Talk:Autobiography sales and chart positions

For a December 2004 deletion debate over this page see Wikipedia:Votes for deletion/Autobiography sales and chart positions

Chart data presentation

This is a proposed attempt to deal with the complaints about the opening paragraphs in this section. While it actually takes up more screen space than the text, I think that readers might find it more approachable. In the retained text, the term "ironic" was changed because that is not an example of irony.

I will not be touching the article page for now, this is only a proposal. iMeowbot~Mw 22:03, 24 Dec 2004 (UTC)

(I moved this section to Talk:Autobiography sales and chart positions/Chart1 to make this page a little less unwieldy for discussion.) iMeowbot~Mw 18:35, 25 Dec 2004 (UTC)

This looks pretty good. --Tony Sidaway|Talk 22:29, 24 Dec 2004 (UTC)

Thanks. Some   cruft is added now to deal with MSIE rendering bugs. It would be great if everyone (especially Everyking) could confirm that I put all the external links in the right rows. iMeowbot~Mw 22:42, 24 Dec 2004 (UTC)

No offense, but I think it's rather ugly. It's all those blank spaces that'll never be filled in, you know. I don't object in principle to the idea, I like moving some of this kind of info into charts and I've been in favor of that before, but I just don't think this particular table is very good. Everyking 03:36, 25 Dec 2004 (UTC)

No offense, but constructive criticism would be far more helpful. There has been widespead agreement that the existing presentation is unreadable, and something needs to be done with it. What specific changes do you suggest to improve that section? iMeowbot~Mw 03:42, 25 Dec 2004 (UTC)

Have people actually said this? I don't want to ignore anybody's opinion, but I don't know of anyone saying it's unreadable. I think it's OK. I want to add more of an intro to this at some point, and I think it will all work itself out, more or less. I think the table format as we have it now works fine. Some data is just better left to the prose. Everyking 03:54, 25 Dec 2004 (UTC)

Yeah, that portion has come up repeatedly.

The amount of chart stuff is exactly what Johnleemk keeps harping on. If chopping it isn't an option, the alternative would be to try different presentation forms and see if they fly.
Worldtraveller cited the same spot: "A whole paragraph describing the movement of the album in the charts over several months is mind-numbingly boring to someone who's not an obsessive fan."
In peer review, Drstuey wrote "There is far too much detail on the chart position and sales that could easily be summarised, we absolutely don't need to know that the album rose from No. 19 to No 16 in its 12th week on the chart for example."
Calton had not-very-nice things to say about the very same paragraph in peer review. It was also near the top of my mental list when I remarked that the article needed a major rewrite.

I know there is more of the same out there, and I can poke around all the scattered wiki pages to find them if you want, but it's definitely a popular point for shooting down the article.

So, we have this dilemma. If I understand your position correctly, you don't feel that putting the detailed chart information on another wiki project would be acceptable; do I have that right? We know that there is plentiful opposition to this section/article as it stands, and that is going to be a sticking point if you ever want to see this thing featured. Even off in a separate article, we know that people are going to continue to harp on it. So, if the details have to stay, they need to become more palatable.

That's what I'm trying here. Not that table in particular, but experimenting with different presentations. Maybe it would work with text in a different form. Maybe a 3D animated glowing orb would do the trick. Okay, forget the orb, but we're going to need to do something differently to address the criticism.

Anyway, sorry for rambling. Would it be worth my time to experiment with and propose other ideas? Do you want to try coming up with some? Yes, I will accept "no" to those questions if that is what you would prefer. I'm only offering this as one possible way out of the ongoing disagreements, but I won't force the issue if you'd rahter handle it differently. iMeowbot~Mw 05:22, 25 Dec 2004 (UTC)

I like the idea, certainly. But I'm not sure that it can be done any better than the way we have it. It looks rather like we've already tabled all the data that makes sense in a table. What I like to do is summarize in the prose, and provide a few extra details and external cites there, and provide greater numerical detail in the charts. Everyking 14:15, 25 Dec 2004 (UTC)

Er...if we keep the detailed boring chart data, I'm not going to accept it as prose. Tabular makes a lot more sense, IMO. Johnleemk | Talk 14:59, 25 Dec 2004 (UTC)

I very much prefer tables if we're going to include such quantities of chart data, but I feel that detailed charting information is dubiously encyclopedic at best, and possibly falls under source data in Wikipedia:What Wikipedia is not. There's a reason we have external links; just because something is factual does not make it encyclopedic. External links provide more in-depth specific information than what an encyclopedia would provide, just like further reading. Johnleemk | Talk 12:12, 25 Dec 2004 (UTC)

In general I would have to agree that there is too much raw chart data in there. I will even go so far as to say that the weekly numbers may be of interest to the reader, but they have no value here (yes, I'm going to back up that assertion). It's obvious that there isn't going to be any compromise on the amount of chart data, and that's why I'm asking about trying a Plan B that focuses on presentation.

