Design by committee: notes from the latest Willamette River bridge meeting [UPDATED]

Last Thursday the Willamette River Bridge Advisory Committee (WRBAC) had its first meeting since February, and the committee made a recommendation that TriMet go forward with a cable-stay bridge. It was not a unanimous decision, but the committee in recommending the cable-stay has rejected the hybrid form bridge design fashioned by Miguel Rosales.

I was exceptionally disappointed by the decision, but with some qualifications.

In the Rosales hybrid concept, Portland was getting a uniquely tailored bridge that took its formal, geometric cues from the adjacent Ross Island and Marquam bridges while also borrowing some of the St. Johns Bridge's natural elegant curves.

There were two traditional cable-stay options that TriMet and the WRBAC committee were considering, and those would have been a particularly bad choice for the city: oversized, bland, ubiquitous: a cookie cutter bridge plopped into the central city. However, the "refined cable-stay" option that the committee recommended is significantly better. It has more of the hybrid's grace, compactness and elegance - not all of it, but much more than the previous cable-stay options.

So even if the Rosales hybrid concept has lost out, at least its introduction and the support it had in City Hall, the public at large and even on the WRBAC committee has helped TriMet and its designers fashion what I think of as a much better cookie-cutter.

[Note: this paragraph added later.] If you're reading this post and wishing there were more images or renderings of what these two bridge prototypes (the refined cable-stay and hybrid) would look like, so am I. When the WRBAC committee voted last Thursday, they were using the same single image per bridge type that I posted previously ("Willlamette bridge: examing the options"), or at least they were for the hybrid. I think more needed to be done to visualize graphically both these bridges. The committee didn't have enough visually to make their decision, and as a result the focus understandably became more about budget.

As the committee members cast their vote one-by-one, each seemed to say something along the lines of, "I'm no design expert, and I can't tell enough difference between the hybrid and cable-stay to justify the extra cost."

In attendance at the WRBAC meeting was Donald MacDonald, the Bay Area architect who was selected over Rosales's firm. It seemed clear in the meeting and from  talking informally with TriMet representatives that MacDonald was chosen largely because of his firm's experience with the design-build process through which the bridge will be built. I heard the phrase "design-build" and "experience" used a lot in reference to MacDonald.

MacDonald and his firm also seem to have a particular design talent when it comes to the details of bridge design. I don't see MacDonald as a big thinker like Rosales. He wouldn't come up with an all-new type of bridge like Rosales did with the hybrid. But MacDonald will be able to tinker with the refined cable-stay design and its details to make the bridge a little less of a cookie-cutter and something a little more uniquely Portland.

The WRBAC committee is full of decent people with a broad array of public and private interests. The committee is an inclusive wide tent, and in a democratic sense TriMet should be proud of that effort.

But the committee seems to include only one or two people from the design community, and those voices are diluted by the many other interests represented on the committee. There are people from Ross Island Sand & Gravel, for example, as well as construction companies, nonprofits, other businesses. It's in their interest, but not in their expertise or experience, to be deciding about the nuances of design. 

I don't think the WRBAC committee or any other huge committee, however transparent and open to public involvement the TriMet process is, will produce anything close to a great design. There has to be some extra leadership that sees enduring design as more important than staying under budget. Or at least an array of images to make it more apparent to everyone.

This is why a project this big needs a design competition to produce a great design and a great designer. Look at the Portland Aerial Tram. That project had big problems because of the ridiculously low $15 million initial estimate used to sell the idea. But because of the design competition that produced architect Sarah Graham (of AGPS Architecture), the inevitable budget cuts and engineering did not compromise the quality of the finished product. And it's really the product that matters as much as the process, or more.

