Hey, looks good! A few minor comments about Aleph/Manifold: > Manifold pre dates clojure.core.async Afaik core.async is older than Manifold, not the other way around. The first commit of core.async is 10 months older than the first commit of Manifold. > these data structure can be used as the body of a response and Aleph will return their contents as soon as they are available and will close the connect ion as soon as these sources will be closed True, Aleph will delay cleaning up necessary resources until the underlying deferred/stream for the body is done, but to be precise, the network connection isn't one of them (except in the rare case when keep-alive is false). For the most part, given the high overhead of creating connections, we try to keep them around and reuse them for multiple requests/responses. > ... see how it is used in Aleph's examples here. http://Aleph.io is a cool way to present docs, but it's getting out-of-date, and unfortunately, we no longer have access to it. The most up-to-date docs are at http://cljdoc.org, and the current examples are in the examples/ subdirectory in the repo. > (def _server (atom nil)) Slight stylistic note: I usually see underscores prefixing things that are meant to be ignored, like a param you don't need.

Hi @U10EC98F5 🙂 Thanks for your feedback! > Afaik core.async is older than Manifold I've based my assumption on the copyright statement of both projects • https://github.com/clj-commons/manifold#license: Copyright © 2014-2022 Zach Tellman • https://github.com/clojure/core.async#license: Copyright © 2017-2023 Rich Hickey and contributors Could it be possible that the repo in clj-commons project has more recent initial commits than the original project? > the network connection isn't one of them Probably I have to make it clear that I am referring to the deferred or stream and not to the network to avoid confusion, thanks for pointing this out! > The most up-to-date docs are at http://cljdoc.org, Thanks again! I will adjust the link to point to cljdoc > Slight stylistic note: I usually see underscores prefixing things that are meant to be ignored, like a param you don't need. You are right! I always confuse the ignore case from the derefferable case, is it it common to use an underscore in a postfix position or something else?

I don't know where the 2017 core.async copyright year comes from; I based that on the initial git commit timestamp. But even assuming Hickey sat on it before announcing it, he gave a talk on core.async in 2013, as did Tim Baldridge.

> I always confuse the ignore case from the derefferable case Deref-able? I'm not familiar with any notation to indicate deref-ability. Have you checked the https://guide.clojure.style/#naming?

> Thanks again! I will adjust the link to point to cljdoc Unfortunately, the stuff in examples/ is NOT yet in http://cljdoc.org... 😔

> he gave a talk on core.async in 2013, as did Tim Baldridge. mmm at this point I am inclined to remove that sentence, maybe pointing out that both came out almost at the same time > Have you checked the https://guide.clojure.style/#naming? I didn't find anything specific there, maybe it is my memory of the coding style of a previous job, I'll adjust it so that at least id does not conflict with the idiom of something to be ignored > Unfortunately, the stuff in examples/ is NOT yet in http://cljdoc.org... 😔 I will add the examples in this case 🤝

> Could it be possible that the repo in clj-commons project has more recent initial commits than the original project? No, because the clj-commons repos are forks, not new repos, so the history is the same.