Beefy Boxes and Bandwidth Generously Provided by pair Networks
Do you know where your variables are?

I think Casey West is right

by tilly (Archbishop)
on Jun 14, 2001 at 07:43 UTC ( #88284=perlmeditation: print w/replies, xml ) Need Help??

There has been some discussion here of whether Casey West's criticism of PerlMonks is fair and justified.

I happen to believe it is. I don't want it to be, but I don't disagree with him. And as an example of the kind of things I see that bother me, I would like to point to Software Development Testing.

No matter how much Perl Andy Lester and the 4 other monks who agreed with him may or may not know, I don't think that node deserved reaping. Contrary to what was claimed, that node had both a reasonable question and either is, or should be, relevant to many people who use this site. It not only deserved to exist. It deserved to be answered, answered well, and start a conversation that I would want to read. The kind of conversation that makes me interested in coming back here and putting in my own two cents from time to time.

But no. That wasn't to be. It got reaped because it was useless and not about Perl.

First some background. In my view the purpose of SoPW is not to be a help desk. Like the rest of the site, its purpose is to generate interesting, useful, and relevant discussion. Hopefully people get great answers from it. But it contributes a lot more to regulars than just technical assistance in a time of need.

So into this section enters a brief post.

Anyone know where I can find information about software development testing? I'm looking for something to test a web-based interface....
I read this and I see the following. Someone out there has a web based interface that they are working on. It needs to be tested. Beyond having trained monkeys sitting there and clicking on the site, how do you smooth the process?

Therefore this person with a rather common and real problem, wants to learn more. The person isn't asking for a full answer, in fact probably knows or suspects that it is a hard question, but wants some references. And if possible wants to see it discussed in terms of the web.

So, Andy and 4 monks notwithstanding, this looks like a real question. Namely, Does anyone have any recommendations (books, links, or products) about how to test software, particularly in a web environment?

Now that we know it is a question, is it appropriate for this site? Well yes. First of all it is likely that the web based application in question is written in Perl. Secondly Perl is a natural tool to reach for when you need to script an interaction. It is no accident that people do things like write Perl scripts to test for broken links in websites. Thirdly many people here work with websites, and the question of how you test the darned things is most definitely relevant for their lives. (If the idea of testing your code is not relevant to your life, well it should be.)

More specifically when I look in Saints in our Book, here is what I see. vroom is heavily involved in a web application you probably know about. merlyn has written columns on things like using Perl to test your site for broken links. He undoubtably will write future columns on automated testing of various kinds as well. I like collecting things for my grab bag of useful items. I think that a good reference on how to test web sites is something I don't have which would fit. chromatic develops something that is used in web-based applications, and is a fan of extreme programming - a programming style that places a huge emphasis on automated testing. Ovid also works with the web and has expressed an interest in the past in questions about maintainance and security. Automated testing tools are certainly applicable. And so it goes.

OK. So we have a question that to me looks reasonable. A topic that is relevant to a lot of people. A direct request for where to get started in learning more.

Surely someone could name a website that discusses it? Perhaps mention that it is hard and say that you can automate some very routine testing, but other aspects are hard to test? Possibly point out that the XP folks have thought about this and wonder aloud what they have to say on it? Maybe mention that LWP can be used to drive automated web interactions?

Of course some people don't have the patience to actually engage in that kind of conversation. I can understand that. But if that sounds like you, then don't ruin it for people who not only tolerate it, but find it enjoyable to spark such conversations. Reaping was introduced for a reason, to give the site some way of handling offensive trolls without taking vroom's energy. It is not meant for eliminating questions that you think are too trivial to be asked. And in my opinion, using it that way is an abuse of the system.

I added a link for XP since the reference was unclear, and I fixed various typos that people pointed out to me.

UPDATE (much later): I've left this node intact because it is historically important, but before drawing negative conclusions about Andy Lester, please read People learn - an acknowledgement.

