+1
And I agree all the boot loaders should have the serial option. It cuts
both ways: in some cases people _will_ want to compile it out, so I
think it should be #ifdef’d, but there are many valid use-cases. It
doesn’t seem too hard to accommodate both.
Sterling
On 8 Jun 2016, at 11:41, Kevin Townsend wrote:
Why not both? I don't see Mynewt having massive appeal to the larger
Maker community since the community is very focused on Arduino for a
number of valid reasons, but it will appeal to a certain type of
person in that extremely diverse community. It's the right tool for
some jobs and not for others. It won't replace Arduino and I don't
think that is anyone's goal.
Coming from the maker community myself, I appreciate the production
focus on Mynewt which isn't present in Arduino. Many of the design
decisions that went into the platform are clearly based on real world
experience shipping and maintaining devices, such as version control
and unit tests and simulation. I don't think your average Arduino user
is the target here (feel free to correct me, and I'm only thinking out
loud expressing my own perceptions), but more for the 0.1-1% of people
who want to move up to something they can maintain and produce or
sell.
I'm approaching Mynewt primarily as a potential user and product
designer attracted by a fully open source BLE stack, but I'm also
interested to see if some of the more adventurous or curious customers
take the bait and dig into it themselves. Mynewt allows me to design
products that are fully open, which gives people in the maker
community options they might not have with proprietary code from the
various silicon vendors.
I'm trying to make a case for Serial for both sides of the equation,
though. I can integrate USB CDC from free (with an MCU that has a USB
PHY on board such as the SAMD21) to under $1 with a dedicated
USB/Serial converter (which would be necessary with the nRF5x chips
with no on chip USB). Free to under $1 de-bricking and flashing is
very compelling for a significant reliability boost, even on cost
sensitive devices, and worth the extra 2-3KB flash. I see this solving
real world production issues since you are less likely to have bricked
devices. A nice side effect, though, is that the cheap to manufacture
HW also becomes more maker friendly just in case people do want to
poke at it themselves! It seems win win to me, and can solves problems
for both camps.
On 08/06/16 20:17, David Moshal wrote:
Thanks Wayne, that's very interesting, I think it helps explain the
disconnect here.
I understood David Simmons to be suggesting that Makers (like me) are
the target end-users,
i.e: that MyNewt would be an alternative to, say, Arduino (and why
not).
However, in your use-case, you are the end-user of MyNext, and Makers
are your end-users, no?
i.e: a layer lower in the stack. So, in that case MyNewt would be
positioned as an alternative to an RTOS (or no RTOS) for embedded C
programmers building devices for Makers.
So, if I may, I'd suggest that we step back and ask Sterling who
exactly he envisages as the end user of this project:
Makers, or embedded C developers building devices for Makers?
David
