When is a design sufficiently open source to be considered a "development" in the wiki and how to handle edge cases.

Whoops, now it happened. Worked all day to do some clean ups in the Wiki an @MattMoses came around just to undo them. This upset me (as well?), so I undid the undos. I calmed down now, sorry for this spontaneous reaction.

The underlying questions appears to be:

1. Do we want as many designs as possible in the wiki or do we want to try to reduce the number to the essential ones? The latter could mean we remove or stuff away designs which are apparently changed for the sole purpose of gaining some marketing advantage. Because these "designs" gain no advantage from the developer perspective. If they come with a gain, it's not visible.

2. Is the (my) choice to demand "comes with files sufficient to make a copy" a good choice, do others agree? See [reprap.org]

3. How do we prove or request proof that design files are sufficient to make a copy? Current case is "Prusa i3 Acrylic", which comes with a blob of STL files on a private share of the vendor. I downloaded them and found print files only. No drawings of the sheets, no vitamins, no CAD files. Even if this were complete, I'm not so keen to hunt down files stuffed away in various places somewhere in the net and trying to check them for usefulness for 50 or 100 designs.

4. Are STL files "design" files? I not sure about this, even if they're good resolution. STEP files would be better, of course, yet better OpenSCad or FreeCAD.

See also [reprap.org], [reprap.org] (talk page) and [reprap.org] . The latter design was started solely because the user couldn't find design files for the i3 Acrylic either.

All together my current goal is to bring RepRap forward towards being a collaborating development community. Just collecting commercial designs is no gain, I think we should put emphasis on collaboration, incremental improvement an knowledge. In a year or two commercial 3D printers will be just as available as 2D printers, guess what happens with RepRap then. If we don't adjust focus or if we do.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Hi Traumflug, sorry to make you upset. I did not mean to offend you personally, rather I was trying to lead by example and "edit first, ask permission later".

After you explained that the dxf files were missing for those machines, it is obvious that they do indeed deserve the "Not Open Source" tag. I have undone my undoings on the wiki and everything should be back to the way it was earlier today. The great thing about the wiki is that it is easy to go back if it turns out we don't like what we changed.

In regard to your questions, here are my personal opinions:

1. In general, I think we want on the wiki as many designs as possible. The more the merrier. (Unless this costs us dearly in terms of operating speed, server space, infrastructure strain, etc.) However, on high profile pages like RepRap Options we should have only a short list of the best and most interesting machines. How to precisely define "best and most interesting" I do not know...

2. I think this definition is excellent.

3. Unfortunately verifying a design takes work, and if one is lazy about it (as I was today) then we end up with designs that don't have all the necessary information public. Verifying designs is exactly the sort of thing an engaged wiki-editing community could do, if we could get enough people interested.

4. STL files are the bare minimum, but if we are using the bare minimum definition of open source (see my opinion to question 2, above) then STL files are acceptable.

another question is what information is enough to call a electronics board to be called open source . Some boards ( RADDS ) just share their schematic as a PDF and call themselves open source.. Aren't source files/gerber files necessary to call yourself open source?

Quote
ekaggrat
another question is what information is enough to call a electronics board to be called open source . Some boards ( RADDS ) just share their schematic as a PDF and call themselves open source.. Aren't source files/gerber files necessary to call yourself open source?

I take the implication of opensource to be very literal, if you produce something and claim it's opensource then realistically the sources should be released and made available, seems logical enough to me.

stl files are a resulting file produced by software based on some kind of source file eg scad , solidworks, freecad ... etc

pdf export files of schematics and pcb layouts aren't really a source as you cannot reasonably expect to produce a board from it nor can you produce a variant of it from those pdf files easily.

this kind of behaviour has regrettably become quite common on the wiki and quite a few people have gotten away with it for quite some time, if people want to do this kind of thing that is fine, they should just not be surprised
when content gets removed,

to me the RADDS page on the wiki looks like just another advertising page for a closed source product claiming to the open and there are quite a few of these pages,

there is discussion about wether we are obligated to keep these pages around or if we should in the interests of creating and maintaining high quality content.

Quote
MattMoses
1. In general, I think we want on the wiki as many designs as possible. The more the merrier. (Unless this costs us dearly in terms of operating speed, server space, infrastructure strain, etc.) However, on high profile pages like RepRap Options we should have only a short list of the best and most interesting machines. How to precisely define "best and most interesting" I do not know...

