Wikipedia talk:SweeTFA proposition

Excessively complicated

Why do we need to use something like pi, where we must either write a program to calculate the thing every day or have someone remember exactly what digit we're up to when we could just give Raul 4 10-sided dice of the 0-9 format instead and let him roll numbers? In a decade or so when we hit 10,000, we buy him another d10. --tjstrf talk 22:46, 26 March 2007 (UTC)

Because that is not transparent. Thats the beauty of it. We create a page with the first 10,000 digits of pie (10kb) and every day someone goes through and crosses out the ones that have been used, everyone will be able to see what was done from the edit history. If there is a simpler way though that would be even better - Pi is not the point here, the point is random and transparent. David Spart (talk · contribs · logs · block user · block log) 23:48, 26 March 2007 (UTC)

True, what we would need is a random number generator, preferably where numbers are generated by the substituted transclusion of a template. This job could be done by a bot, probably. The use of a "Pi" page would create a ridiculously long page. Nihiltres 00:21, 27 March 2007 (UTC)

Need for subjective criterion

How do you factor in all the subjective criteria in the FA selection process? At the moment there are some unwritten rules that would be easy to factor in to the FA selection process (no self-references- eg Jimbo Wales or Wikipedia) but how about stuff like the need for topical variety? At the rate the Indian and Hurricane wikiprojects are churning out FAs, a truly random selector would result in an Indian/Hurricane FA more often than it would if Raul were running it. Also, there are other considerations, like some topics (schools, pop culture) being slightly less prominent than "encyclopedic topics"- I believe that this is proper, and there are plenty of complaints on the main page too when an obscure pop-culture article makes the main page . IMHO, randomly selecting the FAs will result in more complaints of bias, not less, since the articles that receive the most complaints on the main page (indian articles, pokemon etc) will get more representation since they are no longer suppressed. Frankly, as a contributor who has contributed fairly extensively to the FA process (3 featured articles promoted to the main page) I like the way the current system works. and imho this proposition isn't needed. Raul654 is doing an excellent job. Borisblue 23:05, 26 March 2007 (UTC)

I think it may even just make the appearance of bias worse. Once we get two hurricane or India topics in a row (it's bound to happen if we're picking randomly), then any bias, real or not, is immediately perceived. ShadowHalo 23:43, 26 March 2007 (UTC)

Both of your objections are adressed by the "override" and the "proximity" rule. There can be overrides for topical events if there is support for it, and similar topics are skipped if they are selected in close proximity. David Spart (talk · contribs · logs · block user · block log) 23:45, 26 March 2007 (UTC)

"Solution" to the problem?

Although you mention in the proposal that "Many users become aggrieved when User:Raul654 does not pick the article they have worked on for days", I don't see the problem with this system as compared with your proposition. Won't this have the same problem? People may be even more frustrated that there is no person to ask regarding topic selection - while Raul654's (probably reasonable) claim to impartial topic selection may be unverifiable, it is probably a good thing that there is a human who can arbitrate such a decision, or change it if necessary. The "override option", the analogous process in your proposition, would produce the same problems that you cite regarding daily discussions: "a daily debate on what FA should be tomorrows TFA would be a disaster of Pompeiian proportions. A daily train-wreck that is would pit user against user and create endless bitterness." Since the "override option" is a necessary prerequisite for this proposition, I question its feasibility. Nihiltres 00:37, 27 March 2007 (UTC)

The override option is specifically only for time-relevant events that are connected to an article, and only one a week at most, to get around that problem. People will be happy to know that their article stands an equitable chance. Right now, every day (this is a simplification) 1,299 people are annoyed that there one wasn't picked and 1 is really happy. With this everyone will be happy since they know that it is a lottery, and that might get lucky any day. The more FAs they write the bigger their chance of "winning". I don't think people are going to sit around moping about how a random number generator is screwing them every day. Once the system is explained to them , they will think "fair enough!" and that will be it. David Spart (talk · contribs · logs · block user · block log) 00:44, 27 March 2007 (UTC)

