Beefy Boxes and Bandwidth Generously Provided by pair Networks
Clear questions and runnable code
get the best and fastest answer
 
PerlMonks  

Summing up recent ideas into a concept: Code vs. Prose

by deprecated (Priest)
on Apr 29, 2001 at 20:30 UTC ( #76482=monkdiscuss: print w/replies, xml ) Need Help??

Folks, I've been mulling this about in my head for two weeks now. mothra noded something about the quality of perlmonks vs. irc vs. newsgroups. I initially thought, "pfff, maybe /s?he/ doesnt get very much out of perlmonks, but I sure do..."

So let me tell you, then, where this started really becoming foremost on my mind. I'm a real unix nut. I know several different flavors, I'm fluent in shell, and even know the internals of the various config files pretty darn well. My skillset and knowledge has pretty much kept me employed as a Unix/Networking weenie. Through being a Unix guy, I learned perl. My resume includes mentions of Unix, Perl, Shell, Apache, Cold Fusion, DB2, and MySQL. So when I go out looking for a new contract, I get offers from lots of places. Recently, I was brought on with an organization to do a number of things, but mostly, I am a perl hacker.

This is where mothra's post comes in. I was at work, taking a break from some code that was hurting my brain. I thought I would go and slip into some perlmonks, and also get some new code into my head, hoping for some insight or a novel way to approach what I was doing.

I didn't find a thing. I'm a pretty good perl programmer. I get what I need done, done. I even manage to sometimes come up with novel ways to do things myself. I'm at a precipice as a programmer. I have all the books I could about perl, and I'm starting to pick up books on Computer Science and other languages instead. I can't find any higher-level information. I am not getting any deeper into programming, perl, or computer science through Perl Monks.

Instead, I find that I'm stagnating here. I'm reading nodes, and theyre almost all prose. Only two of the Best Nodes of last week are actually code. I'd like to be electrified like the first time I read Lincoln Stein's code. I want to read nodes that shake my brain up and insert new ideas for how to code and ways to approach things.

I see a good parallel in WebHick's excellent recent post, Perl Enlightenment and Personal Journey To It. I'm knee deep in the water, it feels great, and I want to go scuba diving. The harder I look, though, the more I see that I have to go looking to other places to find new things in programming. For example, I've got these books in my shopping cart on amazon:

Maybe the more astute among you noticed that I'm looking at Lisp (scheme actually) and Haskell. The reason for this isnt that I dont like Perl. I love perl. I really do. But I dont find anyone who can start shoving programming theory down my throat who also codes in perl.

I'm already a professional. I dont have the luxury of going back to school full time to learn more about programming, and to learn "deep code." So I rely on the community. I'm getting nowhere. I had hoped perlmonks could push me past this speed bump. I learned some new things, but I'm still just a perl hacker. I'm still just somebody who writes code. Heck, I even dream (literally) solutions for code problems I have. But its just putting legos together, not envisioning the whole picture.

I discussed this with several people in the CB. While le correctly mentioned that this is something that gets tossed around a lot. Signal vs. Noise. I've been on USENET and BBS's and Mailing lists for 11 years now. We hear this over and over. I dont know what can be done about it, but I am venting particular observations in the hope we can do something about it this time.

While I was whining, Petruchio brought up an interesting term, "reputation inflation." This is, I think, to a large extent true. We see a lot of posts with HUGE reputations. This is due to the massive population growth of the monastery. As an abbot, I get 25 votes a day. I've only been here for six months, and made a bit less than 100 posts. I didnt really mean to, but I have made a whole lot of pretty useless posts. I try real hard to make sure that the post is useful and on-topic, but when I look back, I see that I havent contributed much of anything.

Perhaps Ovid's recent post, Stubborn as a Saint is indicative of this. I dont know what specifically got his attention, that he felt compelled to tell people to read his code more carefully. In fact, I dont think a lot of people here are reading code at all. We read conversation and discussion because its interesting. It is, however, junk food. We have no incentive to post code, nor to review code. "Code" posts get very low (usually < 35 ) reputations. This post, while full of rhetoric and interesting thought provoking ideas (okay, maybe I flatter myself), will probably score at least that within a couple days.

Why is this?

I dont know. I wish I had a solution. The thing is, we have had zillions of good suggestions proposed in Perl Monks Discussion, but few of them are ever implemented. This isnt a slight to vroom, but I think we could be fixing this problem and making better programmers.

We dont need to revamp the Voting/Experience System, because we can't agree on how to do it. And besides, thats a symptom, not a cause. People like me should not be getting to level 7 in six months. Its not because the voting and experience system is off, its because we're all posting these saccharin, brain junk food nodes.

