AMF, really open and ready for adoption?

[Daid]

Hi,

I'm David. Developer of Cura, the OpenSource 3D printing software package made for Ultimaker and other RepRap 3D printers.

I've added some very basic AMF loading to the latest release of Cura, build from the examples and information on wikipedia. But as hardly anything exports AMF yet, there isn't any real added value for the users.

Now, with the next big release of Cura, I'm planning to add a much easier interface to make plates of prints, or constellations as they are called in AMF. So it would be interesting to save these as AMF, as AMF supports what I need to properly store all information. (Without inventing something odd as the Makerbot .thing format)

As I understood that AMF is an Open Standard (according to wikipedia, and most sources)
However, when I went to look for information in more detail, I hit a paywall or a dead-link wall. Wikipedia points to http://www.astm.org/Standards/F2915.htm, which asks $47 for the standard.
Wikipedia also links to http://enterprise.astm.org/filtrexx40.cgi?+REDLINE_PAGES/F2915.htm but that download link isn't working.

So I'm questioning how "open" this open standard really is. And I'm starting to wonder if I really should go trough the trouble at all.

[Hod Lipson]

Hi Daid,

 

First, thanks for implementing AMF (even if basically). A number of larger companies (both CAD and equipment manufacturers) have assured me implementations are in the works.

If any other standards are a reference, adoption takes time.

 

As for openness of the standard: We're trying to make it as open as possible: Anyone can influence the development of the standard; Anyone can vote; The free Wikipedia entry has all the important information, and the reference implementation is also open-source and license-free.  All done by volunteers.

 

Yes, Unfortunately ASTM does charge $47 for the formal spec document. Initially I was unhappy about that too, until I realized that it’s probably better than the alternative, of having ASTM be sponsored by donations from large corporations. This way (and that’s true for all standards), the standard development is supported and influenced equally by all of its stakeholders.

 

For your purpose you could probably do with the Wikipedia article and the reference code just fine. If you have specific questions not covered by the Wikipedia article, please feel free to contact me directly and I will add it to the article for your benefit and the benefit of others.

 

Hope this helps!

--hod

 

Hod Lipson

Associate Prof. of Mechanical & Aerospace Engineering and Computing & Information Science

Cornell University, 242 Upson Hall, Ithaca NY 14853, USA

Office: (607) 255 1686 Lab: (607) 254 8940 Fax: (607) 255 1222

Email: Hod.L...@cornell.edu

Web: http://www.mae.cornell.edu/lipson

Administrative Assistant:  Craig Ryan  cd...@cornell.edu

Calendar: http://www.mae.cornell.edu/lipson/calendar.htm

--

You received this message because you are subscribed to the Google Groups "STL 2.0" group.

To unsubscribe from this group and stop receiving emails from it, send an email to stl2+uns...@googlegroups.com.

To post to this group, send email to st...@googlegroups.com.

Visit this group at http://groups.google.com/group/stl2?hl=en.

For more options, visit https://groups.google.com/groups/opt_out.

 

 

 

[BobC]

Unfortunately, "open standard" has different connotation to "open source". Rarely is the text of a standard copylefted.

I was challenged by Hod's statement to think of a standard that is free of charge but does not have corporate sponsorship, but none spring to mind. I've come to accept the practicality that standards are IP that I have to pay for, in common with books, music, software etc.

If someone wants to create an alternative open source (copyleft) standard that would be great, but in the absence of that the best option may be to seek donations for open source projects. I am sure people would be prepared to contribute, myself included. I appreciate this doesn't address the principle of free access though.

[Reinoud Zandijk]

On Sat, Mar 30, 2013 at 12:53:00PM -0700, BobC wrote:
> Unfortunately, "open standard" has different connotation to "open source".
> Rarely is the text of a standard copylefted.
>

> If someone wants to create an alternative open source (copyleft) standard
> that would be great, but in the absence of that the best option may be to
> seek donations for open source projects. I am sure people would be prepared
> to contribute, myself included. I appreciate this doesn't address the
> principle of free access though.

