FAQ
Morning,

I'll preface this by saying: This email is written because i am facing a lack of information and am feeling upset about it. If i missed something really obvious, please do tell me i am an idiot.

That said, i am here to lament a lack of information i am facing.

Recently there was YAPC::NA. At that event things were talked about and announced, things about perl core development. I've been reading vague mentions, allusions, rumors. Apparently Jesse Vincent announced something.

"Fortunately, Jesse Vincent's plans for 5.16 and beyond represent
the kind of change of narrative Perl 5 needs, regardless of what
happens with Perl 6."

From twitter, mostly, i pieced together that apparently he intends to change how `use 5.xxx` works? In order to make it apply some sort of fixed list of things, instead of the accumulation of everything that came before? I'm not entirely sure. Details are thin on the ground and hard to find.

So, an aquaintance of mine wrote this blog post recently, borne out of frustration about Perl: http://me.veekun.com/blog/2011/07/22/perls-of-wisdom/

I disagree with him on many things, but i also agree on many things. Most importantly though, he asked this:

"Do the core Perl maintainers care?"

And as much as i want to say "That question is dumb, of course they care, they're working on things!", i have to say: Thanks to a seemingly complete and utter communication failure of the Perl community at large the question seems to be entirely valid to ask. There's no way he could have feasibly known this.

I spend almost all of my free time with Perl. I release things to CPAN, i weigh in here when i think i can provide something valuable, i provide patches, badger authors into release, contributed to EUMM and many more. I follow over 100 perl people on twitter, spread interesting perl links on reddit and hackernews and make a lot of effort to keep up with what's going on.

But yet i find myself unable to find any information about the things the core people are planning. Not for lack of looking. I did. The only reason i know they actually did talk about something was twitter. Yet i did not see any blog posts, announcements, press releases, recordings or even just talk notes released. Of YAPC::NA i've seen a grand total of 3 packs of slides released.

I have to honestly say that right now i feel the perl core is telling me:

"Spend the cash, travel 6000 miles or fuck off."

To which i have to say: I'd love to. I just moved to another city, took up a new job and did talk to my new boss about whether he'd support it, but he did not want to that early in the work relationship. So, i'm effectively too poor to be allowed to know what the perl core is planning.

So, yes, it is entirely understandable for people who don't spend their entire life on Perl to think Perl is dead. From the outside the perl core looks like it is just coasting along and planning nothing.

Is this how things are wanted?

--
With regards,
Christian Walde

Search Discussions

  • Christian Walde at Jul 23, 2011 at 10:59 am
    Alright, so i have been told by the unimitable mst in his usual fashion that i am an idiot.

    And in some ways he is right.

    The answer to my email is fairly obvious: There is little communication because a lot of work is being done and needing to be done and there is not dedicated person to handle these things, so things end up a muddled clusterfuck.

    As @kentnl points out, even PHP has a means and venue of communicating things: https://wiki.php.net/todo

    And something along those lines is what p5p needs as well. One or more persons to act as go-to points for busy people to drop information to be published along with guidelines on when and what manner. A central place to store these things and other information about p5p. And lastly a person who does spreading of information, by posting them to appropiate places and to make things get out of the echo chamber.

    I'm willing to make these things happen.

    Please provide input.

    --
    With regards,
    Christian Walde
  • Léon Brocard at Jul 23, 2011 at 1:43 pm

    On 23 July 2011 11:59, Christian Walde wrote:

    And something along those lines is what p5p needs as well. One or more
    persons to act as go-to points for busy people to drop information to be
    published along with guidelines on when and what manner. A central place to
    store these things and other information about p5p. And lastly a person who
    does spreading of information, by posting them to appropiate places and to
    make things get out of the echo chamber.
    Fab!

    I'm not a huge fan of wikis, they get out of date very easily.

    A similar project is the Python Insider blog for "Python core
    development news and information":

    http://blog.python.org/

    Such projects tend to start off well and then tail off. You need a good plan.

    The p5p weekly summaries were a good way to keep up with things, but
    were a lot of work to create. How about something similar (but perhaps
    not so often) which doesn't try and summarise every thread but keeps
    focused on the bigger picture?

    Good luck, Leon
  • Léon Brocard at Jul 23, 2011 at 1:52 pm

    On 23 July 2011 14:43, Léon Brocard wrote:

    The p5p weekly summaries
    Ooh, I wasn't aware of this archive of the summaries:

    http://dev.perl.org/perl5/list-summaries/

    A lot of detail indeed! Leon
  • Christian Walde at Jul 23, 2011 at 3:19 pm

    On Sat, 23 Jul 2011 15:43:03 +0200, Léon Brocard wrote:
    On 23 July 2011 11:59, Christian Walde wrote:

    And something along those lines is what p5p needs as well. One or more
    persons to act as go-to points for busy people to drop information to be
    published along with guidelines on when and what manner. A central place to
    store these things and other information about p5p. And lastly a person who
    does spreading of information, by posting them to appropiate places and to
    make things get out of the echo chamber.
    The p5p weekly summaries were a good way to keep up with things, but
    were a lot of work to create. How about something similar (but perhaps
    not so often) which doesn't try and summarise every thread but keeps
    focused on the bigger picture?
    Absolutely. One of my main issues with the current situation is that it is:

    - hard for people in the know to figure out what's going on (i can find out what i need by talking to mst, but not everyone has my thick skin and there's only one mst)
    - impossible to outsides to know what's going on here

    The former is a problem, but not a critical one. The latter is one, because it leads perceiving Perl 5 as stagnating.

    As such the thing i'm thinking about would be regular, maybe monthly, posts, explaining the highlights of p5p that happened in that time in such a way as to be suitable for consumption by the general public. These would then be posted outside of our echo chamber, meaning programming.reddit, hackernews, linkedin groups, slashdot and other general news outlets (suggestions highly welcome). Heck, it could even include highlights on perl/cpan projects that would be of interest outside of perl.

    Then again: I'm sure others have brilliant ideas too. I'm not fixated on one thing and would love to hear what others think might help.

    --
    With regards,
    Christian Walde
  • Nicholas Clark at Jul 24, 2011 at 10:40 am

    On Sat, Jul 23, 2011 at 05:19:30PM +0200, Christian Walde wrote:
    On Sat, 23 Jul 2011 15:43:03 +0200, Léon Brocard wrote:
    On 23 July 2011 11:59, Christian Walde wrote:

    And something along those lines is what p5p needs as well. One or more
    persons to act as go-to points for busy people to drop information to be
    published along with guidelines on when and what manner. A central place to
    store these things and other information about p5p. And lastly a person who
    does spreading of information, by posting them to appropiate places and to
    make things get out of the echo chamber.
    The p5p weekly summaries were a good way to keep up with things, but
    were a lot of work to create. How about something similar (but perhaps
    not so often) which doesn't try and summarise every thread but keeps
    focused on the bigger picture?
    Absolutely. One of my main issues with the current situation is that it is:

    - hard for people in the know to figure out what's going on (i can find out what i need by talking to mst, but not everyone has my thick skin and there's only one mst)
    - impossible to outsides to know what's going on here
    As in, the monthly development release which *ships working code* along with
    a short description of what's changed doesn't do this?
    The former is a problem, but not a critical one. The latter is one, because it leads perceiving Perl 5 as stagnating.

    As such the thing i'm thinking about would be regular, maybe monthly, posts, explaining the highlights of p5p that happened in that time in such a way as to be suitable for consumption by the general public. These would then be posted outside of our echo chamber, meaning programming.reddit, hackernews, linkedin groups, slashdot and other general news outlets (suggestions highly welcome). Heck, it could even include highlights on perl/cpan projects that would be of interest outside of perl.

    Then again: I'm sure others have brilliant ideas too. I'm not fixated on one thing and would love to hear what others think might help.

    I agree with your reasoning about what the problem is, and why it's a problem.

    But as you're suggesting a monthly cycle on this, and we already have a
    monthly development release announcement, it feels like building on that
    is the way to go.

    Nicholas Clark
  • Christian Walde at Jul 24, 2011 at 11:13 am
    Still pretty tired and i'll answer emails later on in more detail, but this one i want to answer right now:
    On Sun, 24 Jul 2011 12:40:15 +0200, Nicholas Clark wrote:

    Absolutely. One of my main issues with the current situation is that it is:

    - hard for people in the know to figure out what's going on (i can find out what i need by talking to mst, but not everyone has my thick skin and there's only one mst)
    - impossible to outsides to know what's going on here
    As in, the monthly development release which *ships working code* along with
    a short description of what's changed doesn't do this?
    No, and here's why:
    Subject: Perl 5.15.1 is now available
    I don't even bother to read those because i have little time and want to do things. I don't *ever* see those outside of p5p.

    I don't want to put down the people working on those. I think it's absolutely awesome that there's a regular release schedule!

    So where's the problem? No draw. It doesn't look interesting. I'll not waste many words and instead demonstrate how it could be done better:
    Subject: Regexes can now be sped up on multiple strings at once - Monthly Perl Dev Release - 5.15.1
    Subject: Core slimmed down and Unicode Regex fixes - Monthly Perl Dev Release - 5.15.0
    Subject: Array creation sped up and windows bugfixes - Monthly Perl Dev Release - 5.13.11
    These i would have read and i am sure many others would have too if they had been linked outside of p5p.

    The text of the announcements could be improved some too, there is a lot of boilerplate in there (the 5.13.11 announcement was *entirely* boilerplate) which will want people to quickly close the email again. Put the exciting stuff at the top! Also include links to the bug tickets so if i see an interesting thing mentioned i can go and read more about it.

    Right now they look like documentation to be filed away in folder H87/b in stall 19 in the second sub-basement, created out of duty and duly forgotten afterward. If you want other people to read them, they need to start looking like the frontpage of the Perl Vogue.

    Also, please: Be excited about your own work. It's not a shame and it will make other people excited too. :)

    But as you're suggesting a monthly cycle on this, and we already have a
    monthly development release announcement, it feels like building on that
    is the way to go.
    That is an incredibly good point and i agree completely.

    --
    With regards,
    Christian Walde
  • David Golden at Jul 24, 2011 at 11:22 am

    On Sun, Jul 24, 2011 at 7:13 AM, Christian Walde wrote:
    So where's the problem? No draw. It doesn't look interesting. I'll not waste
    many words and instead demonstrate how it could be done better:
    Subject: Regexes can now be sped up on multiple strings at once - Monthly
    Perl Dev Release - 5.15.1
    Subject: Core slimmed down and Unicode Regex fixes - Monthly Perl Dev
    Release - 5.15.0
    Subject: Array creation sped up and windows bugfixes - Monthly Perl Dev
    Release - 5.13.11
    These i would have read and i am sure many others would have too if they had
    been linked outside of p5p.
    As I said in my other post, see http://perl.com/ right now for what
    chromatic did for 5.14 features.

    If you want people to read it, you have to get it off email and into a
    blog/news feed somehow.

    Are you volunteering to write four or five of those (say, one a week
    between releases) based on the dev perldelta?

    -- David
  • Nathaniel R. Reindl at Jul 24, 2011 at 12:20 pm

    On Sun, Jul 24, 2011 at 6:13 AM, Christian Walde wrote:
    Subject: Regexes can now be sped up on multiple strings at once - Monthly
    Perl Dev Release - 5.15.1
    Subject: Core slimmed down and Unicode Regex fixes - Monthly Perl Dev
    Release - 5.15.0
    Subject: Array creation sped up and windows bugfixes - Monthly Perl Dev
    Release - 5.13.11
    +1

    I particularly like it when an open source project gets the marketing
    aspect right and highlights the enhancements and improvements right at
    the very beginning of it all. My first thoughts typically range from,
    "Man, I hope they fixed $foo," to, "Another release?!" I'm sure
    others feel the same.

