the anomaly on the AMF subcommittee ballot

[Jenny Buck]

All, Hod has verified for me that I was his only negative vote on the
recent ballot. Worse than that, I know it's a dissenting vote from
the peanut gallery (if that doesn’t translate well, they were the
cheapest, rowdiest, least cultured and/or educated seats in old
vaudeville theatre), as I'm a fringe player at ASTM and an external
element to this direct discussion. I greatly appreciated Hod’s
request, though, that I share those concerns with you all for
discussion +/-; I am well aware you all know this area one heck of a
lot better than I do. I know engineering, product design,
manufacturing, & issues with nascent technologies, but I do NOT know
software. Below with a few followup edits, is the comment that went
in with my negative response. There is one additional piece that
wasn’t included in that comment. It’s that I know Hod’s thoughts and
strategies are in the right place on what he’s doing & how he’s going
about it; I think it’s crucial to remember, though, that this is a
volunteer activity, & there are no “for life” positions. Although
it’s easy to say the same faces will be sitting around the table one,
five or ten years from now, it’s known that others will arrive, & it’s
just as likely that the people and possibly the perspectives could be
very different in time than we envision they “should” be from any
vantage point today. It is harder to predict progressions over
timelines for an entity like ASTM than it would be for a university, a
company or a personal project. Apologies for the verbosity; a lot of
this comes back to philosophy vs. short sweet technical needs.

Best,
j

[...]
Many of the options envisioned for the future of this document, such
as geometric tolerances, may turn out to be crucial to AM’s growth;
although I’m disappointed not to see many of these yet in this
document, I know designers and industry will call out loudly enough
for them that there will likely be a pull-through effect soon on
them. I think this document is choosing a very risky path pursuing a
singular as yet not fully tested form of representation, though,
regardless of whether future provisions “may” provide other
representational methods. That by no means signifies I have anything
against meshes, triangles, or the curved triangles in particular – as
an engineer, I’m incredibly intrigued by the curved triangles, and am
going to carefully word this comment not to attack them or attempt to
strike them from the document; my concerns are external to my interest
in them.

I’m concerned the jump to what appears, whether we intend for it to or
not, to be a preferred representational approach (in particular the
optional curved version of the triangles) without basing it on a
deeply proven history or having researched and comparatively tested at
least a wide swath of options and possibilities has issues:
~ We’re risking ASTM’s credibility, possibly even somewhere down the
line making ASTM’s input obsolete if we can’t respond to problems that
develop with it in a timely manner, which will affect us on the next
iteration of this standard, possibly more generally within this field
~ Representational development takes more work than determining that
designers/ industry/ the public is ready for a yes/no option like
whether texture to be added; by spec’ing the base document as we
have, with any mention of other possibilities all the way back in non-
mandatory appendices, not deeply embedded in the philosophy and
intention of the document we are potentially generating a setback to
innovation in this area and for the industry at a time when it’s
growth may be critical to the industry as a whole
~ It should be noted that the setback to innovation could one day
potentially even lead to the expenses for undertaking such development
being more conducive to a non-open source approach to make it
financially worthwhile even to undertake. As Hod has clearly
articulated, even the current curved triangle is impossible to
realistically test today without getting it into this document and
getting it out there for people to start working with. The route will
be harder, not easier, for subsequent new representational
approaches.

I do think this is a wildly important moment in manufacturing history,
& these decisions could easily have more implications that we would
envision. It may not be our lifetimes, but very likely by those of
the next generation, AM is potentially going to have one of the widest
impacts on humanity of any recent technology besides the computer, and
it potentially may be greater. We don’t have any way of knowing what
is optimal on the representational side of AM files yet – it’s just
too early. Although I’m aware this isn’t necessarily in the ASTM
guidelines, I think our role is to nurture exploration & possibilities
– particularly in such an unusually varied field with so many
unrealized directions it may soon head. We need to pack the cards
properly to get the greatest output from what is probably currently
best described as a technological primordial goo. I’m aware there have
been concerns about getting the “everything but the kitchen sink”
version of this standard approved; strangely, I think it’s even more
critical in a strategic sense for the industry not to have
possibilities get cut back or even cut off because of the creation of
the document. There is some piece of stepping back & looking at the
human benefit first, then the nurturing of the growth of the industry,
that I’m not sure we’re quite nailing on the head yet.

I feel it’s critical this document DOES, not “MAY,” support multiple
representations thoroughly FROM ITS INCEPTION, particularly since we
are all well aware that the optional curved form of the triangles is
not yet by any means tested. Regardless of what the format
requirements are for ASTM documents, I think the goal of this document
continuing to grow to support new approaches in this industry should
also even be written into the opening paragraphs. Multiple
representational routes signify that we’re providing a standard
outlining options to be explored, chosen between, and optimized for
various specific uses, NOT that we’ve picked what we believe (we have
no way of knowing yet) will work best. Providing one singular
approach by default IS signifying, whether we choose for it to or not,
that we believe we have identified that “best choice.” If we are
truly to put forward a “best choice” we OWE it to industry and to the
public to have FULLY TESTED and PROVEN that the mesh/ triangles/
curved triangles really ARE that best choice, and to have a thorough
grasp of their limitations & pitfalls, something that is impossible in
any foreseeable future without the recursive predicament of having put
them out there through a document like this to get them tested... I
DO have a problem with placing untested curved triangles out into the
world singularly without substantial further proof behind the choice;
I have far less concerns with placing untested curved triangles out
into the world if they are one option within a series. That approach
leaves it in the hands of the user to identify each options benefits
and limitations, & choose between them. Philosophically, practically
and strategically, those additional options in my mind need to be
clarified and explorable, though, from the start.