Read The Unix Time-Sharing System paper by Ritchie and Thompson (reading #6). This is a classic (ancient) paper about the precursor to today's Project Athena (and many other variants of Unix). However, don't assume that just because you are a Linux hacker, you already know the material in this paper. Unix of the 1970s was quite different from that of the 2000s. Another difficulty in reading such an old paper is that principles that are well-established today may have been controversial when first introduced. If you notice places where the authors seem to belabor an obvious point, then the point may not have been obvious at the time!

As usual, start out by reading the abstract, introduction, and conclusion. You will find that this paper lacks an outline (which would typically be found at the end of the introduction; a one-sentence overview does appear in the abstract) and a conclusion. Therefore, scan the paper for section and subsection titles, to give yourself an idea of how it fits together. Then, read the whole paper.

This paper presents an early perspective on the design of a system that has stood the test of time and remains popular 30 years after its introduction -- admirable longevity for a software system. Therefore, there is much to learn from it. As you are reading, consider what you can learn both from the authors and from the perspective of hindsight. For example:

• Ritchie and Thompson were minimalists. What are examples of some features that they explicitly left out? In what circumstances are these omissions justified?
• The authors claim that Unix was "not designed to meet any predefined objectives." Do you believe this claim? If not, what were some objectives, and how were they achieved? If so, how did the lack of constraints affect the development and evolution of Unix?
• The Unix philosophy encourages creation of many small utilities that users compose with pipes and filters. What constraints might have inspired this solution? What are its advantages and disadvantages?
• (For those who are familiar with modern variants of Unix:) Give a few examples of how Unix has changed in the last 30 years. What principles or design decisions enunciated by Ritchie and Thompson have been abandoned or changed in the meanwhile? Do those changes indicate that Ritchie and Thompson made a mistake, or that changing circumstances required rethinking the design?

Write down your thoughts for at least one of the issues above and bring to class for discussion. Also, bring in your answers to the exercises in Hands-on 2.

If you are interested in learning more about Unix, there are several other good papers and books, listed in section 11.3 of the Suggestions for Additional Reading (pages 10-39 and 10-40 of the class notes).