Vasilis Papavasileiou

> Also, I don't think this option is currently available for signals embedded in the DOM via Eliom_content.Html5.R and friends. Indeed, we do not expose the strong stops for the...

Warnings are just warnings, and we have looked at the remaining ones. We would need to do ugly things to get rid of the type inference warnings. I personally prefer...

Sure, the IP doesn't make much sense, we live in a world with dynamic IPs and whole groups behind a NAT. I will look into how to implement this. (I...

I can't really predict how hard it is before I try to do it. Should I, or is the hack trivial to implement?

Great. In that case, I will try to do it on top of the cohttp version, to prevent unnecessary divergence.

I like the idea. Implementing this may be a bit of work (I am not sure we have clean-enough interfaces to query the routing data structures), but I will take...

We know, there is a warning. Replacing all the remnants of the Camlp4 HTML syntax is an extremely mundane task that I haven't had the courage for. Contributions welcome :).

The up-to-data HOWTOs are under the `/tuto/` path: https://ocsigen.org/tuto/manual/ The `/howto/` stuff should be removed. For RPCs, see https://ocsigen.org/tuto/6.1/manual/how-to-call-a-server-side-function-from-client-side . I just made a few minor fixes (currently only visible...

Right. I thought I had fixed this. I pushed a fix to the `service-split-att` branch (cd0b7ab7c32796170aaf6bdc0d7912ab42ae00fd, #359).

Presumably all network issues that have prevented WebSockets adoption still apply. Fragile interaction with proxies, an entirely separate protocol that firewalls may not like, etc. But implementation-wise, we should be...