WikiHouse – An Open Hardware Building System

“WikiHouse is a contribution to the debate on Open Hardware and Open Design by 00:/, Momentum Engineering, Espians, Beatrice Galilee… and a global community of designers, including YOU! WikiHouse will be shared via a Creative Commons license for anyone to adapt and improve. A WikiHouse is fabricated from locally sourced plywood cut on a CNC mill from openly shared template files, and assembled with minimal skill by local people. The first WikiHouse will be constructed in South Korea at the Gwangju Design Biennale 2011. We are now looking for architects, furniture designers, product designers, and craftsmen from around the world who are interested in contributing to the WikiHouse process. If that’s you then please drop us an line on [email protected]!” *

A commentary on this interesting project by Eric Hunting:

“WikiHouse is a project developed by London design group 00:/ intended to explore the prospect of Open Hardware development in architectural design and housing. It’s objective is to establish a series of Creative Commons licensed housing designs that are engineered for production from plywood using CNC and assembly with simple puzzle-fit construction, bolts, and screws. These designs are intended for on-line distribution as Google SketchUp data allowing free access and customization. The designs draw heavily on the previous work of MIT in CNC-based housing construction, using much the same kind of puzzle-fit structural system. MIT’s own work on this concept turned to commercialization of on-demand production services through proprietary software doing procedural CAD/CAM generation rather than truly hackable Open Hardware development. So here is an opportunity to really put this technology into the public’s and the open tech community’s hands.

Six simple structures are shown on the site, based on a stressed skin structural system using clear span bay sections linked by slot-fit cross members and rigidized by skin panels. Most are single floor cottage or shed scale. This form of construction has been demonstrated sound in a great many other projects to date and well suits the use of flat bed CNC and simple CAD drawing. It has become increasingly popular with a lot of student architecture projects because of its versatility, though it has some limitations with window and door installation. However, 00:/ ‘s designs appear rudimentary and experimental. None are shown in a finished state and only one small portion of a structure seems to have been prototyped so far.

Though strong and very flexible in the range of modest scale forms possible, this kind of structural system tends to favor ‘throw away’ architecture because the finished products are very limited in adaptation and post-construction customization -even more so than conventional stick frame construction. Plywood itself varies greatly from country to country and manufacturer to manufacturer in long-term resilience and in the use of potential toxic binder materials and wood preservatives. These structures are potentially somewhat demountable in core structure but not very reconfigurable because the design is locked by the absolute form of the unit profile sections and they ultimately rely on ‘destructive’ assembly using self-embedding screws, nails, glues and conventional non-removable interior finishing. Repair and renovation are surgically destructive processes. This is generally overlooked as a ‘problem’ with these structures as they rarely get used permanently and because conventional housing construction is generally the same. But it does limit the potential economy of this housing to little better than conventional because most of the costs of housing construction is in finishing, not primary structure. So as long as one outfits the interior in a conventional way, the economy of these kinds of structures comes only from the relatively small net reduced time and possible sweat equity in the primary construction. This, however, is much improved where buildings are modest in scale and the interior design exploits the virtues of the clear-span volumes through open plan design. This is particularly well suited to quick-build relief housing where finishing is kept minimalist or mostly eliminated by utilitarian pre-finished/non-finished materials. In that context, this approach is a real boon with a potential to exploit truly low cost materials like laminated cardboard or recycled materials.

Despite the limitations, the very easy ‘hackability’, simple materials, ready suitability for flat-bed CNC, and freedom of experimentation with minimum CGI skills at the pre-construction level makes such structures an appropriate choice in an Open Hardware context and this project’s idea of establishing freely accessed/customized plans on-line has great potential. Unfortunately, this design team doesn’t seem as yet to be doing very much to realize that potential. There are no CAD or model files for any of their structures available on the above web site. No detailed descriptions or renderings of any designs. No step-by-step instructions. No attempt to characterize the structural system as with a standard design methodology. Either we have caught this project in a very early stage of development, or they have among them a very nebulous notion of what Open Hardware is about, how you communicate with the larger community, and what designers’ roles are in this context. They note that they will be showcasing their first full house at a design conference in Korea in September, which is nice, but that sort of traditional venue is irrelevant to Open Hardware. For that, the Internet is your showcase venue and the ‘source files’ -the CAD files and instructions- what you showcase and share. The whole point to Open Hardware is open participation and the ‘source files’ are the essential medium of that participation. If they’re ready to go for september, shouldn’t there be a complete design to show on their web site now? They must have the CAD and modeling done for that much.

