Re: [Spud] SPUD's open/close are unconvincing

[Yoav Nir <[e-mail]>]

> On Apr 9, 2015, at 9:40 PM, Tom Herbert <[e-mail]> wrote:
> 
> On Thu, Apr 9, 2015 at 10:46 AM, Toerless Eckert <[e-mail]> wrote:
>> Glad you didn't ask how SPUD can solve world peace, because i am
>> still researching that aspect.
>> 
>> There is a lot of work going on wit security models of device deployment,
>> you may want to look at 6LO, 6TISCH, ACE, DICE, ANIMA and there is a
>> lot more outside the IETF. I would see that work as prerequisites
>> to move forward.
>> 
>> IN general, i think there is one good and sane device operations method,
>> in IoT which is to NOT upgrade the OS after initial device deployment,
>> but just upgrade/modify applications. And with SPUD, that means you
>> now can upgrade/fix/improve transport layer.
>> 
> Well, there's already over a billion smartphones in the world which
> are already regularly updating their OS and this hasn't yet led to
> mass anarchy. You can certainly make avoiding OS updates and avoiding
> OS in general a design point for your products, but this is not a
> relevant design point or requirement for a networking protocol. As I
> said, it's a red herring wrt SPUD.

There are also over a billion smartphones stuck in an old version of the OS. Android 2.6 is very popular as is the latest iOS version your phone supports (lots of iPhone 4 out there, and they’re not getting iOS 8).

And don’t get me started on desktop operating systems. By NetApp’s numbers just under 17% of desktop and laptop systems are running Windows XP. The majority are running the already outdated Windows 7.

And those phones and desktops are things people are always looking at. Lots of people do upgrade the OS on their phones and computers when given a choice. How many people are going to want to upgrade the OS on their refrigerator, car fuel-injection system, air conditioner, or industrial sensor? Those things are deep into “if it works - don’t fix it” territory.

Yoav