Re: [NonGNU ELPA] New package: sweep

[Philip Kaludercic]

Juri Linkov <juri@linkov.net> writes:

>>>> I agree with RMS (<E1odDGZ-0005aw-79@fencepost.gnu.org>) that it might
>>>> be nice to have a more indicative name, or at least something that
>>>> includes "Prolog".  Mor eso because "sweep" makes me think of something
>>>> that cleans.
>>>
>>> I've given it some thought, and all in all I would like to keep the name
>>> `sweep` for this project.  I do see however why this name may be
>>> unhelpful for Emacs users who are not familiar with SWI-Prolog, or who
>>> are looking for some "cleaning" package...  As a possible solution, I'd
>>> be happy to change the name of the Elisp package to `sweeprolog` while
>>> still referring to the project as a whole as `sweep` (e.g. in the
>>> manual).  Does that sounds alright?
>>
>> I think that would be nice, and there shouldn't be a need to rename the
>> entire project.
>
> I wonder where this requirement to put all tags in the package name comes 
> from?
> While looking at the long list of package names, it's hard to find a package
> whose name immediately hints at what it does.  This is a non-problem
> that is solved in all package managers by using tags and full-text search.
> And indeed list-package allows both ways, e.g. when someone is interested
> in Prolog, then it's easy to type '/ k' (package-menu-filter-by-keyword).
> Or try to use '/ d erlang RET' (package-menu-filter-by-description)
> and see that almost no package has it in the package name,
> yet these packages can be easily found:
>
>  auto-complete-distel    melpa  Erlang/distel completion backend for 
> auto-complete-mode
>  company-distel          melpa  Erlang/distel completion backend for 
> company-mode
>  company-erlang          melpa  company backend based on ivy-erlang-complete
>  distel-completion-lib   melpa  Completion library for Erlang/Distel
>  edts                    melpa  Erlang Development Tool Suite
>  erlang                  melpa  Erlang major mode
>  erlstack-mode           melpa  Minor mode for analysing Erlang stacktraces
>  ivy-erlang-complete     melpa  Erlang context sensitive completion at point 
> using ivy
>  lfe-mode                melpa  Lisp Flavoured Erlang mode

The package list does help, but having a name that can be understood on
its own seems like a good thing.  There is no reason to make package
discovery more difficult.

Tough I might be a bit extreme on this topic.  E.g. I remember when
starting out with *nix, not wanting to use grep when searching for a
program to filter a file, just because of the name.