I recently received a mail from Alex Mamalyha, web community manager for i.materialise, announcing the launch of a new service from Materialise NV. i.materialise is a rapid manufacturing service aimed at designers, and the beta site gives a good idea of the way the service will work. Obviously there are many web-based rapid manufacturing services these days, and the announcement of a new one is a fairly regular occurrence which I usually just ignore. But given the extent to which Materialise have supported and encouraged designers’ use of RM technologies through their .MGX initiative, I thought this was one service that deserved further investigation.

The ‘manifesto’ of i.materialise claims the service makes “3D printing as easy as printing on paper”. Obviously such claims owe more to hyperbole than fact, but the i.materialise interface is presented in a relatively simple and obvious way. A workspace in the centre of the screen visualises the model once it is uploaded, and a number of drop-down menus to the right give the choice of materials, surface finishes etc.

To test the service, I used a model I made previously for Nina Pirhonen, a Finnish designer and creator of the PomPom character and series of books. The model was originally created in Solidworks, but in order to upload it to the i.materialise site it first needed to be converted to .stl format.

3D model of PomPom © Nina Pirhonen

It’s here that some of the limitations of the i.materialise service first begin to show. Whether it’s to simplify the operation, or because limits don’t in fact exist, there’s no information regarding maximum file sizes or number of triangles/polygons. This is fairly basic information that anyone, designer or not, with the skill to create a 3D CAD model will want to know, since it has a fundamental effect on the quality of the manufactured model, and most CAD packages allow the quality of the .stl file to be easily determined. In this instance I used a relatively coarse setting, giving a triangle count of 20,024, and a file size of 1,001,284 bytes. I also imported the file into Rhino in order to export as .3ds, .obj and .wrl formats.

Exported .stl model of PomPom © Nina Pirhonen

Uploading the file is easy – click the upload button and choose your file – and quick; a 1.5Mb .3ds file took about 20 seconds to upload, with a further 8 seconds for the i.materialise software to analyse the file. Exactly what the analysis involves isn’t clear, but I assume the model is being checked to ensure it’s a closed volume. No errors or warnings were given about the model, which suggests it isn’t being checked in terms of the feasibility of it actually being made – the ears, arms and feet/ground of this model would definitely throw up problems, particularly in some of the more fragile material options. Once the model is uploaded it appears in the workspace of the user interface, and can be viewed from different angles.

i.materialise interface © Materialise NV (click for larger image)

Another limitation of the interface is that there’s no option to change the build orientation of the model. Since most rapid manufacturing technologies have different resolutions in their horizontal plane and vertical axis, this can be an important choice, affecting which surfaces have the smoothest finish. For a service aimed at designers it’s definitely an option I would expect to see.

On the right hand side of the interface there are a choice of materials, including ABS, polycarbonate and polyamide, alumide, and multicolour composite. As different materials are chosen the price automatically updates, along with the surface finishing options. Most materials can be painted and some can be mechanically smoothed (similar to tumble polishing). There’s also an extensive ‘library’ of information about materials and manufacturing techniques.

Multicolour Composite material properties © Materialise NV (click for larger image)

Given the nature of the model I was testing the service with, I was particularly interested in the multicolour composite option. Materialise use a Z Corp Spectrum Z510 for this process, which prints at 600 x 540 dpi. The problem was that, as far as I could tell there is no way to specify the colour of the surfaces. Usually Z Corp Spectrum printers use VRML files, but when I uploaded in this format (the system accepts and recognises .wrl files, even though they are not listed as a usable file type) no colour information was retained. The same was true when I tried an .obj file. Looking around the site and through the FAQ’s didn’t give any clues as to what file formats I should be using or whether this option is functional, but obviously this is something that needs to be fixed before the service comes out of beta testing.

All in all, I’m not really sure what to make of the i.materialise site and service. It’s stated explicitly that the service is aimed at designers, though it’s not made clear whether that means just design professionals or includes consumer-designers. Either way, for those with the experience and skill to create their own 3D models it seems a bit simplistic. There’s none of the control you get setting up a model for printing yourself, particularly deciding what orientation the parts should be printed. To be fair it’s possible to contact i.materialise direct, but then the service becomes little different to using a local RP/RM shop (although admittedly, the range of materials and processes is much greater than most shops are able to offer). And of course it should be kept in mind that the site is still in beta testing, the whole point of which is to iron out the glitches. In this respect at least, i.materialise have done a good job – the UI is easy to understand and the whole process of uploading a file, choosing a material and ordering is easy to follow.

Finally, one tantalising option, which isn’t possible to review but which is suggested in the FAQ’s, is the future possibility of ordering some of the .MGX designs. Quite how this will work is unclear – will it simply be possible to choose from a catalogue and hit ‘print’, or will it be possible to modify the design? But this is obviously one area where the i.materialise service can offer something unique, over and above similar web-based services or a local RM shop.

Update: Since posting this article I have swapped a few emails with Alex Mamalyha; my questions and his answers are reproduced below:

1. Is there a maximum file size, and is there a limit to the number of triangles or polygons in a model?
2. How are colours specified when choosing the multi-colour composite option (using the Z-Corp Spectrum printer)?
3. Will you attempt to make any model, or will you advise if an uploaded model is unsuitable?
4. What is your returns policy?

Finally, I noticed that it’s possible to upload a VRML (.wrl) file and the system will recognise it, but this isn’t listed in the ’supported formats’ list.

1. There is no limitation on the file size or number of triangles in the model.
2. Colors should be stored within the uploaded file, so if you choose Z-Corp we will print it in color (I realize it is a bit confusing since the preview generated images are in single color, but we will improve that bit shortly).
3. We did some testing with Blender users. If the uploaded file is not suitable for printing we will fix it ourselves, unless it requires severe design changes (after all we don’t want to print cube when the person is ordering sphere) in which case we will contact the designer and explain what has to be changed. We will not attempt to print something that is not printable. Additionally, we are developing plug-ins for major CAD programs that will provide designers with the info on problems with the file before they even upload it.
4. Once we print and ship the model, it cannot be return for obvious reasons (usually it’s one of a kind design)

About VRML support. We are still in a pretty early BETA and some info may not be consistent in all parts of the website, but we will do our best to provide support for a number of different file types.

Are there plans to add functionality which would allow the designer to choose the orientation of the part in the build chamber? Or maybe make it clearer which is the Z-axis so the part can be oriented in the modelling application?

We are working on functionality, so that preview window on the website will allow designers to rotate the model, rather than have screenshot generated (as it does now). It does not matter how the file is located during the upload process, we have support engineers, who check all incoming files and position them, we don’t expect people to know how the model should be oriented for printing purposes.

This post deals with the results and conclusion of the user trial discussed earlier. The findings of the study can be divided into two main areas: the results of the drawing exercise and the success of developing the drawings into a 3D CAD model, and the results of the two CAD modelling exercises. It’s important to stress that in both cases the objective was not to judge or analyse the quality of the design, but rather to gain subjective feedback from participants about which activities they enjoyed or disliked, and which approach resulted in the product they were most happy with.

Drawing exercise and 3D CAD model development

The table below shows an analysis of the drawings returned by the participants:

All drawings were analysed and coded according to the following criteria:

I(i). Number of sheets – How many sheets of paper were used in the exercise?
I(ii). Number of drawings – The total number of sketches made during the exercise, including sketches of ideas which were rejected.

II(i). Understanding of Safe Model concept – Did the participant understand and follow the instructions regarding the images of the safe model?
II(ii). Inconsistency between drawings – Did sketches exhibit inconsistent or contradictory information?
II(iii). Understanding of orthographic projection.
II(iv). Use of annotation.

III(i). Evidence of design iteration – Did the participant develop and test the validity of a design through sketches?
III(ii). Number of different designs drawn.
III(iii). Final design identified – Did the participant make obvious which was the final design?

IV(i). Functionality – Did the participant design a functional element in addition to the basic functionality of the USB memory stick?

Attach (A) – A method of attaching the product
Grip (G) – A feature which allows the product to be held more easily
Retain (R) – A method of keeping the cap in place
Other (O) – Any other form of functionality

IV(ii). Functional detailing – Did the participant include functional details such as screws or split lines in the design?
IV(iii). Cosmetic detailing – Did the participant include cosmetic details such as fillets or chamfers in the design?
IV(iv). Colour and Texture – Did the participant include details whose colour or texture were specified?
IV(v). Manufacturing constraints – Did the participant consider details imposed by manufacturing such as draft angles or material wall thicknesses?

V(i). Existing Designs – Did the participants look at the envelope of existing designs before, during or after the exercise?

VI(i). Degree of interpretation – a measure of the degree to which the CAD operator had to interpret the participant’s drawings in order to build the CAD model. Measured on a scale of 0-5, where:

0 = no interpretation needed, the drawings were accurate and fully resolved;
2 = some interpretation needed, the drawings were accurate but some details were unresolved
5 = significant interpretation needed, the basic idea was communicated but details were unconsidered or unresolved

The three most important findings, highlighted in orange in the table above, relate to design iteration, functionality and the degree of interpretation required to translate participants’ sketches into 3D models. The sketches showed that only four participants drew more than one design option. Even fewer (three) engaged in any form of design iteration, i.e. a process in which a design idea was modified. The most common form in which drawings were returned was a single idea, drawn from multiple viewpoints. As such, the ability of the participants to engage in design exploration through sketching was extremely limited.