Being an FA-contributor myself, I just want to say that waiting months for my articles to appear on the main page is not as traumatic an experience as you describe. Most featured content contributors do their work just to improve the encyclopedia, not for the perceived "glory" of a main page FA. And even if it was a big deal to me, I really don't see why I would feel better about being screwed by a random-number generator than by Raul654- and if I were really desperate for some reason to have my article on the main page at a certain time, I would prefer to take my case up to a fairly reasonable and respected wikicontributor than to wade through the bureaucracy created by this proposal. I'm not adamantly opposed to your idea, I just see it as adding a lot of complexity for no real benefit. Borisblue 04:48, 27 March 2007 (UTC)

Yeah, there aren't 1,299 people miserable about their article not being chosen. In fact, there are only about 400 articles at Wikipedia:Featured articles that have not been Today's featured article. And I can say that I'd like to see Hollaback Girl on the Main Page, but I realize it's going to take time, I trust Raul654 isn't part of some anti-whatever conspiracy, and I'm not going to roll over and die if it's not on the main page tomorrow. ShadowHalo 07:05, 27 March 2007 (UTC)

I still think this 1,299 issue is missing the point anyway since as I have suggested before, it seems likely to me that quite a number of FAs involve several contributors who work on several FAs. As have also pointed out, I personally suspect that it is more 'obscure' FAs which are likely to have only one or two primary contributors and by removing the human element and making it nearly random, this means these people are less likely IMHO to see their FAs on the page so if anything will be more discouraged. Those who work on several FAs e.g. as part of a wikiproject such as cricket, hurricanes, Final Fantasy, pokemon are already more likely to see one of their works as a FA at some stage due to the fact they probably concentrate a reduced amount of work on a larger number of articles. When talking about discouragement, it's the ones who work tirelessly on single articles because they're one of the few with the knowledge, resources and interest in that article that will be more likely to be discouraged. Finally, as I've also said before, let's not forget our primary concern with TFA should be the reader. While obviously acknowledging the contributions of our editors is important, it should not be the primary concern when it comes to TFA Nil Einne 08:17, 29 March 2007 (UTC)

if it ain't broke...

Don't fix it! Really, I am not sure this is needed. Unless and until we can see that Raul654 has a problem, do we really need to create a solution? It is impossible to create any situation where people will think like you want them to. Simply because you want people to be satisfied with the TFA process does not mean they ever will be. I fail to see how creating a random number generator to do the job will actually work? Besides, anytime you create a system that is designed to be automated, you open up the system to abuse. Someone will come up with a way to game this system. They always do. You can't game the system now, since Raul is a real person who can be trusted to make good decisions ( WP:AGF ). Again, this seems to be a rather WP:CREEPy proposal, and I am not sure we need it. --Jayron32|talk|contribs 05:22, 27 March 2007 (UTC)

This may have been mentioned already, but the this machine won't pick FA articles and show them on the main page on dates relevant to the article will it? Like showing the FA on the... i dunno... 1966 world cup on the date of the anniversary of the final, or something.

SGGH 13:57, 27 March 2007 (UTC)

ARghh, if you read the propostion to the end you will have your answer! David Spart (talk · contribs · logs · block user · block log) 01:18, 28 March 2007 (UTC)

This whole proposal is bollocks, but...

it has a real clever name. Nice work. 203.97.51.149 01:15, 28 March 2007 (UTC)

Another issue

In the 24 hour period you mention, how are users supposed to sift through the article, make sure that it's of FA quality, find a good free use image of it, write up a summary for it, and get everything ready for the main page? I can think of a few former featured articles that might have been put on the main page under this policy before they lost their status, and they would have been horrible examples. In short, I think this is stupid, utterly useless bureaucracy. Ral315 » 08:24, 28 March 2007 (UTC)

OK fine, so the articles could be picked a week in advance! David Spart (talk · contribs · logs · block user · block log) 20:49, 31 March 2007 (UTC)

Nice thinking, but no.

