User talk:Kiac/Tables

Just a couple of points need to be re-phrased

I must admit that this way the table looks much neater. But I think some statements may need to be re-phrased.

• 1) The database we have for Australia is from '97-'09, but I contacted them just recently to find out whether they've had changes in their certification-award-levels and they relied that no changes have been made to the certification volumes since 1989. So we need to make the establishment year for Australia 1989, but in the parenthesis mention Note that the provided database may not illustrate records released prior to '97.

Done. But do they even publish any certifications at all from before 1997? The online database definitely begins at 1997 doesn't it? Or do they publish others which were certified earlier within those tables?

• 2) Also may not be correct to say Certification numbers might overstate sales if stores order more albums than they are able to sell, because regardless of how many CDs record companies ship out, records will not get certified unless soundscan confirms that a particular record has reached a gold/platinum status. That statement would be correct for the system they had before soundscan came into the picture in 1991. Probably, it would be best to emphasize the years.

I think we need a clear explanation on how certifications and sales are accumulated. Because articles like this are pretty clear in saying that they are determined by shipments, not sales. So I think this should be duly noted, if we are going to use certifications to determine/estimate "sales". How does SoundScan explain this, do they say that they record all actual 'sales' or something, is that what the issue is?

• 3) And finally Record labels must also pay a fee to certify shipments, which is not always done, perhaps it would be better to say that Record labels must pay a fee; therefore, they may not always apply for a certification each time a record reaches a gold/platinum status, meaning certifications might not be visible in the databases for more than a short period of time after an album is released.

I think we have to emphasise 'sales' not so much 'certifications' in that last sentence. If the record label never registered to be tracked at all, then i can't imagine the association would track the album, award a certification, then remove it because they didn't get their fee paid. Or do labels just have to pay to receive the actual certification? That would seem a little odd.

• 4) Also can we avoid using the article by Wall Street Journal? The article seems to generalize all foreign markets. The statement they have, would be correct if addressed to developing countries, it would be incorrect to say that sales in countries like UK, Germany, France for example is nonexistent or difficult to track. --Harout72 (talk) 15:13, 9 September 2009 (UTC)

My main issue with the previous notes (and the Constraints section) is that it is seriously lacking in verifiable sources, or it's basically original research - that's why I made an effort to use the sources such as WSJ. I agree it is quite pidgeon-holed, but it gets the same sort of point across that I believe ou were aiming for, which is basically that there is issues with tracking certs and sales. I agree better sources are required. k.i.a.c (talktome - contribs) 13:18, 11 September 2009 (UTC)

For point 1. I personally don't know of another database for Australia. Honestly, I didn't know when they'd started issuing certifications until I contacted to ask them about whether the volumes have changed or not.

For point 2. That's odd that they still rely on shipments, I don't know why they do that when they have the Nielsen Soundscan.

For point 3. Well, since we are providing sales through certification-databases, I think it's probably best to use the term Certification, but it's up to you. It looks like that major record companies not so much labels must pay the fee to get the records certified. RIAA explains that here. Note that it says If multiple sales levels are simultaneously certified, only one audit certification fee is charged. Therefore, record companies sometimes wait until the sales of a record hits its end and that way they only have to pay the fee once, but get it multi-certified. That's why I was suggesting to state Record labels must pay a fee; therefore, they may not always apply for a certification each time a record reaches a gold/platinum status, meaning certifications might not be visible in the databases for more than a short period of time after an album is released. And we can source that statement using that the same page coming from RIAA (up to you).

For point 4. How about we use the WSJ's article by stating in a footnote (for example Sources have been known for publishing inflated figures as they often misinterpret or misrepresent total sales or something like that. Because if we say that sales reliable information is spotty in the U.S., and often nonexistent overseas, then what's the use of all the certifications, I mean if it's spotty in US and nonexistent in foreign markets, then what are we trying to prove by certified sales? I mean, that may arise another dispute. What do you think?

Overall, good stuff though.--Harout72 (talk) 00:04, 12 September 2009 (UTC)

• I was actually just thinking back to work out if I'd come across Australian certifications from prior to 1997, and I have. Myself and User:Dan arndt spent quite a while digging up Certifications for the Midnight Oil discography, I had put it up for FLC, and only missed out because of a missing couple of music videos. The sources are reliable enough to be valid in proving a certification system did exist, from music journalists, etc. Just ARIA has not published them, so perhaps these Certifications prior to 1997 do exist. Also I'm not sure where these figures come from, but there's this article List of best-selling albums in Australia, which is made up of certified album's prior to 1997. I can't imagine John Farnham's Whispering Jack has done 1.6 mil in the past 12 years! So there's something missing here, someone must have recorded the sales/shipments or at least made claim to certifications?

• I'm going to support point 2 with the So What Is a Shipment? and The Process references as well, there's an actual explanation in them, and then we have an outside publication in WSJ there too.

• Makes sense, Certifications it is.

• Completely agree, we'd just be stomping on our own playing field. I still think though, we need to establish some kind of weight on the fact that certifications are seldom entirely exact/accurate and complete.

• I've also removed the line Also note that 95% of the music markets have experienced a decline in their Certification-volumes due to music piracy which began in 1999-2000. As it is a pretty big statement to include unsourced, and I've been unable to find a source. Whilst searching i found the other 95% figure, which I think sufficiently covers the piracy issue. You think we're okay to implement now? k.i.a.c (talktome - contribs) 08:40, 13 September 2009 (UTC)

We could use the following links to support the statement Also note that 95% of the music markets have experienced a decline in their Certification-volumes due to music piracy which began in 1999-2000.

IFPI 2003 report states that music piracy is the main factor for the decline in sales. As for the changes in certification levels, we could source some of these: table shows changes in the German Certification-award-levels (scroll the 90% of the page down, you will see a table Gold und Platin Auszechningen gold and platinum awards), changes in award-levels in Finland, changes in Switzerland, changes in Argentina, changes in Brazil. I think it's important to state that the certification-volumes have declined, in case some editors want to verify our calculations. They could at least verify some of our certification figures as half of the available markets do post the changes in their web sites.

Other than that, we're good to go.--Harout72 (talk) 20:18, 13 September 2009 (UTC)

Haven't been able to find some of the relevant parts in them sources (massive PDFs!), but I'll take your word for it. All done now, implementing. k.i.a.c (talktome - contribs) 14:23, 15 September 2009 (UTC)

Is France 1968 or 1973? Because the Infodisc source that is used says 1973. k.i.a.c (talktome - contribs) 14:39, 15 September 2009 (UTC)

See where it reads Quelques Précisions (some details) in Infodisc, immediately below that it states Les Informations tiennent compte de la période du 1er Janvier 1968 au 31 Décembre 2007 (The information reflects the period from 1 January 1968 to December 31, 2007). It seems a bit confusing because also at the top it reads Les Certifications depuis 1973 (Certifications since 1973). We should probably go with 1968.--Harout72 (talk) 15:27, 15 September 2009 (UTC)

Perhaps it is complete after 1973, but has some certifications from 1968? Or something along those lines. Whatever, I'll leave it as is. k.i.a.c (talktome - contribs) 15:31, 15 September 2009 (UTC)