2. I think this definition is excellent.

3. Unfortunately verifying a design takes work, and if one is lazy about it (as I was today) then we end up with designs that don't have all the necessary information public. Verifying designs is exactly the sort of thing an engaged wiki-editing community could do, if we could get enough people interested.

4. STL files are the bare minimum, but if we are using the bare minimum definition of open source (see my opinion to question 2, above) then STL files are acceptable.

1 - I agree and disagree , there is nothing wrong with as many designs as possible, however it's the quality of the content that is under question.
looking at the reprap options that the page has fallen victim to those whose only real interest is advertising their printer and where it can be bought

3 - it's actually not that difficult, you don't necessarily have to build it, just verify that usable valid sources exist and are accessible in a obvious way and the build notes and instructions are somewhat available

4 - stl files really unless made directly in blender or some mesh editor, are really a resulting file generated from the source , i would encourage the release of actual sources as well as stl files

strictly speaking some design variants which claim to be GPL but not releasing the modified sources are really breaching the license and if anything we should be putting a bit of pressure on the authors to release the sources and maybe remove their design from the wiki if they don't comply within a reasonable amount of time, eg Some of the gpl printer designs have been on the wiki for up to a year and being sold for a similar amount of time but no sources have been released and with no obvious plans to do it any time soon

---

-=( blog )=- -=( thingiverse )=- -=( 3Dindustries )=- -=( Aluhotend - mostly metal hotend)=--=( Facebook )=-

Glad to see this works out so well.

I take that this NotOpenSource template needs improvement. At least one additional field for the reason, like "design of sheet parts missing", "no sufficient vitamins description" or similar.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Quote
thejollygrimreaper
pdf export files of schematics and pcb layouts aren't really a source as you cannot reasonably expect to produce a board from it nor can you produce a variant of it from those pdf files easily.

this kind of behaviour has regrettably become quite common on the wiki and quite a few people have gotten away with it for quite some time, if people want to do this kind of thing that is fine, they should just not be surprised
when content gets removed,

to me the RADDS page on the wiki looks like just another advertising page for a closed source product claiming to the open and there are quite a few of these pages,

there is discussion about wether we are obligated to keep these pages around or if we should in the interests of creating and maintaining high quality content.

I could make arguments the opposite way. If we were in the software world, I don't think it would be even questioned if I released source code for a library or even complete program under an open source license but did not provide a binary object, installer, etc. It would be up to the end user to determine how to properly compile it and make use of it complying with the terms of the license. Providing the PDF of the schematic I see as the minimum necessary to have it "open". Yeah it's not in a very usable format, just as providing a source code as a jpeg would be of limited usefulness.

In the case of the RADDS design, it's already covered under a CC-NC-SA license. Any derivative could not be of a commercial nature so a complete open "source" including layout would be of limited value financially. Anyone who is going to modify the design is likely going to know Eagle (or similar) and could duplicate the layout and perform their own routing. But I don't think requiring the layout is necessary just to call it "open source".

Also, if you want to go down the road of requiring a particular level of openness, does that mean we'll start excluding non-open source items in favor of the open source alternatives? Now I'm not talking just about designs, but information in general. So talking about or providing information netfabb Studio? Or Kisslicer, SolidWorks or Eagle? All of those are closed source programs that may have free versions but also have commercial offerings that have a financial motivation.

My $.02 is that if the information is there for real informational purposes, it's ok regardless if it's fully open or closed or somewhere in between. If it's just there advertising the cdru MakerPrusaMax90Pro model that is available exclusively on my website and nothing else, then toss it. Whatever information is on there though should be clearly indicated what type of license that it's covered under as well as any notes if applicable if what is available is not compliant with what the license would indicate.

Quote
Traumflug
I take that this NotOpenSource template needs improvement.

Done: [reprap.org]

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Quote
cdru
If we were in the software world, I don't think it would be even questioned if I released source code for a library or even complete program under an open source license but did not provide a binary object, installer, etc. It would be up to the end user to determine how to properly compile it and make use of it

I'd compare PDFs/JPEGs more with binaries. Software coming as binary only is certainly not considered to be Open Source.

Quote
cdru
In the case of the RADDS design, it's already covered under a CC-NC-SA license. Any derivative could not be of a commercial nature so a complete open "source" including layout would be of limited value financially.

Does financial value matter for open source? Does it matter for RepRap? With open source software it doesn't, at least not directly. People do business by using open source, not by creating or distributing it. I could imagine similar businesses with open source hardware.

