Talk:List of classic rock songs

Table format

A sortable table format for this list might be helpful. Below is a example of a new format that could be used should article not be deleted.

Year	Title	Artist
1968	All Along the Watchtower	Jimi Hendrix
1971	Baba O'Riley	The Who
1976	Blinded By the Light	Manfred Mann
1975	Bohemian Rhapsody	Queen
1971	Brown Sugar	The Rolling Stones
1974	Can't Get Enough	Bad Company
1969	Come Together	The Beatles
1976	Crazy On You	Heart
1973	Dream On	Aerosmith
1977	Dreams	Fleetwood Mac
1973	Free Bird	Lynyrd Skynyrd
1978	Hot Blooded	Foreigner
1976	Hotel California	Eagles
1978	Just What I Needed	The Cars
1983	Legs	ZZ Top
1967	Light My Fire	The Doors
1973	Money	Pink Floyd
1976	More Than a Feeling	Boston
1983	New Year's Day	U2
1983	Photograph	Def Leppard
1979	Refugee	Tom Petty and the Heartbreakers
1978	Renegade	Styx
1975	Rock and Roll All Nite	Kiss
1971	Stairway to Heaven	Led Zeppelin
1973	The Joker	Steve Miller Band
1981	Tom Sawyer	Rush
1978	Wheel in the Sky	Journey
1978	You Really Got Me	Van Halen
1980	You Shook Me All Night Long	AC/DC

Piriczki (talk) 18:14, 15 January 2016 (UTC)

Piriczki, the table format is much more encyclopedic. I would add a column for sources (see List of blues standards, for example). It can easily accomodate a number of different citations without making a "mess". —Ojorojo (talk) 15:24, 25 January 2016 (UTC)

@Piriczki:: Agreed, a sortable table seems perfect for this. I like the particular order of the columns as well: Year, Song, Artist, it's neat, with the small column first, and maybe more useful that way for a particularly release date-oriented list like this. A final Notes column should be sufficient for citations and any other notes - less clutter than Notes + Sources columns, and a few less characters in a long list.

I would continue to use the superscript form for the Hickey source (and any other sources with a good number of cites) as the normal cites generate a linkback like ^a for each item and they add up in the References section - unless Wikipedia has an alternative cite method to handle many cites to a single source.

Are there any significant performance issues with a large table - do we have similar examples?

You are likely familiar with regex text replacement to automate the conversion, but just in case you're not, I'm checking: are you? Conversion shouldn't take more than half an hour. Cheers! --Tsavage (talk) 18:14, 25 January 2016 (UTC)

A column for sources is an idea but the number of sources used for the list of blues standards seems excessive. If these are widely recognized songs ("classics") we shouldn't need multiple sources, one should suffice. Otherwise it might attract link spam.

I asked at Help talk:Table and apparently a very large table is not a problem. I put together a table with 200 rows at User:Piriczki/sandbox/List of classic rock songs and it sorts instantly no problem. As far as converting text to a Wikipedia table, I'm not familiar with that. Just 200 rows was very tedious work.

Regarding the Hickey source, of the 2200 songs listed over 600 had only one play. One play could indicate a familiar "classic" that just happened to not be in any of the stations' rotation that week or it could be a one-shot, seldom-played rarity which would be a good reason to not include it. Some of the obscure or early Beatles tracks and Led Zeppelin tracks that only had one play could have been on "Get the Led Out" type features where seldom played deep cuts are aired. Again, not very good candidates for the list. I also noticed most of the modern rock songs on the list were played by only one station (KUFX) and there were some other anomalies like seven songs by the Clarks that were only played on their hometown station WDVE. "Drinkin' in the Driveway" (2012) by Michael Stanley was played on WNCX, probably by Michael himself on his afternoon drive shift. Cutting everything post-1989 might take care of some this but there still needs to be a lot of pruning, maybe hundreds of songs. Piriczki (talk) 19:56, 25 January 2016 (UTC)

Cool:

• I meant NO extra Sources column, Notes should be sufficient for cites and anything else that may come up.
• I hadn't looked at the raw data. It would be great to trim the stuff you mention - yes, the Beatles Xmas stuff highlighted the single-play type of situation - we just need a source-based rule, in that, especially if the list gets POV-pushed, it should have clear sourcing and inclusion standards. Let's think about how? Maybe it's enough to trim based on number of plays, although, considering it's a one-week sample, one could argue that a one-off novelty song would still have one play, but a low rotation classic would come up regularly over a longer period, and we wouldn't want to exclude that... To think about...
• After Afd, I can do the conversion. Look up regular expression (too detailed in WP article, look on web for simple how-to). It's a search and replace option in text processors and word processors that gives some characters special uses, so you can make up search strings to find specific patterns. Like this lil guy in the search/find box:

^\(.*\)(\([0-9]*\))

...will go line by line, basically find all numbers between (), like a year, then this in the replace box...

|\2 |\1

...will drop the parentheses, flip the order, and add a vertical bar/pipe at beginning and between, like this...

