Something I experienced even more acutely with The Black Swan, editors were not content in rejecting the book as a wise businessperson would reject an investment, by saying something like it may be great but I do not wish to take the gamble or politely appeal to caution. I remained undeterred by the insults in the rejection letters. And I wanted to do my own version of what is called literature. Nor should literature have institutions formalizing and commoditizing things. But literature should not have explicit boundaries: the confines of the subject are internal and may remain elusive and hard to express in words. I had written a specialized treatise in mathematical finance, and was embarking on a career writing scientific papers, which must be as narrow as possible. Now I have no problem with focus and precision: I do not want my vacuum cleaner user manual to be a stream of consciousness. Not unexpectedly, the reaction of publishing houses had been unanimously dismissive: why would anyone mix finance, Solon, and Proust? And why these mathematical discussions? The good thing about inserting “randomness” in the title is that it allowed me to write about anything that crossed my mind, given the ubiquity of chance and, worse, the lack of awareness of it. “What’s wrong with confusing the reader?” was my usual answer. “Why are you confusing the reader with both Nero Tulip and yourself”, I was often asked by those who did not find the mixture too uncanny and continued reading the text. It included parables with fictional characters, one of whom seemed to resemble me, which appeared to be confusing since I also had explicit autobiographical episodes. The official “subject” (an essay is not supposed to have such a constraint as a subject) was a random mixture of autobiography, philosophy (of probability), mathematics, inductive logic, musings on historical events and financial markets. It was a continental style meditative essay and the Anglo Saxon world, in spite of their infatuation with Montaigne and Umberto Eco, were about five hundred years late to the genre. The system knows from the read() call which data are to be read, and data must be read before the call returns (unless they are already buffered).ġWhat comes closest to such a system call is probably (as already mentioned by Tsyvarev) fadvise(2): Give advice about file access - Linux man page:Īllows an application to to tell the kernel how it expects to use a file handle, so that the kernel can choose appropriate read-ahead and caching techniques for access to the corresponding file.By then I had written the first volume of the Incerto, Fooled by Randomness, a book that was practically impossible to publish. To read data from disk to kernel buffers, having a dedicated system call is not needed 1. To write data from kernel buffers to disk, fsync() can be called, but it doesn't have to, if it suffices that the buffers are flushed eventually, which will happen anyway sooner or later, unless the system crashes or is reset. The pwrite() function shall be equivalent to write(), except that it writes into a given position without changing the file pointer.Īlso fsync() write data from kernel buffers to disk, so which system call read data from disk to kernel buffers ? For pwrite(), besides lseek, it also call the fsync()?
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |