Blog / Twitter

Bible Gateway recently shared their most-popular Bible verses of 2014, and I wanted to discuss this chart a little more:

The chart stems from the idea that if someone is equally likely to see a verse on any day of the year, each day should have 1/365, or 0.27%, of a verse’s yearly popularity. This chart shows days when there’s a spike in pageviews for each verse for a particular day (whenever it was over 0.4% of the annual total).

The theme of the chart is that people follow certain paths through the Bible during the year; I labeled a few of them on the chart. But there are definitely a few patterns I can’t explain:

1. At the beginning of the year, two lines emanate from Genesis that look like they're on track to read the full Bible in a year, but one of them is faster than the other. Why are there two?
2. At the bottom right of the chart is a shallow line that looks like it involves reading Genesis and Exodus starting in May and ending in December. There's a similar line in the New Testament running through Matthew from June to November. What are those?

I was curious whether the same patterns would appear in Twitter for the year, so I ran a similar analysis on the 43 million tweets this year that mentioned Bible verses. The answer is that, yes, you can see many of the same paths in both charts:

They even include the same two (or three or four) fast readings of the Bible at the beginning of the year and the slow reading of Genesis and Exodus in the second half of the year. You can see similar peaks around the Passion stories leading to Easter and the Nativity story leading to Christmas. (Christmas is the last day that appears on this chart.) The Twitter chart more clearly shows the weekly rhythms of the devotional life, with vertical lines just barely visible every Sunday. The main difference is that there’s not as clear a path through the New Testament.

The Twitter chart also shows some horizontal bands where sharing is pretty light. These “sharing shadows” appear in the opening chapters of Numbers, 1 Kings, and 1 Chronicles.

Prolific Verse Sharers

A quirk of the Twitter chart is that some Twitterers tweet (and are retweeted) a lot. I suspect many of them are bots, but it’s hard to say whether they constitute “Bible spam”–many people do appear to find them helpful by retweeting them. The top fifty or so Twitterers are responsible for 16 million of the 43 million tweets this year. The chart doesn’t look too different if you remove them (mostly, the frequent repetition of Matthew disappears), but that just could be because I didn’t remove enough users to affect the results meaningfully. For all I know, this chart mostly just shows how Twitter bots share the Bible during the year. The consistency with the Bible Gateway data (in which I have more confidence), however, leads me to think that this picture is reasonably accurate.

Here are the top non-bot (as far as I can tell) sharers of Bible verses–these people tweeted the most Bible verses (and, more importantly, were retweeted most) throughout the year. Some of these people I recognize, and others… not so much. The “tweet” numbers reflect only tweets containing Bible verses and include others’ retweets of their tweets.

1. JohnPiper (105,836 tweets)
2. DangeRussWilson (87,382 tweets)
3. WeLiftYourName (52,638 tweets)
4. JosephPrince (50,889 tweets)
5. BishopJakes (49,109 tweets)
6. siwon407 (48,994 tweets)
7. RickWarren (42,637 tweets)
8. JoyceMeyer (39,703 tweets)
9. jeremycamp (32,003 tweets)
10. DaveRamsey (28,173 tweets)
11. RCCGworldwide (26,731 tweets)
12. AdamCappa (25,976 tweets)
13. Creflo_Dollar (24,422 tweets)
14. sadierob (20,068 tweets)
15. Carson_Case (19,846 tweets)
16. TimTebow (18,303 tweets)
17. Kevinwoo91 (17,230 tweets)
18. levimitchell (16,355 tweets)
19. jesse_duplantis (15,755 tweets)
20. kutless (14,806 tweets)

Most-Popular Verses

Here are the most-popular verses shared on Twitter in 2014:

1. Phil 4:13 (613,161 tweets)
2. 1Pet 5:7 (261,417 tweets)
3. Prov 3:5 (218,019 tweets)
4. John 14:6 (212,883 tweets)
5. John 13:7 (207,084 tweets)
6. 1Cor 13:4 (197,379 tweets)
7. Matt 28:20 (187,407 tweets)
8. Ps 118:24 (183,475 tweets)
9. 2Tim 1:7 (182,758 tweets)
10. Ps 56:3 (180,139 tweets)

You can also download a text file (411 KB) with the complete list of 2014’s popular verses.

John 13:7 (“Jesus replied, ‘You do not realize now what I am doing, but later you will understand.’”) is the oddball here, but it turns out that it’s mostly from over 100,000 retweets of a single tweet in April. (Since it was a one-off, I omitted him from the list of top sharers above, although his tweet count of 163,497 would put him in first place.)

How do the year’s most-popular verses compare among Bible Gateway, YouVersion, and Twitter? The answer: there’s a good deal of variation. Below are the top ten from each service; only Proverbs 3:5 appears in all three lists, and YouVersion and Twitter only have one verse that overlaps, which surprises me (given that they’re both based on sharing).

If we look only at Bible Gateway and Twitter, the average verse differs in its ranking by about 3,000 places, or nearly 10% of the Bible. The largest differences in rank: 1 Kings 20:14 is much more popular on Twitter (rank 4,380) than on Bible Gateway (rank 27,119), while Ezra 5:14 is way more popular on Bible Gateway (rank 13,995) than Twitter (rank 30,018).

