You can also view my gallery at Designs en Bois.

Monday, June 22, 2009

Can furniture design go Agile?

As my last post indicated, I am working a new job now an rather than analyzing the software industry, I am now a part of it. I am now a product manager at a small software firm, and I'm heavily involved in designing our brand new next-generation product. The company I work for subscribes heavily to the Agile development methodology, which basically means you estimate and code in smaller iterations, using the best information you have available at the time rather than building an entire requirements document up-front, and then setting the programmers free.

So why am I boring you with this propellerhead detail? I'm wondering if the same principals could or should apply to furniture design and construction. My current project (which I owe you guys quite a bit of video on) has taught me how difficult it can be to anticipate everything you will encounter, especially on a more complicated longer term project. Bad assumptions in planning or design, shortages of lumber, and even actual mistakes in building the piece all lead to in-process changes, requiring creativity and problem solving. For example, I will post a video shortly on one such unexpected problem that led to a fairly significant change to the entire look of the face of my casework, but I made lemonade out of the situation.

y point is, spending weeks in advance fine tuning joinery in Google SketchUp may help visualize the construction, but is the time spent in front of the computer worth the extra time if you have a jig already set up to cut those tenons? Why model what you already know, or more importantly why model what you DON'T already know. For instance, what if the species of wood you chose has really difficult grain making a through tenon almost impossible to chop and you need to switch to blind tenons? All that time modeling those through tenons was wasted. Now I'm not saying you should just grab a stick of figured walnut and start cutting joinery without a plan, I'm just saying that maybe we're better off designing only the key details up-front, and then working out the specifics as we go along. I think there is a happy medium here, and I feel like this last project has pulled me closer to that point. But I'm curious about everyone else's experiences an opinions. Can furniture design go Agile? Or is Agile perhaps what differentiates craft woodworkers from the big production shops? I think it very well could be.


Anonymous said...

I think you get to move to a new higher level woodworker Rob. The details being shoulder to shoulder lengths being paramount and equal between like parts for M&T joinery rather than meeting a plans dimension to a 1/64"

jlsmith said...

I wasn't familiar with the Agile concept so I did a little research
( and ) and while I think I understand the issue you are trying to raise, I am not at all convinced that the Agile concept provides a good model for design/construction of unique physical objects. It will take a craving out of some of my time to respond more completely but I wanted to post a response, however limited, just to let you know I found your post interesting.

Rob Bois said...

Thanks jl for posting the link to agile. I should have done that. But for the record, I'm not convinced that agile totally works for woodworking either, since agile is designed to get smaller chunks of software to the market more quickly. You can't exactly deliver just the doors to a cabinet and expect to get paid. But I do think we can learn something from the concept. I think as woodworkers we like to be in complete control of our projects, but especially for more complex pieces that just isn't realistic.

TheYurtingYeti said...


I think woodworking already uses/used the Agile concepts. Just think about a few years ago before using sketchup or CAD drawings was the woodworking norm.

All woodworkers did was draw out basic plans with potential notes for how they wanted things to go together. Then they went forward adjusting as necessary. If you think about the days when the purchase of wood wasn't such an important step, these guys could go forth with less planning because they had enough wood to go around.

With the new technology we have almost taken a step back from the artists from years back because we plan right down to the last joint. If it wasn't for the joy of human error and variation in wood/grain software programs could take a lot of the "fun" from working on the fly to produce true art.

Granted, I love the software as it allows me to visualize what I want to make. But if it wasn't for the fact that I don't want to waste wood, most of my ideas would do just fine scrawled on a napkin with my mind to piece together the rest as the project took form.

jlsmith said...

Well I must admit I don't quite see how what I read on is simplified to "estimate and code in smaller iterations, using the best information you have available at the time", but I take your point that you are not truly suggesting that woodworking should go 'agile' but simply using the concept to explore the relationship between CAD and woodworking.