    This is what really differentiates the open source bazaar from the
    commercial marketplace from the end user's perspective. At the end of
    the day, that's what pays the bills.

    /n
  • Renee Bäcker at Jul 24, 2011 at 1:55 pm

    On 24.07.2011 13:13, Christian Walde wrote:
    Subject: Regexes can now be sped up on multiple strings at once -
    Monthly Perl Dev Release - 5.15.1
    Subject: Core slimmed down and Unicode Regex fixes - Monthly Perl Dev
    Release - 5.15.0
    Subject: Array creation sped up and windows bugfixes - Monthly Perl
    Dev Release - 5.13.11
    +1
    These i would have read and i am sure many others would have too if
    they had been linked outside of p5p.

    The text of the announcements could be improved some too, there is a
    lot of boilerplate in there (the 5.13.11 announcement was *entirely*
    boilerplate) which will want people to quickly close the email again.
    Put the exciting stuff at the top! Also include links to the bug
    tickets so if i see an interesting thing mentioned i can go and read
    more about it.
    Not everybody is a good writer and I think release engineers should not
    worry about the announcements too much - they have enough to do to make
    the release happen. So it would be good to have other people than the
    release engineers to write announcements.

    - Renée

    --
    Perl-Magazin: http://perl-magazin.de
    Perl-Nachrichten: http://perl-nachrichten.de
  • Brian d foy at Jul 24, 2011 at 12:05 pm
    For what it's worth, I try to go through meaty examples of the most
    interesting features of each new Perl release over at The Effective
    Perler:

    http://www.effectiveperlprogramming.com/blog/category/perl/new-features

    I usually have most of them done before the maintenance release is
    official.
  • Dave Rolsky at Jul 23, 2011 at 3:26 pm
  • Jesse Vincent at Jul 23, 2011 at 3:38 pm
    Hi Christian,

    I'm sorry.

    Clearly, I screwed this up more than I intended.
    I have to honestly say that right now i feel the perl core is telling me:

    "Spend the cash, travel 6000 miles or fuck off."

    I _did_ present a talk at YAPC::NA about my vision for the future of
    Perl 5. It did get a positive reception. The stuff I talked about isn't
    yet policy (no matter how much I wish it was.)

    It's not yet policy precisely because it's something I've presented at
    a single conference and talked to a relatively small subset of the
    core development community about.

    Next week, I'm giving the same talk (with corrections, refinements and
    updates) at OSCON. Two weeks after that, I'm giving the same talk at
    YAPC::EU. Unfortunately, I don't think I can pay out of pocket to
    attend YAPC::Asia or YAPC::SA to present this, this time around :/

    Without having presented the talk to those 3 very different audiences
    and had the post-talk discussions with them, I've been hesitant to
    announce as anything that loks even a little bit like policy.

    I'm still new at this Pumpking thing and even newer at the "and where
    are we going" part of it. Please accept my apologies for any pain or
    frustration caused.

    My slides from YAPC are at this URL:

    http://fsck.com/~jesse/tmp/2011-07-05/221c115e-a9cb-43fa-b949-6a58192e91e7/slides-for-yapc-na

    I know the slides don't have enough text to stand on their own, but I
    hope that they're better than nothing. As soon as the YAPC::NA organizers
    post the video of my talk, I'm happy to link to it, no matter how much
    listening to myself makes me cringe.

    Again, I'm sorry for having upset you. The blame is mine alone. Please
    know that it was because I was trying to do the right thing and screwing
    up, not because I was trying to do the wrong thing.

    Best,
    Jesse
  • Christian Walde at Jul 23, 2011 at 4:00 pm
    Jesse, we talked in private and i'll say publically what i said there:

    There is no need to apologize. Everyone had good intentions, but things did not happen optimally. Your reasoning for what you did makes perfect sense to me and i hope it gives you the sort of information and confidence you need to go forward with what you're planning to.

    I think it is very good for perl to have a person who wants to do things and such a person needs our support. Especially since the things you are planning to do, at least from what little i've heard, are great ideas.

    Thank you for taking the time to answer here, and also thank you for the time on IRC. :)

    --
    With regards,
    Christian Walde
  • Richard Foley at Jul 24, 2011 at 8:08 am
    Let's see:

    1. Jesse is pumpking and wants to let people know what he (and p5p) are up to.

    2. Christian wants the world to know what Perl is doing, AND to promote the
    idea that something IS being done (as it is - we know this, the rest of the
    world clearly doesn't).

    3. Dave thinks monthly summaries might be optimal.

    4. We all want Perl promoted.

    The danger is that any *regular* summary runs the risk of being ignored,
    because the information may not always be world shattering, similar to the
    process whereby people have to think up things to say in silly daily
    performing-seal stand-up meetings at work.

    How about Christian taking Jesse's pronouncements and having a regular posting
    spot on perl.org or perl.com or news.perl.net, or wherever is most practical,
    and us ensuring that these posts are pointed to from other perl places,
    perlmonks, perlbuzz, programming forums, etc. etc. Christian is responsible
    for pushing the water uphill, and ensuring everyone links to his postings. We
    need to back him, and Jesse too, by ensuring we have these links in place to
    that core news/PR site for Perl, whatever form it finally takes.

    Ciao

    Richard
    --
    Richard Foley
    Ciao - shorter than AufWiederSehen!
    http://www.rfi.net/books.html

    Jesse, we talked in private and i'll say publically what i said there:

    There is no need to apologize. Everyone had good intentions, but things did
    not happen optimally. Your reasoning for what you did makes perfect sense
    to me and i hope it gives you the sort of information and confidence you
    need to go forward with what you're planning to.

