HackerTrans
TopNewTrendsCommentsPastAskShowJobs

pfitzen

no profile record

comments

pfitzen
·3 yıl önce·discuss
It depends on the user and the usage. In some cases reading these books doesn't yield the best return for the investment of your time. Consulting the literature that Knuth refers to, rereading and working through the examples, writing your own code could give you so much more.

Volume 2 is within arms reach on my desk. After reading the book and additional literature over the course of several years, the book is filled with sticker notes. Over time my usage shifted from reading TAOCP to rereading (and changing) my notes.

> Would you still recommend this book, and if yes, in what circumstances?

Everyone in math/cs should at least know about the existence of theses books. A rough, cursory knowledge of the content is essential, so that one can look up things if there is the need to really understand a specific topic. TOACP should be considered as one of the best and most important/influential science books of the last century.
pfitzen
·3 yıl önce·discuss
Passing a pointer doesn't always solve the problem. The idea behind std::piecewise_construct is to control the time at which the non-moveable object is constructed. Simply passing a pointer means the object is created outside the container; before the container is created. With std::emplace the container is already (partially) created and gets control over creating the objects it is going to contain. The container gets the power to change the arguments to the constructor of its contained objects.
pfitzen
·4 yıl önce·discuss
A book in german. Elstrodt: Mass- und Integrationstheorie With notes on the people/history of the development of this field.