Open Source ShopBot Software & Hardware (was ShopBot Software)

[Nathan Cravens]

Ted,

 

I'm delighted to see someone in Industry, in a highly viable fabrication method, interested in our group. To get tools we discuss out the door we need more people like yourself in this discussion to flesh out Open Source business models for Open Manufacturing. I look forward to having that discussion with you.

 

With that said, try this on for size:

 

Begin by going Open IP by releasing schematics and CAD files of the ShopBot.

The company, Bug Labs, is doing just that and provides the model for this: http://bugcommunity.com/wiki/index.php/Hardware_Specifications

 

I've suggested before that we must do more than schematics and CAD files by using detailed step-by-step video documentation. 

 

By going Open Hardware, means you will have more freedom to communicate how to produce your product freely in this visual method. The benefit to is that the workforce is informed and buyers will come forward and increase product sales, that is, when this medium is presented properly.

 

The short of it would involve video segments, organized in a way for immediate viewing on a variety of methods of business procedures, including production task methods of interest to our community.

 

Freely distributed video documentation benefits your firm in many ways:

 

Production:

A manager has more time to refine production process while the medium itself does the training, ideally all of it. New hires are encouraged to train themselves in other roles off the clock, becuase training materials are freely available for those roles.

 

Video of other company procedures make a potentially more informed employee, less suspect to cynicism, and therefore more involved in refining production methods than riding them. Being Open shows you care in more ways than a paycheck, and when people know they are cared for, they will care for you more than just working a job. They will make you a more competitive producer.

 

Consumer as Replicator:

To save labor cost, if the ShopBot is not already sold as a kit, video can be used for the newcomer to assemble the product. The video content would end the current training program, which your firm may receive profits from, but ending the program will not be a loss for this reason:

 

Video as Marketing:

Freely distributed video shows how easy it is to build and use a Shopbot. Sales will increase when well presented.

 

It may be tempting to sell the video instructional, but I believe it wiser to distribute video freely so potential customers can become familiar with your product. Your product earns more than videos I suspect.

 

In sum, the video option provides a way to both keep members of the organization informed while tempting buyers of your machines by giving them the knowledge of the product and its useful applications.

 

This is how the Freeconomy can generate higher dividends for ShopBot while additionally benefiting the Open Manufacturing community and beyond.

 

 

Nathan 

On Sat, Feb 21, 2009 at 4:46 PM, Paul D. Fernhout <pdfer...@kurtz-fernhout.com> wrote:

Smári McCarthy wrote:
> Ted, I totally agree with this. There are two benefits to be had from
> this:
>
> 1. There is a significantly large technical user base that can and
> probably will contribute to your software, which is to your benefit.
>
> 2. There are groups working on free software for CAM devices, such as
> the Fab Lab's (rather disappointly slowly progressing [partly my
> fault])
> Kokompe project. There is a general consensus that a piece of software
> that is capable of reading all major formats and talking in a
> meaningful
> way with all major devices is a good thing to have. Currently a very
> large number of CAM devices are well understood and documented, and
> Shopbot is in my opinion the most valuable piece missing from the
> puzzle
> at the moment.
>
> Please liberate the Shopbot!

Here is another suggestion based on what Ted had previously written:

 > > The Windows thing also bears on your other other comment. The USB
 > > protocol is not being used to control the machine in the sense of
 > > passing vector data to an outboard motion-control board, as would be
 > > the case with a more traditional CNC tool and as I think it may appear
 > > on first inspection that we do. Rather, nearly all the motion
 > > processing work is done on the PC, with only an extremely low level of
 > > timing and step info streamed to the outboard control card. Doing it
 > > this way is much less expensive than developing specialized boards.
 > > Over the years that data stream has evolved into a relatively complex,
 > > convoluted, and highly specialized set of transactions (I won't say
 > > kludge) designed to optimize the motion in these particular tools. By
 > > not publishing it we're not trying to protect it, as much as simply
 > > feeling it is so arcane we can't imagine fully documenting it in a way
 > > that makes sense out of the context of the primary software on the PC.
 > > More importantly, because all the real motion processing happens at
 > > the level of the PC software, access to our protocol would not get you
 > > much in terms of running it from a different OS. For example, a
 > > vectored move is not sent at the level of the USB pass.

Ted could build an interface box to go between ShopBot and a computer (or an
ethernet network).

So:
ShopBot -- usb -- interface -- usb -- PC.
or
ShopBot -- usb -- interface -- ethernet -- PC network.

The interface might need to accept vector paths and return queries as to the
ShopBot's status. If the interface had an ethernet port, then anyone could
send data to it or check on it. It would be nice if the interface box had
extra USB ports or parallel ports so one could stream extra data like images
from a local USB camera to the network or control add-ons. (There are
security issues there, obviously, if anyone on the network could start up
the ShopBot.)

