Where are the source files?

The website says open source at least a dozen times and yet I can't seem to locate any actual source files for the software. I read of a possible SDK someone wrote about in forums here but even then that isn't open source. Am I simply looking in the wrong area? I bought the developers edition assuming that I would be able to design my own software suite to my liking but thus far have been pretty underwhelmed. I don't want to come off as condescending or rude but if you pepper open source all over your website I assume it means the source files are available to the public or at least those who bought the developers version. Is there a git-hub setup where the community can submit software changes etc? I see that there are the files to build the hardware and a fork of TinyG2 but nothing specifically for the handibots software.

Hello DomAAmato,

I agree with your sentiments 100%. I purchased my 'Developers Edition' back in Feb with the expectation of having extra "Developer" resources to design apps and possibly a new front end GUI to my liking but found out that is just not the case.

The "Developers Edition" is just a standard Handibot that cost us about 2 - 2.5 times what a "Standard" Handibot will cost everyone else and has nothing extra in the line of developer resources. Everyone else has access to the same programming information.

There is supposed to be a new API framework "coming soon" but no date has been set. But the downside of that is that we will probably need to purchase a new control board to use it.

Even to use the new jigs that are supposed to be "coming soon", we will need to pay an additional $149.95 just to get the jigs, base plate and rail for an additional .3" of cutting area. All that on top of the $2800+ we aready paid for the "Developers Edition".

Thing is, I purchased the Handibot to get familiar with CNC and possibly develop some apps before I purchased one of the larger Shopbot tools, but now I am re-thinking that because of how slow things seem to be happening at Shopbot.

Yes, I'm under whelmed also. I don't think anyone from Shopbot bothers to check this forum on a regular basis. Seems they are at shows a lot and don't have time evidently.



Edited 3 time(s). Last edit at 06/12/2014 09:06PM by gclayton.

Hey guys, check out this response from Ted Hall on this thread:

[handibot.com]

Also, the Handibot, like other CNC machines is controlled by GCode which is totally open. It's literally a text file with move commands in it. I think there are huge opportunities for you guys to create apps that can dynamically build GCode on the fly for different projects.

I myself have been toying with the idea of having a simple web interface that can allow someone to build shapes and objects, press a button and have the code built and sent to a ShopBot.

That is something that you can do TODAY. Nearly every function that is in the current ShopBot control software can be called via GCode, and the ShopBot control software can call files via the command line in Windows, so you could write an app that would work just as I described.

If you created something that dynamically built GCode, it will work with not only the new G2 system, but any other CNC machine too so you're not limited to just ShopBot.

If I was a better Ruby developer I would be able to take my concept of modeling something in Sketchup and exporting it out to GCode all right in the program...

Sorry, if I seem to be repeating my points.

I understand that you can create gcode files or Opensbp files and send these to the Handibot or any other Shopbot tools machine. However, in order to do this, you have to use the current sb3 software.

Those of us that purchased the "Developers Edition" expected to have access to the developer resources that would allow for development of custom applications and to be able to develop a completely custom front-end GUI to suite our needs. This really cannot be done with the current resources.

This is from the description of the Handibot Developers Edition located on the Handibot Store:

Quote

Handibot® is a community-supported, open source development platform.

We are sharing design files and other development resources with the Handibot community and encourage contributors to use these resources to rip, remix, burn and share their version of Handibot and accessories. Use Handibot as a starting point or use it just how it is, fully assembled and ready for work.

However, none of the developers resources currently available offer this openness. We don't even know when these resources will be available.



Edited 1 time(s). Last edit at 06/13/2014 12:05PM by gclayton.

Regardless of whether or not you can parse gcode commands to the machine is irrelevent. Shopbot promised open source in their advertisment of this product and it hasn't delivered on its promise. The code itself could use work too, the UI is just outdated and confusing not to mention it doesn't even adhere to Keyboard UI standards

Microsofts Keyboard UI Guidelines

It would probably be in in shopbots interest to just make the source available on github, I would be more than happy to fork it and submit pull-requests that would at least address these problems.

Open Source isn't an advertising phrase, it's a set of standards with varying licences to deal with varying distribution situations. Open Source Initative

Read the definition from the link, Shopbot does not comply with the standards set forth and so until the time when they do, they are technically falsely advertising their product.

I am happy with the hardware but that is a moot point in all of this.