There is, and thats the Wikipedia entry. If all information is essentially
there, then its a copyleft document of the standard. It might even become the
formal declarative one.

Reinoud

[Matthew Peters]

If you guys want this to be globally used, having a fee-driven way of getting at the specs isn't the way to do it. There's a lot of people who are getting into this play field (that of Additive Manufacturing) who are coming from the Open Source mentality. In fact, having a fee-driven spec method is very counterproductive as it becomes an elitist way of guaranteeing that only companies willing to pay will adopt the spec.

Hod Lipson, you're view that either having the specification paid for rather than being corporate sponsored is incorrect. There is nothing that required you to have only those two methods. I'd like you to back up your statement.

I'm another developer similar to Daid who will not be receiving any form of payment for the software I'm working on beyond donations. I'm not interested in paying for something that will not give me any return on investment and it really makes me suspect that the specification is entirely run by volunteers if there's payment required to view it. It's in my own interest now to build a new spec that I will release fully documented on RepRap Wiki as well as Wikipedia that will fulfill the same role as AMF but without the fee required to view the specification.

Thanks,

      Matthew

--

[Hod Lipson]

 

This spec is ALREADY on wikipedia. And reference code is ALREADY open source and free.

That's why we put it there - so you can access it freely. If there is any question you have pls let me know and I will do my best to answer.

 

I probably can't just distribute the formal document, but if anyone wants to help edit or check it, pls let me know and I will send them the current draft directly.

 

However, you are right that since this standard is likely to be used substantially by non-profit users, perhaps we can work to convince ASTM to release this freely for non-profit use.

 

I'd be happy lead this. Anyone else supporting?

 

--hod

[Hod Lipson]

All:

 

I contacted ASTM to formally request that they officially release this standard as open-access.

I will let you know how it goes.

 

Meanwhile, if there are ASTM officers on this list, perhaps you can see what you can do also.

 

--hod

 

 

 

Hod Lipson

Associate Prof. of Mechanical & Aerospace Engineering and Computing & Information Science

Cornell University, 242 Upson Hall, Ithaca NY 14853, USA

Office: (607) 255 1686 Lab: (607) 254 8940 Fax: (607) 255 1222

Email: Hod.L...@cornell.edu

Web: http://www.mae.cornell.edu/lipson

Administrative Assistant: Craig Ryan, cd...@cornell.edu  (607) 255-0992, Upson 258

Calendar: http://www.mae.cornell.edu/lipson/calendar.htm

 

From: Hod Lipson
Sent: Saturday, March 30, 2013 10:14 PM
To: st...@googlegroups.com
Subject: RE: AMF, really open and ready for adoption?

 

 

This spec is ALREADY on wikipedia. And reference code is ALREADY open source and free.

That's why we put it there - so you can access it freely. If there is any question you have pls let me know and I will do my best to answer.

 

I probably can't just distribute the formal document, but if anyone wants to help edit or check it, pls let me know and I will send them the current draft directly.

 

However, you are right that since this standard is likely to be used substantially by non-profit users, perhaps we can work to convince ASTM to release this freely for non-profit use.

 

I'd be happy lead this. Anyone else supporting?

 

--hod

 

 

 

From: st...@googlegroups.com [mailto:st...@googlegroups.com] On Behalf Of Matthew Peters
Sent: Saturday, March 30, 2013 5:36 PM
To: st...@googlegroups.com
Subject: Re: AMF, really open and ready for adoption?

 

If you guys want this to be globally used, having a fee-driven way of getting at the specs isn't the way to do it. There's a lot of people who are getting into this play field (that of Additive Manufacturing) who are coming from the Open Source mentality. In fact, having a fee-driven spec method is very counterproductive as it becomes an elitist way of guaranteeing that only companies willing to pay will adopt the spec.

