ArchivesSyndicate This Blog |
After Darwin: Should Mendel be a specific 3D printer or a technology toolbox?Friday, January 16. 2009In which your narrator suspects that we are using the wrong model of
evolution...
Darwin has been a roaring success by virtually any measure you'd care to name. In April of 2008 we had exactly five reprap and repstrap printers more or less operating. Last month, Dr. Bowyer estimated that several thousand were either operating or under construction. That's not exponential growth. It's hyperbolic. I, personally, have no doubt that in 5-10 years Dr. Bowyer is going to find himself on the Queen's annual Honor's List. Mind, in my opinion he richly deserves it. That said, I wanted to share a few ponderings that I've been nursing on the nature of change in the reprap movement for the past few months. Basically, Reprap has two models of evolution going on in parallel. In the core team, we see what I will call a punctuated animal model of evolution while the larger reprap community is a very vibrant bacterial model. Back at the end of 2006 and the beginning of 2007, the Reprap Core Team had several prototypes for a first release Reprap machine; Da Witch in New Zealand,  A.R.N.I.E. at Bath  and Tommelise 1.0 in the US.  By March of 2007, Darwin, which was to become the first official release reprap machine had grown out of the A.R.N.I.E. prototype...
 Basically, Darwin morphed into a fully industrial product. It began with the controller boards being outsourced for production by the Reprap foundation and has culminated with a shippable kit purchasable for US$1,100 {shipping not included} requiring little more than the sort of assembly you'd be expected to apply to something bought from Ikea. What is getting built out there in its thousands, to use Dr. Bowyer's metaphor, is 100% vitamins - 0% replicated parts.
A few tentative attempts have been made towards coming up with a second generation mainstream reprap machine. The most notable effort in this regard has been undertaken by eD at Bath university.
We've found ourselves increasingly sliding towards designing with industrialisation in mind from the onset. Instead of designing simpler electronics that would be easy to cobble together, we have been designing boards with parts like surface-mount chips friendly to outsourcing and short production runs. Design, for absolutely the best of reasons, has been getting more and more centralised and unified. I think that this has been happening in large measure because of the model of evolutionary development that the core Reprap team have been using. While we have been giving lip service to Darwin's original theory of natural selection, in practice we've been using Richard Goldschmidt's Saltation theory in which very large changes happen from one generation to the next.
The only place that sort of thing can economically happen is on a factory production line variously defined. That's why there are factory production lines and that's why it makes sense to make something at one point on the planet and then ship it everywhere.
I think, and this is again my own, very personal opinion, is that the core team needs not so much to dictate the technologies used as manage and facilitate the interstices/interfaces between the different parts and systems while letting the community at large provide the microenvironment-specific solutions to problems.
If we do that, as I see it, we have a chance of creating a prototyping/production capability that truly can diffuse into the furthest reaches of the planet. As it stands now, we're developing technology that is easy to make in a factory, put in a box and ship worldwide. In my opinion, what we have slipped into doing now is the antithesis of what we set out to do so many months ago.
Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
Hi Forrest, I think there is a lot of truth in what you are saying. However I look at these RepStraps as "seeds".
Before you can harvest a crop, you have to plant your seeds. If we are successful, we'll be making genuine offspring AND will have to plant new seeds all the time.
Though, it would be preferable to make 'mutations' more like 'modules', so all techniques can be combined. In order what materials are most available.
Any ideas in how to do that? (Is it a matter of better wiki-style documentation?)
Hi Batist! Until very recently, I felt that way, too. The problem with the "seeds" that are being developed now is that they are less and less accessible to developers. They're a bit like the genetically engineered crops that companies like Monsanto are developing these days. You can grow a nice crop, but the seeds that the crops produce are sterile by intent. I don't think our guys mean to do that, but it's increasingly working out that way.
I have been following the project for sometime, however, have been hesitant to start building one because of the various directions the project has taken. It seems like every time I read the blogs or forums I read about something new that has caused me to take pause and see how it plays out. But in the end, whatever I was reading never seems to have made it past that blog/forum post.