Ugh. I cant come up with anything more to say useful on the subject. I think I made my point clear. I wanted to post this (despite the obvious irony of bitching about this specific kind of post) because we have users here who are capable of providing solutions for the monastery and maybe, just possibly, solutions for me. Yeah, tell me where to find enlightenent.

*sigh*

brother dep.

--
Laziness, Impatience, Hubris, and Generosity.

  • Comment on Summing up recent ideas into a concept: Code vs. Prose

Replies are listed 'Best First'.
Re: Summing up recent ideas into a concept: Code vs. Prose
by converter (Priest) on Apr 29, 2001 at 22:36 UTC
    Often when I come to the monastery it's to take a break from working on code and the last thing I want to do is read more code. I suspect a lot of us visit during breaks from coding and feel the same way when we fire up the browser and click the perlmonks bookmark.

    Maybe we just need to accept the fact that a lot of monks don't want to stare at huge chunks of unintentionally obfuscated code when they visit, and try to find an approach to the subject of learning Perl that is more palatable to fatigued perlers looking for a break from work.

    I was thinking about my Perl studies yesterday, and trying to think of something that I could do at least once a day to learn some new and interesting Perl concept. While I pondered, I remembered fondly my old copy of The C Puzzle Book and wished there were a similar volume of Perl puzzlers available. Perhaps a Perlmonks "Perl Puzzlers" or "Perl Quizzes" section with short but meaty chunks of Perl code would be a good start? (nysus had a good start on something like this, I think, but I haven't noticed any new posts lately).

Re: Summing up recent ideas into a concept: Code vs. Prose
by merlyn (Sage) on Apr 29, 2001 at 20:34 UTC
    How ironic that I'm upvoting you on this post, and it'll probably end up in the top 10, displacing yet another code post. {grin}

    -- Randal L. Schwartz, Perl hacker

Re (tilly) 1: Summing up recent ideas into a concept: Code vs. Prose
by tilly (Archbishop) on Apr 30, 2001 at 01:43 UTC
    First of all I find it interesting that your reading list includes a book that was recommended here not long ago.

    But as for the rest, I understand and sympathize. While I have found a lot of people in the Perl community who taught me a lot about how Perl works, how the things it interacts with work, how Unix works, etc, I can only point to one who really taught me much about computer science per se.

    However what you learn elsewhere applies to Perl. And by taking the energy to translate from Scheme to other languages, you undoubtably will understand the underlying principles better.

    As for PerlMonks filling this purpose, I don't know that it really can. The problem in my eyes is that there are too many audiences here. The Monastery is growing by leaps and bounds. It contains a pretty good mix of discussions. Aimed at a pretty broad mix of people. Potentially valuable conversation that doesn't fit the basic format and doesn't hit a broad audience doesn't happen to the extent we might like. Plus, as with most value added suggestions, trying to cover those issues takes a lot of energy.

    Now I do have an idea for how we might solve most of these issues and get you your content. Unfortunately the content that is at the top of your reading list might take a while to arrive though. (See an upcoming discussion for details...)

    UPDATE
    The discussion is at Should PerlMonks have official courses?.

(jptxs) Re: Summing up recent ideas into a concept: Code vs. Prose
by jptxs (Curate) on Apr 29, 2001 at 23:31 UTC
    perhaps this could be addressed, in part, by YAPMS (yet another perlmonks section). I'm at a loss as to what it would be called, but its contents would be theory talk. I too find myself wanting a bit more these days. I think I still get a lot out of the Q/A in all its forms here, but I also don't think I'm as good of a perl hacker as many others. I could envision a section where code led discussions can take place. Basically someone posts some hairy code and then it is refined and discussed in CS and formal programming model terms. I see some of this in the Obfuscation side of the house as some of these concepts fall out of a good japh just by accident. Formalizing it could benefit those of us who are approaching the glass cieling and also those who've been making mouth-blowfish on it for a while wondering where the door is.

    Of course, it's all fine and good for me to propose this as I'd be the one reading and learning. The real burden would fall on those among us who already know that sort of stuff. The question is, do they want it? And do they feel like populating it for our benefit? I face the same challenges at work where, as scary as it is to me, I am considered one of the most technical people in our group. There have been many times where people in the group have yelled that they need info and or help with topics, and, though I agree that it's a great idea, I end up doing all this work for it and then dealing with the complaints and short-comings when these people do not recieve instant enlightenent from my words. Don't get me wrong, I love sharing my knowledge. But when you're not a teacher by trade, one can only spend so much time teaching and not on the trade which made you a teacher in the first place.

    Two side of the coin. Y'all can flip at will =)

    "A man's maturity -- consists in having found again the seriousness one had as a child, at play." --Nietzsche
Re: Summing up recent ideas into a concept: Code vs. Prose
by clemburg (Curate) on Apr 30, 2001 at 12:33 UTC

    I'm knee deep in the water, it feels great, and I want to go scuba diving.

