I would share my thinking about ScopedPromises, those over a connection with wire management of remote assets, whether in-memory or over a network. I’m concerned about QoS of the eventual send. 

When an eventual send goes out, with a resolution expected, Three registrationsexist, during ::

• pre-existing scope table registration to the remote asset, 
• promise to the result, over there.
• an export of the local resolver pointing back at the promise of the result of the send, in Bobs Cat.

I aim to add a 4th registration ::

• a pointer out to the actual message send, as it goes from Bob to Alice, over there. 🫵

With this a QoS eventualContext can be used to do a variety of tasks:

• status, including expectations <queuing time>
• cancel / fail
• adjust priority

I would extend Promises so.

••• rabbit ❤️‍🔥🐰




On Fri, Jul 7, 2023 at 22:22, rabbit via Cuis-dev <cuis-dev@lists.cuis.st> wrote:

Good evening, I am remiss for failing to humbly acknowledge all the hard work others offered in establishing the core of Cryptography. I wrote ASN1, X509 and ReedSolomon, but others did the real core heavy lifting. I am a harvester and integrator. I am so grateful, it;s coming together into one package. Now <Crypto-rabbt.31.mcz>, weighing in at 6,500,910 bytes! Yummy!

Test results :: <465 ! 420 ! 4 ! 4 ! 37 ! 0>

My latest:

Eye candy:


On 7/7/23 19:00, rabbit wrote:

#GREEN without the FAILs Um...I brought in everything. My life's work. Crypto now includes all my work over the past 22 years. I did not realizze, it is 6.5 MBytes in size. Heh.

Issues:

I regressed, or are failing

I introduce a name change from SchedulerVat to ECar. I added am EUICat for addDeferredUIProcess: scheduling. I also added a EHyperCat, which instantiates 3 ECats and adds a EUICat, for 4 VirtualizedSchedulers.

SharedNear requires wire management over a channel (SharedQueue), so we need a Scope. This is a Good Thing, allowing Scope debugging locally.

My attention is on getting EParrotTalk working.

••• rabbit ❤️‍🔥🐰
--
••• rabbit ❤️‍🔥🐰