    I think it is very good for perl to have a person who wants to do things
    and such a person needs our support. Especially since the things you are
    planning to do, at least from what little i've heard, are great ideas.

    Thank you for taking the time to answer here, and also thank you for the
    time on IRC. :)
  • Renee Bäcker at Jul 24, 2011 at 8:20 am

    On 24.07.2011 11:13, Richard Foley wrote:
    The danger is that any *regular* summary runs the risk of being ignored,
    because the information may not always be world shattering, similar to the
    process whereby people have to think up things to say in silly daily
    performing-seal stand-up meetings at work.
    Why not having regular summaries and - whenever something "world
    shattering" happens - write some kind of press release? The summaries
    could be posted on news.perlfoundation.org, blogs.perl.org and the press
    releases to the more public websites like reddit,... The press releases
    should have links to the summaries in it.

    I try to keep the German speaking people informed by posting news about
    (developer) releases on perl-nachrichten.de [1,2,3] and other forums and
    I'd like to support Christian whereever I can.

    - Renée


    [1] http://perl-nachrichten.de/index.cgi/tag/Perl%205.11
    [2] http://perl-nachrichten.de/index.cgi/tag/Perl%205.13
    [3] http://perl-nachrichten.de/index.cgi/tag/Perl%205.15

    --
    Perl-Magazin: http://perl-magazin.de
    Perl-Nachrichten: http://perl-nachrichten.de
  • Abigail at Jul 24, 2011 at 8:29 am

    On Sun, Jul 24, 2011 at 10:20:12AM +0200, Renée Bäcker wrote:
    On 24.07.2011 11:13, Richard Foley wrote:
    The danger is that any *regular* summary runs the risk of being ignored,
    because the information may not always be world shattering, similar to the
    process whereby people have to think up things to say in silly daily
    performing-seal stand-up meetings at work.
    Why not having regular summaries and - whenever something "world
    I think that if anyone creates regular summaries, noone will object, and
    most people will be pleased. But someone has to step up and actually do
    it. We can't say "let there be regular summaries", and then they will be.
    shattering" happens - write some kind of press release? The summaries
    could be posted on news.perlfoundation.org, blogs.perl.org and the press
    releases to the more public websites like reddit,... The press releases
    should have links to the summaries in it.
    Can you give an example of a "world shattering" event happening on p5p?

    There were press releases related to the Perl5 core maintainance fund,
    and the donations, but I'm wondering, "for whom?" Is there anyone interested
    in such a fund that doesn't already know about it, but scans press releases?



    Abigail
  • Renee Bäcker at Jul 24, 2011 at 8:45 am

    On 24.07.2011 10:29, Abigail wrote:
    I think that if anyone creates regular summaries, noone will object, and
    most people will be pleased. But someone has to step up and actually do
    it.
    I think Christian has volunteered ;-)
    Can you give an example of a "world shattering" event happening on p5p?
    A new stable Perl release, a new feature that was publicly discussed
    (MOP seems to be such a topic), ...
    The person responsible for the PR has to decide on a case by case base.
    I think that could be also discussed with Mark Keating who is doing the
    PR stuff for TPF.
    There were press releases related to the Perl5 core maintainance fund,
    and the donations, but I'm wondering, "for whom?" Is there anyone interested
    in such a fund that doesn't already know about it, but scans press releases?
    It's not enough to post them on the TPF blog but to send it to news
    sites etc. Big news sites cite when Wikipedia is doing a fund drive etc.
    we should come to that point, too. And I think it's great that Mark
    Keating has taken the PR part of TPF as he is doing a great job (TPF is
    more visible since then - e.g. on Twitter, Facebook, ...).

    - Renée

    --
    Perl-Magazin: http://perl-magazin.de
    Perl-Nachrichten: http://perl-nachrichten.de
  • Nicholas Clark at Jul 24, 2011 at 10:30 am

    On Sun, Jul 24, 2011 at 10:44:54AM +0200, Renée Bäcker wrote:
    On 24.07.2011 10:29, Abigail wrote:
    I think that if anyone creates regular summaries, noone will object, and
    most people will be pleased. But someone has to step up and actually do
    it.
    I think Christian has volunteered ;-)
    I was very keen on summaries before, and saddened when they stopped, but
    I was well aware how much work it took to keep them up.

    Trying to figure out *why* I liked summaries of a list that I read anyway, I
    felt that it was the summaries reached a wider audience than the list*, and
    demonstrated that activity was ongoing, at a time when there wasn't much
    else *external* happening. A summary popping up every week as a front page
    story on use.perl.org indicated activity, even to people who didn't follow the
    link to actually read it.

    We now have regular monthly development releases.
    (Sufficiently regular that they aren't actually news)

    If use.perl.org were running, that would now be a monthly story, which would
    fill the same role. So I'm not sure if the (incremental) benefit of *regular*
    summaries is as great as it was, but the cost is just as high.
    Can you give an example of a "world shattering" event happening on p5p?
    A new stable Perl release, a new feature that was publicly discussed
    (MOP seems to be such a topic), ...
    We already make announcements of new stable releases outside p5p.
    Whilst such an event would fit as part of a summary, a summary would have
    to be more.

    I'm distinctly uncomfortable about a *proposal* for a new feature being
    externally newsworthy. Publicising something at an early stage runs a big
    risk of failure of expectation management. That sounds like a plan of
    regularly announcing vapourware, given that not all proposals would be
    viable.

    Additionally lot of the new features that *have* got implemented have turned
    out to have significant problems. (That they have problems is a different
    problem with how things get done, mostly not related to "communication").
    Features that seem good well beyond first glance have turned out to have
    nasty interactions with other parts of the language.