Also troubling while watching the WRBAC meeting was the information the committee had to make its decision. The refined cable-stay and the hybrid came with two very different estimated price tags: about $89 million for the refined cable-stay and about $112 million for the hybrid. It became clear during the discussion, though, that many of the extra costs piled onto the hybrid may or may not be legitimate, listed as "requirements risk" and "design development risk". The construction costs for the hybrid were also listed at $75 million compared to about $62 million for the cable-stay, even though the materials would be about the same.

"This is a little bit riskier, a bit more innovative," TriMet's Rob Barnard said of the hybrid. That's probably true, but it seems like TriMet, although not trying to deceive anyone, may have loaded up the hybrid's extra risk-based cost. I'm not an engineer or an estimator, of course. This is entirely supposition. But I do know that the cost estimator TriMet used, the National Contractors Group, is really just one retired engineer who works from his home and has ties to TriMet's engineer.

Oh, and while we're talking about budget, note that TriMet's willing to spend somewhere around $135 million. So technically the agency could afford either one.

After the meeting, one WRBAC committee wrote me an email saying, "I'd like to have a Ferrari, but I can't afford it." The hybrid, he implied, was the Ferrari.

I'd venture that the choice here is more like that between a Geo, a Toyota, and a Mercedes. The traditional cable-stay is the Geo: cheap, ugly, bland. The refined cable-stay is the Toyota: not bad looking, and affordable, but not unique or special. The hybrid is the Mercedes: more expensive, but gorgeous and timeless.

Notice that I said a Mercedes, however, and not a Ferrari. I think the committee believes the hybrid is a Ferrari: something too ridiculously expensive and unrealistic to choose. It's too bad, because the Mercedes is really the better metaphor: something Portland would have to stretch a little more to have, but would really love.

Ultimately a great building project, whether it's architecture or a piece of engineering like a bridge, needs a great client and process. I think TriMet and its WRBAC committee are just a Toyota client, and there's nothing wrong with that per se. Maybe it's too naive or idealistic for me to want a Mercedes for Portland and not a Toyota. But if you were keeping your car for a hundred years, and the financing for that Mercedes was absolutely doable--within your allotted budget, in fact--what would you go for?

I know metaphors are merely stand-ins for the real bridges, and you shouldn't have to take my word for it (nor anyone else's) that the hybrid is better than the refined cable-stay. And besides, the two are much closer now, especially since the hybrid has lost a little of its design grace since Rosales left the project - it's now missing the suspension bridge curve.

The WRBAC committee's recommendation of the refined cable-stay does not mean the decision is final. The City Council will hear from TriMet's steering committee tomorrow, and vote, I believe, on June 17.  On June 22 comes a 'request for adoption' from the steering committee. There is also a Portland-Milwaukie Light Rail Project Open House on June 11 where you can make your voice heard.

Ultimately I think MacDonald's bridge very well may be pretty good. Perhaps how the rest of you feel about it will depend on how much importance you place on this bridge being better than just pretty good and instead being something great.

Comments

[Lance Lindahl]

So now that things are relatively settled with the Wilamette River Bridge, does this mean that the design community will finally speak up in regards to the other half dozen smaller bridges that are a part of this project?

Things have been awfully quiet out there so far. Quality design matters at Riverplace, and in Southeast Portland and Downtown Milwaukie too right?

Are those crickets and blowing tumbleweeds that I am hearing?

It would be sad to see the rest of this rail line composed of nothing but cookie-cutter, pre-cast, generic, freeway ramp style bridges.

[David Benson]

One difficulty that I have in fairly comparing the cable-stay bridge vs. the hybrid is lack of detailed drawings. The only representations I've seen so far are images from from the perspective of 1000 feet distant or more.

If the non-design community is to be persuaded, we must be educated with more than metaphors and generalities. We need help in learning how to see the differences in quality that justifies higher costs. We need not just judgments, but the detailed, specific arguments, along with illustrations. Was such a critique made available to the committee and is such a critique available to the public?