"Song 1" Artist1 (1999)

"Song 2" Artist2 (1912)

...into...

|1999 |"Song 1" Artist1

|1912 |"Song 2" Artist2

Like magic - click Replace all and the whole file is done! I just use 'em a bit, you can get real good at making up expressions, I just run a series of little ones that make a couple of changes each. In any case, it's dead easy when each line is exactly the same format, like, there's only one pair of () per line, etc. Anyhow, you can learn, or I can do. --Tsavage (talk) 23:24, 25 January 2016 (UTC)

Table format (cont'd)

Here's a sortable table: User:Piriczki/sandbox/List_of_classic_rock_songs_2. May I suggest this replace the current list? This is the Hickey list from 1964 to 1994 with songs with only one play removed and all the years filled in. It still needs a little cleanup but it's mostly ready for further editing (wikilinks, etc.) Piriczki (talk) 17:26, 6 February 2016 (UTC)

It looks great, however, I don't think we should use it to replace the current list for two reasons:

1. it loses all of the links and footnoting, which is a substantial amount of work, the disambiguation alone takes a while (it would be way easier to add and delete from the existing list)

2. not sure how excluding the single play items fits with NOR

Regarding the second point, cutting out the single plays, I agree that a song like Beatles' "...Saturday Club..." is almost certainly the result of some sort of novelty or one-off play, however, other one-play tracks may simply be low rotation. In other words, if the sample was not one week, but one year, would it be the same songs at the single play level (or whatever low cutoff relative to the rest), or would some have risen significantly, so that a new bottom cut-off group emerged? And, per content policy, what is the justification for changing the source in this way, it's logical but it is also a form of cherrypicking?

Also, why the new 1964-1994 range (I would think the more recent we go, the more we drift from the core classic rock songs)?

Thanks! What do you think about the above? --Tsavage (talk) 19:17, 6 February 2016 (UTC)

UPDATE: I think I have a solution for trimming the single plays. I started a new section call "Criteria for inclusion," copying the idea from List of sovereign states. I think we can work out a rationale that complies with content policy. I think it's important to make the inclusion criteria as clear as we can, to avoid any sort of...shenanigans. I'll start a new section for it. Lemme know what you think. --Tsavage (talk) 21:05, 6 February 2016 (UTC)

Completely revised list

The original, unsourced chart I've replaced with a subset of the chart from the Hickey source. A quick comparison of the original indicates that it was pretty much included in Hickey. Here's what's there now from Hickey:

• all dated songs up to and including 1989
• all undated songs with a "year needed" tag (songs after 1989 can be removed when dated)
• each song is cited to the Hickey source, using ¹ instead of the usual ref link because of the number of items, the References section would be a mess.

Hickey seems to be a reliable source for this: it's quite the comprehensive survey of American classic rock radio, he describes his method in detail in the article and footnotes, along with the raw data, and it's published as a bylined article in an ESPN group title, FiveThirtyEight, so it seems reasonable to assign it the reliability we give to other major news media.

Citing each song individually can be useful for managing any new additions to the list, makes it easy to see what is sourced where. --Tsavage (talk) 00:36, 25 January 2016 (UTC)

Year of release?

Do we list the release year for the album or the single, when they're different? --Tsavage (talk) 23:32, 26 January 2016 (UTC)

Since not all songs were singles, I would use the album release year for consistency. Piriczki (talk) 17:23, 5 February 2016 (UTC)

Yeah, that's what I've been doing, for that reason. --Tsavage (talk) 19:56, 5 February 2016 (UTC)

Criteria for inclusion

Started a section in the article, Criteria for inclusion, with some initial text. The idea comes from List of sovereign states. I think it's important to make inclusion criteria as clear as possible, to help avoid any future inclusion or deletion problems.

In the main source, the Hickey playlist of 25 classic rock radio stations, logged over a one week monitoring period, some songs receive only a single play. We can perhaps reasonably consider these novelty songs, broadly speaking, songs not part of the regular rotation that are played for a special occasion or purpose, and eliminate them from the list. That should comply with WP:RS and WP:NOR, in that we are not arbitrarily excluding certain items from a set, there is a logical reason and a specific exclusion criterion, consistent with the scope of this article. Comments? --Tsavage (talk) 21:19, 6 February 2016 (UTC)

"Reasonably consider{ing}" songs played once to be "novelty songs" and therefore not "classic rock songs" is original research. Explaining your opinion and saying it is logical does not change the fact that it is your opinion.