[Daid]

Only the explanation of the spec is on wikipedia. Details are missing. (just an example, what are the possible values of the "unit" attribute?). The open-source implementation is just 1 interpretation of the spec, there could be things missing ignored or wider then the spec specified.

I also do not know if there are parts of the spec that are "required" and parts that are "optional". If numbers are allowed to be stored with scientific notation. If there are limits on the <rx> <ry> <rz> values in constellations. There are all kinds of details that can cause problems when we want to exchange AMF files with other programs. Without the formal spec we are running into a minefield blindfolded.

To post to this group, send email to s...@googlegroups.com.

Visit this group at http://groups.google.com/group/stl2?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Charles Overy]

Hello,

Mathew and Daid,
I am happy to purchase a copy for you of the spec if it will be useful to
your work. Please contact me directly and I will make it happen. If the
$50 is a problem for small developers, I would challenge others on this
list to sponsor the distribution. A standard is only as useful as its
adoption.

We are not a large company but having wrangled stls for many many years,
and having participated in a small way on the standard, I think it is an
extremely important I agree sponsorship does not solve the
underlying philosophical problem of copyright but the history of STL2 has
been of well thought out compromise. Before that many people tried to
develop various flavors of new file formats that were corporate sponsored
or sponsored open formats.

Mathew and Daid, Thank you for taking the time to implement AMF.

Charles

LGM
cwho at lgm3d.com

[Hod Lipson]

Thanks Charles.

Meanwhile, I have asked ASTM to see if they will agree to make this open access.

--hod

-----Original Message-----
From: st...@googlegroups.com [mailto:st...@googlegroups.com] On Behalf Of Charles Overy
Sent: Monday, April 01, 2013 12:22 PM
To: st...@googlegroups.com
Subject: Re: AMF, really open and ready for adoption?

[Rob Gilson]

As the developer of the P3D Javascript AMF parser implementation, I would like to lend my support to both Matthew Peters and Daid requests for a freely available AMF standard.

As a volunteer myself I do not feel that I should have to pay a third party's fee to contribute to an open project.

Hod, as I understand it this is a volunteer led standard, if the development of the standard is being done voluntarily and without the financial support of ASTM then what is the value added of ASTM with regards to the AMF standard? What are our sponsorship needs?

- Rob

[Rosen, David W]

Hello,

It seems that the AMF standard is at an interesting cross-roads.  By way of introduction, I am the F-42 Design Subcommittee Chair, the group that oversees the development of AMF.  This means that I am the lead volunteer for the design subcommittee.  I do not work for ASTM and cannot speak for them.  But, I have been in meetings with ASTM staff where we have discussed similar requests for open access to some standards.

ASTM operates a pretty lean organization and they do not sponsor the development of standards in the sense that they do not provide funding to groups who want to develop standards.  The advantage of developing standards under the ASTM umbrella is that ASTM is a well known, well respected, international organization.  

For the AM industry, particularly users of AM technologies and OEMs that want to include AM-fabricated parts in their products, the development of standards has been a top priority for many years.  A variety of paths converged several years ago when the ASTM F-42 committee was formed.  At that time, the idea of a STL2/AMF was proposed and incorporated into the set of AM-related activities.  Everything made a lot of sense at the time and almost everyone involved was very pleased with the AMF that resulted.  Having the ASTM "stamp of approval" gives AMF a certain stature in the industry and has helped to get AMF into some commercial products.

Back to my opening comment about the cross-roads.  The open software community is an important constituency and is likely to continue to grow in importance.  How important is the ASTM "stamp of approval" to this community?  I think that the commercial industry and academia will continue to find ASTM involvement of significant value for the foreseeable future.  I do not think that ASTM will allow their AMF standard to become open/free.  Their business model (this is my opinion) is to make a little bit of money from many people (annual membership is $75, most standards cost $30-50), enabling standards to remain very low cost.  As a consequence, they typically do not give away anything.  

