What memyra actually does
When something needs a piece of data it has fetched before, it asks memyra
first. If one of your devices is holding it, the answer comes back in
milliseconds. If not, memyra says so plainly — a miss — and the caller
fetches it from wherever it came from originally and hands it back on the
way past, so the next request is fast.
That is the whole idea: memyra is a cache, not a store. Nothing here
is permanent, and nothing here needs to be.
What a lending device can see
Nothing. Bytes are sealed before they leave the hub, and the name each
piece is filed under is a one-way function of its content — so a device
cannot read what it holds, cannot tell what it is, and cannot tell that the
phone next to it is holding the same thing.
It also never touches the disk. Not a temp file, not a cache directory, not
a spill under pressure. When a device disconnects, what it held is simply
gone, and the hub is told so it stops pointing there.
Adding a device
Sign in, pick the kind of device, and memyra gives you a link. Open it on
the device — that is the whole setup. Phones and TV boxes install the app;
a Mac or Linux machine runs one line. The link works for 72 hours and can
be revoked at any time.