With all the news going around today about the National Transportation Safety Board safety recommendation stemming from the 35W bridge collapse, it seemed to makes sense to link to the recommendation itself. At five pages, it's fairly short and to the point, and overall a remarkably easy read.
While the larger investigation is ongoing and "no determination of probable cause has been reached," the recommendation strongly suggests poorly-designed gusset plates--gusset plates surprisingly undersized for the role they played in holding the bridge together--likely played a significant role in the bridge's collapse. The recommendation also gives some somewhat discouraging explanations as to why the error wasn't caught before the bridge was built, why engineers working on subsequent modifications to the bridge didn't recognize the size of some of the gusset plates as an issue, and why regular inspections didn't bring the problem to light.
From my completely civilian understanding of this, once the bridge was going up, its design wasn't questioned. That makes sense when you think about it, especially considering the considerable resources that would be required to add regular reviews of the designs of bridges and other infrastructure to the top of the inspections and other work that's already done. It just wouldn't be practical. Still, it's kind of unsettling to think that while the 35W bridge may not have been doomed to collapse, once its designs were finalized and approved, it might as well have been.