    IIRC [correct me if I'm wrong] one year OSCON had a series of
    "state of the $foo" talks. Guido's talk on Python was mostly explaining
    carefully why a particular feature proposed for Python was not actually
    as good a fit as was first thought, and why it ultimately was rejected.
    Whilst a fine technical explanation, to most of the audience the entire
    talk was very boring. "Tell me about something that exists, not something
    that never existed".

    Talk is much cheaper than implementation, and a good, thought-through
    implementation is turning out to be far more expensive than most of the
    ones that we have had (so far). I think that fixing that is going to be
    helped by more thinking and technical discussion on list, and to do that
    we're going to need to concentrate on having fewer, but *better*
    proposals. Heck, more than proposals we need working prototypes, so that
    it's easier to explore the problem areas.

    I fear that (for "proposals") what will make for good readable contents of
    summaries will soon (necessarily) tail off on the details, and find the
    next proposal a more journalistically appropriate area to report on.
    Which would be the direction I think we specifically need to avoid.

    Summarising discussion about bug fixing, or refactoring cleanup, is boring.

    People have been asking me (for some years now, off and on) to give
    conference talks about the core, and I explain that I can't see how to give
    a *good* talk on it. Most of the stuff going on is pretty boring unless you
    already know a lot of the details, and incomprehensible unless you already
    know at least some of the details. And one doesn't know what level of
    detail the audience already know (or more appropriately, the spread of the
    level across the individuals in the audience) so it's hard to know what
    level to pitch the start of any explanation at.
    The person responsible for the PR has to decide on a case by case base.
    I think that could be also discussed with Mark Keating who is doing the
    PR stuff for TPF.
    PR is useful. What are we trying to achieve from our PR? Getting across
    that people attempting to paint Perl 5 as "special biologist word" are
    wrong? And that they shouldn't be treating the impending zombie camel
    invasion as impossible?


    I'm not convinced that *regular* summaries of the week to week ins and outs
    of this list are the best thing to be using for PR, particularly given that
    we don't have an infinite supply of expendable interns to throw at it.

    Nicholas Clark

    * Note that there are about 600 subscribers. Most of whom seem to be totally
    passive. I wonder how many of those 600 automatically filter the list into
    a mailbox that they never read. What I'm getting at is that that "600"
    figure is deceptive - the meaningful number of *participants* on the list
    is considerably lower.
  • Daisuke Maki at Jul 24, 2011 at 10:47 am
    just my $.02

    From somebody who runs a possible PR outlet (Japan Perl Association),
    I would love
    to make announcements if I can get at least a list of major points
    that should be broadcast to a wider audience.

    I'd be willing to look for the necessary information bits, but I don't
    think I can spend the
    time watching the list on a regular basis to find what's going on all the time.

    Regards,
    --d
  • Richard Foley at Jul 24, 2011 at 12:11 pm
    A small group of semi-dedicated PR people is probably a lot better than none.
    In fact I'd go so far as to say a group of PR people *IS MUCH* better than
    just the one. Mutual support, reminders, peer motivation, etc. etc. Perhaps
    an "internal" mailing list of potentail PR info, which is periodically cherry-
    picked and disseminated widely?

    And I don't mean just to the news.perl.org (or whatever it is that we who
    already know, read), I mean to forums and external news outlets, non-perl
    specific conferences, and the like.

    Ciao

    Richard
    --
    Richard Foley
    Ciao - shorter than AufWiederSehen!
    http://www.rfi.net/books.html

    just my $.02
    From somebody who runs a possible PR outlet (Japan Perl Association),
    I would love
    to make announcements if I can get at least a list of major points
    that should be broadcast to a wider audience.

    I'd be willing to look for the necessary information bits, but I don't
    think I can spend the
    time watching the list on a regular basis to find what's going on all the
    time.

    Regards,
    --d
  • David Golden at Jul 24, 2011 at 11:10 am

    On Sun, Jul 24, 2011 at 6:30 AM, Nicholas Clark wrote:
    We now have regular monthly development releases.
    (Sufficiently regular that they aren't actually news)

    If use.perl.org were running, that would now be a monthly story, which would
    fill the same role. So I'm not sure if the (incremental) benefit of *regular*
    summaries is as great as it was, but the cost is just as high.
    Not only do we have a monthly release, we have a monthly perldelta,
    which is a pretty good summary of what's new. Unfortunately, it's
    pretty dry, too, and suffers from the too-long;didn't-read syndrome.
    But it could be source material for a more publication-ready summary.
    What it wouldn't include that the old summaries had was a summary of
    significant/surprising bugs (e.g. recent smart matching issues) and
    significant new areas of discussion/debate on p5p.

    But something is better than nothing and I'd be pretty happy to see a
    more readable summary of the perldelta (perhaps with examples, as
    appropriate) posted to perl.com. chromatic's summaries of 5.14
    features there are a good example of what I'm talking about could be
    done on a monthly basis for each dev release if someone were to
    volunteer to do so.
    Can you give an example of a "world shattering" event happening on p5p?
    I'll go further in challenging this mindset and say that I think
    Christian's original message sounded like it was about giving
    visibility to the Perl community more than it was the rest of the
    world. I don't think we need press-releases or puff-pieces on HN,
    reddit, etc., to keep our own community informed of developments.

    It may benefit Perl "marketing" to do more/better press releases or to
    get articles on HN, reddit, etc., but internal development discussions
    and progress updates don't need that. Let's not confuse the two goals
    or make one harder to get done because we're trying to kill two birds
    with one stone.