This finding is supported by existing research into the manner in which designers use the activity of drawing. Designers tend to use drawing in two ways: firstly as a means of ‘exploration and manipulation,’ and secondly as a means of communication. The first is a creative activity in which multiple sketches are used to develop a design from the first idea to the ‘best’ idea. Such sketches do not need to be accurate or even realistic provided they offer an insight into the problem or possible solution. A communicative sketch, by contrast, is a method of explaining a (partial or full) design solution.

Drawings returned by participants appear to show an inability to utilise sketching as a method of exploration. Instead most participants attempted to draw the ‘correct’ design immediately, i.e. they tried to communicate a final design without testing whether it was, in fact, the best solution. There is a ‘randomness’ within design exploration which can be attributed to a lack of inhibition among designers to the act of drawing. Designers are often taught that mistakes when drawing have value and can lead a design in new directions; in contrast a number of participants’ drawings showed evidence of the use of an eraser to remove ‘wrong’ sketches. This inhibition or discomfort with drawing was further borne out by responses from participants, nine out of ten of whom preferred the process in Part II where drawing was not involved.

Sketch returned by one trial participant showing a lack of design iteration

Although their sketches showed a lack of design exploration, the need for design iterations was implicitly recognised by all participants in their reactions to the CAD model representation of their design. Initially all participants believed the CAD model to be an accurate interpretation of their drawings. However, all participants subsequently accepted the invitation to modify the CAD model, and all believed that the design was improved by this process of modification. Participants perceived the CAD model as a ‘sketch’ or work-in-progress which required development, and recognised that design iteration was necessary to arrive at a better design.

The results clearly demonstrate the value which participants attached to their self-designed products: a clear reason for this was the ability to introduce additional functionality to the product. All ten participants added additional functional elements to the USB memory stick, for example, details which enhanced grip, or methods for ensuring the cap was not lost. One participant shaped the device such that it could act as a bottle opener, whilst another attempted to decrease the possibility of the product being knocked and broken when plugged into a computer. This value placed on functionality rather than just aesthetics, together with a preference for a design which more closely fits the consumer’s needs, also coincides with findings from mass customisation literature.

Design with a method of retaining the cap when removed

The degree of interpretation required to translate a participant’s sketch to a 3D model is also a factor which needs to be highlighted. The drawing below shows a not-atypical sketch returned by one participant.

A ‘final’ design sketch from a trial participant

The drawing shows no indication of the preferred design, nor does it show design ‘refinements’ such as fillets or curved faces. Thus as the trial went on, it increasingly became obvious that one of the tasks, when translating a sketch to a 3D model, was to ‘read between the lines’ and second guess what the participant might want, rather than attempt to faithfully reproduce what was drawn. When presented back to the participant, the model which was built from the sketch showed a hook feature, as well as small fillets and slightly curved side surfaces. However the model was built and constrained in such a way that it could be updated to ‘close’ the hook, thus making the feature a loop. This was, in fact, what the participant had intended, and after modifying the model the final design is much more rounded than the original sketch.

The 3D CAD model produced form the sketch above

The final design after reconsideration and manipulation by the participant

3D CAD Model Modification

Results in this area refer to the tasks within both Part I and Part II of the trial which involved modifying the CAD model. The results largely consist of a comparison between the design process of Part I and Part II, and which process yielded the most favoured design. Opinions of participants were recorded during and immediately after the trial, and are summarised in the table below.

Discussion of Results

The most significant finding from the trial comes from the paradox highlighted in the table above. As previously mentioned, nine out of ten participants felt uncomfortable with the drawing exercise, preferring to modify a CAD model representation of a design. However every participant believed the drawing exercise ultimately led to the best design. When questioned, the main reason given was, as might be expected, that the design more closely matched their needs and wishes than the pre-existing model. Three participants stated that they would like to imagine their design was unique, and did not feel certain a pre-existing model would not be modified in similar ways by others. One participant said he would be proud to show the product to friends and explain to them why he had designed it in the way that he had. Thus the trial shows that the best results, in the participants’ own opinions, came from the less enjoyable process.

Conclusions and Future Research

The trial clearly shows that participants placed significant value on the ability to design their own USB memory stick. Products which were self designed were valued more highly than those which were customised from pre-existing designs, despite the fact that most participants felt uncomfortable with the process of self design and preferred the process of customisation. Participants were generally unable to engage in design iteration through sketching, and used drawing as a method of recording and communicating a design rather than exploring it. However, when presented with a CAD model representation of their own design, participants recognised the value of developing that design through iteration in order to arrive at a better solution. Participants placed a high priority on the ability to incorporate additional functionality into the basic usage of the memory stick.

The trial raises a number of questions which would benefit from further research. Of most interest to me is the question of how to resolve the apparent paradox between the preferred design process and the preferred outcome. Participants unanimously favour self designed products over modified or customised pre-designed products, however a clear majority did not enjoy the drawing task required to initiate the self design process. In a setting outside of a user trial it might therefore be expected that consumers would not engage in self design at all, and thus never arrive at a point where they were able to assess the value of their self-designed product. Future research should therefore investigate ways of capturing consumers’ design intent without requiring the consumer to sketch or draw those ideas.

A further question stems from the issue of how to interpret a consumer’s design intent, particularly when that intent is not well explored or developed, or when it is physically impossible to meet all criteria. In the trial this was overcome by the use of a designer/CAD operator, who was able to use experience and intuition to ‘second guess’ what the participant wanted to achieve with their design. In a commercial setting this scenario would likely be impossible, as it would require the input of a professional designer for every consumer created design. This would suggest the need for an automated process which would replace the designer’s intuition, or more realistically, which applied certain rules to constrain and condition the consumer’s design. Such a system, whilst inevitably limiting creative freedom to some extent, would also give the consumer confidence that the self designed product would always be manufacturable. Such a system, therefore, could be considered an advanced form of customisation toolkit, one which enabled the consumer to move beyond configuration and engage in freeform design.

In general participants understood the need for design exploration and iteration, even though this was rarely displayed in their sketches. Within the CAD model manipulation tasks, some participants were helped by the use of Genoform, which was able to suggest new directions for exploration which the user was otherwise unable to see. It may be that an automated iteration would be of benefit to some users in improving their designs.

A final question arises from the value which participants placed on adding additional functionality to their design. The trial was able to recognise that a participant valued their design of, for instance, a grip detail. However it did not ask the participant to compare that grip detail with, for instance, a method of retaining the cap when removed. The results do not record whether the participant considered a cap retention method but rejected it as less important than a grip detail, or failed to consider a cap retention method at all. Thus the trial could not disclose whether participants saw value in added functionality in general, or only in the added functionality that they had designed. Such knowledge would be valuable, particularly if a future trial were based on a more complex product.

From Configuration to Design: Capturing the Intent of User Designers is the title of a paper I recently presented at MCPC 2009 in Helsinki. It details a user trial conducted as part of my PhD research, which sought to understand the extent to which non-professional user-designers are able to engage in design exploration and to communicate design intent. The paper itself, together with the presentation given at the conference, can be downloaded from the Papers and Presentations page of this site. Much of the background argumentation to the study has been made in previous posts, therefore what follows is an edited version of the paper, focusing on the design, conduct and conclusions of the user trial. This first post deals with the design and conduct of the study, a follow-up post will concentrate on the results and conclusions.

Design of the Study

The focus of the trial was the design of a USB memory stick. This was chosen as a relatively simple product whose functionality was easily recognised by those who took part in the study. The trial was intended to investigate two main research questions:
What is the best method for consumers to conduct design exploration?
How well are consumers able to communicate design intent?

It built on the observations of a number of researchers with regard to the way designers and architects use drawing as a way to generate and evaluate design solutions, but sought to place such observations more specifically within a mass customisation scenario. It also sought to understand the practical difficulties of expecting non-designers to use drawing in the same way that trained designers do. The intended outcome was to better understand what future tools will best enable consumer-design, which will form a major part of my future PhD research.

I should make clear from the beginning that within the user trial, neither modelling exercise (in Part I or Part II of the trial) was intended to test or replicate a co-design exercise. Within design research, co-design can be described as a subset of user-centred design (also called participatory design), in which the user takes part in the actual design of the object in question as part of a design team. This contrasts to user-centred design itself, in which the user contributes experience or opinions, but the designer carries out the design task. In both user-centred design and co-design however, it is the designer who is perceived to hold ‘expert’ knowledge and who has the ultimate power of decision, and the user is relegated to the position of contributor. In contrast, this trial sought to investigate a situation in which the participant had the ultimate power of decision over his or her design. In all situations where the participant worked with the CAD operator (described below), care was taken to ensure the CAD operator did not offer opinions as to the value of any design or decision. Advice was given only on the capability of the CAD software to achieve a desired outcome, rather than the value of that outcome. Thus the role of the CAD operator was that of a facilitator between the participant as designer and the requirement to create a 3D CAD model.

