[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [tlaplus] Re: CSP vs. TLA+



Yeah. The "different" understanding between Dr. Hoare and Dr. Lamport, about a concurrent system, or , generally say, a system,  makes me really curious. 

From CSP, a process can be STOPPed or SKIPPed, doing nothing like a zombie. a finite steps/events...

From Temporal Logic Action(TLA), with the [] operator everywhere, any system is infinite. If doing nothing, it is because it is stuttering....(or we say, recursive looping).

From my feeling, "x' = x" stuttering is a recursive iteration. I have been thinking that CSP's STOP should br re-defined as recursive doing nothing. And I wrote a paper about this idea and got rejected by peer-reviewers...

Huailin

On Fri, Apr 23, 2021 at 5:03 AM Andrew Helwer <andrew.helwer@xxxxxxxxx> wrote:
In TLA+, I usually model a process stopping or terminating by it entering a state where the only enabled action is one where the variables are all left unchanged.

Andrew

On Friday, April 23, 2021 at 1:58:33 AM UTC-4 hua...@xxxxxxxxx wrote:
folks,

As a fan for both CSP and TLA+, I realized something pretty fascinating and 'd like to share my thoughts. 

For a CSP **Process**, e.g., the famous Vendor Machine, which is being used as an example in Hoare's CSP book, a process can be broken/halt. 
Hence, STOP and SKIP process are explicitly defined in CSP to describe that a process get terminated accidentally(STOP) or successfully(SKIP).
In other words, in CSP, Tony Hoare think that a process's behavior can be FINITE steps.

However, 

For TLA, Lamport proposed that a behavior is always infinite by introducing the stuttering step. 

From my perspective, the suffering step of a TLA+ spec is nearly equivalent to the recursive behavior for an CSP process. 

I am very curious to find a consistent mapping between CSP and TLA+. For example, how we gonna describe a STOP process by using TLA+ logic.

I am working on this problem these days(I've been thinking that STOP is NOT simply STOP.

Huailin

--
You received this message because you are subscribed to the Google Groups "tlaplus" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tlaplus+unsubscribe@xxxxxxxxxxxxxxxx.
To view this discussion on the web visit https://groups.google.com/d/msgid/tlaplus/2fce71ec-47a9-4724-84d5-7bac4c9ccab8n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "tlaplus" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tlaplus+unsubscribe@xxxxxxxxxxxxxxxx.
To view this discussion on the web visit https://groups.google.com/d/msgid/tlaplus/CAE7Z%3D%2B413Xnh6HaPqpf50H0Fxs5AkVcBku62aLf87cA28EJtAA%40mail.gmail.com.