Here, Thomas Teger talks about the challenges designers are facing in today’s digital product development process and product lifecycle management, and how existing tools are addressing their needs only partially at best, if at all. Thomas is co-founder and CPO of swatchbook and has over 20 years of experiencing in building software solutions and bringing them to market. swatchbook is a material lifecycle management platform in the cloud that fundamentally changes the exploration, visualization and sharing of materials.
I’ve been in this industry for almost 25 years. I visited companies and design studios all over the globe. Companies that design some of the most beautiful and desirable products in the world. Products where design is at the core (I’m trying to avoid saying where design is everything). Products that live and fail by the design. Yet when I look at the tools that are being used and the processes that are in place, it is sheer dumbfounding how detached design and the whole design process is from the rest of the product development process.
Design(er) ≠ Engineer(ing)
Of course, in order to put this article in the right perspective, I have to define what I mean by “design”. I am talking about the aesthetic aspects of a product.
“Design”, whether it is noun or verb, and “designer” are 2 of the most mis-used (abused?) words in the English language. Being a native German speaker, I never found this problem in the German language.
First and foremost design IS NOT engineering. Over the years I found that many engineers call themselves “designers”, but I’ve yet to meet a “designer” who calls themselves an “engineer”. Design and engineering programs at universities are completely different, with completely different objectives. And even design demands further distinction. Depending on the industry, there are industrial designers, automotive designers, fashion designers and more. When talking about engineering in this article we refer to “mechanical engineers”.
So for the sake of the discussion, “designers” are the people that make objects look pretty and work well, while engineers turn these ideas and concepts into products that can be manufactured and sold.
Yes, I understand that this is extremely simplified, but it helps move the discussion along.
It is clear that one can’t live without the other. Engineers might think differently, and of course that depends on the product. But whenever there are aesthetics involved, they just can’t. And it shows.
The fact that the terms designer and engineer are being used so interchangeably in today’s industry leads directly to the problems that we are still facing in software.
Design is an afterthought …if you are lucky
Digital tools for engineers have been around since the 60s, at least for 2D, and started to go 3D in the 70s. These tools are used for “CAD” – Computer Aided Design. See the problem? “Design” here really means “engineering”. So correctly you would call them “CAE tools”. But applications in that category are actually used for simulation and analysis of products. Confused yet?
These CAD tools have been built from an engineering perspective. Starting with prismatic shapes, boolean operations to add and subtract it finally arrived at parametric, feature-based solid modeling in the late 80s with the creation of the product Pro/ENGINEER (known as PTC Creo today). Yet again, as incredible as these applications were and still are, it left the designers behind. It didn’t give the flexibility a designer needed, didn’t cater to the non-structured thought process a designer is used to, and was always, always (and still is, decades later) too difficult to use.
Digital tools for designers started with the 2D paint application with the birth of the Mac in the early 80s. While there were some 3D applications out there – who remembers Form-Z? – it really wasn’t until the early 90s that dedicated 3D tools for designers came around. These tools were termed CAID (computer aided industrial design) or CAS (computer aided styling) solutions.
Alias was the company and the product that grew to an amazing adoption in the industry, and is still today the standard design tool used in Automotive design studios around the world. A competitive product to Alias was CDRS by Evans & Sutherland. While not significant from a market adoption standpoint, it is worth mentioning it to create a reference for later in the article.
While Alias carried a hefty price tag, a much cheaper and quickly very capable product came around in the late 90s which took the industry by storm: Rhinoceros (aka Rhino). These tools addressed the needs of designers by allowing them to build organic shapes much more easily, without the usual restrictions of CAD tools.
Another product that found rapid and huge adoption in the design space was SketchUp, a tool originally developed for architects. We will talk about this later.
With the invention of new technologies, like mesh modeling and sub-D surfacing, other tools came to market, giving designers (creative people) a different tool set that allowed them to “sculpt” their designs in the computer, just like they would do in real life. Tools that dominate the market here are Z-Brush, Mudbox, and Freeform.
Now capable of building your 3D digital models, you want to be able to “visualize” them, i.e. be able to make pretty pictures. CAD products never went past phong shading. In order to be able to visualize, and really create a “pretty picture”, you had to turn to the dedicated design applications that allowed users to create a “photorealistic” image of their designs.