It would be nice, from the prospective of a new builder, to have some place to see all of the options. Then I could pick and choose the different subsystems that meet my needs.
That's more or less what I've proposed. It's something that we've very little experience in making happen, though.
Hi Forrest,
I can see why you're concerned. It is very possible that the darwin may just become an opensource hardware project like the arduino, in which the convenience of having the dedicated community make designs and sell cheaply mass produced kits or pre-assembled items outweighs the desirability of per-user customization and experimentation. Unfortunately, not every user will be interested in contributing, and it will be those users who will be most interested in just getting a cheap 3d printer.
I agree with what you said, that if the core team could set up an official 'api' and 'interface' definition for each piece of your hierarchy, then people could focus on making modifications in their own area of expertise or interest, and leave the rest up to others in the community. Being able to use "RepRap compatible" hardware and software would make things much simpler, instead of having to carefully redesign the whole system when you want yours to be arduino based instead of pic based, or use a gantry robot instead of the darwin.
Having the interfaces down would allow the RepRap to evolve as a single species, instead of a cloned individual or a whole genus.
It's an inevitable consequence of the original decision to make a machine which cannot fully replicate itself, instead incorporating elements which are 'readily found' or prebundled electronics packages -- the cost of the other components are minimal, so when all of it, the bolts, electronics, and pieces of metal pipe are all offered together, why not? It saves a ton of effort and costs little more money. Until or unless reprap actually is what it promises to be -- a machine which can fully replicate itself -- it's not going to do much replicating.
I'm a big fan of free will, myself. Invoking inevitability is just an excuse for not trying.
I suppose inevitable is a bit of a strong word. I should have said that this outcome, while perhaps not what was originally desired, certainly shouldn't have been entirely surprising.
Really? Personally, I'd imagine the halving of price when one builds a reprap from cheap repstrapped parts rather building or buying a repstrap would be irresistible to most any confident in their own skills or low on funds. It's sure attractive to me, and it's going to just keep getting cheaper.
I agree with Ralith, I'm excited about the prospect of getting others involved with this project at half the cost I had to put into.
I'm somewhat in the same boat as Jeremy. I've been following the project for a while and ended up building a 3 axis robot based on a sherline mill, the 1.1 stepper controllers, a sanguino and my own ethernet board. I also run my own firmware and host software. Many of the parts I built for myself could benefit from horizontal gene transfer but lacking clear interfaces my design looks monolithic. That doesn't mean there aren't some interfaces just that the emphasis hasn't been on specifying them. Looking at your stack I think we have good interfaces in gcode, and maybe toolheads (or they are easy enough to custom fit). It would be nice to have a standard on the firmware PC communication, the connections from the controller to the robot (steppers) and maybe the toolheads. For now I'm building on my own standards but if the standards were broader it would be easier to exchange designs.
I think we need to start talking about what those interfaces need to look like.
I think the industrialization that has occurred with bits from bytes is more akin to the linux system. Almost no one uses the main linux kernel that is updated and released by linus and the developers around him. Everyone uses ubunto or some sort of distribution like that because it's much easier to use with extra's on top than the use the raw "guts" of the system. But ubunto etc is dependent on the main kernel's development with idea exchange going back and forth. Most development on the main kernel as done by epople making stuff for their own distro's and then it feeds back into the central version for all to use later. At the same time,all those distributions profit from making sure they stay as a swarm that works with interchangeable parts with the main original release.(ie my bits from bytes system I can replace any plastic part with a standard repraped part. The community develops most of the stuff, but adrian (analogy to linus) provides a single spot where everyone can agree on that all ideas should be heard for discussion)
I think the number by adrian is a bit off in terms of in the wild and actually done and complete. The idea of lots of people who aren't advertising their repraps isn't completely correct. I think the majority of people who are making machines or have them running are either involved in the forums, on a blog or is in some other way out there in the community. Getting a number of sales from zach and ian would make it all a bit clearer. Even then remember that lots of people will buy the kits but not complete it.
