Also, the fashion of physics changes. Back in the day, physicists seemed much more comfortable writing long intuitive arguments with lots of words, and were always happy with a strong connection to classical physics which they obviously had a strong understanding of (since at the time, that was all physics). These days, it's far more common to prefer a more mathematical oriented approach; people prefer to see an equation rather than paragraphs of intuition dumping. Also, we don't care so much about a connection to classical physics, since most physicists are now quite comfortable with quantum mechanics.
Even as a physics researcher it can be painful and difficult to go back to the original papers for things. This is also reflected in the popularity of wikipedia among physicists. It's much more likely that I'm going to understand the explanation of (for example) Bell's theorem on wikipedia than reading his original paper "On the Einstein Podolsky Rosen Paradox." People have come up with better examples, emphasised important points better, and refined the understanding.
We also can't forget the impact that LaTeX (or equation typesetting in general) has had on physics. Reading some of these old typewritten equations can add unnecessary cognitive load.
Now, this is unique to physical sciences. None of this can be relied on with philosophy (at least, continental philosophy). Sure, there are good modern summaries like on SEP, but you can never really be sure that your interpretation of the author will be the same.
