On 25/09/13 09:02, the wrote: > Hello > Does absence of warranty mean that I can't trust results derived with > the help of gsl?
No. You will find the same absence of warranty on almost any software, including, for example, Microsoft Excel. > Do I have to read the source code first to determine if the library > does what it is expected to do? No. But you can read the source code if the library does not do what you expected it to do. This is a big benefit over closed-source software. If GSL doesn’t work you can fix it. If closed-source software doesn’t work you can’t. > Does this all mean that if I'm not willing to read the source I am not > able to use this library to get trustworthy scientific results? No. The results will nearly always be trustworthy, as with any good software. When I use GSL I seldom check the source code. But I do test the results against other independently-written software. The benefit of GSL is that if there is a discrepancy, I can check the source code. -- John D Lamb