Replies are listed 'Best First'.
Re: I think Casey West is right
by tinman (Curate) on Jun 14, 2001 at 11:43 UTC

    I saw the node, I didn't answer it... but I'm sorry it got reaped... in mitigation of sorts, I offer the following

    Web Test tools..

    You can also try the (expensive!) mainstay of QA people, LoadRunner.. I do know that it has good automatic scripting capability (I loadtested some of my work on it recently)...<short aside>mod_perl rocks</short_aside>
    For load tests, a free alternative is Velometer, a personal favourite of mine... or you can use the venerable ab, bundled with the Apache webserver...

    While I was trawling around, I found this link, which I thought was interesting.. if you're not a QA person (I'm not), it makes for really informative reading...

    Most of the links that I mentioned above don't really have anything to do with Perl, I'll be the first person to admit that :o) but from my personal experience, I've tested a lot of Perl applications using some of the tools... and I'd be happy to see someone else benefit from my hours of trawling the net trying to locate a free testing tool..I have to admit that the title of the root node touched a nerve... I've found Perlmonks to be much friendlier than any other online community, and I'd like to do my bit to see it remain so....

    /me ceases the self-righteous chattering and walks off..

Re: I think Casey West is right
by voyager (Friar) on Jun 14, 2001 at 09:39 UTC
    I saw the question, thought it was vague and uninteresting to me. But I agree that it would be useful to others and in no way polluted Perl Monks. Should not have been reaped.

    It has always astounded me that for every reply to a (purportedly) stupid question excoriating the poster to RTFM, there are several replies offering contructive help. If you don't think think the post is worth your time, hit the "back" button and let people who are willing to help, do so. It's not that hard.

    Is this a place where you race to answer the hard questions and show your elitism by slamming the newbies? Or is this one more resource for the Perl community, helping whoever get a resolution to a problem?

    I have thick skin and will put up with a lot for the valuable help available here. But are we advancing the cause by jumping down the throat of so many people so often?

Re: I think Casey West is right
by lemming (Priest) on Jun 14, 2001 at 09:50 UTC

    You have some good points. Sometimes, we have to look at the question and find how it relates to Perl.

    That node was reaped, while a node about baby names got attention. We were able to bend our minds around Perlish names for a baby, while a node that could of sparked a discussion on programming and testing was nuked.

    I think there may be room for both.

Re: I think Casey West is right
by coreolyn (Parson) on Jun 14, 2001 at 18:27 UTC

    It was my understanding that reaping was to eliminate 'line noise' and trolling and as such I very much appreciate use of the reaper. However, in this case it seems that the reaper may have begun to be utilized as a censor which is another ball game.

    Negative voting and gentle chiding on a node (even an anonymous one) can have more influence on shaping the education of new users than reaped ones. It provides an opportunity for less expirienced monks to participate in the maintenance of nodes and to show off what they've learned about posting here. It also allows for monks with more votes to give approval to nodes that were well 'chided' and to inspire other monks to handle the off loading of routine user guidance.

    Additionally if PM can't flirt the edges of all the other aspects of the industry it gives an image of monks as myopic thinkers that border on nuerotic.

    Perhaps devising a way of PM to proudly post the XP's of Anonymous Monk (Assuming the line noise is reaped) would be a way of quantifying some aspects of the Casey West Factor. (Possibly by factoring in the XP's of responses to Anoymous posts?)

Re: I think Casey West is right
by xphase_work (Pilgrim) on Jun 14, 2001 at 16:40 UTC
    I agree with most of the points brought up already in this discussion, and I personally think that it is a large grey area. In the case of the Software Development Testing node mentioned above, I think that more time was needed before that node was reaped. With a node such as that one, we should wait to see if any discussion is generated before reaping the node. If an intelligent discussion is formed, then then node should be left, but if there is no discussion, then reap away.

