Looks like Temporal and Prefect had a baby. I hope this gets traction and best luck to the team.
Genesis of a New Programming Model
11–18 of 18 posts
Re: Genesis of a New Programming Model
#12Looks like Temporal and Prefect had a baby. I hope this gets traction and best luck to the team.
Thanks!
Re: Genesis of a New Programming Model
#13Sorry, but what does this actually do? I normally solve the problem described here by wrapping the fetch function with a resolving while and a timeout.
Re: Genesis of a New Programming Model
#14Sorry, but what does this actually do? I normally solve the problem described here by wrapping the fetch function with a resolving while and a timeout.
To summarize, our scheduler maintains state to allow retries to happen even if the application moves between compute instances, which is quite common when running in the cloud (e.g., deploying on autoscaling containers, or serverless functions).
Re: Genesis of a New Programming Model
#15Programming is dead, prompt AI
Re: Genesis of a New Programming Model
#16> the operation is now guaranteed to run to completion.
unless the backend breaks somehow (server changed ip/ssl/imap password...) in which case it will result in an endless useless retry loop?
Re: Genesis of a New Programming Model
#17> the operation is now guaranteed to run to completion. unless the backend breaks somehow (server changed ip/ssl/imap password...) in which case it will result in an endless useless retry loop?
We can run it forever but it’s rarely useful. Usually, all operations have an associated expiration time so if it’s impossible to succeed it will eventually abort.
Re: Genesis of a New Programming Model
#18Programming is dead, prompt AI
I so wish AI could create systems that deal with the complexity of problems that Dispatch solves for. Until then, we’ll still have to write a few lines of code.