Perhaps we can reasonably consider songs that sources do not call "rock" and were played fewer that 100 times to be "local favorites" and logically exclude them. Perhaps we should reasonably limit the list to songs that are from the "classic rock era" ("1960s to the late 1980s") and logically exclude anything outside of those dates. All of these are opinions. They might be logical or reasonable, but they are still our opinions. - SummerPhD^v2.0 14:24, 5 July 2017 (UTC)

Limiting the list to 100+ spins would shorten the list to just 40 songs. Since this article was restored (it was dormant as a redirect for 10 years) dozens of songs have been added without references based on editors' opinions, and the creation of a "criteria" by one editor represents original research. Personally, I think the results of the AfD were misinterpreted. The votes were fairly evenly split between keep, keep as a redirect, and delete, and there was no consensus for "keep.". I believe "keep as redirect" was wrongly interpreted as "keep" when the intent of those votes was to not have this list. Perhaps another AfD is in order. Piriczki (talk) 15:19, 5 July 2017 (UTC)

Limiting it to 100+ would also permanently limit the article to that one source. Also, it would compound the issue of the criteria being OR.

I've ranted a bit more below. If there aren't any good answers, I will take it to AfD. If anyone wants to take it there before me, they are certainly welcome to. - SummerPhD^v2.0 16:58, 5 July 2017 (UTC)

Citations needed

As expected, various users have added dozens of their own favorite songs that, in their opinion, are classic rock—all without references. Piriczki (talk) 18:47, 2 January 2017 (UTC)

Inclusion criteria

The list currently has roughly 1900 songs, all but two of which are sourced. Of those that are sourced, 100% of them cite Hickey, with a handful also citing Cardillo. That's enough for the "once source" tag.

For some reason, the number of songs citing Hickey has varied over time. The source lists 2,230 songs. At first we included roughly 2,000 songs cited to that source, roughly 90%. Now we have about 1,900 of them. Songs have been removed, others have been added.

I have zero confidence that we are accurately citing the source. There are certainly songs excluded and there are very likely songs citing the source that are not in the source.

Our inclusion criteria? After the list, we have a self-referential section called "Criteria for inclusion". Wikipedia does not refer to itself. Yes, we need to directly state what the inclusion criteria are, but it is in reference to the list, not the article. It seems rather glaring by its placement that this is an after thought.

"Listed songs appear either in the playlist of a radio station that is programming in the classic rock format, or in any other source that is referring to songs played on classic rock radio." Those are not the criteria we are actually using, of course. Instead, we have "Listed songs appear in one book, minus those played only one time, with a few other additions and removals as various editors have seen fit."

Oh, but let's not forget our second source: "30+ Classic Rock Songs I Never Want to Hear Again". How many of those 34 songs do we list? 79, over 200% of them. That's outstanding.

Yes, there are "classic rock songs" that virtually anyone would agree with. "Stairway...", "Layla", "Revolution", etc. Pick a largish American radio market and at some point they had a station that self-identified as "classic rock" (not so much currently, as the title is stale). At some point in the 80s or 90s, they had some kind of promotion where listeners were asked to vote for their favorites. The station then made a big deal out of playing the "100 greatest rock songs of all time!!!" or some such. Some lists were heavy with Zepplin, Floyd and such. Others had more Bob Seger, 38 Special, Allmans, etc. All had local favorites. A recent list in Philadelphia included more Hooters, Hall & Oates, Todd Rundgren, etc. than you would find outside of the area. Bob Seger's "Turn the Page" wasn't in the top 50, which would be shocking 100 miles south of here. I wouldn't be at all surprised if there were "classic rock" stations excluding Metallica, Black Sabbath, etc. or mixing in Carol King, James Taylor and some softer Simon and Garfunkle.

In the 80s, it was common to have "classic rock" defined in such a way that a song had to be at least X years old to qualify. Others hinged on the age of the band. Still others were more flexible, using mostly material from a core time frame and a handful of earlier and later songs mixed in.

^But alot of stations play stuff like Green Day(2000's era) and Pearl Jam so it doesn't matter how old a song is. — Preceding unsigned comment added by Mr krabs 993 (talk • contribs) 00:58, 7 December 2017 (UTC)

A local (Philadelphia) classic rock station used to play Monty Python's "Bright Side of Life" every Friday morning. Does that make it a classic rock song? Classic rock stations play "Happy Birthday", some end their days with "The Star Spangled Banner", others include various non-classic rock songs with news reports (especially sports). In the 80s, more than a few played "We Are the World", "Do They Know It's Christmas" and "Hands Across America" on occasion (more than "once").