This proposal, although completely offered in good faith and constructively, seems to be written from a stance that fails to appreciate the finer points of how the TFA system works, how FAs are promoted and the democratic process that lead to Raul being voted in as TFA director. Systemic bias is inherent in promoted FAs and a truly random system would only exacerbate the complaints (random isn't random enough). Raul does a remarkable job of coordinating scheduled TFAs, distributing the small pool of promoted FA subjects to reduce the systemic bias as well as bear the occasional torrent of abuse from those perceiving purposeful bias and conspiracies where there is none. The poor bastard ;) --Monotonehell 11:06, 28 March 2007 (UTC)

Exactly. I've already tried to make this point, glad that I'm not the only one who sees it Nil Einne 08:14, 29 March 2007 (UTC)

I don't see a problem here.

It seems that most of what's on this talk page is criticism, but maybe you haven't stopped and contemplated the whole proposal. Currently, one user has all the power and say (regardless of whatever other new user gives a "suggestion" that week). It's ridiculous that this would be, in a "society" where we're not supposed to be practicing any form of government! If you have problems with some of the specifics of the plan, address that specifically, but do not bash the whole proposal. The whole thing is well thought out, in my eyes, and so I (and probably David) would appreciate more specificity to your problems so that we can fix the proposal to the way that would work best for everyone. (This is the way Wikipedia should work; not the TFA way!) _└^Jared_┘^┌_talk^┐  11:21, 28 March 2007 (UTC)

WP:NOT#BUREAUCRACY- the reason this proposal is getting bashed is that that it gives no real benefit to the TFA system while adding substantial amounts of complexity. KISS. Also, I'm not sure if random decisions is the way wikipedia "should" work. I think the gist behind this proposal is that "process" is more important than "product". Even David agrees that Raul is doing a good job, he just thinks that the process should be changed. Personally, I believe that if a process that looks crappy on paper produces a good product( product being TFA selection) then we should keep the process. This is actually consistent with how WP works as a whole- we have a God-King, and we let him keep his nondemocratic powers simply because he makes good decisions for the benefit of the project. If it ain't broke, don't fix it. Borisblue 13:37, 28 March 2007 (UTC)

The whole "Raul is doing a fine job thing" is (no offense) just a way to be nice. This may be coming off rudely, but I think, while doing an OK job, the position he's in has so much power, and he fails to see that he's doing anything wrong, or sees it and just doesn't want to give up his rights as the TFA coordinator. Because of this whole TFA fiasco he's involved in, he seems to think he's better than everyone else, or atleast has all the say (and that's just from my contacting him on a few occations and looking in on others). All I'm saying is, this is a simple proposal that, with support, will make WP more of what it's supposed to be, in my eyes. _└^Jared_┘^┌_talk^┐  02:07, 29 March 2007 (UTC)

From this talk page, I think we may conclude that his "not wanting to give up his rights as the TFA coordinator" is in line with community consensus. WP is supposed to be an encyclopedia- not an experiment in social democracy. Unless you can point out specific flaws with the TFA selection distribution (beyond "I don't like Raul654") I find it difficult to take this proposal seriously. Borisblue 03:24, 29 March 2007 (UTC)

Not at all - this is just wiki-politicking, inertia and nonsense. Not a single substantive criticism of this proposal has been made. Who wants to royally piss-off one of the most powerful users in wikipedia? David Spart (talk · contribs · logs · block user · block log) 20:38, 31 March 2007 (UTC)

problems? What problems?

• It is not particularly fair to other users, that User:Raul654 has so much power as I am sure he will agree? Why is this considered 'power'. It's a thankless task. Not power.

• On the other hand the idea of a daily debate on what FA should be tomorrows TFA would be a disaster of Pompeiian proportions. A daily train-wreck that is would pit user against user and create endless bitterness. Where is the daily train-wreck? Never had one soo far.

• 3.1416...? Should older FAs get featured earlier? Gotta use them before they degenerate, as some do. (That's why we have FAR.)