Over the years more and more dedicated rendering applications came to market, whether they were plugins to modeling applications, or standalone applications catering also to the entertainment market. Products that found great adoption were 3DS Max, Maya, Cinema 4D and others. We are still talking ‘90s.
Newer developments that made an impact in the industry were V-RED by PI-VR (acquired by Autodesk), and KeyShot by Luxion.
The UI experience
Most CAD, CAE, CFD, CAID, CAS, sculpting and visualization solutions over the years suffered and still suffer from the same problem: they are extremely complex.
There are 2 parts to complexity: the complexity of the task, and the complexity of the tool. In order to be good at creating digital designs, you must be able to master the tool, and be able to master the task. This is still the number one challenge today. And just because you are good at mastering the tool doesn’t mean the results are beautiful.
Being able to model something is not easy. Especially when it comes down to advanced surface modeling, it is a complex task that requires years of practice.
Visualization is the same. While these applications are incredibly capable they require significant of time to master the complex user interface and all the options.
While these tools are catering to the needs of individuals in their different roles in the product development process, the user interface is extremely complex. Most of it has to do with the fact that over two decades worth of features and functions have been added to the programs to make them more powerful, while paying the price of giving up ease of use.
And as a designer you don’t think about this when you go to school to get your design degree. You don’t envision your job being something where you flip through menus trying to find the right function, or filter, or setting to create that awesome class-A surface, that killer visual effect that makes your rendering pop.
Over the years there were several products that broke down the barrier of complexity and provided an easy to use solution that catered to the masses for quick adoption. 2 products that stand out are SketchUp for modeling, and KeyShot for visualization. Both products took the approach of “less is more” by hiding the complexity and rather putting the tools in the hands of the designer, allowing them to focus on the objective and the goal, supporting the creativity of the designer rather than dictating the quality of the outcome by how well you mastered the tool.
Bringing it together
Aside from the task complexity and UI complexity of all these tools, there is yet another challenge: the complexity of interoperability. In other words, how do you get the data from one application to another? Or, from the design application into the engineering application, or from the engineering application into the visualization application?
Data exchange is a whole industry in itself. Data standards such as IGES, STEP, JT, 3DXML, OBJ, FBJ, ALEMBIC and more have been developed to make exchange of data between applications easier.
Companies developing these application have also recognized this need, and are providing APIs or SDKs to allow 2 applications to “talk to each other” with minimal loss of data. There are even third party companies that specialized on data exchange and built a whole business around it.
While these standards help, they are by no means perfect. It is an ongoing challenge, and may never be perfect. But it certainly helps tremendously to break down yet another complexity in the product development process.
How to design a better product? Kill the one you have. Start fresh. Re-think the approach.
BMW does it. Apple does it. Adidas does it. Every product and fashion company does it. They are starting from scratch (for the most part) and develop a brand-new product. Each and every time.
Software companies? Not so much. Many of the mature 3D modeling and visualization applications that have been around for 20 or 30 years are just adding feature after feature from release to release. This phenomenon, known as “feature-creep”, no longer serves the masses but rather the individuals who scream the loudest or pay the most money. And in the end it does one thing and one thing only: it makes the product more complex and harder to use.
This is true for every product out there. Whether it is a CAD application, a modeling application, a rendering and visualization application. Eventually even the easiest ones will get harder and harder to use due to more features and functions.
Taking the approach of “starting over” for established companies with a mature product is not easy.
Why don’t the CAD and modeling companies do it?
Companies that built CAD products that are now dominating the industry such as Dassault Systèmes, Siemens (formerly Unigraphics and IDEAS), Autodesk, and PTC grew into billion dollar companies just by building CAD tools for engineering and manufacturing with 100s of thousands of users. So why not go after design if design is so important? Everybody talks about it, shows pretty pictures on their website, but invests so little.
Fundamentally, it is a business problem. While the importance of design in all areas is ever increasing, there are only so many designers that are needed when compared to the number of engineers that bring it all together and make into a workable and manufacturable product. The ratio between designer and engineers is at least an order of magnitude higher for engineers.
As mentioned before, many of these systems have been around for over 2 decades. 2 decades, millions lines of code resulting in many decisions, features and workflow to solve very specific problems create a complexity that makes it very difficult to introduce different paradigms that are suitable for different needs.