    Part of seeing more of the ocean is to accept that you have to wear different equipment in different areas. The stuff that feels free and easy when diving in Florida will kill you when you fix oil rigs.

    Less pictorial: I don't think Perl can teach you all about programming. No programming language can do this. It is neither reasonable to expect this of Perl, nor of this community.

    Christian Lemburg
    Brainbench MVP for Perl
    http://www.brainbench.com

Re: Summing up recent ideas into a concept: Code vs. Prose
by Masem (Monsignor) on Apr 30, 2001 at 17:38 UTC
    Part of the problem is that with perl, I would guestimate that somewhere between 70-90% of it's use is for CGI scripting, which means that 70-90% of the questions and related discussion is about "how do I access a DB?" "how do I do secure transactions?" etc. These topics tend to either be answered (correctly) by RTFM or a discussion on how to do them; code really doesn't help. Certainly there's nothing wrong is us answering these questions (and attempting to reduce their numbers by helping newbies to RTFM first), but these don't lead us to improving knowledge for the more advanced monks.

    ObfCode is a potental way to learn for intermediate to advanced monks, but this is more for tricks with perl rather than improving programming style.

    So you're looking at more like the article contents at perl.com for advanced users. Here, we sorta have that between the Tutorials and Reviews sections, though neither is a full substitute for that. What I suggest is similar to what some print jounrals do for programmers: a month in advance (or more) a topic is selected, and numerous articles are sent in to cover that topic, with an editor selecting the best ones. Translating it to something here, I would add a new section "articles", with general topics (eg "network sockets programming", "threaded processes") offered up for suggestion, and any monk wishing to write an article may do so. The topic suggestions should be limited to only higher level monks (6th or above?) to avoid having topic pollution ala Q&A/SoPW duplication. "Articles" can be as fundamental as a tutorial, or robust as pure example code with commenting, but it ought to end up being a combination of prose and code so that a message on how to improve ones programming is taught. Which means the message shouldn't be spoonfeed, but nor should it be so vague as "simply read the POD for everything you need". These articles could go on to include psuedo-lessons: after reading the article, the reader could be challenged to modify the code to add new functionality and the like.

    (From a PM programming standpoint, there's a few things that would have to be done, but nothing too drastic that I can think of).

    Having such a feature would help distinguish PM from USENET or other web forums, which end up being more Q&A instead of learning tools.


    Dr. Michael K. Neylon - mneylon-pm@masemware.com || "You've left the lens cap of your mind on again, Pinky" - The Brain
      I would guestimate that somewhere between 70-90% of it's use is for CGI scripting,
      Most of us that have to make a living providing support services to the Perl community at large put that more down at the 40% mark.

      It just seems higher because CGI is much more complex than it first appears, and is usually undertaken by people who are light-duty or non- programmers, and therefore show up more in forums trying to get help.

      But many people I know have been hacking Perl heavily for years and have never even put up their first CGI program.

      Of course, "how do I access a DB?" is a question that is valid whether or not you're in a CGI space, so maybe you've miscategorized some questions. {shrug}

      -- Randal L. Schwartz, Perl hacker

        I'll take your word on it with the years of experience that you've had in perl'ing. But I think the statement that I was trying to make, that most of the questions that we are asked in SoPW/Q&A, are "uninteresting" or "unchallenging" as they seem to represent uses of perl that a majority of perl programmers are involved, and thus already should have a good base of knowledge to go on (eg FAQs, USENET, PM archives). Looking at the current front page of SoPW, there's only 1 out of 10 questions that I would consider to be 'challenging' to find the answer for, in that I would expect to learn something new when helping this user out; the rest I can answer with basic knowledge.

        And that is the atrophy that the original thread here to aiming at; there's too many 'uninteresting' yet still useful questions being asked that doesn't help those on the high end of the XP scale. I have no problem answering the uninteresting questions, as long as I have the oppurtunity to learn from others on more advanced questions, and while I still find that this is happening for me, I can see how others a bit more advanced in perl would find the lack of new knowledge deterimental to the PM experience. Which is why I think having Articles aimed beyond the beginning user would be a perfect addition to beat the PM atrophy.


        Dr. Michael K. Neylon - mneylon-pm@masemware.com || "You've left the lens cap of your mind on again, Pinky" - The Brain
        Translating it to something here, I would add a new section "articles", with general topics (eg "network sockets programming", "threaded processes") offered up for suggestion, and any monk wishing to write an article may do so. The topic suggestions should be limited to only higher level monks (6th or above?) to avoid having topic pollution ala Q&A/SoPW duplication. "Articles" can be as fundamental as a tutorial, or robust as pure example code with commenting, but it ought to end up being a combination of prose and code so that a message on how to improve ones programming is taught. Which means the message shouldn't be spoonfeed, but nor should it be so vague as "simply read the POD for everything you need".