    I think that some people are quick to reap nodes like that due to past experience. For me, and probably some others here, I had to find the answers to questions like this one by myself, by digging through books, man pages, and search engines. I do feel somewhat miffed when people ask a basic question that is easily answered by themselves with little searching. Now this doesn't mean that all nodes like the above should be reaped, but it would be nice for these questions to foster more discussion.

    I would have thought the Software Development Testing node to be more interesting and better for discussion if it were worded like so:

    I searched google for Software development testing, and came up with a couple ideas like <Whatever the person found on google>. What do I need to make sure I test? Are their any particular traps I need to watch out for? What makes anyone of these ideas better then the rest?

    I agree with tilly that the purpose of the site is to generate discussion, and as long as a node does this, then I think it is worthwhile for that node to exist.


(ichimunki) Re: I think Casey West is right
by ichimunki (Priest) on Jun 14, 2001 at 23:42 UTC
    Having watched the NodeReaper for a few months now, and often dutifully considering nodes, I still wonder what purpose the Reaper really serves. The nodes don't go away (although it appears now that editors can make some of the really bad parts of it go away). Doesn't this all just draw more attention to it? I prefer to see the Reaper used to suggest edits, since poor titling and formatting make nodes less valuable.

    I think it's too bad if PerlMonks has a reputation that makes it not the ideal first choice for a place to refer newbies. I certainly find this website more inviting than wading through newsgroups and less invasive than mailing lists. It's certainly the first place I turn whenever I get the "ask ichimunki, he's a Perl guru" questions from coworkers (I fooled them into thinking this by posting my BrainBench Perl certificate). And right after I tell them to upgrade to a Perl version higher than 4 I tell them the answer on found here.
Re: I think Casey West is right
by Masem (Monsignor) on Jun 14, 2001 at 15:30 UTC
    I think another reason this got reaped away is that this really is a question that showed no attempt to do any bit of research, and was only looking for the quick and easy answer. (FWIW a quick search gives me 273,000 hits, with the first page of hits being very relevent). It's hard not to call it elitism, but when the good ol' days of the net were here, this person would possibly have gotten flamed for not doing a good amount of background research, including FAQs and the like, before posting. Add to the fact that it was posted as AM without very other many details and somewhat poorly worded, and it's hard to take the question seriously.
    Dr. Michael K. Neylon - || "You've left the lens cap of your mind on again, Pinky" - The Brain
      There was a time when this community prided itself on handling FAQ questions with answers like the following:
      Google is your friend. If you have a general question the first thing you should try is a quick search.
      As for it being hard to not call it elitism, I agree. I also find it stupid. Consider the merits of reaping vs the above short post. Reaping takes more work. The person whose post was reaped gets no feedback, their post just disappears, so you are liable to get a post along the lines of, "I already asked this but PerlMonks seems to have eaten my question..." By contrast a short post pointing at how to use a search engine will encourage the person who asked to use a search engine next time. Better yet, other random people who happen to browse PerlMonks will see that people on PerlMonks would happily start with a search engine for vague, general questions and are more likely to do that rather than post themselves. Finally giving the friendly answer makes the site nicer for everyone.

      Why then would people reap? It makes no sense to me. It seems like a recipe to get frustrated and get even more opportunities to get frustrated down the road.

      273,000 hits -- Now that is exactly why I WOULD go and start asking others. 273,000 hits, even if I could examine 1 hit a second that would take a little over three days to examine them all.

      Almost every question posted on this site can be answered with a "bit of research", so do we reap them all?

      Not everyone writes beautifully, could it be a second language issue? Maybe AM is used to chinese, and had some difficulty posting in english.

      I think what this person was looking for, and I think rightfully so, was a way to cut through the 273,000 pages of marketing and junk to find out what conclusions experienced developers had already made. Maybe AM was doing research, but decided to work in parallel. Ask the question, and while responses come in go out and start sifting through the 273,000 hits.