    -- David
  • Brian d foy at Jul 24, 2011 at 11:57 am
    In article <20110724103007.GJ23881@plum.flirble.org>, Nicholas Clark
    wrote:
    If use.perl.org were running, that would now be a monthly story, which would
    fill the same role. So I'm not sure if the (incremental) benefit of *regular*
    summaries is as great as it was, but the cost is just as high.
    The website doesn't matter. It could be blogs.perl.org or perlnews.org.
  • Renee Bäcker at Jul 24, 2011 at 12:35 pm

    On 24.07.2011 12:30, Nicholas Clark wrote:
    I'm distinctly uncomfortable about a *proposal* for a new feature being
    externally newsworthy. Publicising something at an early stage runs a big
    risk of failure of expectation management. That sounds like a plan of
    regularly announcing vapourware, given that not all proposals would be
    viable.

    Additionally lot of the new features that *have* got implemented have turned
    out to have significant problems. (That they have problems is a different
    problem with how things get done, mostly not related to "communication").
    Features that seem good well beyond first glance have turned out to have
    nasty interactions with other parts of the language.
    Good point! I hadn't thought about that...

    But it could be nice if we have a wider audience (maybe not the
    "managers/decision makers", but more than just p5p) when a brand new
    feature is implemented and developers should test it. E.g. MOP is
    introduced and the next developer release (BTW: thank to all the release
    engineers making the monthly release happen) is released. Some special
    announcement should be made and a "Call for testers". That would help to
    find the problems *before* the next stable release is happening.

    But I think that the normal "5.15.x is released" announcement is not
    enough to attract people to play around with the new feature.
    People have been asking me (for some years now, off and on) to give
    conference talks about the core, and I explain that I can't see how to give
    a *good* talk on it. Most of the stuff going on is pretty boring unless you
    already know a lot of the details, and incomprehensible unless you already
    know at least some of the details. And one doesn't know what level of
    detail the audience already know (or more appropriately, the spread of the
    level across the individuals in the audience) so it's hard to know what
    level to pitch the start of any explanation at.
    I think one *talk* is not enough. What I dream of is a Perl 5 Core
    Hackathon/Workshop where all the details are explained. I really would
    like to help to work on bugfixes. But very often I don't know how to
    find the places where I have to change the code. And what are the
    "relations" between the different files in the tree? And what to do
    after I've changed something? Run the core tests, but is that enough?
    What files are generated when?

    To get a better understanding of these things I'd like to start a series
    of articles for "$foo" (and I'd like to chat with about that at YAPC::EU ).

    But that's off-topic now ;-)
    Nicholas Clark

    * Note that there are about 600 subscribers. Most of whom seem to be totally
    passive. I wonder how many of those 600 automatically filter the list into
    a mailbox that they never read. What I'm getting at is that that "600"
    figure is deceptive - the meaningful number of *participants* on the list
    is considerably lower.
    I'm mostly passive, but very interested. Very often I just think "Wow.
    These guys do an amazing job! I wouldn't know how to fix this or
    that..." and sometimes I think "Ok, good to know this." ;-)


    - Renée

    --
    Perl-Magazin: http://perl-magazin.de
    Perl-Nachrichten: http://perl-nachrichten.de
  • Steffen Mueller at Jul 24, 2011 at 4:33 pm

    On 07/24/2011 12:30 PM, Nicholas Clark wrote:
    * Note that there are about 600 subscribers. Most of whom seem to be totally
    passive. I wonder how many of those 600 automatically filter the list into
    a mailbox that they never read. What I'm getting at is that that "600"
    figure is deceptive - the meaningful number of *participants* on the list
    is considerably lower.
    Don't underestimate the number of people reading along on nntp.perl.org.
    Incidentally, I am one of them.

    --Steffen
  • Dagfinn Ilmari Mannsåker at Jul 25, 2011 at 10:00 am

    Steffen Mueller writes:
    On 07/24/2011 12:30 PM, Nicholas Clark wrote:
    * Note that there are about 600 subscribers. Most of whom seem to be totally
    passive. I wonder how many of those 600 automatically filter the list into
    a mailbox that they never read. What I'm getting at is that that "600"
    figure is deceptive - the meaningful number of *participants* on the list
    is considerably lower.
    Don't underestimate the number of people reading along on
    nntp.perl.org. Incidentally, I am one of them.
    Or news.gmane.org (which I use).

    --
    ilmari
    "A disappointingly low fraction of the human race is,
    at any given time, on fire." - Stig Sandbeck Mathisen
  • Paul Johnson at Jul 25, 2011 at 11:36 am

    On Mon, Jul 25, 2011 at 12:00:05PM +0200, Dagfinn Ilmari Mannsåker wrote:
    Steffen Mueller <smueller@cpan.org> writes:
    On 07/24/2011 12:30 PM, Nicholas Clark wrote:
    * Note that there are about 600 subscribers. Most of whom seem to be totally
    passive. I wonder how many of those 600 automatically filter the list into
    a mailbox that they never read. What I'm getting at is that that "600"
    figure is deceptive - the meaningful number of *participants* on the list
    is considerably lower.
    Don't underestimate the number of people reading along on
    nntp.perl.org. Incidentally, I am one of them.
    Or news.gmane.org (which I use).
    I'm sure the docs used to say that if you were running bleadperl then
    you really should subscribe to p5p. However, I can find that reference
    now, even in some older versions. But I'm pretty sure that was the
    reason I first subscribed. Well, that and I used to enjoy reading
    Larry's posts.

    However, even now Porting/pumpkin.pod says:

    If you're interested in all the latest developments, you should
    definitely subscribe.

    Just hypothesising on why there may be a number of passive subscribers
    to the list. (I count myself amongst that group nowadays.)

    --
    Paul Johnson - paul@pjcj.net
    http://www.pjcj.net
  • Richard Foley at Jul 26, 2011 at 6:35 am

    Just hypothesising on why there may be a number of passive subscribers
    to the list. (I count myself amongst that group nowadays.)
    Among other reasons is having a family and other commitments taking priority,
    also having an itch to scratch might bring one into p5p, and then having
    another itch to scratch might take one out -> remaining passive but
    interested. Every now and then the itch comes back, I think that happens a
    lot.

