Thanks for the comment and especially for GSLL!

Yes, you can include this in the documentation of GSLL. I will be very happy for it. This post was just my attempt to start learning GSLL and document for myself and anyone who might find it interesting. For a future chapter on a tutorial it still needs some work but that can also be done.

Now, let me just answer your points, specifically 1). I guess I expressed myself in the wrong way or I was not clear enough. My intention was that the function “random” from the CL standard is different from the ones in GSLL since you cannot pass a seed number. In “random” you pass a state object, not a seed number. Just this. I didn’t mean that GSLL changed the implementation’s conformance, in fact, it never crossed my mind that paragraph could imply that. I hope it’s clear now!

As for point 2) you are completely right. I was looking at GSLL source and looked at the functions I could use.

I plan to write a little more about GSLL as soon as I start learning a bit more of it. In the near future I will be looking at the random distributions.

All the best,
Jorge

There are a couple of small points:
1) What do you mean by “Contrary to the standard of Common Lisp”? There is nothing in GSLL that changes the implementation’s conformance with the standard. Do you mean that GSL’s random generator implementation differs from the CL standard’s specification? If so, perhaps it’s better to say that.

2) Note that copy-making-destination and copy-to-destination are not intended to be called directly by the GSLL user, and the symbols are not exported from the GSLL package. The proper interface is #’gsl:copy; it will select between the two based on the presence or absence of the second argument.

Thanks for the great job.

Liam