So, it's relatively quick and easy. But what about the quality assurance of peer review? What if someone requires anonymity to feel safe to critique the work of a powerful person? The flip side of which is, how do you protect against public, durable, anonymous nonsensical criticism (of the sort that plagues the internet these days)

which is to say, the status quo is that this is done in private

what exactly is a cell? Pretty sure it's not a neuron.

what kind of metaphor even is this

it's not clear that this is an example until I inspect the code - make it easier to understand what the example is

I think you need to tread carefully here. Making journals obsolete is kind of an unintended side effect of what your real goal is here. If you make that clear, you avoid having readers react viscerally to an attack on the journal system on which their entire career and sense of self-worth might be based.

The systems are not the problem. The problem is that there aren't such systems yet. Or put another way, the systems are the unmet need (with the problem being that the need isn't met)

that have been applied to this pipeline?

typo

now and today are redundant

I don't understand this paragraph

I don't understand what this means

This should name and link to the "Shared Knowledge" section, as well as name and briefly describe the detour you're about to take. Roadmapping is critical in a document this long and complex.

sharing of data

delete

each time what ?

I

this is a good place to point out that by now I'm a little lost about where we are in the overall structure

this seems at odds with the machine-learning layer having to "putty over" everything

of course, you run the risk of politically or idealogically alienating some readers here. One way you could overcome a kneejerk response is to provide an follow-up additional concise, simple less controversial example, such as a drug that is later discovered to have adverse effects (e.g. thalidomide)

it's not clear why it's "dangerous" and I'm not sure if you come back to this idea later, but you should be explicit here

I think this paragraph is meant to say "what about security, is it dangerous to have my precious data exposed to the internet?", and that briefly no, but you'll address this issue more fully below in section ###

From this simple description I really don't understand the concept of pinning or how that avoids the problem of loss of all peers. If it's just mirroring an entire server, that seems prohibitive, but I'm probably not understanding it. I don't think it's worth a complete description, but this paragraph needs to be clearer and also point to a source for a complete description.

hard to understand what this means

this is really important, where you bring together the parts you have already discussed and motivate the rest of the paper. As such, it should be prominent and very clear. As written it looks unimportant and would be easy to gloss over.