Re: [arch-d] New Version Notification for draft-nottingham-avoiding-internet-centralization-02.txt

[Spencer Dawkins at IETF <[e-mail]>]

Hi, Christian,

I'm mostly reading this thread with interest, but on your question below
...

On Wed, Feb 16, 2022 at 3:16 PM Christian Huitema <[e-mail]>
wrote:

>
> In contrast, we have examples like gmail in which a single free service
> pretty much blocks competition from anybody but Microsoft. This is a
> case where the barrier to entry is rather high, largely because the high
> investment required to deal with spam and other security issues. The
> operating cost do raise with the number of users, but not quite linearly
> -- spam control, for example, becomes easier when a larger number of
> users enables better gathering of information. The theoretical example
> of that is the pure software market, in which the marginal cost of an
> additional user is near zero. In that case, the dominant player has more
> revenue and can innovate at a faster pace than competitors with fewer
> customers.
> * I wonder how many of our network protocols have that characteristic?*
>

 I'm not sure what you're thinking of, re: network protocols, but if
transport protocols are included, I would think that QUIC (and the
constellation around QUIC, like MASQUE, etc.) might have that
characteristic (among many other characteristics).

Is that relevant to what you're thinking?

Best,

Spencer