Did you see this:

ShopBot open initiative

I can tell you first hand that they are actively developing this system. It's not easy as they have to be backwards compatible with nearly 20 years of hardware.

They aren't going to release something until it's good and ready for developers... Remember that were talking about something that controls a high speed cutting tool.. Not just another app.

Also let's not forget that the entire design for the Handibot is on GitHub... I know that's not what you want.. But it shows that they are truly embracing open source.

This group is one of the first people to get a brand new machine. When you're beta testing stuff things don't always go as planned.

I would suggest that rather than be upset that something is taking longer than you want it that you embrace this opportunity to be one of the first.

I had the chance to meet the entire ShopBot team and they are a fantastic, small group of people that actually listen to their users. These messages you post here are read by them.

I was lucky enough to be able to try out an early Handibot. I would suggest that rather than complain... You innovate. I'm not a coder so my innovation has come in the way of design tools and jigs. Because of what I did ShopBot changed the accessories and documentation of the Handibot because of what I found.

They made things easier for me.

Just like your hands are a bit tied because of the lack of an SDk doesn't mean you can't do something... Even I write gGCode.

I had trouble with the machine when I first got it because I had no jigs and no documentation on how to square it. Now we have that and it works great.

Get out there... Do something awesome with what you have and you'll be rewarded for being a trailblazer. I certainly was.

Is it really possible to be backwards compatible with 20 year old hardware? Is ShopBot Tools going to give the new control board to all the previous owners of the 20 year old hardware?

No one is asking anyone to release anything before it's ready. But it seems that the new control board has been in planning/development for some time now, hasn't it?

It's not easy to embrace an opportunity/innovate if you don't have the resources necessary to do so. You have to understand that the things that us developers want to do has nothing to do with putting together geode to cut a part.

As for your statement about beta testing stuff and things not going as planned, the Developer Edition Handibot was not advertised as being a beta machine.

"We" still don't have the jigs to make using the Handibot easier. But that is not what we are wanting in this case.

I am not sure what exactly the plans are for the older machines, I am sure they aren't going to give out new control boards, but it would be an awesome upgrade to not have to use an old machine with outdated hardware and software... That's for the people at Shopbot to articulate though.

I can tell you from beta testing stuff for ShopBot, and other companies that they do listen to the feedback. They are actively reading these posts. They are actively developing the new boards and interface. I can tell you, because I talked to the people that are actually doing it that it's no easy task.

And you do have the resources to build awesome stuff today! So many software companies make software that creates GCode to control machines. In fact, there's one on the App Store now that you can sketch out lines and have them turned into ShopBot GCode.

What's stopping you from writing a lighweight web app that could make some killer GCode? I was at NC Maker Faire a few weeks ago and there were a bunch of people doing just that, they did some really awesome stuff. I could name off 15 things off the top of my head that I would build as a lightweight webapp for my Handibot. All they'd have to do is make GCode.

There was a guy who made a simple webapp that had nice big touch buttons on it that you could sketch things and have them passed directly to the machine. That alone would be fantastic on my Handibot with it's little touch screen because it could run right in a browser.

I've had a ton of other ideas too: A door drilling app. Something where you could setup a particular backset and diameter of a door knob hole that needed to be drilled. The web app could show it nice and visually, and show you how to mount a fence on your bot. Once you've made your choices, BAM, it builds Gcode for you that you can cut.

A "drill press" would be awesome. There's a ton of potential right there.

A radius corner app. Guys that fabricate countertops would love to be able to radius corners with just a few taps. You could totally do that in a web app.

Going even further you could take a Sketchup Ruby script and have it build Gcode based on something you drew. I did up a whole concept on that when I first got my bot.

Barring web apps, if you prefer something like VCarve where they do a lot of the heavy lifting of toolpaths for you, they've got Lua scripting. You can write some crazy awesome apps in that.

The guys at VCarve, per my request made me a Lua script that will allow me to take an entire Wikihouse model right from Sketchup into Vcarve and automatically make all of the toolpaths. I can literally print houses from Sketchup using a ShopBot (or any other CNC)



Here is more information on writing gadgets: VCarve Gadgets

Now I'm not a coder, at all. The last language I knew how to program in was Pascal. I am envious for what you all could potentially do. There more than enough to get started.

