I already screwed up my uploaded package by using a plural and then later read that the style guide suggests singulars. Embarrassing.
On Fri, Jun 23, 2017 at 5:57 PM, Neil Van Dyke <n...@neilvandyke.org> wrote: > Jack Firth wrote on 06/23/2017 04:47 PM: > >> Pot-stirring question: do you think it would make more sense to name the >> collection `math/glpk` instead of just `glpk`? >> > > You're exactly right, it's a controversial question. :) > > In general, for a flexible, decentralized ecology of third-party packages, > I think one usually wants either a flat namespace of package names, or > (like PLaneT or Java) a namespace that's qualified by the package's > origin. (Qualified by one of the developer's identities, and whatever > additional qualification the developer chooses to add. In the case of > Java, from the very start, they elegantly outsourced the namespace > qualification authority to the Internet domain name registrations [*], but > PlaneT's simple method was also reasonablee.) > > One mitigating factor, in this particular case, is that the author is a > core Racket person, and there's an ancient legacy in core Racket, of > sometimes using hierarchical package names for purely topical taxonomy > purposes. IMHO, this is suboptimal, but it's ancient precedent, so I > wouldn't try debating it if core Racket continued to do this. > > Just know that, every time I gaze upon a new hierarchical package name > used for purely topical taxonomy purposes, a single tear rolls down my > somber face. > > [*] Java used the domain name registration system only to the extent that, > if you owned a domain name, you were socially considered to have authority > to name packages qualified with that name and subdomains. They didn't use > DNS itself at all, although you could imagine how DNS *could* later be used > to implement package signature authentication in a distributed way, atop > existing DNS infrastructure. (Not that I'm advocating using DNS that way > for Racket -- there some other ways of implementing authentication that > make more pragmatic sense, given where we are. And I'm not going to argue > for Java-like use of domain registration authority in Racket, at least not > unless it's part of implementing a trust system for packages.) > > -- > You received this message because you are subscribed to the Google Groups > "Racket Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to racket-users+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