    --
    Ciao

    Richard
    --
    Richard Foley
    Ciao - shorter than AufWiederSehen!
    http://www.rfi.net/books.html
  • Richard Foley at Jul 24, 2011 at 9:39 am

    "for whom?" Is there anyone interested in such a fund that doesn't already
    know about it, but scans press releases?
    Yes indeed, there's no point in pointing to a tree in the forest and saying:
    "what a beautiful tree, surely everyone will notice it!" Any serious PR for
    Perl needs to take the message out to the decision makers in industry, who are
    currently opting for Java over Perl in almost every new project being
    developed. I'm a developer and am *constantly* asked to maintain some old
    rubbish until it's replaced with a shiny new java thingy. Developers on the
    ground can do a lot for promoting Perl, but only so much against the downward
    press of finance and white-collar control. The powers that be in the Perl
    Foundation might be excited by the academic brilliance displayed by their
    papers and round-table mutual .../discussions, but the managers who decide on
    how their company money will be spent, and which tools will be used, clearly
    aren't.

    Keep your eye on the target.

    Ciao

    Richard
    --
    Richard Foley
    Ciao - shorter than AufWiederSehen!
    http://www.rfi.net/books.html

    On Sun, Jul 24, 2011 at 10:20:12AM +0200, Renée Bäcker wrote:
    On 24.07.2011 11:13, Richard Foley wrote:
    The danger is that any *regular* summary runs the risk of being
    ignored, because the information may not always be world shattering,
    similar to the process whereby people have to think up things to say
    in silly daily performing-seal stand-up meetings at work.
    Why not having regular summaries and - whenever something "world
    I think that if anyone creates regular summaries, noone will object, and
    most people will be pleased. But someone has to step up and actually do
    it. We can't say "let there be regular summaries", and then they will be.
    shattering" happens - write some kind of press release? The summaries
    could be posted on news.perlfoundation.org, blogs.perl.org and the press
    releases to the more public websites like reddit,... The press releases
    should have links to the summaries in it.
    Can you give an example of a "world shattering" event happening on p5p?

    There were press releases related to the Perl5 core maintainance fund,
    and the donations, but I'm wondering, "for whom?" Is there anyone
    interested in such a fund that doesn't already know about it, but scans
    press releases?



    Abigail
  • H.Merijn Brand at Jul 24, 2011 at 9:00 pm

    On Sun, 24 Jul 2011 10:20:12 +0200, Renée Bäcker wrote:

    Why not having regular summaries and - whenever something "world
    shattering" happens - write some kind of press release? The summaries
    could be posted on news.perlfoundation.org, blogs.perl.org and the press
    releases to the more public websites like reddit,... The press releases
    should have links to the summaries in it.
    We did have a mailing list for summaries, it worked great

    perl5-summary@perl.org

    the last post ever received on there is hereby attached as reference
    material to the brave that wants to do this

    --
    H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/
    using 5.00307 through 5.14 and porting perl5.15.x on HP-UX 10.20, 11.00,
    11.11, 11.23 and 11.31, OpenSuSE 10.1, 11.0 .. 11.4 and AIX 5.2 and 5.3.
    http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/
    http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/
  • H.Merijn Brand at Jul 24, 2011 at 8:51 pm

    On Sun, 24 Jul 2011 11:13:10 +0200, Richard Foley wrote:

    3. Dave thinks monthly summaries might be optimal.
    So do I, but I also remember how big a drain it is to do for the one
    doing it.

    I've been away from (any form of) internet for a week, and already
    lagging over 320 mails in p5p only. A summary would really help not
    to ignore all that is slightly out of my focus area.

    --
    H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/
    using 5.00307 through 5.14 and porting perl5.15.x on HP-UX 10.20, 11.00,
    11.11, 11.23 and 11.31, OpenSuSE 10.1, 11.0 .. 11.4 and AIX 5.2 and 5.3.
    http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/
    http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/
  • Stevan Little at Jul 23, 2011 at 4:21 pm

    On Jul 23, 2011, at 11:38 AM, Jesse Vincent wrote:
    I _did_ present a talk at YAPC::NA about my vision for the future of
    Perl 5. It did get a positive reception. The stuff I talked about isn't
    yet policy (no matter how much I wish it was.)

    It's not yet policy precisely because it's something I've presented at
    a single conference and talked to a relatively small subset of the
    core development community about.

    Next week, I'm giving the same talk (with corrections, refinements and
    updates) at OSCON. Two weeks after that, I'm giving the same talk at
    YAPC::EU. Unfortunately, I don't think I can pay out of pocket to
    attend YAPC::Asia or YAPC::SA to present this, this time around :/
    FWIW, I have submitted a talk for PPW on (mostly) the same subject as Jesses talk (however with a more MOP-in-p5 oriented spin on it) and will likely do another one at OPW. I also know that a few other people are planning to discuss core-related stuff at PPW as well.

    I will not pretend to speak for Jesse or the large core-hackers group, but personally I think a slow buildup would be beneficial on this subject. I say this because (as Jesse pointed out) it is not policy and not everyone in the community is in agreement that it is a good thing. I am of the opinion that we should "market test" this within our own echo chamber for a little longer before we go shouting it from the mountaintops of the internet.

    <my-two-cents subject="marketing">

    Too often lately in our industry (which is a very fashion oriented industry) has hype-inflated some new tool/language/framework beyond reasonable reality. While this may be a recipe for short-term success, but is not a recipe for longevity. Perl is all about longevity! It has been around for decades and has consistently re-invented itself over the years (the programming language equivalent of Madonna if you will). Any marketing efforts we do should look for longevity and not short term hype-fueled gain (I think this is especially important given the marketing drama that has surrounded Perl 6 all these years).