So, should the open software community stick with the ASTM-affiliated AMF activity?  Should this community abandon ASTM AMF and develop a replacement, open format?  Should the AMF activity leave ASTM and become an open standard?  Or, are other paths worth developing, exploring, and discussing with ASTM?  There are serious implications for each of these options.

I don't think either Hod or I want to dictate to this group.  It would be great to reach some mutually beneficial agreement that balances the concerns of the open software community with those of all other involved communities.  I encourage more discussion among the group and we will see if a consensus emerges.

Dave Rosen

david...@me.gatech.edu

---

From: "Rob Gilson" <thatot...@gmail.com>
To: st...@googlegroups.com
Sent: Monday, April 1, 2013 5:50:49 PM

[Leonid Raiz]

It seems that ASTM is driven by bureaucrats who live in the 19th century. How long this group is going to put up with this? I suspect that intellectual property of AMF standard belongs not to ASTM but to this group which is self-governed. Can someone tell me what bad things would happen (besides ASTM loosing their $$) if the current version of the draft is attached to a posting to this group?

[Hod Lipson]

Here is an alternative idea: We pay ASTM to make AMF an open-access standard for the next three years. The payment should be reasonable compensation for lost future income.

 

The idea is borrowed from academic journals: These days most publishers offer authors the option to pay the publisher an upfront sum in exchange for making a paper open-access -  in a way compensating the publisher for lost future income.

 

I would suggest that ASTM would calculate reasonable expected income from this standard in the next, say, three years, and we can consider this. If the amount is reasonable, we will be able to raise the funds.

 

--hod

[Leonid Raiz]

I do not quite follow the publisher's analogy. In case of academic journals authors receive from publishers the benefit of publishing. Our group may use Google infrastructure to publish drafts for free. What justifies paying ASTM 'protection' fee and how much is it worth?

- LR 

Sent from my iPhone

[David K. Leigh]

Leonid,

 

I appreciate your desire to have an open source software standard.  I also understand the benefit of open source.  I also understand that most people who work in the realm of open source understand that the “in-kind” contribution and cost is carried on the shoulders of volunteer time and an “unselfish” desire to benefit the community.  The reality is that in an effort to evangelize the additive manufacturing world, we have taken to groups like SME and ASTM to help in developing education, curriculum, and standards.  These organizations have an associated cost and it’s the price we as a community have chosen to bare.  I would council you that a continued effort to denigrate some of the organizations and individuals with whom we have chosen to partner will not help your cause.  We get your message and will work to that end.

 

David

[Hod Lipson]

I realize that different sectors of the market have different values and that we must cater to ALL stakeholders.

Therefore, we WILL find a way to make AMF freely available to you and any other non-profit and/or individual.

Please bear with us as we look for a solution - we do not want to abandon ASTM/ISO.

 

The publisher analogy is as follows:

 

Anyone can publish a paper online, but being published by a recognized, high profile journal (e.g. Science, Nature) with a selective review process adds credibility to the paper (at least for academia).

 

Similarly,

 

Anyone can publish a standard online, but being published by a recognized, high profile standardization organization (e.g. ASTM, ISO) with a systematic standard development process adds credibility to the standard (at least for government and industry).

 

Most publishers today allow authors to pay to make their article open-access. That benefits both the publisher and the community. I am sure ASTM will recognize the benefit of doing this with standards.

 

--hod

 

Hod Lipson

Associate Prof. of Mechanical & Aerospace Engineering and Computing & Information Science

Cornell University, 242 Upson Hall, Ithaca NY 14853, USA

Office: (607) 255 1686 Lab: (607) 254 8940 Fax: (607) 255 1222

Email: Hod.L...@cornell.edu

Web: http://www.mae.cornell.edu/lipson

Administrative Assistant: Craig Ryan, cd...@cornell.edu  (607) 255-0992, Upson 258

Calendar: http://www.mae.cornell.edu/lipson/calendar.htm

 

[Daid]

Hey Charles,

That's not needed. I work at Ultimaker. While we are only a small company, we could shell out $50 for the standard. So it's not really a money issue for me. It's more an moral/principle issue. We are pretty keen on keeping our software open, and have as much interchangeability with other pieces of software as we possibly can. Having free open standards helps for this.