I agree with Yeti. What Yeti speaks to is the long tradition of woodworking as it relates to the making of free standing objects. The free standing object leaves a lot of room for improvising and allows for the development of what one could call the artist craftsmen (ie Maloof, Nakashima, and Krenov). You can see Maloof's demonstrate this tradition by watching a series of videos at

So if Yeti is correct, then the question is why do you think differently? Perhaps it comes down to experience or lack there of. Being exposed to an experienced woodworker, (someone that learned woodworking before computers) or someone that was trained in traditional methods, one would quickly see just how important (not very) Sketchup (or any CAD program) is to their method of woodworking. A good online example of this is Tommy MacDonald (young but traditionally trained) at

Now it might sound like I am one of those anti-CAD woodworkers but in fact I use CAD (Sketchup, AutoCAD and Cutlist, etc.) in my own woodworking. Most of my work revolves around architectural built-ins (I am an architect with extensive CAD experience and I build a lot of my own designs). Built-ins don’t have the same amount of freedom that free standing projects have. CAD allows me to model complex existing conditions that can then be used to design to in CAD. If I choose I can even model how square, level or plum the existing conditions are. CAD is very useful in these types of projects but I am not a slave to it allowing it to become an end instead of a means to an end. It’s simply another tool and like any other tool you need to learn how to use it effectively. I think that maybe you have come to realize this. If you have this knowledge will serve you well as you continue to pursue the art of woodworking.

Rob Bois said...

Very thoughtful comments coming in - I love it. I think jl hit the nail on the head in describing furniture design as an art. I don't think many artists have a completed concept or model of what they are setting out to achieve before they put paint to canvas. By the same token, I've found that my best work in furniture design comes from multiple iterations, not from a cool sketchup draft. Sketchup is a wonderful tool for estimating lumber, generating a sense of scale, an roughing out a design, but I fear it also inhibits creativity when people view their model as gospel, not a starting point. As an example, in my most recent project - a drop front desk with curved legs, I rebuilt the stiles for the drop front three times to get a shape that best complemented the legs. I could not have short circuited that process in sketchup. More importantly, I shouldn't have even wasted any time trying to model the front, as it just took more time away from woodworking. I'm still trying to fine tune my use of sketchup, but my primary concern is that its incredible prevalence in woodworking today may actually hinder creativity, and influence people to think of furniture design more like industrial design, and less like an art form. By going "agile", it automatically forces the woodworker to think creatively throughout the process, as opposed to playing a robot in an assembly plant.

As a side note, most projects throw a curve ball at you some time during construction. I will post another video shortly showing how I made lemonade out of lemons on the desk project described above.

jlsmith said...

Ah Rob, I have to disagree with you about how painters produce their artwork. The idea that painters don’t really have a painting all worked out before they start painting is quite a modern concept that basically starts with the impressionist and continues (by some) through today. But even the most casual stroll through any major art museum would not leave one with the notion that painters throughout the history of art have operated with a ‘I’ll figure it out as I do it’ attitude. Pick any ‘known’ artist that lived before the mid 1800s that history has been kind enough to preserve more than just their finished works and there will be a body of work that will show just how much preparation was done for each completed piece. Painting, historically, is a very planned and thoughtful art.

It comes as no surprise to hear that you realize it’s through design development (your ‘multiple iterations’) that your designs improve. This is how design (of anything) always has been. Design (defined as ‘conscious intent’) not only requires a lot of skill, but it also requires even more effort to do well. Anyone that tells you different is simply misleading you (and perhaps themselves) in an attempt to keep alive the oldest myth about the design process, that it is a mystery and unexplainable.

My advice is do not concern yourself with the ‘incredible prevalence’ of Sketchup in woodworking but focus on your own development and enjoyment. If you must think of how Sketchup is influencing woodworking, perhaps you could think of Sketchup this way: For all its limits, there is little doubt that Sketchup has expanded the types and the complexity of the projects of many woodworkers who use it and this isn’t a bad thing.

Neil....a Furnitologist said...

Ahhh.....a bit of seeing the light....thats great stuff, your woodworking will only get better for it.

What I don't see here is that "the design process" does not dominate the path to a completed object. There comes a point where the "building or construction" phase takes over and dominates the "design process". At a point, a dicotomy occurs where the design process path narrows and the building path widens.

Also I see alot of "art" tossed around.....unless you are sculpting a new form in wood, its hard for an art critic to take the term "the art of woodworking" seriously especially a freestanding object (ie: a functioning piece of furniture) as mentioned above as a work of art, its an old argument that tells alot about the level of creating. Of course the penultimate can of worms is what the internet woodworking community is really doing is "craft". Simply, let me just say that the artist doesn't boog him or herself down with technical jargin like the internet woodworker does, craft guy's do that to understand the "skill", an artist is more concerned with "technique" no matter how it is arrived at.

For example, building a Townsend reproduction is not art it is craft, it is an old form and just takes "skill" to closely replicate. The issue is the ego that built the Townsend doesn't want to be classified with clay pots, glass blowing and weaved fiber, but he/she is. Furniture is a decorative art not a fine art, unless it is a new "furniture form", usually sculpted.

Interesting posts.....Neil