Two Methods of Design Exploration

The study was divided into two parts. In part I, the participants were first required to undertake an unobserved drawing exercise, followed by a design modelling exercise with the assistance of a trained CAD software operator. In part II, they were required to choose one of six pre-existing designs which was then modified with the assistance of the CAD operator. Participants were placed in one of two groups; group one conducted part I of the exercise followed by part II, whilst group two conducted the exercise in reverse order.

The first method of design exploration, addressed in part I of the study, could be classified as unconstrained concepting. Participants were free to explore issues of functionality and aesthetics with no constraint other than that the design should be bigger than a minimum volume (the minimum size required for the electronics to fit inside). This meant that the first method was close in scope to the design process of a trained industrial designer. The second method of design exploration, addressed in part II of the study, can be classified as constrained concepting. Participants were able only to modify a pre-existing design within the constraints allowed by the CAD model, the idea being that the task was closer in scope to a MC toolkit experience.

Part I – Sketching Exercise

Participants were briefed as to the task and requirements of the exercise, but conducted the exercise unobserved so that they were in a more natural environment and worked in a less time-constrained manner. Participants reported spending one-and-a-half hours on average on the task, though most reported thinking about the task over a period of days before beginning. Participants were required to complete the sketching exercise within one week of having been briefed.

When being briefed, participants were told to create drawings on an A4 marker pad supplied to them. A number of images were supplied of a ‘minimum volume’ USB memory stick (see the section headed “Concept of the Safe Model” below), and participants were shown how to use these images as underlays over which their own designs could be drawn. Participants were instructed only to use the drawings pads and pens, pencils, markers etc, and specifically not to create designs using a computer. It was also made clear that the purpose of the study was not to judge drawing skill.

Sketch design by one participant in the trial

Participants were instructed to design the body and the cap of the memory stick, and to imagine they were designing a personal product, i.e. not to consider the needs of other consumers. It was stated that participants should not copy an existing design, and that the more personal the design (in terms of either function or aesthetic) the more useful it would be to the research. Participants were also supplied with a sealed envelope of existing USB memory stick images, and told they could open the envelope at any time during the exercise.

Sketch design by a second participant in the trial

Part I – Modelling Exercise

Participants were required to return their drawings by mail such that no verbal explanation could be given about the design. This made sure that their ability to communicate through sketching only was tested. The drawings were used as the basis for construction of a 3D CAD model by a CAD software operator experienced in making industrial design models; the CAD operator was therefore required to ‘read’ the drawings, ‘interpret’ the participant’s design intent, and develop the 2D drawings into a 3D model

Finally, participants were told that whilst they could work on any number of designs, at the end of the exercise they should have one final, favourite design. Participants were instructed to return all drawings, even those of discarded ideas. It was also emphasised that when submitting the final design, participants should consider how well it could be understood by someone looking only at their drawings.

In some instances what was drawn was not physically realisable in 3D. This is sometimes referred to as a ‘failure of reinterpretation’ – where the person making a drawing has failed to reinterpret and understand the logical implication of their own drawing. The image below shows such a failure of reinterpretation: the indentation in the top surface is shown as breaking the side surface in one view but not in another. In these cases the 3D modeller’s task became one of intuitively judging the participant’s intent, rather than trying to faithfully reproduce was what drawn.

Sketch design showing a failure of reinterpretation

3D CAD model from the sketch above

3D models were built using Solidworks 2007 (Service Pack 2.0) CAD software. Solidworks is a hybrid (it allows the use of both solid and surface modelling techniques) parametric CAD modeller, in which features are primarily created from constrained, dimensioned sketches. One of the skills of the CAD operator lies in understanding how to constrain sketches such that dimensions can be altered and the model will update. Sketches which are not appropriately constrained will cause the update to fail, which can then result in significant time spent “debugging” the problems. The software and version were determined specifically by the use of Genoform, an automatic iterative design program that formed part of the study (see the section headed Genoform below).

Participants were invited back to conduct the modelling exercise approximately one week after having submitted their design. They were asked to review the model and to comment specifically on how well it captured their design intent (i.e. did it look the way they expected). Attention was drawn to specific aspects of the model, particularly where the CAD operator had interpreted a difficult-to-understand drawing or feature. Participants were then asked whether there was anything they would wish to change about the model either to improve the design or to correct mistakes in the interpretation of their drawings.

When the CAD model had been modified to a state the participant felt reflected their aspirations for the design, the Genoform software program was used to generate alternative design options. Initially ten options were generated but participants were free to generate more if they wished. Options which were liked or perceived as interesting were imported back into Solidworks; these reimported options were then compared to the originator model. In a majority of cases the participant requested changes to the originator model, based on ideas stimulated by the Genoform options, however in no instance was a Genoform option chosen as a ‘most favoured’ design.

Part II – Modelling Exercise

In this part of the study, participants were shown six pre-designed CAD models. The reasoning behind each design, e.g. why it was a certain size or contained certain features, was explained; the extent to which it might be modified was also made clear. Participants were then asked to choose one of the six models as the basis for the rest of the exercise.

6 pre-designed models presented to the trial participants

Having chosen a model, participants were asked which aspects of its design they wished to change. Where it was possible to modify the model by changing a feature’s dimensions or parameters this change was accepted. However any request which involved adding new features was not accepted. For example: with the ‘grip’ feature on model 1, the number of grip details could be modified, however a similar grip feature could not be transferred to any other model. In such a way participants were deliberately constrained in their ability to influence a given design. The CAD model was again modified by the CAD operator in front of the participant according to his/her instructions.

When the chosen model had been modified to reflect the participant’s intent, the Genoform program was again used to generate alternative designs. Ten options were generated initially but participants were able to request more options. Again, those options felt to be interesting were imported back into Solidworks and compared to the participant’s own modified model. In this exercise Genoform was less able to suggest new ideas or directions, and a majority of participants preferred their own modified model to the Genoform derived options. In those who found the Genoform options useful, none chose a Genoform option as the ‘most favoured’ design, preferring instead to modify their own model further.

Concept of the Safe Model

Most industrial designers understand the safe model (sometimes also called a ‘keep away’ model) concept. It’s used to understand and visualise the minimum possible size of a product, whilst taking account of internal mechanisms and electronics, thickness of materials, tolerances, etc. A safe model of an MP3 player for example, would be created by ‘expanding’ the dimensions of the internal electronics by an amount equal to the thickness of the materials used in the outer casing, plus the distance required between the electronic components and the inside of the casing. It can also incorporate considerations of safety, ergonomics, marketing, etc; thus the safe model for a family car would be affected by the need for crash crumple zones, headroom in the passenger compartment and size of boot. A safe model does not dictate the final design of the product (though it does influence the final design), rather it indicates the absolute minimum volume a product can be when all other requirements are met.

Safe model of a touch-screen mobile phone

Example of a safe model highlighting a potential problem. In such a case the designer would need to redesign the product in this area, or ascertain whether the internal components could be moved

The concept of the safe model was used in two ways in the user trials. Firstly, having calculated a safe model for the USB memory stick, images of this safe model were given to participants during briefing of the drawing exercise. Participants were shown how to use these images as underlays which acted as guides during design. Provided the participant’s drawings were not smaller than the images of the safe model, their design would be realistically manufacturable. The safe model was also used in the two modelling exercises. By modelling the safe model inside Solidworks, any design could be superimposed to check if it satisfied the minimum volume requirements (Figure 4). Furthermore, when setting up the parameters for the operation of the Genoform software, the safe model placed lower limits on the extent to which Genoform could modify the design.

A participant’s ‘final’ design, with the safe model also shown

Genoform

Genoform is an iterative design exploration tool which operates as a plug-in module to Solidworks. It is produced by Genometri, a design technology company which develops specialised software, which was created as a spin-out company from the National University of Singapore. Genoform works by varying the dimensions of a Solidworks sketch; the designer can assign which sketches Genoform can manipulate, which dimensions within those sketches, and the degree to which the dimension can be varied. Thus it is possible (for example) to instruct Genoform to vary a dimension of 10mm by plus or minus 25% (i.e. a range of 7.5mm – 12.5mm). It is also possible to set maximum or minimum values, thus the designer may decide that the 10mm dimension can never be reduced, but can be increased by 45% (i.e. a range of 10mm – 14.5mm). In this way Genoform will run through the structure of a Solidworks CAD model, altering dimensions by a random factor within limits decided by the designer, and creating new iterations of the original CAD model. Genoform will create between one and one thousand variants, as the designer decides. The image below shows variants of a single design created by Genoform. My earlier post goes into more detail about how Genoform works, and you can download a trial copy here.

Design iterations created by Genoform. The original model is in the top left

Trial Participants

Ten participants were recruited from within the postgraduate student body of Loughborough University in the age ranges as shown in the table below. Participants were required to be computer literate as defined by daily engagement with five out of seven of the following activities: web browsing, e-mail, social networking, chat, VOIP (e.g. Skype), Microsoft Office software, other software. Participants were also required to self identify as “being interested in design and new technology”. As such, the profile of participants fitted with the findings of e.g. Bauer et al. (2007) and Füller and Bartl (2007) regarding the types of consumer most likely to engage in mass customisation. Furthermore, the trial excluded participants who had trained or were working as industrial designers.

The results and conclusions of the trial are discussed in the following post.

