Wikipedia:Requests for arbitration/Pudgenet/ContentDispute

From Wikipedia, the free encyclopedia

[edit] From (Resolve content dispute)

Again -- getting exasperated here! -- I have not edited Perl in about a month. None of the other editors of Perl were notified about this arbitration. This could not be more off-topic and inappropriate. Further -- not that I think it matters -- his compromises are of the sort of "let's add 'some people think that 2+2=5, and there's evidence to suggest it's true." It's inappropriate no matter how it is worded.

Specifically, the Garshol link is full of incorrect information. This page has been around for years, and has been widely debunked. The Shlomi Fish article, apart from being rambling, poorly written, and a general mess, is off-topic, as it is not about Perl, but communities. I don't know about the other link; maybe it would be appropriate; if one of the actual participants in the discussion had been notified about this artbitration, maybe we could find out. The TIOBE data he wants in has been widely and utterly debunked as offering any actually useful meaning to the article. The quote from Randal about the Perl conferences he himself claims was taken out of context, and Barry is factually incorrect when he says that, even if true, that it is a sign of declining popularity (as it could merely be a sign of lack of growth) (and, further, O'Reilly last week expanded its Perl offerings for this year's OSCON ... I know because they told me, and asked me to give a talk). As to benchmarks, I absolutely deny Barry's assertion that the chart is "meaningful." I think benchmarks comparing languages can only be meaningful given a wide variety of platforms (hardware and software) and code examples, and this was not. It's far better to find a detailed benchmarking web site and link to it, since there's no way to meaningfully incorporate all the necessary data into a Wikipedia article. Finally, his statement about Windows is false. He says users are limited to pure Perl, even though he himself concedes this is false in the next sentence; and further, most common modules that need compilation are available as binary distributions themselves, using ActivePerl and PPM, so they are not limited to pure Perl even if they cannot compile on their own.

I add all this not so you will judge this dispute, but to briefly show that all of this has already been gone over, time and again, and that the decisions to not include those edits by Barry were reasonable consensus decisions. Pudge 17:03, 20 June 2006 (UTC)

"Most common modules" is a big qualification. The huge number of modules available for Perl is misleading. Steve p agreed and used much stronger words than me about the C module issue. There were several modules that I couldn't install because they were written in C and I couldn't find decent instructions for the free compiler that was available, and couldn't get help even on Perlmonks, and I'm not the only one.

Yes, all of these issues have been discussed in mediation, and in more detail than this, and I'll be glad to link to the particular discussions. I might make room on the evidence page for it, but I'm thinking of waiting a day or so to see if the arbitrators have anything to say.

One of the reasons you haven't edited Perl in about a month is because, once mediation began, I abruptly stopped posting the things that had been causing the revert war. Also, other people, like Scarpia, helped your cause by reverting even mediated content. Mediation is over, and arbitration is needed. -Barry- 18:24, 20 June 2006 (UTC)

As to the C modules, I realize there is a problem. I never implied otherwise. But that doesn't make your statement "Users without a C compiler are also limited to pure Perl modules if they wish to add to the module library that comes with Perl" true. It is, indeed, false. Users can often download binary versions, or can get a C compiler and build it (in most cases, even if it is difficult to figure out).

As to the reason I have not edited Perl having to do with you not posting: I ask you to not state as fact what my unexpressed reasons are for the actions I do or do not take. Your statements to that effect are wholly inaccurate (as evidenced by the fact that I stopped editing a week before mediation began, and you say you kept editing for another week after I stopped). However, again, if you want that dispute to be arbitrated, you should have invited the participants to that dispute here. You did not do so, they are (mostly) not participating, and it is irresponsible of you to even ask for a judgment without their input, and frankly, I think it calls into question your whole motive here: in this arbitration that you are purporting is about my supposed misbehavior, you are attempting to leverage it to get your way in a content dispute I am not even the most significant part of, without the input of the editors of that article. Pudge 21:59, 20 June 2006 (UTC)

As you said, the next sentence mentions the compilers, but I'd welcome more clarification in the first sentence. Saying binary versions are available for those without a compiler would be misleading or untrue because PPM, which is the way many people download modules, still doesn't work for all modules. I'd want to be fairly clear about the difficulty some people would have and not just say there's a decent option for everyone no matter the module. Here's the discussion from round one of the mediation. (BTW, round two wasn't the one I didn't have with you. I'm not counting that as mediation. Round two began here, with the actual new discussion here.) During mediation, when I reread a related Perlmonks post, I realized that even if you figure out how to work a C compiler, you might not be able to use all of the C modules, so there are cons that I left out too.

