[onerng talk] review of RNGs

[Philipp Gühring]

Hi,

> > > So how should we design those fields exactly?
> > > OpenSource driver: Yes/No
> > > OpenHardware design: Yes/No
> > > Any other ideas while we are at changing the columns?
> > >
> >
> > Yes, oddly enough things I think OneRNG is good at:
> > * Entropy Sources (e.g. Avalanche/RF)

Hmm, Multiple-Multiple Choice? Or a free text-field?

> > ** Configurable by user (i.e. choose which sources to use)


> > * Raw output available (i.e. mode with no whitening)

In that case, 2 submissions, one for the raw output and one for the
filtered output might be a good idea.

> > ** Whitening method (e.g. CRC16, AES)

von-Neumann filter. Again, multiple-multiple choice of freetext?

> > * Firmware updateable?

> > ** Updatable over USB?

Hmm, that only makes sense for RNGs which have USB. Should we be that
specific?

> > * Firmware signature able to be verified by user?

Before or after the firmware is loaded into the device, or both?

> I really like these in addition to open-source hardware/software.  I
> might
> add "health monitor".

Yes, that´s a good idea too.


> > Collect the official website & contact details, and when you add more
> > columns you'll be able to let manufacturers know.

I have collected website and email address, but I guess half of them are
unuseable.

My idea would be to generate individualized emails to the people, with a
direct link where they can edit their entry.


> > Possibly also keep the raw data files publicaly available on
> something like
> > Amazon S3, so others can analyse them with new tools.

I´ve been thinking about giving access to the random data to other
researchers, yes. Until now, nobody asked for it. Unfortunately, I don´t
have enough ressources to open up the whole database (I just dedicated a
1TB HDD for the random numbers a few weeks ago) to the general public for
download. And I haven´t worked out a concept for how to anonymize the
metadata yet.
I guess I would ask any interested researchers to send me a HDD that I
would fill with the database then. But I would also offer to have their
test-code executed on my machines on all the data.


> Making it a wiki might not be a bad idea, if we don't think it would be
> vandalized.  That would let manufacturers update their entries.

There were a few entries already that had vandalized the result page to
some extent, which I had to redact.

 
> There is an error in my Infinite Noise TRNG entry.  The initial price
> was
> indeed $15, but right now I only sell the ones I make by hand, and I
> currently charge $35 for them.  If anything, I'm more likely to
> increase
> the hand-built price going forward.

Ok, I updated the price.

Best regards,
Philipp Gühring