[Brian Libby]

That is a GREAT point David. I wish there were more detailed drawings and renderings I could show you! Even at the WRBAC committee meeting, as people were voting on which bridge type to approve, it was based on one enlarged JPEG rendering per bridge type -- the same images that I posted last week before the meeting.

I'll completely grant you that my post is stronger on generalities and metaphors than on detailed info. I'll try to provide more of that as it becomes available. But failing the existence of that info, my observations are all I can provide.

Hopefully, though, I will have more posts about this in the future that give the content both you and I are looking for.

[Michael Davalt]

I am sorry to have missed this meeting. I was at the first one, and I was impressed with Rosales's argument for his design, particularly his emphasis on the importance of aesthetic considerations when designing a structure that we must live with for a long time to come.

I haven't seen the alternative, but the Rosales design was a beautiful, graceful structure that would have become a Portland landmark. It was superior to the cable-stayed alternative he included in his presentation. Bridges are powerful symbols, and deserve to be treated as such. Our city faces a beautiful river, and we ought to invest in a structure that becomes it.

I too believe in full participation of the citizenry, but beautiful bridges are not designed by committees. I'm sorry to see Miguel Rosales's bridge rejected.

Michael Davalt

[nathan]

brian, thanks for being the watchdog on projects like these. the po community needs someone like you who's willing to say "wait a second...."

i think, however, that this row about the bridge is a little overblown. honestly, the rosales bridge isn't all that great. and the other options aren't all that bad, though they are, as you have said, bland.

i don't know if i really believe that the bridge decision by trimet is coming strictly from a funding perspective. I believe a major part of it is that they don't want to try out new and untested designs in this financial and political climate. remember, they are currently cutting regular services, and politically it wouldnt look good to combine that with cost overuns on an a bridge that is perceived to be experimental... and that's a valid and realistic concern.

[Grant]

Ubiquitous: How many cable stay bridges are there in Oregon? How about in Washington?

[Steve]

Regarding the lack of detail views, I crossed the Sauvie Island Bridge over the weekend, and the huge tied-arch connections at the bottom of the cables on that bridge are horrible, but a good example of the type of detail to be avoided on this new project.

I found a nice image and reference here:
http://www.portlandspaces.net/blog/the-burnside-blog/2008/12/7/willamette-crossing-s-fork-in-the-road

[David]

Brian,

Do I understand correctly that the review committee made such an important decision with no more than the JPEG images of the alternatives that you posted a little while ago here? If so, I'm stunned. They lacked vital information they needed to make an informed decision. Differences in cost can be simply expressed in numbers. For design competitions good images are essential, allowing us to examine differences in size, proportion and significant details.

Off-the-shelf design is fine, I suppose, if we only aspire to be an off-the-shelf city. But what makes us most miss about our own places when we are away and makes us want to visit other places are those good things unique to here and there. Short-term savings for an average bridge, is horribly short-sighted. Frank Lloyd wright said something to the effect that a doctor can bury his [sic] failures, but all an architect can do is plant ivy. With a bridge, you cannot do even that.

And you made a terrific point about the tram, which went over budget but instantly became sculpture and tourist destination due to its beautify. Now that it's built, who would now prefer to substitute an erector set version that you can see on any mountain slope anywhere in the world?

[Bad design]

I think the framingof the discussion is all wrong. The choice was between a toyota carolla (generic cable stay), Toyota Camry (Refined cable stay) and a Hybrid Camry (Hybrid) options. Nothing proposed here was a luxury bridge and shame on anyone involved in the bridge process for characterizing it as such.

[JR]

We should wait for the Mercedes and build the project in 2025 for $2.4 billion instead of $1.4 billion in a couple years.. transit is secondary to design.

[Brian Libby]

JR, this seems like a nice attempt at sarcasm. It shouldn't have to be one or the other when it comes to transit and design. And I bet you know this. Besides, TriMet is already budgeted for what the 'Mercedes' costs. They're just not willing to take the perceived 'risk' that having a more unique design entails.