It's fine to notify more Perl editors about this arbitration before the arbitrators get to work. I think there are already enough points of view in the mediation pages, and I think enough people already know about the arbitration, but whatever the arbitrators decide is fine with me. -Barry- 22:16, 20 June 2006 (UTC)

So, instead of my original:

Users without a C compiler are also limited to pure Perl modules if they wish to add to the module library that comes with Perl. There's free software that can enable these users to install C modules, however it tends to be poorly documented, especially for beginners.

Something like this would probably be better:

Some Windows users [though I think Windows is implied by the context at the start of the paragraph] without a C compiler may be limited to pure Perl modules if they wish to add to the module library that comes with Perl and if a binary distribution isn't available [though I don't know if PPM can install all binary distributions]. There's free software that can enable these users to install C modules, however it tends to be poorly documented, and it might not work for all C modules.

Or you could look on the bright side and say "it will work for most modules." I don't know the exact numbers. I just don't want it to sound like all modules are easily available to everyone. I've tried using very few modules, and two or three of them were in C and I couldn't install them, and that was a pretty high failure rate for me. -Barry- 07:57, 21 June 2006 (UTC)

Comment by others:

As there was no finding of fact when the Perl mediation was closed out, the undersigned other individuals disagree that the content disputes be resolved in this arbitration.

Steve p 20:03, 21 June 2006 (UTC)

(sign here)

I have a few comments on the Benchmark changes -Barry- would like added. The benchmark chart had already been added to Comparison of programming languages, but was removed based on this discussion. -Barry- brought it up in mediation as well, where the mediator called it "not very good". [1] -Barry- later stated on 4 June 2006 that "if this goes to arbitration for other reasons, I'll make it an issue." [2] I don't see this brought up in the original Request for Arbitration though. Steve p 02:46, 21 June 2006 (UTC)

There was too much opposition on Comparison of programming languages and I was afraid that if I took the time to create and maintain benchmarks for all of the main languages, it might all be reverted some day and be a huge waste of my time. I already had the one for Perl done, but it wouldn't look right to have Perl be the only language with a benchmark on Comparison of programming languages, so I didn't really fight it being removed once I saw all of the opposition. If I was able to get a binding ruling that it wouldn't be removed, then I'd consider committing to creating and maintaining a benchmark chart for the main languages, but I'm not even going to ask for that. I'm asking for the Perl benchmark only, in the Perl article, since it's less work. -Barry- 07:57, 21 June 2006 (UTC)

(Commenting on what is mentioned in the section above) "Most" and "some" are weasel words and unacceptable. "Windows" should not be re-added as Perl users on all operating systems without C compilers cannot compile C-based Perl modules. RPM, apt-get, PPM, and the BSD ports collections appear to be all very well documented, but I do not see the need why the article should be slanted either way. All of this was discussed in either mediation or Scarpia's edit summaries. Please re-read them both. If you have further questions or concerns, which you did not seem to have in mediation, I suggest posting it on the Talk:Perl page where the whole pool of Perl editors can help you, especially since your concerns on the wording have never been discussed there. Steve p 10:09, 21 June 2006 (UTC)

"Most" and "some" aren't weasel words. The MSVC tools aren't well documented. "Audrey Tang told me to try this for using C modules. I'm not even sure there's any documentation for it, and comments in the code don't count. I never tried it though. I assume it's a compiler. -Barry- 17:33, 21 June 2006 (UTC)

• Please re-read Wikipedia:Avoid_weasel_words#Other_problems, especially the bullet point "Some/many/most/all/few".
• The MSVC tools are not Perl, these are Microsoft tools. I don't think the Perl 5 commiters have much control over Microsoft's websites or documentation practices.
• Camelpack is also not Perl, it is an individual module. If it is missing documentation, I suggest talking to the author of that software, or trying something else. There are many suggestions and alternatives within the Perl README documents.

This has gone too far off topic to continue here. As I offered before, if you are having a particular problem with Perl on Win32, please let me know what it is on my talk page, and I'd be happy to help. If you would like to continue the content discussion, bring it up on Talk:Perl.