"Classic rock" does not have an agreed upon definition. We have a list from one source that we are reprinting here. We've excluded some songs based on on editor's opinion. Various other editors have obviously added and removed songs based on their opinions.

As it currently stands, this article needs a dose of TNT. Thoughts? - SummerPhD^v2.0 15:17, 5 July 2017 (UTC)

As it turns out, Hickey does not list 2,230 songs. Hickey's research found that many. The actual article lists 19 songs. We cite that article for roughly 2,000 songs.

This article is beyond repair. I'm replacing the list with the few songs actually listed in the sources. Everything else is unsourced. - SummerPhD^v2.0 20:25, 9 July 2017 (UTC)

I've just finished cleaning out the list, now limited to the songs actually listed in the sources. Next, we'll have to consider how reliable those sources are.

The Hickey's blog entry (not the data file behind it, which is a primary source) is probably the most critical. At the moment, I have little to go on as to how reliable FiveThirtyEight isonso far afield of their normal subject matter: fact checking, accuracy and such. Along with this is the question of how much of an article we can realistically build based on airplay on 25 American radio stations over the course of a week.

Next would the light opinion piece. How carefully are we to suppose USA Today scrutinized whether or not these were "classic rock" songs that the guy was humorously explaining why he wants to never hear them again.

Ive removed several "Further reading" items for a few reasons. Several were about "rock classics" or the "greatest" rock songs, making them off-topic. Several others were published song books whose song selections were likely heavily influenced by what rights they were able to secure, rather than how accurately they were part of "classic rock".

Comments? - SummerPhD^v2.0 01:34, 10 July 2017 (UTC)

Okay...WHAT HAPPENED?

Most of the songs were deleted. I have to ask,"WHY?" Donny (talk) 20:02, 13 July 2017 (UTC)

Talk:List_of_classic_rock_songs#Inclusion_criteria. - SummerPhD^v2.0 01:09, 14 July 2017 (UTC)

Lots of songs are missing from this list, even Layla Stairway To Heaven and Free Bird are missing also theres no beatles songs on this list so please update this it needs a lot of help to be accurate. — Preceding unsigned comment added by Mr krabs 993 (talk • contribs) 22:24, 7 December 2017 (UTC)

Adding additional songs would require additional sources. Let us know what you find. - SummerPhD^v2.0 02:33, 8 December 2017 (UTC)

Well source 1 has over 2230 songs that would be good source — Preceding unsigned comment added by Mr krabs 993 (talk • contribs) 01:04, 24 December 2017 (UTC)

No, it lists 15 and mentions three others. That's source 1, published by FiveThirtyEight, presumably subject to editorial oversight by ESPN, qualifying it as a reliable source. That article refers to a list of 2230 songs played at least once by one of several radio stations during the period the author checked. That list is not subject to any kind of fact checking and, as noted above, would likely include songs that no one considers to be "classic rock": "Happy Birthday to You", the "Star Spangled Banner", etc. While the study wasn't done in Philadelphia, a "classic rock" station here plays "The Bright Side of Life" from Monty Python's Life of Brian five times a week. Lots of stations step outside their format for charity releases as well. - SummerPhD^v2.0 01:42, 24 December 2017 (UTC)

Primary sources

Yes, we have a source[1] from a radio station in Danbury (I assume that's Danbury, CT) listing the 500 songs from their Memorial Day 500.

The idea of a Memorial Day 500 is pretty common: Take the 3 day weekend when the Indy 500 is run and play the "top" 500 songs from your station's playlist. How this particular station chooses their 500 every year, I'm not sure. The method I'm familiar with is a listener vote. Back in the 1980s (and before) this was done by mail, now it's mostly going to be over the 'net (less work for the interns, I suppose).

In any event, is the radio station a reliable source ("with a reputation for fact-checking and accuracy")? That's certainly debatable. Their published list, presumably after fact-checking, proclaims: "...here's is our Memorial Day 500 Playlist." Nice.

Yes, I'm certain that virtually every major market (and many of the medium and small markets) in the US has a "classic rock" station. It's a format that is popular enough to pay the bills. Many of them will have a Memorial Day 500 or something similar, just as they likely have "double shots", "block parties", "four plays" and "side shows" and all the rest. Many of them have had the lists for decades. The question is not whether or not these thousands of lists exist, the question is whether we should use them.

