Beefy Boxes and Bandwidth Generously Provided by pair Networks
P is for Practical

When is it time to stop posting to CPAN?

by perlfan (Vicar)
on Oct 11, 2020 at 10:13 UTC ( [id://11122696] : perlmeditation . print w/replies, xml ) Need Help??

WordList::ID::Fruit::PERLANCAR - seriously, come on dude.
  • Comment on When is it time to stop posting to CPAN?

Replies are listed 'Best First'.
Re: When is it time to stop posting to CPAN?
by Discipulus (Canon) on Oct 11, 2020 at 17:30 UTC
    Hello perlfan,

    while I understand your, even if shortly expressed, sentiments and even more marto's ones, please remeber what CPAN stands for: Comprehensive Perl Archive Network. Not Exclusive nor Elitarian but Comprehensive.

    We are talking about our dear brother perlancar and yes it is the top uploader ever at CPAN with 2083 distributions and he is a good programmer. Ok: a bit exaggerated on the volume of his uploads but CPAN is a vast world and he probably has some tricky Risk like objective to occupy the whole world ;)

    I'm with marto about the possibility to leave out of CPAN some personal project, since cpan clients can download from git online repositories, but this is a matter of taste or social behaviour, not a rule. This happens with the meaningfulness of namespaces too.

    And as ever happens the river is much smaller than the sea of thousands of modules newer, older and leftover. But no one is saying what is good to fit CPAN or not.

    This is the point: freedom always leads to some chaos. Here at perlmonks is the same but I definitely like it.

    The CPAN infrastructure has a cost to host all its mirrors, for the bandwidth and storage, but we are really caring of 14.58Kb of the above mentioned distribution? I'd better point the attention on the CPAN search functionalities. About the river analogy the River stage rank is somehow a useful metric to judge a distribution, or, better, to know the possibility such distro will be updated on time.

    Before uploading advice is good, but it is an advice not a costraint. The document mentions prepan but it is not the real predigestion of CPAN, it is a polite thing to do before uploading, not a constraint.

    We dont like loud people, but we want a law to measure decibels emitted by people? We want a filter in front of CPAN to judge incoming distributions? We need a Quality Perl Archive Network (a lot of work to do)? I vote no.

    Better the chaos of freedom.


    There are no rules, there are no thumbs..
    Reinvent the wheel, then learn The Wheel; may be one day you reinvent one of THE WHEELS.
      I appreciate all the thoughtful comments, placing this RT here since I can tell this response clearly took some time and thought. Thanks again.
Re: When is it time to stop posting to CPAN?
by marto (Cardinal) on Oct 11, 2020 at 12:21 UTC

    Part of my daily routine is to check the metacpan recent page. Often On occasion the majority of new releases (since the previous day) seem to me like they're only going to be of interest/use to the author. Over the years this all adds up, everyone needs a hobby I guess. When reading a post/chatterbox conversation when someone starts talking about putting their pet project on CPAN, more often than not I can see no valid reason anyone would have for using what they propose. This isn't to say that CPAN is 'complete', or that people shouldn't add stuff, but given the modern cpan installers can pull things directly from github that this could be a more suitable place to host some pet projects.

    Update: slight rewording, strike out above.

    Update 2: Related regularly publishes a curated list of what's new on cpan articles.

      Is there a good way in META.json or Makefile.PL for a project to indicate a dependency on a Github-only Perl module?

      Until there is, hosting projects only on Github has a big disadvantage.

        Not directly relevant but similar in concept prior art: there's a package for emacs which lets you install directly from the github repo rather than pulling from its official package repos (MELPA/ELPA) Quelpa. It more or less pulls the source and does the equivalent of a make dist then installs that.

        The cake is a lie.
        The cake is a lie.
        The cake is a lie.

