[<-] [ ^ ] [->] Previous: Use in Teaching Up: The Role of Literate Computing in Undergraduate Curricula TR-CS-94-01 Next: Conclusions
We have addressed the need fairly fully above. Essentially communication skills are important in the real world, employers value them, yet the emphasis on them in the early years of computer science is minimal. The native English speakers need help here, but the issue is more complex for students whose first language is not English. Clearly we should not (and did not) penalise them for poor English. Yet if they do not receive feedback on the quality of their written work it will never improve. If they never have to write anything, they will never receive any feedback.
We were careful to separate WEB from literate programming in general, although at present no other system comes close in power or flexibility. Yet one might ask why, after the widespread acceptance of [LaTeX], WEB has also not spread more widely. Apart from some of the issues we have already alluded to, the availability of fast hardware and postscript printers and display devices is relevant too. The edit-build-test cycle tends to be much more rapid in programming than in document typesetting. Hence it is essential that the extra time taken to tangle the WEB file is negligible. This is just about true on a lightly loaded sparc station. So this requirement has only just been met. On the weaving side, the WEB document itself is hard to follow, thus it must be possible to view the woven document quickly and easily. Again, this requirement has only recently been met. It is interesting that Knuth released WEB a decade ago, yet only in 1992 did he produce a book entitled Literate Programming.