Currently, the i95 list is used for two songs (out of 500...). What does the list indicate? Does it say that every song on the list is an example of "classic rock"? I would argue that the list is a primary source that indicates that it is one of the 500 most popular songs with people listening to a particular radio station in one particular month. That the radio station is one of several thousand calling itself "The Home of Rock 'n' Roll" and says the list is "Classic Rock's Top 500 Songs" is about as notable as them saying what the best movie of 2017 was. No one else really cares. No one (outside of Wikipedia) is citing i95's 2017 list as proof that the Car's "You Might Think" is "classic rock". The radio station conducted a poll (or regurgitated their 500 most requested/most played or whatever). There is no secondary coverage here.

I bring this up because this is a one way street for this article. If we include 2 songs from that list, we should include all 500. If we use that list, we have thousands of others to use as well. If Danbury, CT's 2017 list is reliable, so is San Francisco's 1983 list and Miami's 1997 list and Boulder's 1981 list and...

Comments? - SummerPhD^v2.0 15:20, 26 January 2018 (UTC)

I would argue that we should try to incorporate all 500. Obviously most (if not all) of the songs would be considered classic rock by most radio stations (and arguably most people). I would say that "Classic Rock" isn't technically a genre, but there are huge numbers of radio stations that would classify these as "Classic Rock". Since the "Classic Rock" label was created by, and defined by radio stations, I would say that songs that play relatively frequently on a large number of radio stations should be considered "Classic Rock". If you find a large number of stations playing "Green Day" for example, I would say you should take into account the percent of radio stations that play it. FYI: I'm not trying to make this list unreliable, I'm just trying to make it more complete. Tell me your thoughts, Howpper (talk) 18:18, 26 January 2018 (UTC)

All lists on Wikipedia need selection criteria. The criteria must be "unambiguous, objective, and supported by reliable sources." Currently, the selection criteria being used here call for blue-link notable songs that independent reliable sources directly state are "classic rock".

You seem to be suggesting that we broaden the criteria to include "songs played relatively frequently on a large number of radio stations*". "Relatively frequently" and a "large number" are obviously subjective. (*I assume you mean self-described "classic rock" stations.) I doubt we would find reliable sources listing songs that objectively meet these criteria.

I am questioning whether a radio station is necessarily a reliable source. Wikipedia articles should be based on secondary and tertiary sources. A list compiled by a radio station of its listeners' favorites is a primary source. To qualify as a reliable source, the system** being used by the radio station to compile the list would need to have a reputation for fact-checking and accuracy. (**The station may have a news department, for example, which might be a reliable source, but it is unlikely that department was used to compile the list.)

IMO, the selection criteria we have are about as good as we are going to get. We obviously cannot list every song played by every self-described classic rock station. The list would be a pointless jumble, giving equal weight to "Stairway to Heaven", "Happy Birthday to You", "Hey Jude" and the "Star Spangled Banner".

That leaves the question of reliable sources. For List of signature songs, List of Magical Negro occurrences in fiction and other similar films (lists where membership is subjective), articles in reliable publications (most major newspapers and magazines, websites with editorial oversight and fact-checking, etc.) are pretty much the best we have. It is possible that there are textbooks with some examples. However, the few that I know working in this field tend to use little in the way of traditional textbooks for anything this dynamic. I'm not really sure what else we can realistically expect to find. - SummerPhD^v2.0 19:18, 26 January 2018 (UTC)

Nominated for speedy deletion

I nominated the article for speedy deletion as it fails to indicate its importance and is also a meaningless list. Fefersz (talk) 23:43, 9 February 2018 (UTC)

Fefersz has been indefinitely blocked as WP:NOTHERE. - SummerPhD^v2.0 03:29, 10 February 2018 (UTC)

This is not classic rock, its hard rock!

Round and round shouldn't be here, neither Welcome to the jungle!— Preceding unsigned comment added by 195.245.189.166 (talk • contribs) 00:47, February 13, 2018 (UTC)

1) The songs fit our inclusion criteria, discussed above.

2) Who said a song can't be both? - SummerPhD^v2.0 17:31, 13 February 2018 (UTC)

3) They are played on classic rock stations, so they are classic rock. — Preceding unsigned comment added by Mr krabs 993 (talk • contribs) 12:03, 27 May 2018 (UTC)

As discussed, that would make "Hands Across America", "We Are the World", "The Star Spangled Banner" and Monty Python's "Bright Side of Life" classic rock songs, which they clearly are not. - SummerPhD^v2.0 17:31, 27 May 2018 (UTC)