How about fundamentally changing an existing product? That is also more easily said than done. Most industry leading products have been architected over 20 years ago, and have been built on ever since. Redesigning the application for a better user experience is a massive under taking, and does literally take years to accomplish. And when doing this, you are always running the risk of alienating your longtime users which can lead to long implementation and retraining efforts while running the risk of losing productivity, at least for the short term.
Companies have tried to solve this, though. PTC, Siemens and Autodesk made significant investments over the years in acquiring companies that were offering complementary solutions targeting different markets, in an effort to cover the entire product development process from design to engineering to manufacturing. And some good work came out of it.
PTC, for example, took the modeling approach of a product called CDRS, a surfacing tool targeted towards automotive and industrial designers, which it had acquired in 1995, and re-created this modeling approach inside Pro/ENGINEER (now PTC Creo), while fully integrating it into the parametric, feature based modeling kernel that was at the core of the product. While this creates an incredibly flexible and powerful modeling environment it still suffers from the complexity of the entire application itself, and the UI challenges.
Autodesk took a different approach with the acquisition of a technology provider called T-Splines, which created a new approach to solve complex surfacing tasks. It took this technology, created a dedicated team of designers and engineers and created a new product called Fusion 360, rather than trying to marry two applications. The result is respectable, yet the product itself is not quite there where it needs to be when it comes to solving more complex modeling tasks.
Over the last 10 years we have also seen a number of new CAD companies pop up that started from scratch. Companies like SpaceClaim and onshape are two examples. While both companies developed interesting solutions, they also missed a tremendous opportunity by not addressing a broader market. In the end these companies created yet another CAD tool, not even thinking about tools and workflows for designers. When you look at the founders at each companies, and the team assembled around them, it becomes clear why that is. They are all engineers, and were involved in other CAD companies before. So in essence they are doing the same thing over again, maybe with a new approach, taking advantage of new technologies that are available today. But design is not on their mind, it appears. And if it is, it is, yet again, an afterthought.
The PLM dilemma – further down the habit hole
To take it a step further, there is PLM. PLM (Product Lifecycle Management) applications and systems were developed (I’m avoiding the term designed here on purpose) to manage CAD data and primarily engineering and manufacturing tasks, to build better products in shorter amount of times. These are incredibly capable systems that can cost millions of dollars to implement and maintain. But with this incredible capability comes complexity. And, ironically, none of the PLM systems out there actually cover the “product lifecycle”! Design is not necessarily part of the process. Neither is marketing, the other part of the creative crows in an enterprise. These systems don’t cater to these creative people at all, from both the capability and the user experience standpoint. Creative people will simply refuse to use these systems.
What happened? PLM has been conceived out of a need to manage CAD data which was first known as PDM (Product Data Management) and eventually added engineering and manufacturing specific features to manage the whole production aspect of an object.
When the first PDM systems were originally developed, they were either developed by the same companies / people that developed CAD systems, or by teams of engineers and developers, but not by designers.
The argument why these companies are not addressing is the same as it was made earlier. It is a business problem. Because design is yet again an afterthought, addressing the issue in order to get it right now is extremely resource intensive. And since the problem is not well understood at these companies, the risk of getting this right is high.
When looking at the digital landscape and the tools that are available for designers we can see that there are still huge opportunities for better software solutions that will integrate designers and the overall creative process further into the digital development process, or product lifecycle management for that matter. In particular when looking at opportunities to make 3D easily accessible, providing a simple path to go from 2D to 3D in a meaningful way.
The challenge of building complex products digitally is not going away. With all the VR/AR buzz, digital content on the web craze, guess what you need? A 3D model. A good one.
The problem is that these skills seem to go away. While many tools are taught in schools, they are not taught to the extent that they will produce experts. It is more to complement the degree, than be an integral part of it.
We are living in an app world. Apps, for the most part, are made to do one thing, and that one thing as well as possible. Apps, for the most part, have a very shallow learning curve and can be easily adopted.
Adopting the “app-approach” for the digital product development process is a way of bringing more people into the process while be the same token expanding the process itself to other devices and locations as the requirements may be less stringent and don’t require people to sit in front of the computer.
Bringing the right people together with an intimate understanding of the process, the challenges and the requirements, while adopting expertise from parallel industries like gaming and entertainment would allow companies to do marvelous things.