        Even if CPAN clients can as easily install from GitHub, I like the ease of which a CPAN mirror can be used offline. I do maintain private CPAN-like repositories for personal/work use and should probably "publish" there more often instead of to CPAN for things like these. But I do like the nice things the ecosystem gives me, e.g. RT and CPAN Testers.
      As a side note, my monthly blog post series (such as this one) also began as a reaction to the "What's New on CPAN series" on, which I saw often missed some modules (granted, it's a curated list) as well as included some non-new modules. My blog posts get the data from MetaCPAN API, so it should not miss any new distribution, and they are 99% automated so the effort to create a new post is minimal and there's a higher chance that I can keep up doing it.
Re: When is it time to stop posting to CPAN?
by choroba (Cardinal) on Oct 12, 2020 at 11:54 UTC
    I'd be much more worried about distributions like Net::FullAuto.

    I'd rather ask "When is it time to start posting to CPAN? What's my contribution?"

    map{substr$_->[0],$_->[1]||0,1}[\*||{},3],[[]],[ref qr-1,-,-1],[{}],[sub{}^*ARGV,3]

      Yeah, Net::FullAuto is huge and frequently updated.

      At least the author has now started deleting old versions from CPAN, so it only places a strain on BackPAN mirrors.

Re: When is it time to stop posting to CPAN?
by SankoR (Prior) on Oct 11, 2020 at 14:45 UTC
    That dist is a plugin to the poorly named WordList distribution by the same author. Wordlist itself looks like a system to filter a set of words for games which actually sounds like it could be useful in a general sense but I'd never find it with this namespace choice. I say the bigger problem is deciding not to bundle these plugins with that main dist itself; they've released several lists of words this way in a number of languages already.
      WordList began as a response to Games::Word::*. I wanted a namespace that is more general instead of gaming-associated, and wished that the implementation modules have a faster startup time. Assuming I read your comment correctly, I believe I've cleaned up the namespace so that plugins (roles, base classes) are moved out to WordListRole::*, WordListBase::* and all my WordList::* modules are actual wordlists.
      OK, so I indeed misread your comment. You're saying that it's better that all word list modules are bundled together in a single distribution. And later when I want to add a single new, possibly tiny, word list, I have to release another version of this gigantic WordList distribtution with tens or possibly hundreds of existing, non-changing word lists? That's precisely what I'm avoiding to do. Besides, some of the word list modules are dynamic e.g. WordList::DBI and it depends on DBI. I don't want to make the main WordList distribution itself depend on DBI or other additional modules that are only relevant to one or a few word list modules.
Re: When is it time to stop posting to CPAN?
by perlancar (Hermit) on Oct 12, 2020 at 08:32 UTC

    When there's no longer a good enough incentive to do so, I guess.

    Thanks for the feedback, I'll think twice before posting things like those. Modules like these could be grouped in a larger distribution, but I tend to think that smaller distributions are more efficient overall, e.g. in terms of disk space when there are multiple revisions, or dependencies.

      Actually this reply makes me a little more comfortable knowing that there is a real person behind this prolific body of work. I am not trying to be the CPAN police or stifle contributions. I was concerned about AI/automated CPAN spam. I can't be the only one who suspected this.

        I was concerned about AI/automated CPAN spam. I can't be the only one who suspected this.

        for how long did you suspect?

        I fairly certain it doesn't take but a minute to dismiss such nonsense

        Yeah everybody has covid cabin fever, adjust your public filter accordingly

Re: When is it time to stop posting to CPAN? (When you die, maybe)
by beech (Parson) on Oct 17, 2020 at 00:52 UTC


    Is this what you're really saying?

    When is it time to stop posting to CPAN? by perlfan

    When is it time to stop liking perl/CPAN? by I like perl

    WordList::ID::Fruit::PERLANCAR - seriously, come on dude.

    WordList::ID::Fruit::PERLANCAR - seriously, stop that you like perl

    When is it time to stop posting to CPAN? When you die, maybe




      I'll try to defend perlfan a bit.
      In his reply Re^2: When is it time to stop posting to CPAN?, he wrote: "I was concerned about AI/automated CPAN spam.". Later he admitted that he believes PERLANCAR is to be human, i.e. that this CPAN spam isn't automated or generated by AI.
      So here I would correct your question in this way:
      When is it time to stop spamming to CPAN? by perlquality-fan
      So it is a question about quality concerns, not about disliking perl.

