5 ms·

Cool. But completely not a solution to affordable housing construction in rural areas. What is the barrier to affordable housing construction in rural areas? W

by drsim 7y ago

Cool. But completely not a solution to affordable housing construction in rural areas. What is the barrier to affordable housing construction in rural areas? Where I traveled in Uganda it was availability and affordability of materials and equipment. Cement was expensive as was corrugated roofing. Renting heavy equipment was expensive. We had plentiful cheap unskilled labour, free bricks made from clay soil baked in the sun, lots of hand tools and a foreman with basic construction skills. Transportation was slow but not prohibitively expensive. I hope this 3D printing technique is simply a way of waving a flag to draw attention to the need for better housing in these communities. There are many much better MVPs than an expensive machine using expensive materials operated by an expensive team.

logfromblammo 7y ago
In Uganda, it is legal to use cheap construction methods, like wattle-and-daub using cow dung, and thatch roofing. Appropriate technology there is a hand-press for making TIB-style mortarless interlocking bricks from local dirt. (I have heard that Ugandan brick masons use way too much mortar anyway.) The test site was likely in Mexico because building codes in the US have no provision for establishment and testing of new technology that is too far removed from existing construction techniques, such as inflation forms lined with rebar and sprayed with shotcrete, or 3D-printed corrugated-interior concrete walls. If the machine built 50 houses in the US, none of them could get certificates of occupancy from the inspectors. The end goal is likely not to house the poor around the world, but to establish building codes that advantage the first-movers in printed-concrete home construction, so the patent-holders can build houses in the US that can be finish-ready for $5000, install windows, doors, electrical, and plumbing with assembly-line labor, and then sell for $75000.
- CWuestefeld 7y ago
  If the machine built 50 houses in the US, none of them could get certificates of occupancy from the inspectors. I agree with your concerns over the way regulation gets us stuck in a rut. But from TFA: "The same technology could also help transform the construction of affordable housing in the United States. Earlier this year, Icon printed a welcome center for a master-planned community in Austin, Texas, designed for people who have been chronically homeless, and the company is beginning to print 400-square-foot homes in the community that will be completed in early 2020."
- bluGill 7y ago
  In most states you need a professional engineer to certify that your odd construction is safe. This pulls in legal liability for failure so the engineer will charge a lot of money (and run a lot of calculations) before granting certification. It isn't a big deal to do that if you can figure out how to pay the cost. By contrast though: I can design you a house that follows traditional design rules. So long as I use standard materials under the standard rules someone else has long ago done all the work to ensure the standard is right and so it doesn't need additional certification. It is quite likely I can violate something in the standard and still be safe - but then it is up to me to prove it is safe (I'm not qualified to do this)
  - adwi 7y ago
    Wouldn’t there be economies of scale here though? Presumably the first print would be very expensive from certification standpoint, but wouldn’t the marginal cost going forward be negligible?
    - bluGill 7y ago
      Maybe, it depends. If you build exactly the same house on exactly the same soil type. However every change needs to be re-certified - you can get around this of course, the engineer can decide what changes you are allowed to make. Houses get around this because engineers figured out the limits - a 2x12 can up to (note up to not exactly!) span x distance under these assumptions. The inspector just verifies you met all the conditions and you are good. You can do this with any form of building, but it is much cheaper for one building to just certify that exact design. If you only have a half dozen designs it is probably cheaper to certify each separately, but eventually you have enough different designs that you just figure out the limits and rules and never have to worry about it again.
    - logfromblammo 7y ago
      If you can get the building inspectors to accept the engineer certification from another building of similar construction. You might be able to do that on a single property with 50 identical copies of one design, but one at a time is unlikely. So you build the engineer certified designs and then lobby to get the method of construction included in the building codes.
shadowgovt 7y ago
I have a trite observation, but I think it has merit: Minecraft shows that the cheapest building material will always be that which is nearby. For this solution to scale, it has to be really good at dealing with whatever's available. This seems exactly the kind of "making square pegs fit round holes" problem that is suited far more to localized human ingenuity than automation.
- vorpalhex 7y ago
  Yes but Minecraft doesn't allow your roof to cave or require you to insulate your electrical wires, much less handle severe weather events.

by drsim 7y ago

logfromblammo 7y ago

In Uganda, it is legal to use cheap construction methods, like wattle-and-daub using cow dung, and thatch roofing. Appropriate technology there is a hand-press for making TIB-style mortarless interlocking bricks from local dirt. (I have heard that Ugandan brick masons use way too much mortar anyway.) The test site was likely in Mexico because building codes in the US have no provision for establishment and testing of new technology that is too far removed from existing construction techniques, such as inflation forms lined with rebar and sprayed with shotcrete, or 3D-printed corrugated-interior concrete walls. If the machine built 50 houses in the US, none of them could get certificates of occupancy from the inspectors. The end goal is likely not to house the poor around the world, but to establish building codes that advantage the first-movers in printed-concrete home construction, so the patent-holders can build houses in the US that can be finish-ready for $5000, install windows, doors, electrical, and plumbing with assembly-line labor, and then sell for $75000.

CWuestefeld 7y ago
If the machine built 50 houses in the US, none of them could get certificates of occupancy from the inspectors. I agree with your concerns over the way regulation gets us stuck in a rut. But from TFA: "The same technology could also help transform the construction of affordable housing in the United States. Earlier this year, Icon printed a welcome center for a master-planned community in Austin, Texas, designed for people who have been chronically homeless, and the company is beginning to print 400-square-foot homes in the community that will be completed in early 2020."
bluGill 7y ago
In most states you need a professional engineer to certify that your odd construction is safe. This pulls in legal liability for failure so the engineer will charge a lot of money (and run a lot of calculations) before granting certification. It isn't a big deal to do that if you can figure out how to pay the cost. By contrast though: I can design you a house that follows traditional design rules. So long as I use standard materials under the standard rules someone else has long ago done all the work to ensure the standard is right and so it doesn't need additional certification. It is quite likely I can violate something in the standard and still be safe - but then it is up to me to prove it is safe (I'm not qualified to do this)
- adwi 7y ago
  Wouldn’t there be economies of scale here though? Presumably the first print would be very expensive from certification standpoint, but wouldn’t the marginal cost going forward be negligible?
  - bluGill 7y ago
    Maybe, it depends. If you build exactly the same house on exactly the same soil type. However every change needs to be re-certified - you can get around this of course, the engineer can decide what changes you are allowed to make. Houses get around this because engineers figured out the limits - a 2x12 can up to (note up to not exactly!) span x distance under these assumptions. The inspector just verifies you met all the conditions and you are good. You can do this with any form of building, but it is much cheaper for one building to just certify that exact design. If you only have a half dozen designs it is probably cheaper to certify each separately, but eventually you have enough different designs that you just figure out the limits and rules and never have to worry about it again.
  - logfromblammo 7y ago
    If you can get the building inspectors to accept the engineer certification from another building of similar construction. You might be able to do that on a single property with 50 identical copies of one design, but one at a time is unlikely. So you build the engineer certified designs and then lobby to get the method of construction included in the building codes.

shadowgovt 7y ago

I have a trite observation, but I think it has merit: Minecraft shows that the cheapest building material will always be that which is nearby. For this solution to scale, it has to be really good at dealing with whatever's available. This seems exactly the kind of "making square pegs fit round holes" problem that is suited far more to localized human ingenuity than automation.

vorpalhex 7y ago
Yes but Minecraft doesn't allow your roof to cave or require you to insulate your electrical wires, much less handle severe weather events.