How to ruin a design: the 9/11 Memorial

Update:  After a great comment, I changed my opinion about this.

I rarely use the word requirement – it’s a weenie word. But I know if you royally screw up requirements, regardless of what you call them, no designer’s power can save you. Case in point: this story about the design of the 9/11 Memorial in NYC.

The specific problem in question was how to organize the names of all those who died:

In 2006, Mayor Bloomberg… suggested that people be loosely grouped according to their location that day. And so Arad [the architect] created nine categories. Around the south pool, he’d list everyone who died in the South Tower and at the Pentagon, along with the first responders and the passengers on Flights 175, 77, and 93. Around the north pool would be those who died in the North Tower and on the plane that crashed into it, along with the six who died in the first World Trade Center attack, in 1993.

As is often the case, it’s an executive who issues the problematic requirement, unchallenged by more practical minds, that sends the project into a tailspin.

Chronological is what worked well for the Vietnam memorial (and many others). But they passed on that. So alphabetical perhaps? Noooooo. Too easy. Perhaps by specific incident (Tower 1, Tower 2, Pentagon, etc.)? That’s where they started, but that was too simple  – here’s what they did instead:

But how to group these? Arad and Daniels settled on the idea of a distribution that would seem random, reflecting the chaotic and arbitrary nature of the event itself, but that would have some kind of underlying logic, reflecting the bonds that preceded or came of it. “One of the biggest messages of the memorial and the museum is that the people who got up and did whatever they did that morning, and then died doing it, were no different from the rest of us,” Daniels said. “They were us, we are them.” In 2009, the foundation sent out letters to the victims’ families, soliciting “meaningful adjacencies”—that is, the names of others with whom each victim should be listed.

Random, as a goal? Really? And getting everyone involved in creating requirements? Can you see where this is going?

By the end of that year, the foundation had received twelve hundred requests for adjacencies (and these didn’t include the self-contained adjacencies, such as, say, Ladder Company 7 or Cantor Fitzgerald, which, with six hundred and fifty-eight names, represented the biggest, and most challenging, adjacency block of them all) The reasons for these requests were varied. Sometimes the victims were cohorts, or best friends. In other cases, the families knew, from last phone calls, whom their loved ones had been with in the end—in an elevator, on a ledge—and wanted those people listed together. A same-sex couple and their three-year-old son all perished on Flight 175; their names, certainly, belonged together.

These are moving stories of course. And stories that should be shared. But why did these stories need to define the design of something as simple as a name list? On a project that’s been delayed for nearly a decade? Every other war memorial in the history, including the great ones, didn’t need to go this far.

At this point it was probably too late to simplify, as all the victims had already made their requests. How could they be turned away? But as any good designer knows, if it’s this hard to figure out or explain, odds are no one who visits the place will make sense of it either. But they pressed on:

At a certain point, the foundation recognized that this job could use the assistance of a computer. Even so, the first few computer scientists and statisticians the foundation got in touch with said that it couldn’t be done. “It really did seem insurmountable,” Daniels recalled. But then his chief of staff called Jake Barton, the principal at the media-design firm Local Projects, who took on the assignment, and, with a data artist named Jer Thorp, designed an algorithm that could sort the names in keeping with all the overlapping requests. Before long, they had a distribution designed to please everyone, including Arad.

I’m certain many families would be better honored by finishing the memorial sooner, and making it a welcoming and calming place for all visitors, rather than the micro details of how exactly names are grouped, or not.

I haven’t seen the latest plans for the complete memorial, and I admit the entire experience is unlikely to be ruined by this small set of issues. But I’m also confident the time spent overthinking the list of names earned much more effort than it was worth.

Requirements is a maligned word among designers, but for anyone who sees my point – pick up a copy of the wonderfully potent Exploring Requirements by Weinberg. It will forever improve how you think about problems, designs and working with clients through them both. Had Bloomberg or Arad read it, it would have saved all the victims and NYC much wasted time.

17 Responses to “How to ruin a design: the 9/11 Memorial”

  1. Phil Simon

    Great post, Scott. As usual, too many cooks… The genius of Apple is that Jobs doesn’t run a democracy. Sergey and Larry didn’t crowdsource Google. Bezos took derision from investors and the public at large in building Amazon. Zuckerber focused on an ecosystem at Facebook.

    Any coincidence that these are such powerful, well-run companies now?

  2. Rob Donoghue

    Have you seen the relatively new World War II memorial in Washington DC? Every bit of this article made me think of it.

    -Rob D.

  3. Andy West

    Wow, this is so sad and disrespectful. It’s bad enough when I see this happen to the design of some random line-of-business application that nobody will ever see.

  4. Lynn Cherny

    Huh, when worlds collide. I’ve heard Jer Thorp talk about his tool for the solution to this problem, and seen video of it. A designer himself, I think he found the problem humanly interesting and his solution to the tool design problem, working in those constraints, was really interesting.

    One person’s messed up requirements is another kind of person’s awesome challenge. I’m going to forward him your link, k?

    1. Scott Berkun

      Lynn: I have no doubt the mathematical problem was interesting and challenging to solve (sounded NP complete) – And also, having not seen the result, I realize it might result in a design that is wonderful and amazing to see. But that might have also been possible with a much simpler set of requirements.

      My point is rather that there are so many more important elements of memorial design that likely could have used more attention, attention consumed by solving this problem in a way that was unnecessary.

      If the architects were twiddling their thumbs looking for something extra to do, fine. But it doesn’t appear that was the case. They chose to make the problem much more complex than necessary.