That said, I currently try to keep this licence discussion out, because there's currently no golden solution. As i3 Acrylic shows, tacking "GPL" somewhere on the web page doesn't make it automatically good open source citizens. I'd consider modifiable sources to be more important than licence restrictions, because if you make your copy at home, licences don't matter at all, much less -NC clauses.

Quote
cdru
Also, if you want to go down the road of requiring a particular level of openness, does that mean we'll start excluding non-open source items in favor of the open source alternatives? Now I'm not talking just about designs, but information in general. So talking about or providing information netfabb Studio? Or Kisslicer, SolidWorks or Eagle?

I certainly see a road towards favoring open source over closed source. This doesn't neccessarily mean closed source designs are to be tossed entirely, but they shouldn't be prominent and also not recommended. Closed source designs are industry items or, in RepRap parlance, "vitamins". Even when they're entire printers or complete software solutions.

For example, gEDA and KiCad are easily on par with Eagle, so there's zero point in recommending Eagle. In fact I have no idea how some open hardware evangelists can seriously do such recommendations.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Quote
cdru

Quote
thejollygrimreaper
pdf export files of schematics and pcb layouts aren't really a source as you cannot reasonably expect to produce a board from it nor can you produce a variant of it from those pdf files easily.

this kind of behaviour has regrettably become quite common on the wiki and quite a few people have gotten away with it for quite some time, if people want to do this kind of thing that is fine, they should just not be surprised
when content gets removed,

to me the RADDS page on the wiki looks like just another advertising page for a closed source product claiming to the open and there are quite a few of these pages,

there is discussion about wether we are obligated to keep these pages around or if we should in the interests of creating and maintaining high quality content.

I could make arguments the opposite way. If we were in the software world, I don't think it would be even questioned if I released source code for a library or even complete program under an open source license but did not provide a binary object, installer, etc. It would be up to the end user to determine how to properly compile it and make use of it complying with the terms of the license. Providing the PDF of the schematic I see as the minimum necessary to have it "open". Yeah it's not in a very usable format, just as providing a source code as a jpeg would be of limited usefulness.

In the case of the RADDS design, it's already covered under a CC-NC-SA license. Any derivative could not be of a commercial nature so a complete open "source" including layout would be of limited value financially. Anyone who is going to modify the design is likely going to know Eagle (or similar) and could duplicate the layout and perform their own routing. But I don't think requiring the layout is necessary just to call it "open source".

Also, if you want to go down the road of requiring a particular level of openness, does that mean we'll start excluding non-open source items in favor of the open source alternatives? Now I'm not talking just about designs, but information in general. So talking about or providing information netfabb Studio? Or Kisslicer, SolidWorks or Eagle? All of those are closed source programs that may have free versions but also have commercial offerings that have a financial motivation.

My $.02 is that if the information is there for real informational purposes, it's ok regardless if it's fully open or closed or somewhere in between. If it's just there advertising the cdru MakerPrusaMax90Pro model that is available exclusively on my website and nothing else, then toss it. Whatever information is on there though should be clearly indicated what type of license that it's covered under as well as any notes if applicable if what is available is not compliant with what the license would indicate.

anyone could make arguments the opposite way and quite rightly have for the better half of the last decade however this has left us with a lot of problems as a community,

if the RADDS developers want to use that license and release their sources in those formats then all power to them, however anyone wishing to contribute to the project has to goto a lot of trouble to do so which for some is probably too much especially if they have the practically redraw the source code, it's ultimately up to them, and it just hinders the development making it less likely to be widely adopted.

from where some of us are sitting the only reason some of these projects have any sort of opensource license on them is for marketing reasons, and the sources that are released is just for show as they don't want people making their own as they are just interested in selling something. it's the kind of thing that has to be assessed case by case.

the only road to requiring openness that we'll go down is asking people to provide usable sources to projects and not just export formats, we have a situation now where nearly 50% of the printers on the wiki are released as stl files and most of the missing critical parts , a good number of that 50 percent have had their wiki pages built to the point that there is basic information provided about the printer and where to get it but outside of that information is scarce and incomplete , some of these pages have been like this for many months while the associated printer has been up for sale the whole time. in short a lot of these designs appear to exist for nothing else than making money and the wiki is used for advertising, it is highly likely we'll just get rid of the pages in the future.

---

-=( blog )=- -=( thingiverse )=- -=( 3Dindustries )=- -=( Aluhotend - mostly metal hotend)=--=( Facebook )=-

So , If I take reference of the RADDS design files, change pin connections, designed my own PCB with routing and made my firmware changes in repetier or marlin, I should be able to make it as a commercial board right?