If there is one thing I've learned about testing things for people, and then getting them to develop what I want is you have to make a case for it. You bought this tool because you want to develop on it and extend it's capabilities. You can do that now, it make not be the way that you want to do it, but there are tons of options. No one can make a better case about how an SDK should be, or what features sure be included better than you trying being successful, and also failing.

I can't tell you how many bits I broke, plywood I damaged, and hours I burned playing around with the Handibot when I first got it. Because of what I, and many other people did I know have a better machine, and so will all of you.

Just like the guys at Vectric, and the folks at Shopbot, when I made the case for something by trying to do a thing with the tool, they stepped up and helped me solve my problem. Go out and do something with what you have, there's already a lot and you could shape the future the way you want it.

To address your original question about create your own GUI, aside from doing that in the web TODAY, I personally am not sure how that's going to work in the future. I am not a developer, so when the developers explain how it all works, it's a little above my head... I would bet that no matter what the interface/SDK looks like it will parse and run GCode so time developing a Gcode generating app isn't time wasted. There's an entire industry built on GCode I don't think ShopBot, or any other 3d printing company would abandon that.

I can tell you though that they do read and listen. Just demanding an SDK isn't going to make it come faster... Like any problem solver, people need a challenge to work towards.

So I'll put the question to you:

What do you want to build?

You should lay that out here, and I'd bet you'd get some definitive answers as to how it will work. Maybe even some direction to start developing today.

If you truly want to develop today, as in right this very minute, web apps, Sketchup Ruby, and Lua scripting are your solutions. Showing those ShopBot will likely drive the course of the SDK.

Who knows, you could make some money too. Ruby scripts in Sketchup sell, and I'd bet you could sell VCarve gadgets.

Eric,

While I admire your defense of the shopbot team I think we are arguing apples and oranges. I am not trying to make an app that writes gcode, I have done that already. Its about sending the gcode to the machine itself. I don't want to use shopbot's interface to do that, like I said earlier and gclayton has mentioned in other posts its clunky and obstructive.

I don't want to make a web app, or use lua, or ruby, this isn't about making money.

Further whether or not the code is "ready" is completely moot. Even if they put an alpha release online that would be fine because we could actively develop for it and submit bug fixes. You talk about how you made suggestions for them to fix but here's the catch, you had the physical device. Imagine trying to make jigs with an imaginary device, that's basically the situation we are dealing in the software world.

I get that we can make web interface, I get we can make gcode but thats not what were asking.

You know what will make the SDK get out the door faster? give developers source code!

You want to see how to do things with open source, look at makerbots github and compare it to shopbots.

Makerbots Github

The firmware is there, the parser to talk to the machines is there, their interpreter is there. Their interpreter alone has 28 issues open for discussion so their software is always changing because people are actively working to improve on it.

I appreciate you trying to defend shopbot and I am sure they are nice guys but I don't think you fully understand what I am talking about or asking for.

I won't even go into the politics of Makerbot being "open sourced" that's another conversation entirely... Shopbot is making something entirely new from the ground up, this isn't about them hoarding code and not giving it to you... Makerbot's open source projects are derivatives of other projects....

I hear your frustration, and they hear it too, they're working on things. Show us examples of what you want to do. At the end of the day whatever they have built will interpret GCode, so weather it's SB3.exe or some other new program, what's stopping you from making something now? Sure, it's not ideal for your program to dump GCode into Sb3.exe and cut it, but it does work. SB3 is proven and reliable. When the new thing comes along you'll get that, and all kinds of new API's to work with the machine.

What is that you wrote already? What would you write if you had A B or C? Tell us why you can't accomplish X Y or Z in Ruby or Lua.

I do fully understand what you're asking and I'm giving you the alleyoop here to make your case.

This will be my last post regarding this issue. We just don't seem to be able to get our point across.

There's more to App development than just creating an app to output some g-code.

As stated before, we know we can using the existing software to produce some g-code to create a part!

We want to be able to come up with our own front end to the Handibot that actually controls how the g-code is managed and submitted to the Handibot. That was the reason I jumped and bought the "Developer Edition" of the Handibot.

What if I want to accomplish X, Y, Z in C, C++, C#, Python etc... Pick a language whatever thats not the point. How about this, this is my python script to run a laser engraver from a Raspberry Pi. A developer could look at that and change it to fit their needs, Thats what I did. I took it from another laser engraver and added in the ability to control the laser using SPI and the Easy Stepper Motor Drivers. Also I wrote a simple Image to GCode python script so yeah I could use that GCode from my python scripts, send to the shopbot sb3.exe and have it cut but thats not what we are complaining about.