      I think the reaping has gotten a bit out of hand. If people don't feel like answering, or entertaining a question then ignore it. Just simply ignore it. They really go away that easily.

        Well, as someone that has had to do a lot of research (not net-wise), you learn that when you get thousands of hits, you have to learn how to prune, typically by going to the most relevent hits and deciding how to modify your search to reduce the number of hits. When I did that search, I saw at least 10 pages in the first page listing that were not ads but pointers to "web based software testing" resource pages and links. I'd use those to see if they answered my question, or otherwise use additional keywords to narrow the search. Yes, it's daunting, but that's how research works. If in that case that did not prove fruitful, I would then approach a forum, but adding that "searching on google gave me more hits that useful".

        And while most questions can be done by searching on the web, there are some times where you don't know the right terms that make the web search 'click'. For example, the question that I'm replying on about header() and redirect() functions of CGI; I don't think the original poster would have hit a solution just by web searching given his confusion. There are times that I'm trying to read up on a perl concept, and even if I know the terms, I don't know what portion of the perldocs that it is stored in; even searching at can be fruitless given the number of times that concept appears in the perl documentation, and then it just becomes a matter of reading anything that matches.

        I think in the case of the indicated post, there is absolutely no indication that the user has done any research. Maybe he had, maybe he hadn't. But because of that, there is a negative tendancy to disregard that question, particular in light of the developing trend of using the speed of replies on the internet to ask many smaller and not-thought-out questions instead of asking well-thought-out, thought-provoking questions. I took a look at how SoPW questions are handled with posting hints enabled, and there is no point where the user is giving a link on how to write good questions. (There's sections on where to post, and how to format the writeups, however). I know the site FAQ has this information in it, but I'd doubt an AM looking for the quick answer is going to read this. What I think we need to do is simply add "How to ask a good question" link in the various submit panes, so that we tell users to do:

        • Stay on topic, if possible
        • Provide as much details on the problem or question
        • Reduce any code to the basic problem code
        and other 'simple' things that, to me, would see to be obvious but not necessarily so obvious to newcomers. It may be just as easily ignored, but it could also be beneficial with very little change to the PM source or functionality.

        Dr. Michael K. Neylon - || "You've left the lens cap of your mind on again, Pinky" - The Brain
Re: I think Casey West is right
by petdance (Parson) on Jun 14, 2001 at 20:17 UTC
    I marked the node because the poster clearly didn't bother spending any amount of time thinking about what he wanted.

    I went and posted a new version, both because I'm actually interested in the topic personally, and as an example.

    I'm all for helping, but I've gotta see some indication that I'm not being asked to do someone's job for them.


    %_=split/;/,".;;n;u;e;ot;t;her;c; ".   #   Andy Lester
    'Perl ;@; a;a;j;m;er;y;t;p;n;d;s;o;'.  #
    "hack";print map delete$_{$_},split//,q<   >
      I saw that you did that, and thought it was bizarre. Why didn't you just expand on the topic in a followup node if you were genuinely interested about the subject? That way, the original poster could have learned something. Instead, you went for the brass knuckles. I think the current system is seriously flawed if relatively harmless nodes such as the one under discussion are removed. Unfortunately, this is far from the first time I've seen it happen. I'd prefer to see the power to initiate this type of action made more difficult to use, or restricted to a smaller group of more experienced & tolerant people.
        In retrospect, I agree that I should have added on to it. So we live and learn.

        I'm not sure that "tolerance", in the sense of "tolerance for people who don't do any of their own legwork", is going to help the quality of the nodebase.

        And while we're talking about "tolerance", what's with people --ing my comments that they happen to disagree with? I ++ed tilly's original post in this thread, even though I didn't agree with it much, and was all but attacked in it, because it was well thought out and well-written. To my mind, ++/-- is all about quality, not battles of personalities.