I recently got back from a trip to New York, having been there during ICFF and all the design week activities surrounding it. I was somewhat surprised at how little rapid manufactured furniture there was within the main show (unless you count laser cutting, which was impossible to avoid and demonstrated little that wasn’t being done five years ago), but outside .MGX was again showing it’s new collection at Moss, this year entitled E-volution. I should say straight off that the curation of this exhibition isn’t particularly clear: some of the pieces on display are from previous collections, and not everything in the new collection is on show. Nonetheless, it occurred to me whilst walking round that the designers and pieces involved fall into three distinct categories of the exploration of design for rapid manufacture.

.MGX by Materialise © Moss

It’s worth making clear at this point that when I say ‘design’ I mean the kind of design that .MGX promote and specialise in. The kind of design that’s sometimes called “Design with a capital D”; the kind of design which non-designers usually associate with the word and which some designers dislike because they feel it misrepresents what design ‘really’ is. This is design whose reason for existence is spectacle, whose aim is to make people take notice through initial observation rather than through extended use. Normally I would say there’s too much of this kind of design, and much of it is rubbish. But Materialise has a clear objective, which is to push the boundaries of rapid manufacturing and showcase what the technologies are capable of. If the result is more designers and manufacturers understanding what RM technologies can achieve, then that’s a good thing.

.MGX by Materialise © Moss

The first category is best termed Design as an Exploration of Production. This category is the largest in terms of the number of .MGX products it contains, and is made up of products whose central interest is an exploration of what rapid manufacturing technologies can produce, which conventional technologies cannot. It is typified by complex detailing on both the interior and exterior of the product, geometries which would be impossible to achieve were any form of tooling required.

Fugu vase by Hani Rashid © .MGX

Many .MGX lights fall into this category. Being able to design and manufacture an internal space of great complexity allows the designer to play with the way in which light escapes from a volume, as well as the shadows it creates.

Tulip lamp by Peter Jansen © .MGX

A further branch of the Design as an Exploration of Production category continues the idea of manufacturing what would conventionally be ‘impossible’ forms, but reduces the complexity to much purer geometries. Bathsheba Grossman has worked with Jiri Evenhuis to up date an earlier version of the Torus lamp, in which two simple donut forms interconect. Another example from Freedom of Creation (of which Evenhuis is a partner,) though not produced through .MGX and thus not on show, the Rollercoaster bowl.

Rollercoaster bowl by Janne Kyttänen © Freedom of Creation

The second category is what I call Design as an Exploration of Craft. The main use of rapid manufacturing in this category is as an enabler of actually getting an object produced, an object which would otherwise be too expensive, or require too high a level of expertise, to be manufactured. It doesn’t necessarily mean that only one object is ever produced, more that, even if the production volume never rises above one, the design is still a success. In many ways, it’s possible to see more evidence of this kind of design in past .MGX products than in the newest collection – it seems the ability to create one-off pieces is no longer the wonder it once was. The Damned lampshade by Luc Merx is from 2007, but was on show at Moss, and exemplifies this category.

The Damned Lampshade by Luc Merx © .MGX

The lampshade clearly refers to classical images of the fall of the damned, such as those of Gustav Doré in Dante’s Divine Comedy. It’s about as far removed from what’s usually considered to be ‘good design’ as it’s possible to get, and it’s difficult to imagine it would ever have been made if rapid manufacturing did not exist. Merx refers to the complexity of 18th Century carved ivory furniture as an influence, a craft which is highly skilled and nowadays virtually obsolete (not to mention, largely illegal). The .MGX website describes how selective laser sintering technology allows this intricacy of form, but this lampshade is not an exploration of that intricacy. Rather it takes the capability for granted, and uses it to explore a very personal vision of the designer. In a similar way, though more whimsically perhaps, Dan Yeffet’s Forbidden Fruit bowl is a take on original sin, as well as a pre-emptor of Shapeways Creator service.

Forbidden Fruit bowl by Dan Yeffet © .MGX

The final category I have termed Design as an Exploration of Design. Here the end product may display traits from either of the first two categories, however the most important thing is the way the designer uses the capabilities of rapid manufacturing to explore new ways of designing. I have written previously about Assa Ashuach’s AI stool, on show at Moss, and the way it was designed using a kind of ‘reverse’ finite element analysis to determine where the stool should be rigid and where it should flex in order to create the minimum possible volume of material. In one sense it can be argued that the designer didn’t design the AI stool, Assuach designed the process, and the process designed the stool.

AI stool © Assa Ashuach

A similar notion, i.e. a design process in which the designer does not control the final outcome of the design, is apparent in the Root chair by Sulan Kolatan and William MacDonald. Based on traditional Asian furniture in which tree roots are shaped to create individually unique pieces of furniture, each Root chair is digitally ‘grown’ within variable parameters. There is little information in the show about exactly how this is done, but this investigation of an ‘evolving’ family of forms is something also being explored by Lars Spuybroek’s MyLight (included in the show at Moss), as well as Lionel Theodore Dean of Future Factories.

Root chair by Sulan Kolatan and William MacDonald © .MGX

As well as the Torus lamp with Jiri Evenhuis, Bathsheba Grossman is also showing the Gyroid lamp. Grossman is well known as an early exponent of rapid manufacturing technologies which she uses to make mathematical sculptures. Her background as both a mathematician and artist lead to the creation of sculptural forms driven by equation and geometry in which the personality of the designer determines the meta-design but is strangely removed from the symmetrical, repeating details. According to .MGX

In nature, the gyroid is found when two immiscible fluids are forced to occupy the same space. These fluids interpenetrate but do not dissolve together. The same is true for the Gyroid.MGX which divides the 3D space it occupies into two regions. These regions are identical, interlocking, and yet remain completely distinct from each other.

Gyroid lamp by Bathsheba Grossman © .MGX

These three design explorations encapsulate the types of creativity rapid manufacturing is facilitating amongst some designers today. They don’t just relate to the work produced by Materialse .MGX, they are also able to accommodate, for example, Front’s Sketch Furniture and Future Factories Ghost chair. In time new areas can open up, but if you can think of examples today which don’t fit, I’d be interestd to hear.

Following on from Shapeways, which was spun out of a Philips research project, another Eindhoven-based company offering consumers the opportunity to design and manufacture their own products is studio:ludens. Started by Wouter Walmink and Alexander Rulkens, studio:ludens’ aim is to give people “the tools to create by using our skills as designers and our knowledge about the production process.” Like Shapeways, and ZapFab and FluidForms before them, studio:ludens have developed a set of interface tools which guide consumers through the creation of a product. Where studio:ludens shines though is in the quality of those tools, which without doubt are the most elegant and polished of all those I’ve seen so far.

Currently two design tools are available, the first, epa:kato, creates individualised drinks coasters, whereas lux:creator (still currently in development) allows consumers to design their own lamps. Both tools are Flash based which means clicking the browser’s back button will take you out of the tool, losing any designs that haven’t been saved. Causing the tool to automatically open in a new window would be an easy way to solve this.

The start point of epa:kato, the coaster design tool © studio:ludens

On starting the coaster design tool the user is presented with a square with a number of control points. This seems relatively familiar to anyone who’s used Illustrator or something similar, although they work in a slightly different way. As the points are dragged the shape deforms, but symmetrically, depending on which of the tiling patterns have been selected. This means that although you are controlling where the points are dragged, the overall shapes that are created are never entirely predictable. Clicking and dragging on the handle between the main control points introduces a new control point, by which means the complexity of the shape can quickly be increased, while double-clicking a control point deletes it.

As control points are moved the shape updates according to the symmetry of the chosen style © studio:ludens

What also increases the unpredictability of the design is the fact that the coasters are intended to tessellate. Whether this is an intentional design feature, or has been driven by the need to reduce waste when the coasters are manufactured, the result is a tool where you have the impression of influencing the design rather than dictating it. Once I got over the need to carefully control the design, I found myself designing in a much more playful manner, pulling points around just to see what happened. The downside though, is that without an ‘undo’ button it’s easy to make a change to the shape and then not be able to get back to a previous state.

Crossing lines highlighted in red © studio:ludens

The instructions for epa:kato explain that thin sections of material and lines which cross (producing ‘orphan’ objects) aren’t permissible. The tool has a nice way of showing where lines cross: a glowing red circle which highlights the offending area. However there doesn’t seem to be any indication of how thin a section can be, and of course even if a section is manufacturable, it may break easily in use. Designs which are submitted for purchase are checked before they are produced, but an indication of too-thin sections would be a good addition to the tool.

Snakes in a Plane, a design in the epa:kato gallery © studio:ludens

Registering at the site allows users to upload designs to a gallery, and is necessary before purchasing a design. It’s also possible to choose another user’s design and either modify it or buy it unaltered. Since launching the epa:kato service studio:ludens have added a feature which allows customers to add laser engraved text in a number of different fonts, further increasing the possibilities for customisation.

Modified Snakes in a Plane, with text added © studio:ludens

lux:sculptor, the second design tool offered by studio:ludens, takes the notion of unpredictable design to another level. On first entering the tool you’re presented with a wireframe representation of what looks like a table lamp. There are two ways to influence the design of the lamp, the first of which is to create ‘waves’ through the wireframe.

