, and we have 4
> billion values available. So no reason to overload TXT like it is done
> in DNS. We can do better ;-).
> 
> If you need help defining a new record type (the process for how to do
> this should be documented reasonably well in the manual), don't hesitate
> to ask!
> 
> 
> _______________________________________________
> GNUnet-developers mailing list
> [email protected]
> https://lists.gnu.org/mailman/listinfo/gnunet-developers
> 

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
GNUnet-developers mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/gnunet-developers

Reply via email to