As far as I am concerned, there are hundreds of disciplines related to scientific computing realizing their calculation via Fortran language. Fortran, originated at 1950s as a short name for formula translation, is considered to be the first computer language which design for pure science calculating. Once, Fortran is dominant with glory and pride, while it is fading away significantly.
Many of us have long hold a misconception that Fortran is the fastest computing language. Frankly speeking, Fortran routine executes fastly as well as cheaply, and that is why it is so popular for more tha fifty years. Apparently, however, it is definitely not the fastest. For example, assembly language is definitely faster than Fortran, though it is complicated in coding. Holistically, the true advantage of Fortran is a fabulous balance between performance and coding cost. In addition to that, “first” itself may be an advantage either. I often hear scientific reseachers who are reluctant to learn a new language such as C or Python said that there are lots of past programs written in Fortran, we should better keep this consitency for tranplatation and some other reasons. In my opinion, the reason is bullshit as well as blindness. Most of the historical “legacy” are illy written(not readable, not aligned, lack of annotations and etc) because the majority of those writers are amateurs, or even noobs, I often got the impulsion to rewrite a new one, a better one. In addition to that, the old-fashioned grammar and structures have brought in other flaws. Thus, the value of those heritages must be revalued, like Nietzsche’s, and there is really little worth of historcial codes.
Here are the Tiobe index(rankings) which denotes the popularity of programming languages, please note that these are average positions for a period of 12 months.
|Visual Basic .NET||9||192||–||–||–||–||–|
As we can see, Fortran fell steeply since 1990, while Python keeps a steady position.
So, wake up, dress up, and python up.