Brian Brenner, PE, F.ASCE, Professor of the Practice, Tufts University
A corporate efficiency advisor, Jim Collins, blogs about ways to achieve excellence at work. He discussed the approach of rinsing one’s cottage cheese before eating it. The idea was to remove any excess fat and salt to better optimize diet and athletic performance. Collins thought that rinsing cottage cheese (both the actual activity and the metaphor) was a way of getting better performance out of cottage cheese and life in general.
Strangely enough, this idea is related to bridge design. But before we discuss that, first let’s further examine the approach of rinsing the cottage cheese.
Cottage cheese is not thought to be an unhealthy food. There would seem to be marginal benefit in rinsing it. Instead, maybe there would be more benefits associated with rinsing a less healthy food like fried chicken. A drawback is that the chicken would no longer be crispy or taste good. This is one reason why most fried chicken restaurants do not offer rinsed fried chicken.
To evaluate this further, I researched issues associated with rinsing fried chicken. To conduct my research, I asked ChatGPT. Here is its response:
Rinsing fried chicken—whether before or after frying—is generally not recommended and can have several drawbacks, depending on when you’re doing it. Here are the key issues:
________________________________________
If Rinsing Before Frying (after seasoning or breading):
- Ruins the Coating:
- Rinsing will wash off flour, breadcrumbs, or any other breading you’ve applied, resulting in a soggy, patchy crust.
- Dilutes Flavor:
- Seasonings and marinades will be stripped away, making the chicken bland.
- Creates Mess and Risk of Contamination:
- Water splashing off raw chicken can spread bacteria (like Salmonella or Campylobacter) around your sink and kitchen surfaces.
________________________________________
If Rinsing After Frying:
- Destroys Texture:
- Fried chicken’s main appeal is its crispy crust. Water instantly ruins this, making it soft, soggy, or mushy.
- Alters Flavor Negatively:
- Water can make the fried surface taste stale or greasy once rehydrated.
- Increases Spoilage Risk:
- Wet fried food is more prone to bacterial growth if not immediately consumed or properly refrigerated.
It is possible that rinsing would be less detrimental to cottage cheese than it would be to fried chicken. This is because cottage cheese is not that appetizing to begin with. It is lumpy, bland and overall it is an acquired taste. So if you rinse it, you would trying to optimize something that is already optimized in terms of taste, or lack thereof.
Moving on to project management issues, corporate advisor Jim Collins described how rinsing your cottage cheese can lead to success. He reflected on Ironman triathlete, David Scott’s cottage cheese rinsing:
“Make no mistake, being the best requires you to ‘rinse the cottage cheese’ and have the will to do whatever is necessary to make being great a reality.”
Can engineers apply the cottage cheese rinsing approach to design great bridges? If rinsing reduces fat content, a starting point may be to define what fat in bridges is. I scoured the literature about the topic of bridge obesity, this time by myself, and I found this article:
“Visceral obesity stimulates anaphase bridge formation and spindle assembly checkpoint dysregulation in radioresistant oesophageal adenocarcinoma.”
It is an interesting article, but anaphase bridge formation is a medical concept not related to bridge fat.
ChatGPT was not very helpful either. When I asked ChatGPT to prepare an image of an obese bridge to assist in illustration for the article, it provided this:

To optimize bridge design by “rinsing the cottage cheese”, we should consider if some bridges are designed with too much fat. “Fat” it this case would not be biological fat, but the use of more material than perhaps is needed.
Current structural analysis and design methods apply more precise analytical modelling and optimization than in the past. In a sense, using a more sophisticated analytical model could be like rinsing the cottage cheese, because it may result in a satisfactory structural design that is slender and uses less materials and resources.
In bridge design, consider the example of AASHTO live load distribution factors. This is the method that assigns percentages of truck loads to particular elements of the bridge. The highway bridge design code was developed before widespread computer usage. Bridge analysis models were mostly 2D frames with simplified assumptions that facilitated use of manual solutions. Large structures like the Golden Gate Bridge were successfully designed this way, assisted by approximate methods like the deflection theory. As strange as it seems today, designers of large suspension bridges in the 1930’s did not have computers, AUTOCAD or finite element models. They did have slide rules.

Traditional modelling assumptions mostly did not rely on global structural behavior, but focused on assumptions of worst-case loading and connectivity to maximize impacts on individual elements. For truck loading, individual girders in a typical girder bridge are assumed to simultaneously receive the largest possible load, with vehicles placed in the worst position on the bridge to deliver this load.
This approach is conservative in that it leads to a more robust design, and it is easier to calculate. But it is not realistic, because vehicles cannot be positioned to load the bridge as assumed. That is to say that the same truck cannot be placed in different locations on a bridge model at the same time, to deliver the highest load to all girders simultaneously.
With more recently developed computer analysis tools, it is possible to prepare more precise 3D-models in which all elements of the bridge superstructure resist loading, not just individual girders. In this approach, the calculated amount of truck loading that an individual girder resists is effectively reduced. The global bridge model now accounts for systematic contribution of all the girders, not just one. This can result in selection of more slender girders with less material than by using the traditional method.
But maybe that is not the best approach, either. The “fat” that may result from applying the traditional approach is available to address longer term concerns such as loss of material due to deterioration. This loss over time is known to be an issue, but it is difficult to calculate. Longer term issues like material loss are more difficult to define and measure in precise analytical models. But still, it can be concluded that a structure with more “fat” has a better chance of lasting longer, particularly if it is not well-maintained.
Following this thread, bridges with extra “fat” can last longer, and so to achieve a robust design, maybe the opposite of rinsing the cottage cheese is called for: eat unrinsed cottage cheese with salted butter on top.
Or, maybe overly rigorous methods in athletics and bridge design should be tempered by application of experience and judgement.