That's why I found it more fruitful to study entire compilers/projects, instead of hunting down papers on memory management, code generation, optimization, eval

That's why I found it more fruitful to study entire compilers/projects, instead of hunting down papers on memory management, code generation, optimization, evaluation models, etc. If you get into the "mind" of the project, you will be able to trace the progression of ideas and see why things were done the way they were. I often read peripheral papers on a given project because I tend to get the big picture from those filler papers, usually written by the lead, and/or coauthored with the entire team. The highly technical papers reserved for the best and most specialized conferences usually skim the details and present the most novel idea in its purest crystal form .. usually under length restrictions and addressing a well prepared audience. I recommend a beginner to memory management start with Wilson's Uniprocessor Garbage Collection Techniques ftp://ftp.cs.utexas.edu/pub/garbage/bigsurv.ps Abstract: http://www.memorymanagement.org/bib/wil94.html http://www.memorymanagement.org/bib/wil94.html