Ranking	Bible Gateway	YouVersion	Twitter
1.	John 3:16	Rom 12:2	Phil 4:13
2.	Jer 29:11	Phil 4:8	1Pet 5:7
3.	Phil 4:13	Phil 4:6	Prov 3:5
4.	Rom 8:28	Jer 29:11	John 14:6
5.	Ps 23:4	Matt 6:33	John 13:7
6.	Phil 4:6	Phil 4:7	1Cor 13:4
7.	1Cor 13:4	Prov 3:5	Matt 28:20
8.	Prov 3:5	Isa 41:10	Ps 118:24
9.	1Cor 13:7	Matt 6:34	2Tim 1:7
10.	Rom 12:2	Prov 3:6	Ps 56:7

Bold entries appear in at least two lists.

Data Source

The Twitter data is from Bible Verses on Twitter. A program connects to the Twitter Streaming API with a query for every chapter of the Bible (“Gen 1”, “Genesis 1”, and so on). I run a Bible reference parser on the tweet to pull out all the references. Then an SVM algorithm tries to guess whether the tweet is actually talking about a Bible verse or just happens to contain a string that looks like a Bible reference (“Gen 1 XBox for sale,” where “Gen” is short for “Generation”).

Sidenote: How I Calculate Verse Views

A note on methodology: I’ve never documented how I determine a particular verse’s popularity; now’s a good time, because you can do it a number of ways to reach different answers. Let’s say that someone is looking at Genesis 1, which has 31 verses. That counts as one pageview, but if you’re looking for the number of pageviews that, say, Genesis 1:1 receives, how do attribute a chapter-length view like this? You could give each verse credit for a full pageview, but then verses in long chapters will appear to have a disproportionately high number of pageviews. Instead, I prefer to divide the pageview into the number of verses in the passage: in this case, each verse in Genesis 1 will receive 1/31, or 0.032 pageviews.

Now, what if someone is looking at, say, Genesis 1:1 and Matthew 1 (25 verses) on the same page? In this case, I divide the pageview by the number of separate passages: Genesis 1:1 receives credit for a full 0.5 pageviews, as does Matthew 1. Each verse in Matthew 1 therefore receives 0.5/25, or 0.02 pageviews.

I feel that this approach best respects people’s intentions whether they want to look at multiple verses, several independent passages, or just individual verses.

This year, “School” topped the list of things Twitterers are giving up for Lent, up 44 places from last year. Remaining in the top ten from last year are Swearing, Alcohol, Soda, Social Networking, and Fast Food. Chocolate, Twitter, Sweets, and Lent round out the new additions to the top ten.

I don’t have a great explanation for why School is #1 this year–it could be that Ash Wednesday is later this year, so spring break is closer (for some it even starts today). It’s also possible that Twitter’s audience is skewing younger than it used to, or that younger Twitter users are more likely to tweet about Lent.

Timely topics this year are Boosie, referring to rapper Lil Boosie, who was released from prison this week (people joked that the prison was giving him up for Lent); and Electricity, referring to a widespread power outage in South Africa.

This list draws from 646,000 tweets during March 2 to 8 that mention giving up something for Lent and excludes retweets.

Rank	Word	Count	Change from last year’s rank
1.	School	11,757	+44
2.	Chocolate	9,515	+15
3.	Twitter	8,642	+8
4.	Swearing	7,132	-2
5.	Alcohol	6,325	0
6.	Soda	5,446	-3
7.	Social networking	4,197	-3
8.	Sweets	4,188	+8
9.	Fast food	4,088	0
10.	Lent	2,842	+118
11.	Meat	2,790	+26
12.	Homework	2,760	+61
13.	Junk food	2,723	+8
14.	Coffee	2,678	+123
15.	Sex	2,392	+112
16.	Chips	2,129	-10
17.	Bread	2,020	+114
18.	You	2,016	+21
19.	Facebook	1,926	0
20.	Pizza	1,628	+122
21.	Starbucks	1,566	+120
22.	Candy	1,412	+87
23.	Instagram	1,212	-13
24.	Religion	1,147	+104
25.	Virginity	1,143	-18
26.	Cookies	1,053	-14
27.	Work	1,031	+4
28.	Ice cream	1,025	+27
29.	Boys	1,021	+99
30.	Marijuana	1,018	-22
31.	Smoking	994	-9
32.	Beer	939	+104
33.	Life	933	-8
34.	Food	930	+27
35.	McDonalds	926	+20
36.	Winter	853
37.	Netflix	851	-7
38.	College	819	+16
39.	My phone	777	+6
40.	Shopping	748	+103
41.	Stuff	733	+100
42.	Selfies	731	+1
43.	Chipotle	726	+100
44.	Masturbation	725	-30
45.	Sugar	682	+82
46.	Cheese	670	+94
47.	Me	656	+87
48.	Sobriety	655	-17
49.	Wine	652	+94
50.	Carbs	648	+81
51.	Boosie	581
52.	Fried food	574	-23
53.	Caffeine	563	+70
54.	Rice	562	+86
55.	Catholicism	561	-28
56.	Snapchat	543	+11
57.	Coke	541	-22
58.	Procrastination	517	-18
59.	People	516	+54
60.	Snow	506
61.	Desserts	486	-37
62.	Fizzy drinks	480	+81
63.	French fries	475	-29
64.	Takeout	464	-49
65.	Obama	452	+74
66.	Makeup	451	-25
67.	Taco Bell	434	+39
68.	Feelings	434	-32
68.	Porn	430
69.	Nothing	427	+74
70.	My swag	420	-47
71.	Negativity	417	+28
72.	Red meat	396	+59
73.	Diet Coke	390	+69
74.	Sarcasm	380
75.	Breathing	369
76.	Caring	357	+66
77.	Complaining	354
78.	Tea	352	+64
79.	Pancakes	340	+63
80.	Peanut butter	336
81.	Sweet tea	335
82.	Booze	325	+61
83.	Sleep	320	+33
84.	Hope	316	+46
85.	Cake	313	-13
86.	Pasta	303	+57
87.	TV	302	+30
88.	Texting	297	+52
89.	Eating out	275	-29
90.	Exercise	274	-47
91.	Pants	270	+5
92.	Electricity	268	+41
93.	The gym	258	+16
94.	Liquor	245
95.	Church	243	-46
96.	Tinder	237	+35
97.	Tumblr	236	+46
98.	Math	236	+20
98.	Juice	232	+35
99.	Being mean	230
100.	Chick Fil A	228	-38