Steve p 17:59, 21 June 2006 (UTC)

The weasel word page you linked to says:

Some/many/most/all/few. Using weasel words like Some people think... leads to arguments about how many people actually think something. Is it some people or most people? How many is many people? As a rule, ad populum arguments should be avoided as a general means of providing support for a position.

Here's a quote from Argumentum_ad_populum:

An argumentum ad populum (Latin: "appeal to the people"), in logic, is a fallacious argument that concludes a proposition to be true because many or all people believe it; it alleges that "If many believe so, it is so." In ethics this argument is stated, "if many find it acceptable, it is acceptable."

I don't think there need be arguments just because people wonder about the actual numbers, but if there are some, at least we'll know that the article isn't deceptive. Adding "some" or "most" as I want to isn't an argumentum ad populum by the above definition. But if you find a way to weasle out of using those words while still making it clear that not all Perl modules can be used by all people because of documentation problems with Microdoft's free C compiler, and even with the compiler not all can be used, then say it your way.

You say "The MSVC tools are not Perl, these are Microsoft tools. I don't think the Perl 5 commiters have much control over Microsoft's websites or documentation practices." It doesn't matter! Take away those tools and even fewer people will be able to use the Perl module they want.

Whatever Camelpack is, unless you're saying it's a solution to the C module issue and has decent documentation, we still need to say that not all Perl modules are usable by all people, or to explain the difficulty some people would encounter, which as far as I've seen for the last several years is significant.

Just read the Perlmonks thread I linked to in mediation to see the problems another programmer was having with C modules. Whether you call this stuff a fault with Perl or with the documentation or a fault with Microsoft for not making all Perl modules useable, SAY IT so people don't get fooled into thinking that the huge number of Perl modules that the Perl community brags about are all installable with PPM or any other free method with decent documentation. -Barry- 18:32, 21 June 2006 (UTC)

I suggest moving this entire section of dialog to Talk:Perl. Barry is attempting to force his way using the time-honored wiki abuser's tactic, forest fire. It's best to contain it. -- Earle Martin [^t/_c] 18:09, 21 June 2006 (UTC)

[outdenting]

Earle agree completely with the Forest Fire comment. Feel free to move everything after (04:18, 20 June 2006 (UTC)) to a Perl related page. Also from the above you can add:

• pudgenet -- disagree
• Earle Martin -- disagree
• Durin -- disagree (related comments in other contexts)
• Steve P -- disagree
• jbolden1517 -- disagree

That this is part of a Perl related content dispute and perhaps can be part of evidence but debate of this kind should not be part of the workshop. I see no reason to clutter this RFAr with an extended discussion of these issues, particularly on this page. As an aside, after the move you can also add that there was no finding of fact when I closed out the mediation. I think this belongs to the next mediator to address and is nowhere near ready for the arb committee. jbolden1517^Talk 19:16, 21 June 2006 (UTC)

jbolden: Leave it to the arbitrators and clerks to remove, and if they do, then you can move it somewhere else. The motion to resolve the content dispute hasn't been accepted or rejected yet. Let's show the arbitrators what's involved to help them make their decision. -Barry- 19:35, 21 June 2006 (UTC)

The motion is in order (you'll notice I did tell Earle to leave it) the content debate is out of order. This page is for debating motions not article content. If this were ready to go to arb com, there should be pages like this, which have diffs linked to in RFCs or mediation pages, where people reached conclusions, that are cited in the evidence page. jbolden1517^Talk 19:54, 21 June 2006 (UTC)

jbolden1517, I suggest we move the discussions, and having all who disagree sign individually. Whoever does the actually moving from this section and, I'm assuming the neutrality notice section should add links to the diffs as well as where the content was moved (I'm assuming Talk:Perl for now). This will allow the arbitrators to see the original content dispute and allow them to revert if they wish. Steve p 19:59, 21 June 2006 (UTC)

I'm not sure what you just said about the neutrality notice, but the evidence page says "If you are a participant in the case or a third party, please don't try to refactor the page, let the Arbitrators do it" and we should assume that goes for the Workshop page too. I know you all want your own agendas in the spotlight, but refactoring this page when there are about 10 arbitrators that can do it if they think it's needed, is improper. -Barry- 21:17, 21 June 2006 (UTC)

"I know you all want your own agendas in the spotlight"? Talk about the pot and the kettle, Barry. -- Earle Martin [^t/_c] 23:37, 21 June 2006 (UTC)