[Jacob Barhak]

Hi All,

There are three main issues here: price, credibility, openness. 

Openness allows scrutiny which improves quality and therefore credibility.

The world is going that way. Here is a link to a government document that supports openness for federally funded research:

http://m.whitehouse.gov/sites/default/files/microsites/ostp/ostp_public_access_memo_2013.pdf

I can also testify from experience that the NIH asks for papers funded by it to be available for open access. 

In the past it was harder to have truly open publications since those were on paper and not widely available as today. Therefore determining credibility by a team of experts was part of the publication process and involved expenses that someone had to pay. Therefore the price issue is traditionally associated with publication and conflicts with openness. 

Today there are different mechanisms that allow truly open publication without price for the user or authors and the online publisher gets the money by other ways. 

These mechanisms just leave elements such as proof and advertising at the hands of the authors. Yet if enough users access the information or provide good feedback, it gives credibility. This is a modern alternative to the expert selection system that can be biased in some cases. In the new system, the end user is considered an expert.

In the case of AMF, I am not sure what elements are crucial to success. Probably advertisement is crucial at this point. Once there is wide adoption, the problem becomes maintenance and control of the standard. In both cases, if the users agree  to take part in these activities voluntarily, then big organizations that ask for a fee are not required to support the standard - the users themselves form the committee. 

For now it seems AMF is playing in both playing fields by having Wikipedia and a discussion group and also ASTM presence. This seems reasonable to keep everyone happy. This gives ways to get things freely and advancing fast by contacting the community, and there are also ways to go through traditional channels that are more accepted by older generations. 

Perhaps it is a good idea to keep this duality for a while. 

Last word about credibility. The fact that AMF has this open discussion and mailing list where discussions take place openly gives it good points for credibility. Most systems today are typically complex enough to have issues. For example, your operating system on your machine or smartphone requires constant updates. Standards are the same and should be updated with time to cope with new advancements. They should also have an issue list like a bug list in software. Having an active vibrant discussion such as this one only improves the credibility of the standard. 

I hope this is helpful to the discussion. 

        Jacob

--

Jacob Barhak Ph.D.

http://sites.google.com/site/jacobbarhak/

Sent from my iPhone

--

[Rene K. Mueller]

On Friday, March 29, 2013 1:31:48 PM UTC+1, Daid wrote:
> Hi,
>
> I'm David. Developer of Cura, the OpenSource 3D printing software package made for Ultimaker and other RepRap 3D printers.

Hi David,

I'm coding at http://OpenJSCAD.org, and like to add AMF export also (and then AMF import I hope). I also can only implement based on the information I have, and the Wikipedia article seems OK, but an article at wiki.reprap.org would be prefered covering the actual standard in *full*. Since OpenJSCAD is based on JavaScript, one can define properties (color, material etc) for each polygon and polyhedron; and I am about to add this feature now.

AMF has to be open and published for free to be really adapted, and yes, it's not about USD 50 or so, but the principle to pay for a standard to implement; I am just glad HTML and all that we got for free . . . go figure.

Rene K. Mueller
http://OpenJSCAD.org/
http://reprap.org/wiki/User:Spiritdude

[Hod Lipson]

A quick update:

ASTM is open to the idea of making the AMF standard open-access. They will be discussing details internally this week and will review the options with us soon.

I will keep you posted

--hod


-----Original Message-----
From: st...@googlegroups.com [mailto:st...@googlegroups.com] On Behalf Of Rene K. Mueller
Sent: Thursday, April 04, 2013 1:36 PM
To: st...@googlegroups.com
Subject: Re: AMF, really open and ready for adoption?