I don’t think anybody is making that claim. Problem is: if you can only rank values that have a total order (https://en.wikipedia.org/wiki/Total_order), so if you want to rank countries, you have to simplify.
For the judgment of “how deep is a country in debt” it also seems desirable to have the property that splitting a country into equal parts yields smaller countries that are equally deep in debt.
If you want that, “debt” on its own doesn’t cut it. Debt/population and debt/GDP are two metrics that have that property. They also are simple, so do not seem doctored, which correcting for modeled/guessed at/hoped for future growth or correcting for the age distribution of the population, average fertility, etc. easily could.
”GDP today is not a perfect proxy for GDP in 10 years”
Again: I don’t think anybody is claiming ‘perfect’.
Current interest rates can give a somewhat distorted picture for the US. Banks buy US debt because they have to due to regulation, other countries buy US debt as part of their economic policy, the net effect is that the US government is able to borrow money more cheaply that it should based purely on credit risk. Even so, the US is able to borrow money at historically cheap rates.
My understanding is that if current trends is health care costs and economic growth continue then in the long run the US will be unable to service its debt and maintain its current spending programs without raising taxes. Most spending is on the military, Medicare, Medicaid and Social Security, so there will be some difficult political decisions. However, that doesn't mean the US is close to a debt crisis today.