On the idea of info exchange I agree that currently the main and builders blogs are what has become the main method for information exchange.
I always thought that the issue with reproduction has been one of gestation time more than anything. I've had a dozen requests in my home city (which is the most isolated city on earth) for parts alone. The only reason I haven't done it is because it takes a long time to construct a machine from a kit and then make a machine. Vic's post on the reproduction stated it took him a few months to reproduce. With that rate, it's not surprising that reproduction is currently surpassed by seedling versions instead of repraped parts themselves
The Linux system is great and all, but that's not what RepRap's about; this project does not exist simply to design and offer cheap 3D printers.
Keep in mind that the amount of time required to produce a full set of parts is dropping rapidly; the only serious time barrier remaining is the amount required to assemble and optimize a machine. I believe we'll within the year begin to see many more people following in Wade's footsteps with newly completed machines, replicating, and distributing parts. We may be closer to exponential, or at least rapid, growth than it seems.
As a Linux user from the pre-1.0 days, I can tell you that the state of RepRap now is very similar to the experimental Minix-alternative we struggled with back then. I believe RepRap's future will be similar too, very much as Peter described above. In fact, I very much hope so.
The purists can continue on as they have, much as the FSF have continued on with GNU, and in the process, it should also become much easier to "roll your own distribution".
For most of us, though, the RedHat and Ubuntu analogs will best serve our needs.
(So, when will Harbor Freight start selling stamped-steel RepStrap kits? Soon, I hope!)
I think you're right about the lasercut parts and high-integration boards being a step towards something other than the original principle of the project-- the trend of developments from a team trying to make a true wealth-without-money machine should be in the business of helping to make a machine in the simplest possible way *for each specific location* where they are made. That is going to mean a diverse family of machines rather than a single design, and a design ethic that simplifies and diversifies construction rather than concentrating and complicating it.
However, even if all the RepRap foundation accomplished was the creation of an open-source 3D printer, that'd still be pretty impressive.
This is my first post to not only this blog but to any forum whatsoever. I am normally a reader and not a speaker. Like some of the others who have posted on this blog, I have not taken the steps to build a Darwin either, but it seems likely that I will repstrap my first machine and then improve upon the design. As far as the future goes, I believe that both approaches are possible, and probably will occur, whether intentionally or not. The Mendel could very well be the next reference design from which a plethora of repstrap and repstrap 'species' originate. I also believe that people will continue to functionally specialize their machines and create systems that are unique to their particular variables (cost/parts availability), environments and applications. Personally, I would like to see more work done with a larger variety of materials and/or multi-toolhead designs. I think that the possibilities are limitless, especially when pick and place mechanisms and components are added to the toolbox. What we are seeing here is what has been common throughout human history. A fundamental design exists, and people modify the design to make it fit their environment. Once a standard is introduced (think Eli Whitney or better yet, the ATX standard for motherboards), there will probably be less repstrapping and more reprapping. Hopefully within the next six months I can get my own repstrap operational, with that accomplished, be able to contribute to the object libraries and blogs with something more concrete than ethereal. My focus will likely be creating composites and reinforced designs to expand the utility of the reprap platform.
i'd like to pull the "time" card on this one. i don't believe the community is "losing direction" so much as taking a necessary detour, but time will tell.
as a newbie, i *want* to get my hands on 2nd-gen parts. the "karma clause" plan, where each RepRap hopefully makes two complete sets for the community. that's not happening, so everyone is ordering lasercut...
many of which are not now, and may never be functional. of the ones that are, some will be "cheap 3D printers", in that they never contribute to the karma pool of parts.
one problem I see, each RepRap is calibrated for size and density, rate of flow, heat, etc. so parts from one will not 100% match parts from another. maybe major issues with tolerance unless every child part is from the same parent, which is highly inefficient.
|