The start point of lux:sculptor, the lamp design tool © studio:ludens

There are five wave types to choose from, whose size can be changed using the editing tools. Waves are created using the computer keyboard, the top row of letters (Q, W, E… up to P) make make clockwise rotating waves of increasing frequency, the middle row of letters creates stationary waves, and the bottom row of letters creates anti-clockwise rotating waves. What really adds to the unpredictability is that keys can be pressed simultaneously, which will overlay one set of waves on top of another. As you remove your fingers from keys the waves die away, and designs are ‘frozen’ by pressing the space bar.

Waves percolating through the wireframe © studio:ludens

There is something quite compelling about creating the waves and watching the effect they have on the animated wireframe lamp. However as a design tool I found this even more difficult to like than epa:kato. For one thing it’s very difficult to combine the different wave shapes, you need to be pressing keys with one hand whilst also clicking a wave shape icon with the mouse. What’s more, I kept noticing forms that I liked, but the nature of the tool is such that the ‘design’ is already fading away even as you notice it.

The lamp’s profile is changed by dragging control points on the highlighted curve © studio:ludens

The next stage is to shape the profile of the lamp. This is very similar to the idea first implemented by FluidForms, whereby control points on a profile can be dragged to change the overall shape. Personally I found the order of the process (wave forms, then shape) a little illogical, given that the shape of the lamp is a ‘macro’ consideration, and the shape of the waves is a ‘micro’ one, though I’m not sure that this would bother everyone.

lux:sculptor is still in development and so not all features are working fully. As with epa:kato there is a gallery of designs submitted by other users, but currently these can’t be loaded into the design tool and modified. It’s also not completely clear how the lamps will be manufactured; at first I had assumed some kind of additive process such as SLS or SLA, but later it occurred to me that the lamp could be built up from a series of laser cut pieces, which would explain the wireframe style of the design tool.

Whether studio:ludens can build a successful business from this type of consumer design remains to be seen. I can’t stress enough how impressed I was by the quality of the tools – they’re fast, easy to understand and look good –  and I enjoyed interacting with them to create objects. But I can’t help wondering whether consumers will be frustrated by not being able to control the fine detail of their designs. Somehow epa:kato and lux:sculptor feel like experiments in the process of design, which isn’t a bad thing at all, but might not be what consumers are looking for. If that’s your kind of thing though, take a look at this post on the studio:ludens blog: an experiment to create a 3D form based on the ’shape’ of a persons voice.

I♥Sketch has been getting a lot of attention on design blogs this last week, which is hardly surprising because the immediate reaction from anyone whose work involves translating their 2D sketches into 3D models is “I’ve got to try it!”. Unfortunately there’s no word yet of when this might be made publicly available, but the video below, and the paper (pdf) due to be presented at UIST08, give a good idea of how the system works. But whilst it’s designers who have got excited, my interest is also in whether I♥Sketch has implications for how non-designers might interface with CAD systems

I♥Sketch © Department of Computer Science, University of Toronto

In their paper to UIST08, I♥Sketch’s developers make clear that the software is not intended as an aid for non designers:

“Our chief design goal is to wean designers from physical pen and paper with a similar 2D sketching interface and subsequently ease them into a 3D environment where their 2D sketches transition into 3D models used in product design…It is important to note that ILoveSketch is intended for professional designers who are willing to invest some time learning a system in return for improved workflow later on, rather than for walk-up-and-use casual users.”

Nonetheless, it is obvious from the video that for a person with some sketching skill but no 3D CAD experience, I♥Sketch could present a significant reduction in the learning curve needed to get an idea into 3D and thus into rapid prototype or production. It doesn’t attempt to ‘help’ the user to design (more on that below), but with future (already hinted-at) developments, it could be an extremely powerful tool for non-designers to visualise the 3D reality of their 2D sketches.

I♥Sketch builds on a significant body of  previous work investigating ways of translating 2D sketching into 3D curves and models. The UIST08 paper contains 44 references, and acknowledges the extent to which the system is built on earlier work. However none of the existing approaches have had much success in terms of acceptance by the design profession. The reasons for this, according to the authors of the paper, is that previous systems have focussed on “specific curve interaction techniques” rather than the designer’s workflow, and have failed to understand the designers sketching process and practices.

I♥Sketch has two distinct components: navigation, and curve sketching and translation. Navigation is carried out through a combination of pen gestures and button input (the left hand of the user in the video). Curve sketching is entirely pen-based. At times the two components interact as the software deliberately rotates the view to what it thinks is the best drawing angle, based on the last sketch stroke of the user. The overall feel of the system is based on a physical paper sketchbook, which provides a number of metaphors for the software, such as tearing off a page to discard it, and scribbling to erase a line. At times some of these metaphors seem a little forced, and I could imagine becoming tired of having to drag from one corner of the tablet to another (for example) to review sketches, rather than pressing a single key or clicking an arrow icon. But this is a minor point, and one which could probably be easily addressed in future.

Sketches are reviewed by dragging across the ’sketchpad’ © Department of Computer Science, University of Toronto

It’s when you come to watch the way that I♥Sketch generates curves that you appreciate how the developers have attempted to understand the way that designers sketch, and how sketches are built up from ‘exploratory’ marks on the paper. I♥Sketch uses what it calls ‘multi-stroke curve sketching’, in which a final curve is built up as the product of a number of earlier curves. This simulates the designer making faint guide curves, gradually darkening them as the desired shape becomes apparent. The software uses an ‘ink drying’ metaphor – whilst the ink is wet the curve can still be modified, but once dry it has to be kept or erased (shown at 00:28 in the video). Single complex curves are created where the designer draws to connect two existing curves, and splines are automatically created if one curve is drawn tangent to another (00:44 – 00:54 in the video).

multi-stroke curve sketching © Department of Computer Science, University of Toronto

automatic spline creation © Department of Computer Science, University of Toronto

Understanding how curves drawn in 2D are represented in 3D space is probably the most complex part of the I♥Sketch system. There are five different ways in which this can happen. The first two are designated as two-curve methods:

- the designer draws two curves, either side of a given plane; the software assumes these to be symmetrical and creates two curves each the ‘average’ of the two that are drawn
- the designer draws one curve from one view point, then redraws the same curve from a different viewpoint; the software then interpolates the two lines into a single 3D curve

Once the designer has created a curve, three further techniques (sketch surface methods) are possible. These are all essentially similar to ’sketch-on-curve’ techniques with which CAD users will be familiar, in which a plane or axis system is defined at a certain point on a curve, and subsequent curves are drawn on the plane or planes. I♥Sketch achieves this through an ‘axis widget’ which can be placed on any existing curve and oriented using a set of pen gestures.

Axis widget placement on a curve and curve intersection © Department of Computer Science, University of Toronto

To test the viability of the system I♥Sketch used a professional industrial designer, Calen Whitmore. Whitmore received a one-hour instruction session, before spending 1½ hours testing all the features of the software. Once accustomed to working with the system, he spent ½ hour modelling the jet fighter shown in the video, and then 2½ hours designing the car shown below. Whitmore’s detailed feedback is included in the UIST08 paper, but in general felt the system was much faster at generating good curves than traditional software, allowing more time to think about the design rather than how to achieve the design. One distinct negative was the feature which automatically rotates the sketch to the ‘best’ drawing position (similar to how a designer might rotate pad of paper). Whitmore asked for the time delay before rotation to be increased, but even so still gave this feature the lowest mark possible. I can also imagine that this feature, though well intentioned, would be extremely annoying, a bit like someone rotating my sketchpad without me asking, because they thought they knew the best position. Presumably a future development would allow this rotation to occur only after the designer has explicitly requested the action.

3D sketch © Department of Computer Science, University of Toronto

Whitmore’s frustration with the automatic rotation feature demonstrates the fine line that has to be negotiated when deciding how software should make things ‘easier’ for the user. Most computer users have been in the situation where we know exactly what we want to achieve, but ‘user friendly’ software corrects our actions because it assumes we have made a mistake (I find Microsoft Word particularly infuriating in this respect). But the multi-stroke curve sketching is also an example of the software ‘guessing’ what the user is trying to accomplish, and this received top marks from Whitmore. In terms of I♥Sketch being used by inexperienced or non-designers, it seems to me this may be one area that could be developed. Those who are not confident sketching tend to produce hesitant, less fluid strokes than those who are. It could perhaps be possible to generate algorithms which convert these hesitant lines into more flowing ones, or perhaps suggest possible curves to choose from, based on the user’s attempts. Of course this isn’t going to turn a poorly conceived idea into a good one, but it may allow a poorly drawn idea to be realised in 3D.

Some other possible future developments are hinted at in the UIST08 paper, which refers to surface modelling techniques such as FiberMesh. A project run by the Computer Graphics department in the Technical University of Berlin, FiberMesh generates surfaces directly from sketched curves. The original curves remain on the generated surfaces, allowing further manipulation of geometry, and a range of basic modelling operations such as cut, extrude, pull etc are available.

Sketching operations (from top to bottom): creation, cut, extrusion, tunnel © Computer Graphics Department, Technische Universität Berlin

I♥Sketch already generates files in .igs and .ma (Maya) format – the Maya files are available to download from the I♥Sketch website. According to Paul Salvador, who posts on the ProductDesignForums site as zxys, these Maya curves are very clean and allow surface patches to be easily constructed, as demonstrated in his model below. A system which generated surfaces automatically as Fibermesh does, would greatly enhance the level to which the 3D curves could be understood, furthermore it might allow a surface itself to be selected as a drawing surface, allowing curves to be drawn directly onto a curved body.