Okay, so I've just damned those numbers as having "no value." Why do I say that? Well, I've actually used Autobiography sales figures for an article studying fandom (not for here, original research; and yeah, that's why I have an interest in this article). In order to analyze those numbers, I needed to ensure that they were good numbers. That means that they had to come straight from SoundScan. Wikipedia simply wouldn't have been suitable as a primary source; that's explicitly contrary to its mission and the open editing policy seals the deal. The same is true of any encyclopedia, all are secondary (or even more indirect) sources.

For this article, Wikipedia isn't even a secondary source. According to the references, those numbers were pulled from magazines and press releases. Did those magazines report the numbers correctly? How would I know, short of going back to SoundScan/Billboard? I wouldn't know, obviously. For usable, valuable data, it needs to come from the authoritative source. How can I draw conclusions based on inherently untrustable data?

That, I think, is a pretty good illustration of Wikipedia's limitations, and can be a guide to how much detail would ideally go in. Explaining the topic is Good. Working the facts into some kind of narrative so that the reader can understand how the people and events fit together is Good. Providing ample references so that the reader can dig deeper into the side issues is Good. References are also absolutely Good if one is to attempt any sort of serious research on a topic.

Anyway, that is why other articles (and other reference books) stick to the highlights when speaking of things like charts. Doing otherwise borders on irresponsibility, implying a level of authority that the reference can't guarantee.

With all that said, I'm still willing to accept the idea that those numbers tell a story on their own. Without complete sales numbers or dates to tie the numbers to events, I'm not so sure that they do it, but maybe I'm looking at them the wring way.

Everyking, maybe you could offer some examples of how those numbers would be interesting (or even possibly valuable) when presented in Wikipedia? Good examples might help to soften some of the objections, and maybe help to move things along.

Again, sorry for the length, but "It's not useful!/It is useful!" doesn't seem to be getting anywhere. Maybe talking about why/why not will work . . . iMeowbot~Mw 19:36, 25 Dec 2004 (UTC)

I'd like to express my appreciation to you for taking the time to expand greatly on what I've been saying all along. Johnleemk | Talk 15:01, 26 Dec 2004 (UTC)

To me, that's like asking how it's useful to provide the number of people killed in some war or natural disaster. Not in magnitude, obviously, but in essence. It's a statistical detail that adds historical context and broadens the reader's understanding of the subject. Everyking 15:12, 26 Dec 2004 (UTC)

I fail to see how a user's understanding can or will grow significantly if they know that the album was at position X during week Y in country Z. A summary should be well enough to demonstrate whatever the user is supposed to know. Johnleemk | Talk 16:17, 26 Dec 2004 (UTC)

All the article has is weekly chart positions with most of the sales data missing, and there is no context to tie it all together. Data needs context to become information. It carries no meaning in a vacuum. What does that list of numbers mean? That's what all these calls to summarize are really about. It's even admitted in the article that the rise and fall of unit sales doesn't correspond to chart positions, and for most of those weeks there is nothing in the article that needs the chart positions for support.

Some of those numbers might be interesting if they were mixed into the general timeline, to show how various events (heavy advance promotion from the TV series, the post-SNL dip and the recovery with holiday shopping would be examples) made sales rise and fall (but proper sales data would be best for that, relative chart positions are deficient). Otherwise? Well, knowing what those numbers mean (and more importantly, what they don't mean), not much is being said apart from the weeks that album was in the top 5 or 10, and perhaps how many total weeks the album sells well enough before it stops charting. iMeowbot~Mw 16:38, 26 Dec 2004 (UTC)

Well, there was a post-SNL dip. Was that because of SNL, though? Who knows? So I've never added that, because I worry about potentially misleading the reader. You make valid points, and since you voted to keep the article, I presume you have its best interests in mind. So let's say this: if the data can be neatly summarized in a table, we'll keep it; if it can be integrated eloquently in the prose with the context you speak of established, we'll keep it; if it's either a debut chart position or a peak position, we'll keep it. Isn't that fairly logical? But if you ask me, everything already does fit into one of those three categories...do you not agree with that, or do you not agree that the three categories are all good ones? Everyking 17:20, 26 Dec 2004 (UTC)

I do not view the mundane information presented in prose as eloquent; I find that it would be better off in a table than in prose unless it's properly written. Johnleemk | Talk 17:41, 26 Dec 2004 (UTC)

John, why did you take this dispute to the mailing list, where I do not participate? Did you expect I wouldn't find out about it? Everyking 18:15, 26 Dec 2004 (UTC)

I was looking for guidance, not looking to argue the merits of it. To answer your second question, no, and I wasn't trying to hide anything. Please don't attempt to divert the discussion further; if you want to continue this line of argument, please take it to my user talk page. Johnleemk | Talk 18:41, 26 Dec 2004 (UTC)

I already tried that. Everyking 18:55, 26 Dec 2004 (UTC)

You'd know if there was a real SNL effect (there was) if the article contained appropriate sales data, which it doesn't. It was obvious what was happening from the radio spins and audiences, as well as the units.