Computers have gotten so cheap over the last few years compared to when the
early ShopBot controller decisions were probably made. One source on this:
  http://www.linuxdevices.com/
For example anybody can get an ethernet network router that runs GNU/Linux
for $100, and an entire rugged OLPC costs $200. Entire networked inkjet
printers are $300. (And obviously Stamp computers are even cheaper.) Build
these routines into that embedded box, so a PC can talk to the box in some
standard way, and then the box talks to the ShopBot. Hopefully, the
real-time control needs would not be that critical? Probably ShopBot could
sell this interface box for, guessing, $300 or so as an add-on using some
small but robust off-the-shelf embedded computer (ignoring software
development costs, which would drive the cost higher). Or, ShopBot could
make just that one set of software available for people to install on their
own converted ethernet network routers (either as proprietary or open source).

It's nice to think people will step forward and slog through previously
proprietary code to make it better for free, but it doesn't always happen
(in fact, it may rarely happen in a lot of cases). It depends on how well
the code is written, what platform it is on, and so on. Also, often old
proprietary code has various proprietary dependencies, not the least of
which is, what open source developers want to be developing on Windows if
they can avoid it? But an open interface lets anyone write software on their
own computer in any language and on any operating system to drive the
interface. So, then someone might write tools for, say, Blender to send data
to the interface box.

Open Source is great, but an "open interface" is often easier as a first
step for all sorts of reasons.

Anyway, would such an interface box be something that might help, Smári?
What would it have to do for you to find it useful?

--Paul Fernhout
(By the way, Ted, ShopBot looks great for serious shop use, but a $1000 or
so mini system may fit hobbyist budgets better with lesser needs (like
something you stick a Dremel tool in). I know I'd love such a system at that
price point that fit on a desktop just for turning out a few small wooden
toys I could design together with my young kid, or for putting a robot
gripper and a USB camera on for pick and place. There are plans on the web
for making such things yourself, plus there are CNC mini-lathes and mills
for $3000 or so, but for some people it would be nice to just get an entire
simple system with a similar gantry configuration, even if it could only do
small projects with less precision. Granted, it might not be worth the
liability or potential bad PR if people were disappointed, but if I can buy
a $300 networked printer or flatbed scanner, I'm really surprised I can't
buy a gantry crane robot-like thing for $1000.)

[Nathan Cravens]

The video content would end the current training program, which your firm may receive profits from, but ending the program will not be a loss for this reason

 

Let's retract by saying the video will encourage people to join the training program for hands-on experience.

 

Educational software could be used to replace an instructor over time, with a portion of the lab open during business hours, by appointment to prevent crowding. This training could be free to promote sales.

[Ted Hall, ShopBot]

Hi Nathan,

Only have time for a quick response at the moment ...

But just wanted to indicate I either agree or am comfortable with most
of what you're suggesting. And while it is very spotty, much of what
you'd like to see as public IP is usefully public already is in one
way or another (via our support or development sites), including early
attemps at training videos. See:
http://www.shopbottools.com/buddy_videos.htm (these aren't prize-
winners, but you'll see we are working at it; we're also working with
several school groups on video tutorials and there are free tutorials
for all our software; there are many tutorials of one sort or another
on our site and on our wiki site, as well as youtube; and if you want
to see free, user-helping-user stuff in the extreme, just visit our
forum).

[FYI. We only charge a little for training to cover the costs of food
for several lunches and to cover the cost of several people coming in
on Saturdays to do the training. We do training because of requests
for it, largely by people who need help learning how to use computers.
A sizeable number of ShopBotters have not used computers before ... so
a big chung of training is actually just PC 101. There are also
ShopBot user meetings every week or two, all around the country and
these are open, free, and free-wheeling.]

Finally, I don't want to get repetitous with respect to stuff I
covered in earlier posts. But originally, ShopBot was just plans and a
parts list with instructions on how to do it (with cute cartoons, now
in a gallery on our site) -- but people wanted help putting the parts
together. So then we started shipping kits of off-the-shelf parts and
with plans and instructions -- we did not even provide parts for the
table because we felt people should do it themselves. Then we heard it
was too much work to build a kit, people wanted tools that they could
just put to work making stuff. Please realize that at the time this
was going on, I was an academic and not using ShopBot to make money
though I did eventually start paying a couple of people to help put
the kits together. It was all pretty much a labor of love reflecting
my enthusiasm for the technology. I'm not aware of having really heard
anything about open source back in the mid 90's when we started
ShopBot, but that was certainly the spirit of our initial involvement
and excitement. So ... if it does not sound like I'm rushing to drink
the kool-ade, just appreciate that it is already a pretty deeply
embeded aspect of our operation ... sure we make a few compromises
because we are a for-profit company, but please visit us or one of the
user-group meetings and you'll get the idea.

I'm concerned that my comments here are getting a little too self
promoting, but I will try and post a few more specific replies to your
questions and the posts of a few others.

Ted Hall, ShopBot