Surfaces created from Maya curves © Paul Salvador

One aspect which appears to be missing from I♥Sketch, surprisingly given the extent to which the developers have tried to understand how designers work, is the ability to use one sketch as the underlay for another. It seems that this should be possible – the description of Calen Whitmore’s design process of the car details how four circles were permanently displayed to represent the wheels, giving a sense of size and proportion. However the notion of one sketch being traced as the basis for a design iteration doesn’t seem to be implemented currently.

Exactly how useful this kind of system would be to someone not trained as a designer is open to question. It’s fairly obvious that, even though the need for expertise in 3D CAD systems is removed, it’s replaced by the need to be able to sketch confidently and economically. Sketching seems at first to be easier than operating CAD software, but it’s nonetheless a skill which few people get really good at, and it may be a dead end to assume that just because an interface is easier to understand, it follows that it is easier to achieve satisfactory results. This is the mistake that Neil Gershenfeld makes in Fab when he writes

“… CAD tools still share a particularly serious limitation: they fail to take full advantage of the evolution of our species over the past few million years. The human race has put a great deal of effort into evolving two hands that work in three dimensions… A frontier in CAD systems is the introduction of user interfaces that capture the physical capabilities of their users… The most interesting approach of all is to abandon the use of a computer as a design tool and revert to the sophisticated modeling materials used in a well-equipped nursery school, like clay.”

CAD systems take advantage (not perfectly, I admit) of what computers are good at – fast, accurate computation and simulation. Yes, it takes a long time to learn how to use CAD, but it also takes a long time to learn how to sculpt in clay, and get exactly what you are after. The “physical capabilities of their users” might simply not be able to make the 3D form in clay, at which point you end up with an unsatisfactory product and the need to use a computer. The same could be true of I♥Sketch: without significant ‘help’ from the software, a user might end up just making marks on a screen, rather than creating a satisfactory (to the user) design.

Which brings me on, finally, to  a recent development by Ponoko, called Photomake. This system allows a 2D sketch to be used as the design drawing for a product, which Ponoko manufactures by laser cutting a material of your choice. The 2D sketch has to be photographed or scanned, then uploaded to the Ponoko website, where software will interpret the design and create a final design. Essentially the software ‘vectorises’ a bitmap image, but anthropomorphising the process allows it to be thought of as the software deciding what the drawing really means. This is similar to what happens when one designer looks at another designer’s sketches, and it can be amazing how designers, who have been trained to ’speak’ the same language, can read each others drawings, whereas a non-designer will miss much of the information, or else place too much importance on a non-essential detail. It seems to me that unless non-designers are expected to become proficient in either sketching, or some kind of 3D modelling, the only way for them to satisfactorily communicate their design intent will be the development of software which is able to interpret their wishes. This software may have to guide them, and it may insist they are not able to do certain things, but it’s job will be to do what Photomake does, except with three dimensional designs.

Photomake © Ponoko

Monday 22nd saw the official launch of Superstruct, an online ‘game’ devised by the Institute For The Future, which aims to be the world’s first massively multiplayer forecasting exercise. Currently in preview mode, the game is due to be ’switched on’ on October 6th, but some of the responses to Superstruct’s scenarios already show the level of creativity that might be expected.

Superstruct presents a scenario set in 2019, in which the Global Extinction Awareness System (GEAS), a computer simulation that predicts the extinction of the planet’s species, has forecast 2042 as the year in which the human race dies out. According to the Superstruct press release, this forecast differs from previous doomsday predictions in that

“GEAS does not link this extinction to a single factor. No one disease or war or environmental hazard poses a sufficient danger to draw us to this conclusion. Instead, it is the combination of factors, each below the threshold necessary to put our survival at risk. These factors–which we are calling “super-threats”–reinforce each other in substantive ways, creating a set of conditions that we believe capable of ending the human experiment.”

The five superthreats © IFTF. Click for larger image

The superthreats will form the basis of the game – the idea is that players will use blogs, videos, social networking sites etc to imagine and discuss their lives in 2019 and how the threats have shaped those lives. Until the game is launched fully it’s difficult to know exactly how it will work, but the grander ambition is that players will work out solutions to the problems the threats pose, thus feeding into some of the Institute for the Future’s own forecasting work. The ITFT is careful to avoid talking about prediction however; speaking in Discover magazine Jamais Cascio, one of the game’s directors claims that:

” ‘Future forecasting is all about testing strategies – it’s like a wind tunnel’… Now, instead of having that wind tunnel be designed and constructed by what Cascio calls “our little hermetic cabal of thinkers,” the institute is handing the toolbox to the unruly mob.”

Whilst it’s clear what Cascio means, this view has to be balanced by what Superstruct actually presents to those who want to become involved. The five threats are titled Outlaw Planet, Power Struggle, Quarantine, Generation Exile and Ravenous, and are described as follows:

Outlaw Planet: “In 2019, the mobile internet and sensor networks we rely on to hold our societies together are being hacked, griefed, and gamed.”

Power Struggle: “In 2019, we’re all caught up in the ‘alternative fuel’ wars as the world fights over what will take the place of oil.”

Quarantine: “In 2019, Respiratory Distress Syndrome is here, and it’s not going anywhere. Outbreaks are just something we live with.”

Generation Exile: “In 2019, our neighbours are climate refugees and economic collapse victims who are swarming the planet, looking for a place to live.”

Ravenous: “In 2019, the food chain is broken. So we’re inventing new ways to feed ourselves.”

It is of course easy to criticise such scenarios, and given that they are set in the future there’s little a scenario’s author can do to ‘prove’ their accuracy. Personally I think they are pretty interesting, with the right mix of present-day relevance and future possibility. But it’s quite clear that players of the game are not, as Cascio would have it, designing and constructing a wind tunnel. At most they are deciding what goes in the wind tunnel, and what they’re certainly not doing is deciding whether a microscope, a telescope or whatever, might be a more appropriate tool for looking at things.

This is interesting to me, because it touches on one of the issues I am investigating – how much control do you give people to design their own products? In Superstruct, each superthreat contains a number of questions which the players are expected to consider. In Generation Exile for example, it is proposed that climate change refugees will migrate from low-lying areas, forcing others to find ways of accommodating them. This may well be a realistic scenario, but it doesn’t allow players to suggest ways that such migrations might be prevented in the first place.

Superstruct Challenge from Generation Exile © IFTF

There are societies which have found ways of coping with rising sea levels (dykes in the Netherlands, houses on stilts in Bangladesh) or extreme temperatures (Australian aborigines or North African nomads); perhaps it would be possible in the 23 years between 2019 and 2042 to learn something from these cultures. Perhaps also some countries would declare refugees “a threat to our way of life” and use lethal force to prevent them entering. But the game doesn’t appear to allow for such outcomes. In design terms, the brief is too restrictive.

There are sound reasons for taking such an approach; Superstruct is, after all, a game, and good games need strong story lines and rules. But that’s where the idea of a game as a forecasting tool might break down. It’s possible that when the game is finished, the way the Superstruct world looks will be largely a product of the starting scenarios. The other issue is that all five superthreats are quite obviously dystopian views of how the world might be progressing ten years from now. Dystopias are generally much more interesting worlds to play in (how many computer games are built around utopias?) but they are also much easier to imagine. It’s far harder to conclude that in 2019 life will be largely the same as it is today, and then to predict the impact of the things that are different.

This is another factor I’m acutely aware of in the PhD – the difficulty of predicting a technology’s impact accurately, and not getting swept up in either the enthusiasm of its proponents or the pessimism of its nay-sayers. It has become a maxim in forecasting that “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” The designers of Superstruct are undoubtedly aware of the maxim, given that it was first formulated by Roy Amara, a past president of the Institute for the Future. But it seems the significance ascribed the superthreats over a ten year period was again the result of the balance between the need for accurate forecasting and interesting gameplay.

© The Weather Project 2019

Given that the game proper has not even started yet, it has already generated a lot of content from those interested in playing. Jane McGonigal, another of the game’s directors, has eleven pages of responses on her blog to a request to players to imagine where they’d be having dinner in 2019. She also reports receiving a package from the 2019 Weather Project containing eight glass bottles with instructions to fill them with air samples and return for inclusion in a weather tracking system. ARK (Art Replacing Knowledge) have released an official response denying responsibility for the hacking of the Republican Bank of Malaysia reported in the Outlaw Planet scenario. And reBang 2019 is posting images of his sea voyage from Hamburg to New York. What’s also interesting is that when searching for players, as well as turning up sites such as the Daily Dystopian and The Gupta Option, I found some non-Superstruct sites which really looked like they should belong. My favourites are Project 2019, a “Movement by Black Americans to gain educational parity with the rest of America by 2019 (The 400th Anniversary of the beginning of slavery in America)”; and this one, a film by David Bray looking at a “holistic approach to National Security” brought about by the hyperturbulent pressures of globalisation in 2019.