      I really like this idea. Additionally, I like converter's idea of a Perl Puzzlers book. On dalnet #perl, we have a wonderful user (who I do not think frequents this site) named PerlCure. He does something I have yet to see anywhere else: He makes quizzes for the above-average-but-not-stellar perlmongers present. For example:

        <purl> quiz ? [ sound of cheers from #perl crowd ] <purl> quiz --> <purl> BEGIN{ local $@; my $f = 0; 2/$f } print "Done" <purl> bug or feature ?
      (original credit for the quiz apparently goes to bline)

      This sort of thing is interesting, but we could also, like Masem pointed out, have actual articles attached to this. I get something akin to this on the fun-with-perl list at least twice a week. We get golf, but we also get very interesting and informative questions such as:

        I hope abigail doesnt mind the repost...

        Date: Thu, 29 Mar 2001 20:40:16 +0200 From: abigail@foad.org To: fwp@technofile.org Subject: [FWP] So, you think you know Perl? Your exercise for today: 1) Predict what echo "" | perl -ple 'print q' does. 2) Run the above command and compare it with your prediction. 3) Explain any differences. Abigail

      Folks, these are the kinds of things that get me excited about perl, and make me interested in a new function I dont use or underuse, or use in a different way. I get a ton of innovation out of just little things like that. But we dont really see these too often on PM except for in ob-fu. Which is fun... but useless in terms of my really growing as a programmer. Well, almost useless.

      Masem is right. Articles are a great idea. I dont think I'm qualified to make an article, though I'm happy to try. Perhaps we could have a Quest first to see what articles are wanted, and then, as Masem suggested, have articles submitted. Gee, that almost makes us a periodical. (This particular reason is why I subscribed to <sniff> TPJ)

      One other note here, tilly suggested I go have a look at Dominus' website, which is excellent. I suggest that for anyone interested in learning more about perl.

      I am still very interested in seeing what people have to say about this thread, and I'm glad to see what people have had to say thus far. I think the response is mostly uniform. People would like to see more intermediate-upper level content, as a lot of us have moved past the beginning and even intermediate levels of perl programming and are looking to transcend just hackerdom.

      yours as always,
      brother dep.

      --
      Laziness, Impatience, Hubris, and Generosity.

        I dont think I'm qualified to make an article, though I'm happy to try.

        I would think that anyone on PM that considers themselves to be a non-perl-newbie (say, over level 4), has the ability to write a good article. You may not have the stuff to write something as easily readible as merlyn or chromatic or our other published monks, but you probably know something they don't, and that's what my Articles idea is all about. And if the first time you write something, the article doesn't jive well, just simply use any comments you get on it to improve the next one that you write; given that PM is a friendly community, you should not have to fear rejection or humiliation for writing a poor article. And writing such articles could only improve your own writing skills for your career.

        Mind you, we don't need Articles to be come Ms. Slorovski's 3rd grade English class, but I'm sure that this will not happen...


        Dr. Michael K. Neylon - mneylon-pm@masemware.com || "You've left the lens cap of your mind on again, Pinky" - The Brain
Re: Summing up recent ideas into a concept: Code vs. Prose
by Jonathan (Curate) on Apr 30, 2001 at 01:42 UTC
    Not only do I agree entirely with deprecated but I'd go further. Sadly (IMO) this site has been on a downward spiral for even longer. Losing Abigail was perhaps the start of it.
    These days I login, usually just for a few minutes a day and check out the latest postings made by the monks who have made the greatest impression on me... btrott, chromatic, merlyn, dominus and lhoward and then leave.
    Another problem is the number of votes mediocre nodes acquire is often out of all proportion to their real value. So, if I've time on my hands I'll upvote the 'worst nodes' (they rarely deserve it) and downvote the 'best nodes' (they rarely deserve it).
Re: Summing up recent ideas into a concept: Code vs. Prose
by Mungbeans (Pilgrim) on Apr 30, 2001 at 13:44 UTC
    Is this a filtering problem?

    You want to see more code but you're getting distracted by all the non-code posts?

    Maybe attaching keywords to nodes, and having an option to filter by keywords would help? I mean the sections make a good breakdown (cf craft) but keywords would make it easier to search newest nodes etc. I don't know how much work this would involve, however, or how many times it has been proposed already :-)

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: monkdiscuss [id://76482]
Approved by root
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others drinking their drinks and smoking their pipes about the Monastery: (7)
As of 2020-12-03 01:47 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?
    How often do you use taint mode?





    Results (49 votes). Check out past polls.

    Notices?