        %_=split/;/,".;;n;u;e;ot;t;her;c; ".   #   Andy Lester
        'Perl ;@; a;a;j;m;er;y;t;p;n;d;s;o;'.  #
        "hack";print map delete$_{$_},split//,q<   >
      As far as doing someone else's job, we couldn't be more in agreement. I like helping people. I remember years ago when I was trying to write game software and I stumbled upon a BBS that John Carmack and Michael Abrash of DOOM/Quake fame frequented. John actually shared and posted his beta source code for the then in development game "DOOM". It was like Christmas. I think of Perlmonks in the same way only now I feel I am (hopefully) contributing something useful back.

      That said, I HATE when someone doesn't make any effort at all and posts something like "I need a script that emails me a form and a database..". At least make an effort, but as far as the website testing software node, I didn't see it as doing someone's homework. While vaguq, it sounded to me like someone wanting to get information about an interesting topic (to me anyway) that is not something you can't easily find in a book or other resource.. While not truly a Perl question, it's certainly relavant to many monks here. Perl is the "Duct tape" of the Internet after all. Frankly, I was really suprised it was reaped.


      "To be civilized is to deny one's nature."
Re: I think Casey West is right
by nysus (Parson) on Jun 14, 2001 at 15:15 UTC
    I don't know if I want to agree with you on this one. I think it all boils down to whether or not the poster appears to just want to take or do they appear to want to give as well. And I think what perhaps struck people about this post is the poster was the former type.

    First, they post as AM. That sends out a bad vibe right off the bat. "Who is this guy?" Not that I think anonymity is bad but you must admit that it sets up an element of distrust: If I give to this person, how will they ever return the favor if they are anonymous? It's a psychological thing going on here.

    But what really sets them up as someone who only wants to take is that they offer no kind of background on the project, have no greeting, and nothing like a "TIA". Overall, it just comes across someone who just wants an answer and isn't willing to contribute anything back to PM.

    Part of what is important to making a community great is that everyone is congenial, polite and respective of eachother. Being a newbie shouldn't give you license to break these basic social rules.

    Perhaps someone could have probed this person a bit more: What kind of software is it? What is your job?, etc. Maybe he/she would have opened up a bit more shown themselves to be thoughtful and considerate instead of a boor. But as you know, first impressions are usually the last impressions, that's just the way it is with us humans. So there is little doubt in my mind that the poster is largely to blame for what happened to them.

    $PM = "Perl Monk's";
    $MCF = "Most Clueless Friar Abbot";
    $nysus = $PM . $MCF;

      If PerlMonks has become a place where AM exists but won't be treated well because AM is AM, then it has become time for me to find friendlier stomping grounds. Welcoming and accepting AM for me is the difference between having this place have a welcome mat and a sign saying, "Open, join in" or having a closed door and a sign saying, "Private meeting in progress."

      As for give and take, what ever happened to being generous? Do you really think that I ask myself, "What is this person going to be able to do for me?" before I answer a question? Of course I don't. If the question is easy for me to answer, and I have energy, I answer it because answering is fun. Now if that person comes back with 50 detailed questions showing no desire to start trying to learn, then I will get irritated. Likewise a person who has a complex project (eg write this software for me) is not going to impress. But if it is easy to answer, then I am optimistic. And if it works sometimes but not others, well it works out well often enough to keep me happy.

      So I have to disagree with the attitude that, "Well that idiot just had it coming. *shrug*" And I disagree because my gut response on seeing the question was not negative, and had I seen it before it got reaped, it would have been one I wouldn't have minded answering.

        I have to agree with tilly on this one. I just started programming in perl about 4 or 5 months ago. I have taken alot from my fellow monks here at PM. I haven't yet had the chance to return the favor and really contribute back because I don't feel my skills are up to the level of where they should be to give advise.

        This is my 6th write-up. My previous five were mainly opinions just like this (well, one code posting). I truly want to contribute back to PM as much as I can, but for now, all I feel comfortable offering is 'opinions' and responses to simpler questions.