“A small unidentified vessel tailing my ship out of the English Channel as we head into the Atlantic Ocean” © Rebang 2019.

Personally, though I’m going to be following the progress of Superstruct closely, I’m not going to be participating. The reason is that right now I’m working for a client developing future scenarios, and although the timescale isn’t exactly the same some of the issues would definitely present a conflict of interest. When Superstruct is over, it will be interesting to see how closely its world corresponds with the project I’m involved with. I also wonder whether the game will actually ‘end’ when it’s supposed to, I can already imagine that some of the player’s constructs will take on a life of their own…

The previous two years conferences have focussed primarily on the engineering aspects of rapid manufacturing. Although there were again some very technical presentations this year, it also
seemed to be a definite aim of the conference to look at how these technologies are breaking out of R&D labs and getting into the hands of those exploring the design possibilities, and the societal implications, of RM. Frank Piller gave a great presentation on mass customisation and the way in which rapid manufacturing’s ability to create ‘one-off’ products is a natural extension of this. Evan Malone of Fab@Home, and Kathy Lewis of Desktop Factory both gave inspiring presentations on the way in which consumers are taking RM technologies into their own hands. But most interesting for me were the presentations of Assa Ashuach and Lionel Theodore Dean, two designers whose processes are integral to their experiments in pushing the limits of what rapid manufacturing can achieve.

Talking to Assa the night before, he was a bit concerned about how his presentation would go down with an audience primarily made up of engineers and material scientists. He needn’t have worried, most people were fascinated by the way that the technologies and materials they were responsible for developing were being used in ways they had never envisaged. Assa started by showing The AI Light, a pendant lamp which uses sensors to understand its environment, and which reacts by flexing and twisting in response to what it senses.

AI Light © Assa Ashuach

The AI Light is made in nylon using EOS’s laser sintering process. Inside each wing are two actuators, one to control bending and one to control twisting; these allow the light to perform fluid, organic transformations, rather than harsh, robotic movements. The ‘AI’ refers to the way in which the light learns from its surroundings, and allows what Assa calls “training rather than controlling”

“When you first invite it into your home, you have to let it get accustomed to its new environment. Once it is relaxed, the training can then begin. It has five senses that track changes in its environment and slowly it develops a set of behaviours that indicate a new character to each light. The user is also able to interact with the light by playing with it through sounds, light and movements. This smart structure may behave in unpredictable ways if moved to an unfamiliar space.”

AI Light © Assa Ashuach

Assa worked on the AI Light with Complex Matters, a company run by Dr Siavash Mahdavi which specialises in the design of custom materials, often using rapid manufacturing technologies. These custom materials are cellular microstructures, engineered to display different properties in different parts of a product as the application demands; for instance a material might be rigid and stiff in one direction, but soft and flexible in another. It is this kind of structure that allows the AI Light to flex.

Assa first collaborated with Complex Matters on the design of the Osteon chair, which was also shown in his presentation. Assa described the process of design in this project as “finite element analysis in reverse”: first a set of ‘ideal criteria’ were formulated, then the material structure was designed to meet those criteria.

FEA image of the Osteon chair © Assa Ashuach and Complex Matters

Internal structure detail © Assa Ashuach

The Osteon chair was again manufactured by EOS in laser sintered nylon, and can be described as a cosmetic skin stretched over an intelligent internal structure. The result is a continuous flowing curve whose form is unmistakably derived from the tools and capabilities of CAD surfacing software. But the form alone does not tell everything about this chair – one of the most interesting features is that by designing the material specifically to meet the needs of the product, the material needed to manufacture the chair was reduced by 2/3’s. This is significant in any industry where a high strength : weight ratio is required, aerospace for example, but also has implications for the design of environmentally sustainable products.

Osteon chair computer rendering © Assa Ashuach

Osteon chair © Assa Ashuach

The technique of ‘finite element analysis in reverse’ was also utilised in another furniture project, again designed in collaboration with Complex Systems. A custom designed material was developed with the aim of using the minimum volume of material possible to support a load of 120kg at a height of 40cm. Manufactured by Materialize .MGX, the AI Stool is designed to be soft in the areas which which the sitter contacts directly, but rigid in the areas which support the sitter’s weight.

Computer image of AI Stool internal structure © Assa Ashuach and Complex Matters

AI Stool © Assa Ashuach

Lionel Theodore Dean is the driving force behind Future Factories, and was one of the first designers to understand and begin to explore the ability of rapid manufacturing to produce individual, unique products. As with Assa, the processes Lionel has developed to design these products are as interesting as the products themselves. But rather than custom designing materials and forms to meet a specific need or requirement, Future Factories’ processes introduce an element of chance, often relying on software to evolve a shape in ways that the designer cannot fully control.

Future Factories describes these processes as ‘computational design’. Lionel describes this concept in the catalogue for Digital Design Futures, an exhibition recently held with Justin Marshall of Automake at the Hub exhibition space.

Rather than creating a single discrete design solution, a meta-design is created that defines the function and character over a potentially infinite range of outcomes. The aim is to create coherent recognizable designs but with obvious differences between iterations… There is a balance to be found between freedom and control. A random element is necessary to create something unique; too random and the identity is lost.

An example how this kind of process can be implemented is the idea behind the Tuber pendant lamp. The concept was for a website which ran an animation in which the form of the lamp was continually changing, morphing from one shape to another as dictated by software which ‘evolved’ the design. At any point the customer could ‘freeze’ the animation and order the resulting product, which would then be rapid manufactured in a plaster-based composite material.

Tuber lamp design iterations © Future Factories

An interesting question was raised at the end of the presentation, regarding how much control the user should be given over the design of the Tuber lamp: had the possibility of allowing the consumer to interact with the morphing of the design been considered, rather than leave it to software? The answer was yes, it had been discussed a number of times, for example by introducing slider bars which would control different elements of the design. But for Lionel, this wasn’t about consumer-generated design: the Tuber is a ‘designer’ lamp, it comes from the creative skills of one designer, it’s just that each lamp is different.

Lionel began to see limitations in this process however, in that it relies on the manipulation of pre-existing geometry in a CAD model. As such it was only capable of ‘adjustment’, rather than fundamental change. This was addressed in a later project, ‘Holy Ghost’, which combined the notion of morphing with another process, that of ‘building block’ additions.

Holy Ghost iterations © Future Factories

Like most designers I suspect, I had seen images of the Holy Ghost chair previously. Based on Phillipe Starck’s Louis Ghost chair for Kartell, it had received a lot of press since first being shown. What I had never read about though, was the process by which the chair is designed. The back of the chair consists of a number of elements Lionel calls ‘buttons’, and the first step is to decide how many buttons will be used; a computer script then randomly places these buttons within a three dimensional ‘envelope’ which determines the shape of the back. In the second step the script ‘expands’ the buttons in a uniform manner until they touch. Finally the individual buttons expand in a non-uniform manner to take up the available space, this is what results in differently sized buttons. A series of springs link each button allowing the whole of the back to flex, and the part is manufactured in SLS nylon.

Holy Ghost © Future Factories

Future Factories’ latest project is Icon, a limited series of 100 individual titanium pendants. Lionel had experimented with jewellery pieces in the past. Initially these had been made by rapid prototyping in wax and then investment casting (the Perfactory process) but later this was changed to metal laser sintering, a more direct process.

Jewellery in conjunction with the Jewellery Industry Innovation Centre © Future Factories

This change of manufacturing process in turn required a change in design, primarily to avoid support structures which required much more hand finishing. Initially the new pendants continued to be hand polished, on the outside surface only, but later an automated process was adopted, polishing both internal and external surfaces. Since titanium cannot be soldered, the Icon pendants would be virtually impossible to produce by conventional manufacturing methods. But the Icon series also demonstrates, Lionel believes, the possibility of individualised designs which nonetheless retain an identifiable ‘meta design’. If this meta design were understood in terms of design language, it could be a powerful indication of the way in which traditional manufacturers might retain their brand image in a future where a huge increase in variation is possible.

Part of the Icon series of pendants © Future Factories

It’s virtually impossible to be interested in fabbing or mass customisation and not know about the cool stuff Ponoko are doing, both in enabling consumers to manufacture self designed products and in providing a marketplace for those products to be sold. And the news that they are establishing a global head office and manufacturing facility in San Francisco hopefully shows that Ponoko is already doing well enough to start expanding. I know I’m a bit late in posting about this announcement, but what I found especially interesting was the emphasis placed on the environmental benefits of this new set-up.

Ponoko has appointed Graham Hill, founder of Treehugger, to its board of advisors, and writes in its press release that

“Being able to make products on-demand, close to where people live, reduces waste and cuts down on the carbon emissions associated with transporting products to consumers. Our facilities in San Francisco mean that we’re starting to see this become a reality in the United States, and the appointment of Graham to our board of advisors is a huge endorsement of Ponoko’s vision for a more sustainable approach to the way goods are created, made and delivered.”

In the original plan for my PhD I proposed to look at some of the environmental implications of rapid manufacturing. Unfortunately it’s one of the parts which has been shaved off as I focussed the research and got to grips with exactly what I’d taken on; indeed I’ve no doubt that there’s a whole PhD waiting for someone who’s interested in this area. But it’s still an issue I’m interested in, and up until now the positive possibilities of local manufacturing facilities isn’t something that had occurred to me.

