tag:blogger.com,1999:blog-7883111926682865025.post7866841137178156364..comments2022-03-24T10:06:16.524-07:00Comments on Am I Stronger Yet?: In Async Programming, an Ounce of Opportunism is worth a Pound of PlanUnknownnoreply@blogger.comBlogger2125tag:blogger.com,1999:blog-7883111926682865025.post-3117564004812848412011-12-12T08:26:10.595-08:002011-12-12T08:26:10.595-08:00Well, it makes me think of how the approach is whe...Well, it makes me think of how the approach is when you write "recovery" code for a database system. Any previous action can fail unless you've enforced a specific sequence of events, but even so, in distributed systems, less is known, thus the recovery algorithm need to reason about what is know, what is unknown and what the correct action is to get to a "better state", and eventually be "complete".<br /><br />(hmmm, pretty funny, commenting a blog can only? be done in the name of my own blog?)The Authoritive Oztrailian Bloghttps://www.blogger.com/profile/17978146578351610080noreply@blogger.comtag:blogger.com,1999:blog-7883111926682865025.post-53091488130636905022011-06-07T08:54:58.069-07:002011-06-07T08:54:58.069-07:00For attempt of general solution to a problem of or...For attempt of general solution to a problem of orchestration of async calls please see mesh.js library at https://github.com/tatumizer/mesh<br /><br />It's a javascript library, but all the concepts apply to any evented runtime. There's very detailed readme file, but the main idea becomes clear already at page 1. At the end, you can find example of "reliable service", where I consider exactly the same problem (there's even a link to the above article as a source for inspiration)Alexhttps://www.blogger.com/profile/10409303482727867298noreply@blogger.com