--74.13.128.153 15:08, 28 March 2007 (UTC)

It is power - that is what it is. Use your imagination - imagine what would happen if every day e every FA could be the TFA! If an FA degenrates rapidy, it can always be reverted to a prior good state. David Spart (talk · contribs · logs · block user · block log) 02:13, 29 March 2007 (UTC)

I imagine hell raised on the main page talk when three atlantic hurricanes are featured consecutively. Unless of course, we exercise the override option- for which case we get precisely the same TFA selection distribution when Raul is in charge- with the addition of more process complexity. Borisblue 03:15, 29 March 2007 (UTC)

Again, you apparently have not read the proposition to the end. David Spart (talk · contribs · logs · block user · block log) 20:28, 31 March 2007 (UTC)

Fail to see the point of this

I still fail to see why we need to change the way it is chosen. Has Raul654 done anything wrong? Also, you say that, if this policy is accepted, we will have a Proximity Rule. Who will be appointed to be the person who decides what is too close, or if two topics are too closely related? ffm ✎talk 15:59, 28 March 2007 (UTC)

Raul hasn't done anything wrong that I would know of - but it is unfair that one person decides what gets featured.Tourskin 22:24, 28 March 2007 (UTC)

Well, if someone else would like to apply for the task, I am sure he would be more than happy if the person had a reason for it. If you are still worried about Raul654 having a thankless task, then we could hold elections. I think handing the decisions over to Pi would not be a better solution, though. ffm ✎talk 00:14, 29 March 2007 (UTC)

Yeah, I agree- I would be more supportive of putting the TFA coordinator role on election. If the main concern is Raul "abusing his power" this would be a far better solution than the one proposed on this page.Borisblue 03:33, 29 March 2007 (UTC)

Raul doesn't decide what gets featured exactly, the FA process (which involves many editors) decides what gets promoted to FA status, Raul simply schedules the FAs that come out of that process onto the main page, considering date specific FAs, systemic bias and the do-not-feature list. His "power" that some people are perceiving here is quite limited. --Monotonehell 01:20, 29 March 2007 (UTC)

I jolly well doubt it! OK I tell you what - I would like to do it! Cummon! Lemme do it! David Spart (talk · contribs · logs · block user · block log) 02:14, 29 March 2007 (UTC)

Well if you'd like to shadow the process for a while, watch what goes on, get involved in the FA process, gain the kind of respect Raul has in the community, go for a RfA and become the first editor to actually want to do one of the most thankless jobs on wikipedia since its inception in 2004. Then go for it! (no sarcasm is to be read into my statement - I'm being sincere) Most editors who find out what's really involved really don't envy Raul his job. --Monotonehell 05:41, 29 March 2007 (UTC)

This is exactly my point - the idea of wikipedia is that everyone can get involved - yet this most public thing is openly beyond their input or response. This proposal will make the nomination of an article a much more important thing and motivate users to expend their energy getting FAs - not arguing and complaining when their FA isn't TFA. Not to mention it is much fairer on everyone. And if Raul hates his job so much, as you imply, much fairer on him! David Spart (talk · contribs · logs · block user · block log) 20:35, 31 March 2007 (UTC)

Pi

As others have pointed out, if we really want to do this, is Pi really a wise choice? Wouldn't a random number generator be a better idea? Also, has the proposor read the Pi article?

The most pressing open question about π is whether it is a normal number -- whether any digit block occurs in the expansion of π just as often as one would statistically expect if the digits had been produced completely "randomly", and that this is true in every base, not just base 10. Current knowledge on this point is very weak; e.g., it is not even known which of the digits 0,…,9 occur infinitely often in the decimal expansion of π.