The first, somewhat obvious reaction to the notion of consumer-oriented manufacturing is that it has to be a bad thing environmentally. If the means of production are brought closer to the end user, both physically and in terms of when manufacture occurs in relation to sale, production becomes easier and cheaper and so it’s necessarily valued less. As more things are produced, correspondingly more things will be discarded.

There are a number of reasons to question whether this will necessarily be the case though. The first, which seems to be talked about increasingly, is the possibility of using recyclable materials in the rapid manufacturing process. 3D printers such as those from Z Corp already recycle unused material within the machine, but in theory the part itself could be made from materials which can later be recycled, in the same way that conventional plastics or metals are recycled today. However this seems to me to be something of a false argument: whilst recycling discarded waste is a good thing, it’s undoubtedly better not to produce the waste in the first place. As such, arguing that recyclable materials will lead to more environmentally sustainable practices just avoids the issue of whether rapid manufacturing will lead to shorter life cycles for products.

A better argument to my mind, is that whilst some parts of a product might be discarded more often, other parts will be discarded less. In the consumer electronics field where my work is concentrating, consumers will replace products for three reasons – to gain access to new technology, for fashion, or because the product is broken. Where a consumer is replacing a product because they want the newest technology, rapid manufacturing may not have an answer, but in the other two circumstances it may in fact lead to a reduction in waste.

Reuters recently reported on a phenomenon it called ‘Urban Mining’ – recovering precious metals from the circuit boards of electronic waste, which has become increasingly lucrative as the price of gold, silver and copper has risen. According to the Yokohama Metal Co Ltd, a recycling firm quoted by Reuters:

A tonne of ore from a gold mine produces just 5 grams (0.18 ounce) of gold on average, whereas a tonne of discarded mobile phones can yield 150 grams (5.3 ounce) or more.

Scrap metal from discarded electronics products © Thomson Reuters

It’s an almost universal complaint that these days phones have too much functionality, a complaint that’s backed up by a report I remember seeing which showed that 80% of a typical phone’s features are used once or less by its owner. And yet still we’re seduced by the look of the latest model. Imagine if instead of replacing the entire phone you could change the entire look and feel by adding new covers, or other parts. The old ones might be thrown away, but the guts of the product, the technology which contains the lead, cadmium, arsenic and mercury which are poisoning poorer regions of the world, would instead be retained.

A similar argument applies when it comes to products which are broken. Although it’s obviously annoying when a product breaks, the vast majority of certain kinds of consumer electronics (mobile phones, MP3 players, laptops, DVD players) are discarded long before they wear out. The exceptions tend to be those products where for whatever reason fashion plays less of a role in purchase decisions (refrigerators, vacuum cleaners, microwave ovens etc), and so it somehow becomes okay to hang onto them for longer. But even here, it’s rarely that the whole product is broken; the problem is that as a product gets older, the chance that the manufacturer continues to support it by making spare parts decreases. It can even be that repairing the product is more expensive (and almost certainly more hassle) than buying a new one – in the UK, Dyson vacuum cleaners are one of the few household products people take for repair. Undoubtedly this is partly due to the initial high cost of the machine relative to other cleaners, but it’s also encouraged by a five year warranty, together with Dyson’s commitment to continue manufacturing spare parts for all machines back to the first model made in 1993. In future, rapid manufacturing might further encourage products to be repaired rather than discarded, if for example Dyson only needed to make the 3D files for spare parts available for download, rather than keep inventory of old stock. Consumers might print spare parts themselves, or visit repair shops where parts could be printed and fitted the same day.

Dyson DC02 vacuum cleaner from 1995 © Dyson

There’s one final reason why rapid manufacturing may not lead to an increase in waste. In her PhD thesis, Ruth Mugge looked at reasons why people become attached to products, and why some products more than others. She argues that one way consumers develop an emotional attachment is through an involvement in the product’s design, or through customising the product during ownership. A product in which a consumer has invested time, thought and creativity becomes more valuable to that person, and consequently they are less likely to discard or replace it.

This brings us back to Ponoko again. Because in enabling people to design and manufacture their own products, it’s difficult to imagine those products are then discarded with as little consideration as something bought off the shelf. Even those who visit Ponoko just to buy designs that someone else has created are able to engage with the product’s story, and the designer behind it, in a way that is impossible in a normal purchasing experience. I’m still a long way from being convinced that rapid manufacturing will be good for the environment, but I’m starting to see why it might at least be better than the conventional methods of manufacture we have now.

Following on from my previous post regarding consumers designing their own products, one of the first problems is obviously how consumers will do this. It’s likely that the pioneers will be those who aren’t intimidated by the effort it requires to obtain and then master a 3D CAD package. But given the steep learning curve and the frustrations even experienced designers sometimes encounter when working in CAD, it’s unlikely that any more than a small minority of non-professional designers will be willing to invest the time it takes to become a proficient 3D surfacer.

So starting from the standpoint that helping consumers to design their own products is a worthwhile aim, it becomes important to look at the kind of tools which will enable this. Part of my research will look at developments in CAD software – Google SketchUp, Cosmic Blobs, Teddy etc – and ask whether it will ever be possible to produce stylish products using such ’simple’ tools. But I also want to look at other ways that consumers might be empowered to design products, ways which don’t necessarily imitate the processes that professional industrial designers currently use.

I’ve known about GenoForm, a software package from Genometri, for some time now, since a discussion on Core 77. For the most part it has been presented as a tool for industrial designers, one which allows them to explore design variations, though I think it’s fair to say the reaction on Core 77 was mixed. The main argument against seems to be that generating and filtering design variations is an activity that designers do intuitively, and that GenoForm is a ‘brute force’ method, rather than a creative method, of arriving at the best solution. Nonetheless, it seems to me that an iterative design tool may be valuable to those consumers who lack the skills to design their own product from scratch.

GenoForm works as a plug-in for Solidworks, it also requires Microsoft Excel to be installed in order to generate Solidworks design tables. It is available as a trial license through the link above.In this first round of testing, I’ve used a relatively simple vase form, made by revolving a sketch by 360°. The sketch is a spline, constrained at each end and at its inflection points:

After revolving the spline, a base is added and the surface is knitted, thickened and filleted to create a solid. (Note: this isn’t a requirement of GenoForm, which will work with both solids and un-knitted surfaces):

Opening up GenoForm loads the active model – the one which is currently open in Solidworks. The relatively simple UI allows the ‘Creativity’ to be set – basically the extent to which the software will vary the model’s underlying dimensions – and after pressing ‘Go’ GenoForm will generate variations of the original form:

At this point, GenoForm has varied every dimension in the model’s history tree, to an extent determined by the ‘Creativity’ slider. That includes both sketch dimensions and feature dimensions (such as fillet radii, thickness etc). However GenoForm allows much more detailed control over which dimensions are varied, by giving access to what it describes as the models genes:

Here it’s possible to choose which dimensions are varied and which are fixed at the original value: in the screenshot above I have kept the angle at which the spline meets the base constant, as well as the amount the sketch is revolved (360°, to ensure a watertight vase). The slider on the right also allows fine tuning of the variations as a plus or minus percentage of the original value, such that one dimension could be varied a lot whilst another might be varied only a little, and only by increasing a value, not decreasing it. It quickly becomes apparent also that naming critical dimensions in Solidworks is a good strategy.Having set the genes which I wanted to vary another set of variants were generated. GenoForm allows the ones which look interesting to be saved to an album where they can be filtered or exported back into Solidworks. Unfortunately there’s a problem with the implementation of the album, which is that the image is scaled with the window, meaning it’s difficult to know exactly what shape the object is. It’s not so much a problem in identifying extreme differences between variants, but if you were using the tool to assess small changes I can imagine it would become annoying:

After deciding which variants are interesting, the album can be exported back into Solidworks. GenoForm exports the data as an Excel file, which Solidworks will import as a Design Table:

The variants can then be accessed using Solidworks’ Configuration Manager:

and modified in the same ways as a conventional model, for example by going back into a sketch and tweaking the dimensions which GenoForm has assigned:

Although this was a relatively simple test, a number of things struck me about the way GenoForm works. To begin with I felt that some of the criticisms voiced on Core77 were right, in that this is essentially a tool to replace a designer’s intuition. But as I played around a bit more I was surprised by how GenoForm can actually be used in a creative way, if you choose the parameters to vary carefully. It also quickly became obvious that to play to the strengths of the tool requires an approach during modelling which is usually referred to in best-practice techniques, ie think carefully about which dimensions might be varied later, and the relation of features to one another, rather than just their positioning.In terms of the way s that consumers might use such a tool, I can envisage a design process where the industrial designer decides which parameters of a design are fixed, and which might be changed by a consumer. It’s then quite easy to imagine a web-based tool in which a consumer picks some or all of those parameters and then generates variants, as many times as necessary to reach a product the consumer is happy with. It’s arguable whether this is really design of course, but it does seem a valid way of ensuring uniqueness.To take the investigation to another level, the next stage will be to build a more complex model, and try to really understand the decisions needed to get the required level of variation. I’ll be interested to see whether complexity adds to or subtracts from the creativity of the variants with relation to the original.