Thanks! And Clojure actors.

[alandipert]

Hello David! I was so inspired by this project that I immediately dug in, and ported a (probably distorted, limited, toy) version of what I understand of it to Clojure, here: https://github.com/tailrecursion/chasma/

I wanted to drop in to share that in case it's useful to you or anyone else, and to express again my gratitude for your sharing this. I'd run across actors many times, in many contexts, and it never quite "clicked" or seemed industrial-strength until I read this project's README and started to understand the importance and implications of transactionality.

[dbmcclain]

Hi Alan!! Very Cool!! I am delighted that you also see the benefit to Transactional Conventional Hewitt Actors! I use it all the time now. And it is really a great coordination method between GUI code and the body of problem-domain Lisp logic. A really good reference from one of the originators of this whole idea is to read Dale Schumacher’s web postings at https://dalnefre.com/wp/ His illustration language, Humus, is simple enough to translate to Lisp. Enjoy !!! - DM

…

On Nov 8, 2025, at 09:42, Alan Dipert ***@***.***> wrote: alandipert created an issue (dbmcclain/Lisp-Actors#6) <#6> Hello David! I was so inspired by this project that I immediately dug in, and ported a (probably distorted, limited, toy) version of what I understand of it to Clojure, here: https://github.com/tailrecursion/chasma/ I wanted to drop in to share that in case it's useful to you or anyone else, and to express again my gratitude for your sharing this. I'd run across actors many times, in many contexts, and it never quite "clicked" or seemed industrial-strength until I read this project's README and started to understand the importance and implications of transactionality. — Reply to this email directly, view it on GitHub <#6>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYDUAPLAOCKL5RZWVOF7DD33YMPLAVCNFSM6AAAAACLRBYIFGVHI2DSMVQWIX3LMV43ASLTON2WKOZTGYYDGNZUGA4TSMQ>. You are receiving this because you are subscribed to this thread.

[dbmcclain]

The main takeaways that I get from Transactional Conventional Hewitt Actors is, 1. Most people conflate Actors with Threads and Processes (Erlang). You can’t kill a TCH Actor any more than you could kill the SINE function. 2. Most people also make every Actor hold its own mailbox. That goes along with conflating Actors with Threads. 3. TCH Actors are really great for network security, since you can’t fake an Actor address - the messages will just fall on the floor. And Actors don’t have to respond to every message, just correctly formatted ones. Which means that hackers trying to stab at a connection just receive the silent treatment. And for those who make successful connections to an Actor on this side, they might really be talking through a proxy Actor which may have a shutdown channel (a TAG) fed by some security monitor Actor watching for errant behavior. The monitor can signal the proxy to just become SINK, shutting off the attempted malice. The neat thing about TAG is that it simply forwards messages prefixed by its own identity. And so TAG’s have unique identities, since Actors have unique identities. And so if you want a control backchannel to some Actor, have it generate and hand out TAG’s to itself. Using pattern matching with WHEN or “/“ clauses to check that the incoming first arg matches the TAG that you know you handed out is a way of segregating control messages from the general public queries.

…

On Nov 8, 2025, at 09:42, Alan Dipert ***@***.***> wrote: alandipert created an issue (dbmcclain/Lisp-Actors#6) <#6> Hello David! I was so inspired by this project that I immediately dug in, and ported a (probably distorted, limited, toy) version of what I understand of it to Clojure, here: https://github.com/tailrecursion/chasma/ I wanted to drop in to share that in case it's useful to you or anyone else, and to express again my gratitude for your sharing this. I'd run across actors many times, in many contexts, and it never quite "clicked" or seemed industrial-strength until I read this project's README and started to understand the importance and implications of transactionality. — Reply to this email directly, view it on GitHub <#6>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYDUAPLAOCKL5RZWVOF7DD33YMPLAVCNFSM6AAAAACLRBYIFGVHI2DSMVQWIX3LMV43ASLTON2WKOZTGYYDGNZUGA4TSMQ>. You are receiving this because you are subscribed to this thread.

[alandipert]

Thanks much for the references. I realized the first time I learned about actors in this sense was actually Dale Schumacher, at Strangeloop. I remember he had an extremely compelling visualization of the runtime.

I’m pretty sure this was the talk: https://www.infoq.com/presentations/Actor-Interaction-Patterns/

[dbmcclain]

Very cool! Thanks for that video link. Watching him now! - DM

…

On Nov 8, 2025, at 13:43, Alan Dipert ***@***.***> wrote: alandipert left a comment (dbmcclain/Lisp-Actors#6) <#6 (comment)> Thanks much for the references. I realized the first time I learned about actors in this sense was actually Dale Schumacher, at Strangeloop. I remember he had an extremely compelling visualization of the runtime. I’m pretty sure this was the talk: https://www.infoq.com/presentations/Actor-Interaction-Patterns/ — Reply to this email directly, view it on GitHub <#6 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYDUAJITLBKW6XIYWENKYT33ZIY7AVCNFSM6AAAAACLRBYIFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKMBWHA4DENRYGQ>. You are receiving this because you commented.

[dbmcclain]

Here is one of my Actors-based systems. Everything you see on screen was written in Lisp. And Actors play a major role in coordinating everything. This system analyzes star field images, finds all the stars, measures their magnitudes, then extracts a database segment from the Aladin database in Strasbourg, France, to get their actual magnitudes. So the images self calibrate against the database. As you mouse around the star field image, each star shows its measured magnitude, and right click on it shows all the gory details of the measurements. 

…

On Nov 8, 2025, at 13:43, Alan Dipert ***@***.***> wrote: alandipert left a comment (dbmcclain/Lisp-Actors#6) <#6 (comment)> Thanks much for the references. I realized the first time I learned about actors in this sense was actually Dale Schumacher, at Strangeloop. I remember he had an extremely compelling visualization of the runtime. I’m pretty sure this was the talk: https://www.infoq.com/presentations/Actor-Interaction-Patterns/ — Reply to this email directly, view it on GitHub <#6 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYDUAJITLBKW6XIYWENKYT33ZIY7AVCNFSM6AAAAACLRBYIFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKMBWHA4DENRYGQ>. You are receiving this because you commented.

[dbmcclain]