Should we really be using Pi as a 'random' number if we don't even know if 0...9 are equally likely in it? Nil Einne 08:29, 29 March 2007 (UTC)

Fine, the number is not the point - there are a million ways to chose a rando number. David Spart (talk · contribs · logs · block user · block log) 20:31, 31 March 2007 (UTC)

Raul rejects the proposition

Really Raul, you should have asked someone else to do it. I am now completely convinced that the FA of the day should be picked through some democratic process and not by fiat, no matter how much you enjoy doing it. No substantive arguments were presented above against the proposition, it was mainly people trying not to make powerful enemies on wikipedia. David Spart (talk · contribs · logs · block user · block log) 20:27, 31 March 2007 (UTC)

There were numerous very valid reasons presented on this page why this proposal is a bad idea. That you don't consider them substantive doesn't make them so. Raul654 20:29, 31 March 2007 (UTC)

Go on then, what were they? David Spart (talk · contribs · logs · block user · block log) 20:30, 31 March 2007 (UTC)

Just skimming this talk page, problem with this proposal include the fact that: (1) that it is a solution to a non-problem, (2) that it is needlessly complicated, (3) that it is more-likely-than-not to introduce problems into the process we don't currently experience, and (4) that it doesn't actually address the problem of who or what writes the blurb. Raul654 20:37, 31 March 2007 (UTC)

(1) not a substantive point, since even so it would be harmless, (2) it is one of the most simple propositions in wikipedian history, but some people are confused about the math, (3) right, because right now nobody experiences it at all! The whole thing is in your head! It would be much simpler if we led you edit the whole of wikipedia by yourself, but we don't because that is the polar opposite of what wikipedia is. And anyway, we could always return to the old system if this were true, so this is not a substantive reason why we shouldn't try. (4) You can still write the blurb - who cares, that is not an objection to this proposition, which does not touch on that issue.

You know how simple this proposition is, and you know there is no impediment to its installation. You also want to keep your toys. There is a massive COI here. At the very least you must recognize that someone with as much power as you needs a regular mandate from wikipedians. David Spart (talk · contribs · logs · block user · block log) 20:47, 31 March 2007 (UTC)

(1) It is very much substantive. If you cannot formulate a problem that this actually solves, then we will not be doing it. We don't add new processes just to make more work for ourselves. (2) It is most certainly not simple. You have had to include these "override" and "proximity" rules in order to solve the fundamental flaws this system posses. (These rules are, by the way, not fully characterized by this proposal; in other words, it's an incomplete description. So it's only going to get more complicated). (3) As Ral315 pointed out above, this is going to introduce problems. Your response "nobody experiences it at all" is a non sequitur. (4) then the proposal should say that. Raul654 20:57, 31 March 2007 (UTC)

(1) There are many problems it solves and I formatted them. People say they are not a problem - but they don't say why not, so I can't respond. (2) It is simple, and it is fair. 2 simple rules does not complex make, and they are fully characterized by the description - if not, change them to your satisfaction. (3) No no, it isn't. That it may introduce problems is not an objection, since everything about wikipedia introduces problems but that doesnt stop us, and anyway as I say, we can always go back to Chairman Pellegrini if we want. (4) OK fine it will say that. David Spart (talk · contribs · logs · block user · block log) 21:13, 31 March 2007 (UTC)

You misunderstand what {{rejected}} means, Spart. A rejected proposal is simply one for which there is no consensus and no active changes to address the disagreements. (If it met the latter, then it would be back to proposed status.) This page definitely matches those criteria, so it is rejected. --tjstrf talk 21:03, 31 March 2007 (UTC)

I agree. Consensus has been overwhelmingly against this proposal, no attempts have been made to handle the objections for the past week or so. I do agree that there is a slight COI problem with Raul adding the rejected tag, but that is easily remedied. Additionally, I find slightly hilarious your suggestion that I, a trained mathematician am "confused about the math"- when you didn't even realize that pi wasn't a normal number. This is a horrendously complicated solution to a problem that doesn't exist, and I would suggest that you spend more time writing and improving articles (like FAs) rather than fiddling with policy in areas which you have no expertise or understanding.Borisblue 22:08, 1 April 2007 (UTC)

How very unpleasnat and rude. David Spart (talk · contribs · logs · block user · block log) 22:17, 1 April 2007 (UTC)