paul at boddie.org.uk
Wed Jun 19 12:27:08 EDT 2013
On Wednesday 19 June 2013 15:25:16 Werner Almesberger wrote:
> Paul Boddie wrote:
> > Recently, there's been some apparent discontent expressed on the wiki
> > about the NanoNote roadmap.
> Hmm, the Wiki may not be the most efficient place for complaints or
> > that they should be removed in
> > shame by those who didn't deliver the goodies on schedule,
> Maybe they should be marked as "historical" :-) Of course, I don't
> think anybody is happy that all our great plans regarding the
> NanoNote didn't come to fruitition.
I suggested this as a compromise: everyone gets confirmation that something
tentatively scheduled for 2012 isn't being planned any more.
> > Usually, it's Ron who brings up this topic, and I hope he is still
> > reading this list, but maybe it's worth having a "State of the NanoNote"
> > discussion.
> I don't know of any active business or technical development towards
> a NanoNote successor.
> The key to getting anything done seems to be to find enough money for
> it. We've tried the approach of everyone paying for their own upkeep
> and we've seen it fail.
What I've seen by lurking on the arm-netbook list (which is for the Rhombus
Tech thing) is that you need people who will actively offer to throw money to
cover fabrication costs and who don't expect that every last dollar will give
them a very cheap, great value, fully functioning product in "return of
post". Maybe the people donating funds to that are doing so to create
opportunities for themselves - they eventually get something they can build
on and sell - and perhaps this doesn't work quite so well for "finished"
products like the NanoNote.
> It could work if treating this strictly as a spare time project, with
> daytime jobs ensuring a steady flow of income, but that would mean
> very slow development. Even then, prototype runs will take a huge
> bite out of your wallet, especially if the board can't be
I imagine that the various PCB production services that exist now, while
eliminating problems getting boards made, don't deal with the awkward issue
of populating those boards, which is rather problematic especially with small
and surface mount components. I think that it might have been a goal of the
Fritzing people to offer a complete service, but I imagine that the logistics
of taking bills of materials from a bunch of different people, obtaining the
components, and producing final boards (and having them all work) would be a
Amusingly, I saw this when searching for "surface mount PCB prototyping
That looks like "fun"!
Here's another interesting discussion leading to services that will apparently
provide components mounted on such adapters:
I also see various companies who do "prototyping" inviting quotes, but that
usually indicates a relationship based on some inevitable product development
(a large company working towards the final design of some high volume
product), rather than the kind of haphazard, one-off stuff that other kinds
of people like us might want to do.
(It's at this point that I discover quite a few British companies [*] doing
something that people would normally be trying to do in mainland China, which
does make me wonder whether British individuals, and similarly Europeans,
North Americans, and so on, overlook easier-to-pursue options.)
[*] For example, http://www.pcbtrain.co.uk/ came up in a search, and this
leads to some interesting "technological archaeology" as the parent company
apparently made PCBs for the agency responsible for atomic weapons research
and for one of the pioneers of business computing in Britain.
> On the engineering side, the opportunity of using an established team
> that also has a strong presence in Asia has already been lost.
> Sharism seems to have disintegrated, the rest also has other long or
> medium term commitments.
I think quite a few people have built up experience and relationships in Asia,
hard-earned experience in some cases, but people seem to be getting stuff
done. (But at the same time, I do wonder whether some projects couldn't be
done closer to home, as noted above. The OpenPandora people seem to have
managed better by doing things in Germany.)
> On the marketing side, it has been suggested to put less emphasis on
> the openness since openness doesn't sell. This may be the case, but
> then the question arises what else such a project would have to offer
> that would distinguish it from the competition.
Throwing the openness away for a cost reduction is tempting for some people,
but then you have to go up directly against the likes of Samsung and Apple,
and the result is people criticising you for an inferior product anyway.
> On the technical side, I don't think the CPU is the main issue. There
> are several choices for mid-range SoCs today, some of them more open
> than Ingenic's offering, which shouldn't present an excessive
> engineering challenge.
Lots of people are going for various ARM derivatives, some even having Free
Software support, and I think the Rhombus Tech people are doing a reasonable
job categorising the options here. I'm not sure whether the Free Software
support holds up for all SoCs - ARM are getting a bad reputation for this -
and it isn't promising that Imagination Technologies took over MIPS and will
now presumably infect MIPS-based SoCs with stuff like PowerVR.
> If keeping the same form factor (i.e., small clamshell), mechanical
> issues are more important, especially getting a keyboard. In staying
> in line with the openness philosophy, it should at least be possible
> to swap the keyboard for a different model (e.g., when the old one
> gets hard to source, when a different layout is needed, etc.),
> ideally it would even be possible to make one's own from standard
> components and using standard tools.
> Another component that is not easily redesigned is the display. You
> need to find a suitable module you can a) actually source, that b)
> will still be around by the time you go to production, and c) that
> complies with all the technical and openness requirements.
> For the display, you basically need to find a design used in a very
> popular device, so that lots of clones exist, like the iPhone. That
> may require some compromises and getting more pixels than you
> actually need would drive up CPU and memory performance demands.
I think the logistics and obsolescence/turnover problem presents substantial
challenges which people have so far only really solved to some degree by
trying to obtain reliable products in quantity and sticking to them while
stocks last. The problem of only being able to source after-market spares and
replacements for things like iPhone screens instead of the original products
still seems to be a big problem.
Then again, maybe people should be taking advantage of such previous
generation products. The problem then may be in reliably referencing such
products so that others can take advantage of them, as the nature of these
spares is probably somewhat transient (spares are made to meet a demand
during a particular period, suppliers may come and go) and opaque (where the
products really came from may not be obvious).
> > Probably most similar to the NanoNote architecturally is the GCW Zero,
> > which appears to be almost available:
> There's a lot of tablet designs. Tablets are relatively undemanding
> when it comes to mechanical requirements, but then I wonder how
> you'd distinguish such a product from the army of Android tablets.
> Adding game controls is an option. (But you'd be at risk of
> overselling the product's capabilities.) Another would be external
> benefits, as in the Fairphone.
It's unfortunate that there wasn't some cooperation between Fairphone and
existing open hardware projects before they arrived at the technical
specifications. However, there are some common interests, such as the
interest in modularity which would strike a chord with the Rhombus Tech
people if they only knew about it:
> > Does anyone have (or know of) any interesting NanoNote-related projects
> > going on, whether it is just to extend the Ben or to make something
> > similar?
> My UBBs still serve me well :-) I've also prepared a set of patches
> for putting the atben/atusb drivers into Linux mainline. Alas, the
> merge window closed before I had them ready for submission, and I
> still have to return to them.
I'll get back to the 8:10 port stuff eventually, myself. :-)
More information about the discussion