helping Anelok (was Re: Anelok: one more MCU - maybe an idea for later)

Werner Almesberger werner at almesberger.net
Sat Dec 27 03:11:04 UTC 2014


Ron K. Jeffries wrote:
> Much to like about this two-MCU approach.

Glad you like it :) From an engineering perspective, I certainly get
that certain "this feels right" (even if a bit unusual) impression.
Plus, there's the obvious marketing angle.

> What might I do to help?
[...]
> Feel free to email me if you prefer

That's a question I hope is of interest to more than just one
person ;-)

Right now, I want to advance the project in the following directions:

1) Hardware design testing.

2) Get (software) developers to join the fun.

3) Find a way to get prototypes out.

4) Improve "outsider-accessible" material.


Regarding 1), there I'm pretty much on my own. The things that
absolutely need to be known to be good to proceed (especially with
point 3) are the capacitive sensor and operation on battery power.

At the moment, I'm focusing on the capacitive sensor. To properly
test whether it is sufficiently convenient to use, I need a bit more
of the GUI to work, so that I can play with usage scenarios
resembling real life.

At some point in time I'll need more components and also consumables
for my mill. I may be able to find the latter locally, but
components are difficult.

Regarding 2), there is a lot of work that still needs doing, e.g.,

- user interface,
- memory card driver,
- file system (I guess FAT should do nicely),
- password store,
- crypto functions (building blocks; I hope we can use something
  like Curve25519 for the public-key side of things),
- crypto system (putting the building blocks together),
- USB HID device,
- USB host accepting HID (and hubs),
- BTLE link-layer (tricky since the CC2543 hardware doesn't support
  all the bits needed),
- BTLE higher layers, up to ...
- HID over BTLE (or something equivalent),
- etc.

I was hoping the simulator would encourage people to start getting
their hands dirty, but that didn't work.

Regarding 3), that would be the next step a) to motivate prospective
developers (*) and b) to give them access to the entire system (for
hacking device drivers and such).

Alas, I can't really produce more than a very small number of
prototypes. I.e., I'm reasonably self-sufficient but making
prototype devices to send away would exceed my resources, both in
terms of parts as in the sheer amount of time needed to do all this.

(*) Look at all the happy world-wide collaborators after 03:18 in
    this video:
    https://www.indiegogo.com/projects/mooltipass-open-source-offline-password-keeper
    Each of them has nice shiny prototype hardware. If the night
    sky of Buenos Aires is lit up in bright green light, that's
    envious me.

Regarding 4), this is material people not intimately familiar with
the projects can appreciate. This includes:

- the Web page,
- block diagram and similar illustrations,
- technical documentation,
- user documentation (for later - things have to settle a bit before
  we can document them).

While all these things also play an important role for "inside"
communication, they're important for marketing purposes, where
"marketing" also involves pitching the project to investors.

I'm currently beefing up the Web face of the project, to make it look
a bit less drab but also to explore and define the style elements I'd
like to have in the long run. You've seen a first result in this area
already, the interactive block diagram. There's another one coming in
the next days: hoverable content for the wire frame model on
anelok.com


Now to the helping part. With 1), what I need the most is time and a
little luck. If I need components, things will get a bit difficult:
Argentina currently has a restriction on "Internet purchases" in
force that limits such subversive actions to two per year and
person.

Furthermore, my own financial reserves are pretty much depleted and
my current "day job" (Neo900) pays enough to keep me fed but that
won't allow for lavish purchases on the Anelok front.

I'm thinking of asking people who are interested in helping the
project to purchase things and send them to me. That may pass as
"gift" and not evil "Internet purchase" (even though the senders will
of course have bought these things via the Internet ...) But I'll
have to find out first what customs have to say about this plan.

On to 2). The door is wide open, and there's plenty of work for
everyone. So, developers, don't hesitate to let yourself in ! ;-)

With 3), once a "good to go" design is ready, this is also something
that could be outsourced. Making a few prototype kits would basically
amount to

- purchasing components (you need: shopping list, Internet, credit
  card),
- making the PCBs (you need: design files, Internet, credit card),
- populating them with components (you need: either a steady hand
  and lots of time, a place that will do this, based on the design
  files, etc. Note: some assembly may not be automatable.),
- adding case parts (you need: a CNC mill, a bit of acrylic,
  patience),
- send the resulting kits to developers.

Taking care of all this may be a bit overwhelming for a single
person, but a small group of people helping each other could
certainly do it. All the design details would come from my
prototypes.

Finally, 4). I would love to hand things like Web site design and
maintenance off to someone else. This can be broken down into more
manageable parts, e.g, an overall site style with a nice nav bar
would be very good to have. Or let's start simple, with a prettier
"Anelok" than <H1>ANELOK</H1>.

One condition: like with everything else in the project, all the
tools used have to be Open Source. But it doesn't have to be quite
as hard-core as my build process :)

https://gitorious.org/anelok/anelok/source/web/dwg

(Run "make", then load dwg.html with your browser of choice. What
you'll see is the hoverable wire frame model to be. Then you may
want to look at the Makefile and the collection of Perl and shell
scripts that make all this happen.)

Also technical documentation can benefit greatly from co-authorship.


Many of the prototype production processes and such would benefit
from having a company that acts as the single point where everything
comes together. Such a company could also collect money, e.g., in
the form of crowd-investing (crowdfunding would be for later, when
there's an actual product to make.)

Until we get there, handling finances is and will be a problem, and
things like crowd-investing in exchange for equity or such are
outright impossible. However, anyone contributing to the project
could keep track of what they spend and we should be able to turn
this into some form of equity once the project has established a
more stable commercial footing.

Cheers, Werner



More information about the discussion mailing list


interactive