  5. Greg J. Smith

    Hi Scott, I don’t really buy your critique at all and I think the title of your post is melodramatic.

    I was trained in the field of architecture and now (basically) am working as an information architecture. A large percentage of the non-profit design work we’ve done has been based around appeasing non-profits that have multiple funders with varying investments. You would be amazed how much time could go into developing the promotional literature (and web presence) of an organization with 15 sponsor logos (say at three tiers of funding). Even more complicated: I have architect peers that have worked on funding walls and spent A YEAR making sure that the arrangement and type size of a few dozen names appeases all the stakeholders.

    My examples are quite banal and do not involve people that were murdered – why is it so difficult to imagine that the order of names on this monument was a major issue to the designers (let alone the bereaved)? What is a monument if not an attempt to make sense of and bring order to a horrendous event? If I was commissioning a memorial I would want a designer who would take care in authoring EVERY SINGLE detail – rather than phoning in aspects of the job by delivering conventional solutions that we’ve all seen a thousand times before.

    1. Scott Berkun


      Definitely a melodramatic title.

      If I was commissioning a memorial I would want a designer who would take care in authoring EVERY SINGLE detail – rather than phoning in aspects of the job by delivering conventional solutions that we’ve all seen a thousand times before.

      Where we disagree is in how that care is applied. I’m sure the architects are going to reuse typefaces, colors and types of stone that have been used before. Is that phoning it in? Reusing an existing idea doesn’t mean great care isn’t also being used. There’s plenty of conceptual reuse in the Vietnam memorial, and likely in any good architecture of any kind.

      In this case I think the problem isn’t worth the attention it was given, relative to the common failings of many memorials. The name list is a secondary consideration among many more important ones.

      I have architect peers that have worked on funding walls and spent A YEAR making sure that the arrangement and type size of a few dozen names appeases all the stakeholders.

      I understand the constraints are real. But I also suspect on many of these projects many clients and visitors are unhappy with the results, and their unhappiness is tied to the (problematic) requirements they inflicted on the architects. I’m judging the results, not the challenges of the process.

    1. Scott Berkun

      Steve: not sure exactly which element you find powerful. The groupings (building/flight) or the within group order? The former makes good sense and could be done easily, without any fancy computing power. The later seems an unnecessary problem to solve that wouldn’t have been missed.

      Also, if we’re using the web, why not let that provide the rich experience (adjacencies) and not worry so much about it in the memorial construction itself?

  6. Steve

    Ugh, do I have to spell it out for you? It’s powerful that this particular victim and her unborn child were able to be listed next to the husband/father thanks to the thought that went into the design. The ultimate requirement of randomness with underlying structure is, I think, fairly brilliant. Walk down any crowded New York street and you’ll understand. Yes, the whole redevelopment of the WTC site is taking a lot longer than expected due to lots of reasons more complex than you or I can appreciate. However, I don’t think the design decisions required for the placement of names on the memorial is one of them.

    Your post begs the question of how you would have done it differently and what kind of higher concept you would try to achieve. Do you honestly believe that alphabetical order would be a more fitting tribute?

  7. Ian

    It seems like the argument here rests on a couple of ideas:

    1) The concept of name adjacencies on the memorial is a weak concept.
    2) The memorial was delayed because of the difficulty of executing this concept.

    And the conclusion is that because it’s a weak concept, it wasn’t worth delaying the memorial.

    The first is your opinion, which of course you’re entitled to. I have no way of disproving it, but if you do a search for ‘9/11 names’ on Twitter you’ll find a lot of people declaring themselves very moved by it, for example Linda Tischler of Fast Company saying it was the only press briefing that she’s ever cried at. Again, just opinions.

    The second premise is something that the NYer article could certainly lead you to believe, but really when you look at the overall project, the names arrangement was a smallish part of the overall complexity of the memorial and had no effect on delaying its construction. Consider that the memorials are waterfalls, each the size of the footprint of the WTC towers, and they are built on top of a 120,000 sqft museum.

    I think this problem with this post is that it’s trying to make a general point about project management, most likely in the context of software given your bent, but fails to take into account what this thing is. It’s not a website. It’s a permanent memorial to one of the greatest tragedies ever occurring on U.S. soil. It will become a part of the city and if the city is still there in 100 years, so will the memorial. It was worth it I think to go beyond a purely functional organization of the name and try to express something that makes the memorial more moving. If being moving isn’t the core requirement of a memorial, I don’t know what is.

    1. Scott Berkun


      Wonderful comment. You have captured both my assumptions and the sloppiness of how I presented them. Well done. Thanks for taking the time. A better title or a more careful lead paragraph of my assumptions would have been apt.

  8. Ambrose Little

    I have to toss in my hat with Ian here. This project is *all about the names*, the PEOPLE. How bright and shiny the stones are or the angles on the tower are far less important than the names, the people, and the stories alluded to by their adjacencies. Ultimately, the memorial is for the victims’ friends and families, and the adjacencies matter to them.

    There will be plenty of helps to assist the families and friends to find their loved ones, and they will appreciate the extra attention paid to these human interconnections.

  9. Ambrose Little

    Sorry, “tower” wasn’t the right word–the memorial museum building is what I was thinking of…

  10. Mike Nitabach

    There is a very general lesson from this, which is that defining goals–of a design, experiment, project,etc–requires as much attention as the details of implementation, and if you fuck up while defining goals, you are dooming implementation.



Leave a Reply

* Required