Frankly, this kind of thing just isn’t that hard. The average maker on Instructibles would have more work to show by now. If the idea is to keep things under wraps until the ‘coming out’ presentation, one has to ask what one might hope to get at the Gwangju Biennale that you can’t get on Thingiverse? The point is how many of the sort of people who will _use_ an Open Hardware design one can reach, not those who will just gawk at it and do a magazine write-up. Are we looking at another Riversimple Urban Car hyping itself as ‘open source’ yet, years later, there are still no CAD files and production details released to the public? Hopefully 00:/ will soon avail themselves of some real makers who can set them straight on what Open Hardware is all about. There is a great opportunity here, and it would be a tragedy to miss it.”

2 Comments WikiHouse – An Open Hardware Building System

  1. AvatarNick Ierodiaconou

    Hi, I’m one of the designers working on WikiHouse at 00:/.

    Firstly thank you for including WikiHouse here on the P2P Foundation – it is very exciting to see it gradually gaining support and generating debate as it is still in a very early stage of development!

    Eric, thank you for the thorough feedback. Your comments are very astute particularly in relation to the implications of plywood as a building material. One note on this – we have tried to limit the possible wastage caused by future adaptation of structures by reducing the number of screw-fixings as much as possible, limiting them to internal linings as far as possible and avoiding the need for any glue in assembly. Whilst this obviously assumes a degree of care in staged demolition/adaptation it is in theory possible to surgically extract individual sections of the house and adapt primary structure (which simply relies on ‘bolt and wing-nut’ type connectors). We have also been looking over the past few days at solutions which rely on no bolts at all, but simply on plywood connectors wedged together and malleted into place. It is arguable whether this helps to ‘future-proof’ the structure, but it may help to reduce the number of secondary fixings. This is one amongst the many issues we hope will be explored and improved on by an open community of developers.

    Regardless, it is our hope that the principles which WikiHouse embodies (namely the sharing of architectural design ‘codes’; the use of flat-pack components cut via CNC, to allow for maximum adaptability and reductions in procurement time; and the use of intuitive and accessible design tools – all of which you correctly posit above!) can be applied to other materials and that further experimentation will yield other innovative results.

    The MIT work around the Fab Lab House has indeed been an inspiration for us, as have a great many other projects around the space of Open Hardware and Design including the Hexayurt (for its easy of assembly and instructions), the Open Architecture Network (for their aspiration to open access to architectural designs), the Blackfoot CNC machine, RepRap, and similar projects (for their intention to lower the threshold for access to personal fabricators), and a great many others – many of these will be credited and showcased at the Gwangju Design Biennale as part of the WikiHouse exhibit.

    In relation your points about sharing the source files rest assured we have no intention of keeping these under wraps and are working hard to get a proper site and files online within the coming week. WikiHouse has received a lot more interest within the first few days than we had anticipated so the holding page is already proving inadequate and the need to get the content online is obviously pressing. The only reason we have not already shared SketchUp files is because we have been refining our design guidelines and models following lessons learnt from our own prototype, as well as being very busy finishing the website and preparing for the Gwangju Design Biennale generally. We are nonetheless very much prioritising getting these out! The plugin is also still in development, and whilst we can (and very likely will) be sharing content without it, it should help the workflow for anyone wishing to develop WikiHouse.

    We are certainly not planning to launch ‘black box’ at the Biennale in Korea as we share frustrations about supposed ‘open-source’ projects that are anything but… Rather we aim to get content online with some weeks to spare before an open submission/contribution is actually selected for assembly at the exhibit towards the end of September. At the Biennale we will be providing two-three sections of house as a starting point and then waiting for contributions to determine the form of the last parts assembled at the exhibition. After the Biennale we fully intend to help drive WikiHouse forward.

    I hope this helps to address some of the points raised above. Again, thank you for your insights, and please feel free to discuss and critique the project as it progresses because the feedback is invaluable!

  2. AvatarLori

    CAD seems to be a perennial bottleneck of open source. Every few months I do a web search on the terms “open source” and “solid modeling” to find out if any open source project other than brlcad has achieved solid modeling capability, or alternatively, whether anyone has managed to compile and run brlcad…

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.