Re^6: A request from level 21 (feedback)

[tye]

Thank y'all for much of the feedback thus far (just to be clear, no, I wasn't the one asking for it).

At one point yesterday, my 20/20 hindsight told me that things may have gone better had the trial balloon been preceeded by an announcement with feedback coming as replies to that¹. Even at that point, I knew that it was easily possible that pre-announcing was considered and rejected. Two likely reasons that occurred to me were 1) that the available time was such that an announcement together with an implementation was an unrealistic goal (and announcement w/o implementation was somewhat pointless); and 2) that pre-announcements are often followed by quite low-volume and uninspiring feedback, rather thwarting the purpose of the trial balloon (though I'm very much not implying that there must be a cause-and-effect there).

¹ Well, except that it isn't terribly unusual for monks to post nodes complaining about some noticed site change despite there already being a very recent announcement at the top of the section repeatedly documented as being the place for discussing aspects of the web site. But that is quite tangential to my point.

Today, I am not at all certain that pre-announcement would have improved the results. I had originally strongly wondered if some of those complaining might not have seen the (likely) source of the idea and the accompanying musings in support of it. Today I noticed that such was very unlikely (the most volumous arguments just didn't touch on those points much).

At least two things seem fairly clear to me, at this point. We have two senior monks who are strongly pretend-religion intolerant. And we have one senior monk who is rather upset.

Update: downvote if you want but a rose is still a rose and a spade is a spade.

I'm not sure I grok the idiom(s) involved there, but it sounds to me like "many people may object to what I've written, but it doesn't change the fact that I'm right in criticizing what I've criticized". Which is pretty close to "it doesn't matter how many people disagree, I'm still right". I'm not putting those words into your mouth, I'm only drawing some similarity. It reinforces my impression that there is some "heat" behind your writing that might cast light upon your writing were it known. I am certainly puzzled by some of the things you have written.

This smacks more of "do it now and ask for forgiveness afterwords" than anything else

So you suspect ysth dissembled when he stated why he implemented the feature? That makes me puzzle at what hidden motivation you theorize, as no likely one occurs to me.

Those discussions put the lie to your justifications

I'm not certain I grok that idiom, much less understand how you perceive it or intended to use it. But checking definitions and typical uses of it, my impression is that "lie" appearing in it is no accident. Despite the most common definitions being close to "to show to be false", most uses are closer to "to show one's own words to be false" and several definitions include the "to show that somebody is lying" usage.

I think that idiom would be quite appropriate to use had ysth been the one to link to those previous discussions... before implementing the feature... and quite recently.

But those previous discussions don't even "show to be false" ysth's justifications, which were (in part, and with emphasis added):

" Sometimes you have to make changes first to get to the meaningful discussion part. Otherwise you mostly get a lot of comments from people who like to solve social problems, who tend to be the people who would end up being happy almost no matter what "

So, even had ysth read those discussions immediately prior to implementing the feature (I have no knowledge as to that), his justifications would still make sense to me (I refrain from claiming to agree or disagree with them, mostly because I'm not certain whether I do).

And what's with posting with your "Novice" secondary account? Afraid of losing some XP?!?

Wow. That's some punctuation.

A while ago I decided that I strongly preferred to only "speak" from one account. I don't recall other gods strongly expressing concurrence in self-applying such a restriction. I go to extra effort to conform to that self-set standard.

But even if ysth feels a similar motivation, extra effort is indeed required. Some actions require an account that is a member of gods. Mixing such actions among other actions that one exercises extra caution in order to try to ensure that one doesn't accidentally do those other actions from the gods account, is not fool-proof. And there can also be completely different reasons one might decide that it is more appropriate to do some "speaking" as one of the gods proper.

But that you would theorize that ysth would make that choice out of a concern for XP certainly surprises me. If ysth were concerned much with his own XP, he could give himself however much XP he cared to, of course (he could even be quite sneaky about it). I also find ysth's broader history of actions quite in conflict with the idea that he might be trying to maximize his own XP or be simply avoiding the experience of losing XP.

It's sad when a simple suggestion like mine in 694443 that would make Super Search easier to use goes unimplemented

Well, I'll disagree with your wording (quoted above). A minor change to wording (on Super Search) certainly might make Super Search a bit easier to understand. Your suggested change (unless I misunderstand it) does nothing to change how Super Search is used or works.

I'm not sure what your theories are on how some site changes come to pass vs others.

My reaction to that eventual suggestion of yours for a minor wording change was that it seemed like it might well be an improvement but the difference was subtle enough that I was unsure how much harder it actually made it for one to misunderstand the meaning of that particular Super Search feature. Further, my impression of it was that it came rather late and deep in a thread and so was relatively little-seen, which was unfortunate as I'd have liked to read responses to it.

So you feel that your minor wording change should certainly be prioritized above an attempt to help monks better deal with trolls? (not that using the word "prioritize" here isn't likely an abuse of it)

Since my description that I assume lead to this feature was closer to mentioning it in passing than to calling for an immediate change to be made:

" Another idea that occurred to me was [...]. We could "

I expect that ysth's reasons for implementing the feature included him finding the idea appealing (as in, something that might benefit the site and its users). Of course, I also expect that his reasons included the ones that he stated.