        My first couple postings to PM were as an Anonymous Monk. I have to say that responses I got to those particular postings are what caused me to join PM and continue to enjoy learning perl. I thought to myself..self...this is a great place! People here are even willing to help me answer my idiot questions...WOW! I think some postings by AM get a bad rap just because they are that..'postings by an AM' which are viewed as people just trying to take from the system with no plans of giving back. If we shoot down all AM postings, why would those people care to join PM? Had I been shot down in this way, I probably would never have given back either, but since I received helpful responses, I realized what a great place PM was

        Ahh..the odd dog is a strange beast indeed, nobody wants him, but he always seems to be there.
        I suspect that if any of the following in that post wasn't true:
        1. AM post
        2. Way too brief, no sign of a background to the question or the questioner, borderline "do my homework" feel since the possible answers could probably fill a book
        3. Not directly Perl
        then it probably wouldn't have been reaped.

        Personally, I don't think it should've been reaped, but I'm not very surprised that it happened. I suspect that for most people, AM doesn't itself automatically bring out a negative attitude, but does make them wonder a little more consciously about the validity of the post. And like Nysus, I think that's legit, and it's not the same as automatically feeling contempt towards any AM post.

        That said, if you detect a general creeping change in the attitudes here to dismiss AMs off-hand, that's a good trend to halt. I like to think of AMs as potential future monks. People who are shy either from disposition or from just having discovered the Monastery, and who have not yet completely developed good web-board 'social skills', and so I try to give them some benefit of the doubt.

        Here's a thought: Maybe AM posts in SOPW should be rechristened as coming from "Humble Petitioner". AM immediately brings to mind "Anonymous Coward", and the association is almost entirely negative.

        -- Frag.

        But obviously many people's gut response to that post was negative. That's why the post got reaped. The poster should have taken more care to not generate negative reactions. If you are a stranger in a foreign land and you are looking for advice, it's probably wise to try and be polite.

        It's human nature for people to be very leery of one another. And so if there is no indication that you can be trusted, especially if you are a total stranger, you will probably get treated shabbily. So while this guy may have had a valid question about software engineering, the reason it failed is because he failed to practice good social engineering.

        Everything in this world is give and take. You happen to be a particulary generous individual because you really enjoy answering questions. You give advice and take back enjoyment. I extend you heaps of gratitude to you for being that way. You in particular have helped me a great deal. Unfortunately, not everyone has the same disposition as you. All I'm saying the poster should have understood human nature better and that everyone is not as generous as yourself. We may hate or like schmoozing, but that does not change the fact that it works.

        I personally think it would be near impossible to create a more newbie friendly place than PM. I am still astounded to the help I've received here and one day I hope to be able to return all I have taken. One of the major reasons I think this is so is because people take the time to extend the simple courtesies such as "bro", "fellow" or a simple "thanks". It gives PM a real sense of comradery and I think it can be hard to tolerate people who don't seem to want to respect the PM culture.

        $PM = "Perl Monk's";
        $MCF = "Most Clueless Friar Abbot";
        $nysus = $PM . $MCF;

        As usual, tilly, your comment is clear and valuable. And, as often enough, you overstate it through your own impatience.  I agree with everything about the original post except the *title*.  (tinman's response above is sufficient support.)  And again: "then it has become time for me to find friendlier stomping grounds."  Perhaps the monk who considered it is a newbie at considering (like maybe 1 week).  And I know that you and several others have been riding herd on overenthusiastic considering since it started, but still, this is hardly typical (and, in fact, almost unique).


A reply falls below the community's threshold of quality. You may see it by logging in.

Log In?

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: perlmeditation [id://88284]
Approved by root
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others contemplating the Monastery: (4)
As of 2021-12-06 21:35 GMT
Find Nodes?
    Voting Booth?
    R or B?

    Results (33 votes). Check out past polls.