      Someone will agree that particular modules are spam, someone will disagree. Imagine if the amount of "spam" modules increases by mean that many users find less and less percent of modules informative or useful. This will increase a time for every new user to find suitable modules, and many attempts to find them will deliver a disappointment. It will become a question for users: "should I further spend plenty of time to search suitable modules" or "should I code by myself" (or 3*).
      If higher percent of modules are great, then it is a higher chance to spend less time to find suitable one! Otherwise it wants to be implemented: 1. searching modules by community rating (not only by keywords, because bad module can reserve a good keyword), 2. socialnetwork-like ranking, where one can filter by seeing stars of his/her fans. 3. search engine which allows to reject particular authors and remember such choises.

      Perl community usually proudly delivers for the world that CPAN has MANY MANY modules which are ready to USE. But it is like an "r-strategy" to emphasize the amount not a quality.
      (It may suggest a bit about popularity of Perl therefore?)

      *3 - "I have much time and I want better modules here: so I will write a better module. But the name is already reserved. So I will append '2' to an existing name, or I'll choose an uncommon rare english word with the same meaning for the alternative."

      Also OP could be a bit more verbose on why he considers particular module and similar modules being SPAM. And what are suggestions.
Re: When is it time to stop posting to CPAN?
by karlgoethebier (Abbot) on Oct 20, 2020 at 16:56 UTC

    I suggest to reap this whole thread. Let us put a cloak of silence over this embarrassments.

    «The Crux of the Biscuit is the Apostrophe»

    perl -MCrypt::CBC -E 'say Crypt::CBC->new(-key=>'kgb',-cipher=>"Blowfish")->decrypt_hex($ENV{KARL});'Help

Re: When is it time to stop posting to CPAN?
by ikegami (Patriarch) on Oct 13, 2020 at 02:03 UTC

    It's like gay weddings. It doesn't affect you, so why do you care?


    The above is the original post. There appears to be some confusion about what the above means. A message of tolerance Considered for offensiveness? yikes! I don't see the second meaning, so hopefully, the following will make it clearer:

    Gay weddings don't affect you, so you have no reason to try to object to them. Same goes with the module in question. Its presence on CPAN doesn't affect you, so you have no reason to object to its presence.

      Funny how people misunderstood this comment. It obviously means : "if you aren't gay, gay weddings don't affect you, therefore you have absolutely no reason to care about them / criticize them" (thus if you don't use this or that module, it shouldn't bother you at all that it exists). It's a basic appeal to tolerance, but apparently hysterical political misreading / absence of rhetorical generosity is the norm on the internet now, even in a soft, peaceful place like the Monastery. That's quite sad, really.

        One can read it in two ways, and I prefer the "in dubio pro reo" approach.

        Personally I never understood why some straight people are opposing gay marriage, why should it be their business?

        That's how I understand this.

        I find it more troubling that free communication becomes censored by an inquisition of current "political correctness".

        Please calm down everybody, and leave room for lazy wording.

        Cheers Rolf
        (addicted to the Perl Programming Language :)
        Wikisyntax for the Monastery

        That is indeed what I meant. It doesn't mean you or your religion has to support it. If you have no stake in the game, you should respect/tolerate their wishes as you should have yours respected/tolerate.

        Same with the module. Having it on CPAN doesn't affect me in the least, so why would I care if it's there or not?

      I feel disgusted by this association! It is hugely offensive!

      If you don't want to install someone's CPAN work, disable it using distroprefs. In this case it might look like ~/.cpan/prefs/PERLANCAR.yml.

      --- comment: "I personally do not care about their work" match: distribution: "^PERLANCAR/" disabled: 1

      (this is a literal translation for a file I do have installed for a different author but with different comments)

      Enjoy, Have FUN! H.Merijn
      It's like gay weddings. It doesn't affect you, so why do you care?
      Yeah, nice metaphor.
      Edoardo Mantovani, 2020
      I am also highly offended by this and it disgusts me. No need to project your personal politics into this or associated me as OP with devisive non-Perl related topics, and I don't appreciate you sullying what has become a quite informative thread.
        Stop, Stop, Stop.
        He hasn't taunted anyone, specially the LGBT community, instead, he used a metaphor for describing its ideas, he hasn't said anything related to hate or similar.
        In this moment I need the power of KEKISTAN.
        Edoardo Mantovani, 2020