Image to GCode

Laser Engraver GCode Interpreter

My source is available, shopbots is not.

How about this. If I had A,B,C I could write an all in one program that does not require the use of vcarve or sb3.exe. It could import SVG's or STL's and add support for addons that could interpret other 3D formats. Maybe I could make an addon for Blender 3D so that you could make a model and then just print directly from Blender instead of having to export it convert it to gcode then send it sb3.exe.

You even admit

Quote

Sure, it's not ideal for your program to dump GCode into Sb3.exe and cut it

So how about I write a program that doesn't have to. Is that a good enough reason?

I can list other things but perhaps I would like to not list them and develop them myself instead of submitting my ideas to shopbot so they can develop it.

I don't know how you can so matter of factly dismiss makerbots open sourceness when their code is available online and actively developed by community members as compared to shopbots which does not exist. Makerbot has plenty of problems and I'm not putting them out there because I think so highly of them I was using it as an example of code, available to the public, that is actively developed by the community. Which I cannot understand why you are arguing against that, it seems to me like it would be in shopbots best interest to have that kind of support from the community.

To be honest I don't even know what your arguing with us about, we want to develop software for this device. Would you prefer that we don't? Excuse us for holding shopbots feet to the fire for promising open source. Now please stop I am tired of arguing this.

I think you've misinterpreted my intentions here.... I totally get what you guys are saying and I agree. What every one wants here is to get you guys even more than you already have to develop on. Rather than just saying "I want the SDK now" I am trying to get you guys to really describe what you want. To be clear: I agree with you, I am just trying to spur some ideas.

You're developers, and so is ShopBot. I am sure you know very well that amazing things can happen when there is a specific challenge to solve.

My intention here is to get you to start brainstorming... I don't know what it takes to develop entirely new hardware and software to motion control a CNC machine, but I'd bet it's a lot of detail that the engineers are working with. When you get that deep into making sure a stepper motor moves the right way, picturing someone on a jobsite with a Handibot might be a little far off. Brainstorming ideas now I think would be very helpful for all involved. Who knows, maybe an engineer here will guide you on a potential idea that you have, or provide some sample code for an idea that they have...

So, just to reiterate, I do get what you guys are saying, I really do. That's why I am posing the challenge.

I don't have any more access than you do, and I don't even code... But I had an idea of an easier way to print a house. I worked with Vectric, the guys at Sketchup, and put in a ton of my own hours and we can now print a house directly from a Sketchup model. I can literally take a 3D model that I designed in Sketchup, open it in Vcarve, and form there use a Lua script to automatically toolpath hundreds of sheets of plywood at once.

What took days now takes minutes because of this process and it's all because of some simple scripts and ingenuity. Is this process a super slick program that plugs right into a Shopbot? No, but it works. I've shown this to fabbers all over the country and their jaws are on the ground after I do the demo.

And you know how this got done? I was frustrated with the current process and I challenged the developers at these companies with solid ideas and workflows and we got something done.

That's proof that you can do awesome things now.

Sorry to have caused you frustration, that was not my intent at all. I really am trying to help you all out.

Hey guys, this may be just what you're looking for:

[github.com]

No thats just a fork of the TinyG2 project. The current boards are something different, you can tell because they have a USB B female input vs the Arduino Due which uses a USB micro, unless they went through the trouble of fitting a USB B port on the due. I'm assuming that has to do with their future intentions for the handibot controller but they forked it and haven't made any changes to it over the last year. I can understand them keeping up with the stable release but even the master branch says its outdated so there is nothing to go off of in terms of writing our own software. I suppose we can try sending serial packets at it but we don't know how they are formatted and read by the controller let alone the proper baud rate so that's really just a waste of time on our end.

We [the software team] are really close to release an alpha version of the new system on github, and the first version of the SDK.
And we are sure that developers can help us improving it and go faster in the development process, but we needed to have enough information about the new system to be sure to make it coherent.

Jimmy, for the software dev team.

Thanks so much for the update Jimmy! I know there are at least a few people on this forum/thread that will really appreciate it! I cannot wait to see what everyone comes up with!

Jimmy,

Any chance you can gives a brief outline of what is about to drop? Ie language/os/25 words or less on where it lets us hook in ...anything really.

Very excited!

Mitch