    Marketing is often just seen as shallow hype and lies without anything at the core, and in many cases this is all that it is. But really well done marketing (the kind companies pay millions of dollars to ad agencies for) requires a foundation of good branding, and good branding requires two things; a foundation in truth and consistency of message.

    </my-two-cents>

    This means (IMO anyway) that any successful effort to market the "New and Improved Perl Core - Now with More Shiney(tm)" to those outside the echo chamber, will require there to be some level of agreement within p5p as well as the greater community. This is not to say that *everyone* has to agree on *everything*, only that the general plan is known by all (consistent message) and that there is a significant amount of work actually underway (foundation in truth).

    /me throws a coffee mug at the corner of the internet!!

    - Stevan
  • Steffen Schwigon at Aug 9, 2011 at 2:03 pm
    Hi!

    I'm coming late to that thread:

    "Christian Walde" <walde.christian@googlemail.com> writes:
    Recently there was YAPC::NA. At that event things were talked about
    and announced, things about perl core development. I've been reading
    vague mentions, allusions, rumors. Apparently Jesse Vincent
    announced something.

    "Fortunately, Jesse Vincent's plans for 5.16 and beyond represent
    the kind of change of narrative Perl 5 needs, regardless of what
    happens with Perl 6."
    In the thread I only found discussion about communicating things that
    *have* happened.

    I didn't find the other aspect of Christian's original email: finding
    out what is *planned* and *will* (might) come with the next release.

    Did I miss that?

    Is there something like a roadmap or TODO file[1]?
    Logs of discussions about that?

    I already know the lax answer about “no roadmap, but at least we
    deliver to it”. But seriously: maybe there really *is* something
    anywhere, maybe in a non-strict hard “management roadmap” sense but
    still useful prose?

    Thanks.

    Kind regards,
    Steffen

    Footnotes:
    [1] beside that slide deck from Jesse (that I also still did not
    find)
    --
    Steffen Schwigon <ss5@renormalist.net>
    Dresden Perl Mongers <http://dresden-pm.org/>
  • Christian Walde at Aug 9, 2011 at 2:16 pm

    On Tue, 09 Aug 2011 16:03:38 +0200, Steffen Schwigon wrote:

    In the thread I only found discussion about communicating things that
    *have* happened.

    I didn't find the other aspect of Christian's original email: finding
    out what is *planned* and *will* (might) come with the next release.
    I've been silent because i've been busy with a time-critical project alongside my normal work.

    However, to answer this: There is nothing yet aside from whatever documents Jesse holds, but i have a plan. The plan i have *will* address that in a feasible way. The big problem with all this is not to do it, but to find a long-term-feasible way to do it. Weekly summaries and eventually abandoned wikis or rigid documents are not. ;)
    Footnotes:
    [1] beside that slide deck from Jesse (that I also still did not
    find)#
    Andy Lester summarized it and linked to the pdf release: http://perlbuzz.com/2011/08/the-future-of-perl-5.html

    --
    With regards,
    Christian Walde
  • Philip Monsen at Aug 9, 2011 at 2:19 pm

    On Tue, Aug 9, 2011 at 9:03 AM, Steffen Schwigon wrote:
    Is there something like a roadmap or TODO file[1]?
    <snip...>
    [1] beside that slide deck from Jesse (that I also still did not
    find)
    Andy Lester recently provided a nice summary with links to the deck on his
    perlbuzz blog:

    http://perlbuzz.com/2011/08/the-future-of-perl-5.html

    --Phil
  • David Nicol at Aug 9, 2011 at 4:01 pm
    Well I feel caught up now. Anyone want to take over Macrame towards
    simplifying the parser by refactoring syntax into a smaller set of
    primitives using parse-time macros? Should I try to write another TPF
    grant application towards hiring myself to do it? ("do it" here means
    using the Devel::Declare-type hooks instead of Filter::Simple for
    having in-scope Macrame macros consume and replace incoming
    source-code -- should be an interesting week or so for someone moved
    to do it)

    Andy Lester recently provided a nice summary with links to the deck on his
    perlbuzz blog:

    http://perlbuzz.com/2011/08/the-future-of-perl-5.html
  • Jesse Vincent at Aug 9, 2011 at 5:12 pm

    On Tue, Aug 09, 2011 at 04:03:38PM +0200, Steffen Schwigon wrote:
    Is there something like a roadmap or TODO file[1]?
    [1] beside that slide deck from Jesse (that I also still did not
    find)
    (I know that folks have already linked to Andy Lester's summary of my
    talk. Thanks Andy for writing it and others for pointing in its
    direction.)

    http://www.slideshare.net/obrajesse/oscon-2011-perl-516-and-beyond is
    the deck in question. I haven't been waving it all over p5p because
    it's what I'm presenting at this summer's conferences and it has been
    evolving a fair bit as I've been speaking about it and discussing it in
    person with folks from various slices of the Perl community. Thanks to
    that, I've managed to cut a bunch of stupidity out of my plans _and_
    to include a bunch of excellent new ideas and information.

    While I may occasionally wave around the "benevolent dictator" flag, I'm
    under no illusion that any vision I present is anything more than a set
    of suggestions until and unless I convince the many of you who do the
    hard work that this is the right direction.

    My last scheduled presentation of this is next week at YAPC::EU. It had
    been my plan to turn my YAPC::EU slides into something more prose-like and to
    float the plan here then.

    Jesse
  • Konovalov, Vadim (Vadim)** CTR ** at Aug 9, 2011 at 5:45 pm

    From: Jesse Vincent
    My last scheduled presentation of this is next week at
    YAPC::EU. It had
    been my plan to turn my YAPC::EU slides into something more
    prose-like and to
    float the plan here then.
    I can not resist to say that I am eager to see this presentation in Riga soon
    :-)

    Regards,
    Vadim.

Related Discussions

People

Translate

site design / logo © 2022 Grokbase