Take a look at this week's Billboard 200. Autobiography is at 40th place. Some albums selling better at the moment are two from Ray Charles, the latest in Rod Stewart's series of old standards, and a Mannheim Steamroller holiday album. What does this tell us about Autobiography's relative success? Absolutely nothing, because those records are selling to different audiences with little overlap. What is the actual difference in sales between #20 and #40 on that chart? The distribution isn't always uniform, and it changes from week to week, and the 200 doesn't tell us. So again: what information is provided by reproducing all the relative chart positions? iMeowbot~Mw 10:00, 27 Dec 2004 (UTC)

I wish I had the sales info. But I don't. To my knowledge, that info isn't generally available unless the album makes it into the top ten or so. So we just have to live with that. My position is that we should provide the info if it's in the top 50; once it drops out of the top 50, we can stop updating it (but if it drops out only to bounce back again, as it did after the "SNL drop" you hypothesize—fell to 51 on the BB 200, bounced back to 50—then we need to record that). Radio spins: absolutely, we need more info on radio play. I intend to work on that in the future. I'm not sure how much I'll be able to dig up, though, since it's after the fact, but I'll try. It is indeed quite important and the addition of such info would improve this article and also (if added in a summarized form) help get the main article up to featured quality. Everyking 10:08, 27 Dec 2004 (UTC)

Sorry, the point still seems not to be getting through. I'll try being blunt then. The exhaustive list of numbers simply is not suitable material for Wikipedia, because it is not actionable. It really needs to go unless there is real context. Please also read the talk page to that policy page, it goes into more discussion of the difference between information and data. iMeowbot~Mw 13:05, 27 Dec 2004 (UTC)

I'm well aware what your point is, but I disagree with it, so I keep trying to talk in the direction of ways we can compromise. Everyking 13:17, 27 Dec 2004 (UTC)

Sorry, it's not a matter of opinion. It's objectively factual that the included chart data carries no information value, as explained above. End of discussion. iMeowbot~Mw

I think this is one of the strangest things I've ever read. How can it not have informative value? Maybe the value is minor, or minimal, but it's indisputable that it exists. Everyking 18:17, 27 Dec 2004 (UTC)

10:59, 16 Jan 2005

Good progress. I want this to be the benchmark. --Tony Sidaway|Talk 21:55, 16 Jan 2005 (UTC)

Thanks. For the record, I've been spending a lot of time with this, trying to make sure that any text trimmed is still reflected in the tables. Details that don't appear in the tables (for example, number of weeks in some markets, or the flip-flops in top places with other albums) have been retained. References have all been kept as well. iMeowbot~Mw 07:48, 17 Jan 2005 (UTC)

Please use caution when eliding text.

The album was released in the United Kingdom on October 4, 2004, debuting at number 31 on the albums chart. [1] In Norway, it debuted at number 31 and peaked at number 29 on the top 40 album chart, [2] remaining on the chart for three weeks. [3]. In Switzerland, it reached a peak of number 36 on the album chart in its second week. [4]

The above deleted text does include some things that aren't reflected in the tables. I'm going to look at the tables and see if it can be reasonably included there, but may need to restore some version of this paragraph. Brevity is generally good, but let's not be overly aggressive about cutting into the bone. Thank you. iMeowbot~Mw 01:38, 18 Jan 2005 (UTC)

With total populations smaller than several cities, sales for Norway and Switzerland have no material effect in demonstrating the popularity of the album. Since the album only charted briefly in either country, those numbers don't tell us much. Removed. iMeowbot~Mw 02:39, 18 Jan 2005 (UTC)

The table at the foot of the article therefore becomes redundant. Peaks are now indicated in the earlier chart. iMeowbot~Mw 02:50, 18 Jan 2005 (UTC)

Meowbot, you have strange ideas about what constitutes information, I must say. Anyway, this article should be restored. It survived VfD and therefore should not have been redirected. Do I have consent to restore it? Everyking 04:32, 21 Jan 2005 (UTC)

I could care less either way, but redirecting an article doesn't require a VfD vote. Ambi 10:43, 21 Jan 2005 (UTC)

Nevertheless, the community's decision should be respected. Everyking 10:56, 21 Jan 2005 (UTC)

Per the VfD vote, I will restore the article soon unless there's an objection. Everyking 01:44, 24 Jan 2005 (UTC)

You didn't wait very long -- if I hadn't had class until now, I would have objected. I'd also like to note that, while VfD has agreed that this article title should exist, it said nothing whatsoever about changing it into a redirect, which nobody else seems to mind. Madame Sosostris 00:53, 25 Jan 2005 (UTC)