I would also like to point out (to all) that I said:

" I only mention them to note that the current trolling is likely to be anonymous and thus these proposed changes won't apply. "

So I stated that my feature idea is unlikely to have much (if any) actual benefit in discouraging trolling in the near nor medium-term future (not that explictly). I was also coyly noting that I felt that the sum total of ideas in that thread were also likely quite moot over that same theorized time span.

As for this feature's likely eventual effects on "the XP system", I'm not convinced either way. I see potential benefits, but nothing major. I've read dire warnings, but have yet to be convinced by them, either.

I'm quite surprised by the (apparent) assumption that readers of this site will just naturally recognize the relative seniority² of node authors based on the presence of one of the 28 level names. I'm often disappointed by how so many Perl programmers demonstrate a profound lack of knowedge of Perl's precedence table (see perlop, near the top). And that has fewer levels, is rather important when programming in Perl, and has a lot of underlying logic that dictates much of its structure.

² I'll just skip the argument about how "XP" doesn't say how good you are at Perl other than noting it in this thoraxnote, as I find people who feel the need to state that almost silly.

Looking at the list of level names, I think I'm likely to remember the lowest level name and to recognize the two next-lowest ones but have no idea of those two's relative orders. The rest of the level names I might recognize but I'd have no idea what their relative positions would be.

You are clearly much more obsessed with the PerlMonks XP system than I, since you can easily make proclamations about a "lowly" whatever. But perhaps the increased visibility would lead to many memorizing quite a few level names.

I'll note that I agreed that showing level numbers is probably not wise (but also note that I don't foresee near so dire consequences as you, Argel, and do allow that there could be benefits even to that), simply because I believe that was suggested in this thread at least once and likely implied as a suggestion additional times (perhaps even by the root node's title).

I'll also repeat that I think producing an "I'm a troll" badge of honor is not a good idea, for some reasons echoed in this pair of threads and for additional reasons that I wrote about before (since this proposal has been re-suggested in this thread).

If the feature stays, I too would like to eventually have the extra work done to make the title be displayed based on the level of the author at the time the node was posted. I recall that idea being discussed years ago and finding it entertaining (in a good way). But, no, dragonchild, I'm pretty sure that such a change would only impact nodes that had been posted after the feature is implemented (earlier nodes sh/would be unadorned, IMHO). I'm certainly not going to try to reconstruct years of history for hundreds of thousands of nodes (and I seriously doubt jcwren's site has sufficient information even if somebody did want to try to do such).

It has been traditional for new features to be "opt-in". That is, for the default to be the old behavior (understandably). But I've come to appreciate that often a better approach is "opt-out", not so much by defaulting to the new behavior, but by defaulting to "more behavior" over "less behavior", especially for relatively minor/subtle features.

If you default to "more behavior", then everbody who cares to notice automatically gets notified of the potential for that behavior. Monks who dislike the behavior are motivated to find the documentation of the behavior and/or how to disable it. The end result is everybody who likes the behavior gets it, everybody who sufficiently dislikes the behavior avoids it, and those who don't care enough have it present offering them the opportunity to care more about it.

If you default to "less behavior", then you can list the change in Tidings and write about it in the site documentation and even add a checkbox for it on one of our tons of "settings" pages, perhaps even write a root PMD node announcing it. And if you do all of those things, my experience says that there will still be a ton of monks who would appreciate the optional behavior but have no idea that it has been implemented.

[tye]

Although the number of people who I have seen provide any feedback on this feature is quite small, I see a compromsie going forward that seems clear.

I haven't seen any strong arguments for this feature being an important improvement. Several people have expressed that they find it offensive. So my first thought is to just roll it back (though I'd wait for more feedback before doing that).

But there have been several monks who expressed that they enjoyed the feature (in nodes or in chatter), so perhaps the few who really liked it could do extra work to enable it for themselves.

So if we just patch the default site CSS so that the level information is diaplay: none, then the feature mostly disappears but those who want it can rather easily get it back (by modifying their custom CSS settings).

Note that people surfing PerlMonks with a non-CSS browser will still see the level information. Even my free cell phone's browser supports CSS so this would likely only include the few non-GUI browsers that constitute a tiny fraction of our visitors. Making it a user setting would require several more patches and would just worsen the over-crowding of user settings for something this trivial.

I believe that one could even force the content to be a fixed "not a likely troll" label of their own choosing.

Of course, this will also likely force the feature into obscurity. I suspect a large majority of monks will be unaware of having the potential to enable it and that this majority will grow over time. But that seems a valid compormise.

[Argel]

That sounds like a reasonable compromise.

Regarding "hidden" features, maybe a "Common CSS code" node could be created and a link to it added to the display preferences page? Then more Monks would know what options are available to them without the need for a significant patch.

[Gavin]

++ Great idea Argel I'll second that.

[tye]

Sorry, it has been especially busy for me at work so it took an inordinate amount of time for me to get around to this.

I've patched the site default CSS to include:

.attribution-title { display: none; }

If you would like to (again) see each author's level title listed in their node by-lines, then go to Display Settings and add the below line to your "On-Site CSS Markup:"

.attribution-title { display: inline; }

I've also updated Tidings.

[Argel]

I apologize up front since you clearly put a lot of time into your post but I think I have fallen into the "it's really hard to let go" trap I often find myself in. Therefore I think it's time I bowed out (for the most part) from this conversation. I appreciate the time and thought you have given this -- especially the compromise you propose below -- and again apologize for not giving you the proper response you deserve.