Categories

Rank	Category	Number of Tweets
1.	food	62,453
2.	school/work	18,148
3.	technology	17,615
4.	habits	16,616
5.	smoking/drugs/alcohol	12,665
6.	irony	7,319
7.	relationship	6,563
8.	sex	5,483
9.	health/hygiene	3,476
10.	religion	2,784
11.	generic	2,504
12.	entertainment	1,959
13.	weather	1,496
14.	shopping	1,183
15.	celebrity	961
16.	sports	780
17.	politics	547
18.	clothes	540
19.	money	492
20.	habit	393
21.	possessions	217
22.	clothing	62

Historical Trends

This year I added a new Historical Lent Tracker that you can use to investigate Lenten trends on your own over the past six years.

Here are some of my favorite graphs:

Second-Wave Social Media

Tumblr peaked in 2011, and WhatsApp, which Facebook recently paid $19 billion for, doesn’t register highly.

Fast Food Restaurants

Chipotle is much higher on the list than I expected–is that because people love it or because they hate it?

One Direction vs. Justin Bieber

Snack Foods

Congratulations, Hot Cheetos, on being the snack the most people want to give up.

Media Coverage

The Lent Tracker got some media attention this year. In roughly chronological order:

• What To Give Up for Lent 2014? Twitter Reveals Top 100 Choices (Christianity Today)
• Twitter Lent Tracker Reveals Top 100 Things People (Say They) Are Giving Up (Huffington Post)
• 100 Things Christians Are Giving Up for Lent This Year (The Blaze)
• Forty Days Without (Inlander)
• What Are You Giving up for Lent: Question of the Day (Cleveland Plain Dealer)
• What Are You Giving Up for Lent? (Lancaster Online)
• Social Networks Shows Popularity of Lenten Sacrifices (Pasadena Star-News)
• Why You Shouldn’t Tweet About What You’re Giving Up for Lent (New Republic)
• The 25 Most Popular Things to Give Up for Lent Online (Washington Post)
• Giving Up Something For Lent? Shhh. (Slate)
• More Giving up Twitter and Facebook for Lent (WSJ MarketWatch)

Finally, this Wall Street Journal article doesn’t talk about the Lent tracker, but it discusses the fraught phenomenon of Ash Wednesday selfies: Selfies Bring Ashtags to Lent. (This article may or may not be behind a paywall for you.)

See the top 100 things people are giving up in 2014 for Lent on Twitter, continually updated until March 7, 2014.

As I write this post, with about 5,000 tweets analyzed, the new hot topics so far this year are: “Netflix,” “Flappy Bird,” and “Getting an Oscar.” “Social Networking” is currently way out in front, with twice as many tweets as perennial favorites “Swearing” and “Alcohol.” (Last year, Social Networking came in at #4.)

Look for the usual post-mortem on March 8, 2014.

This year saw a lot of churn in the top 100 things people were giving up for Lent.

The pope announced his resignation on Monday, leading many to say that he was giving up “being pope” for Lent. It came in at #1. (Related, at #18, people said they were giving up “the pope” for Lent.)

Specific social networking sites like Twitter and Facebook generally dropped this year, with the generic term “social networking” (#4) taking over as a catchall. Instagram (#10), Pinterest (#52), and Snapchat (#78) were all new to the top 100.

With Valentine’s Day falling on the day after Ash Wednesday this year, it came in at #13. My wife suggests that the timing may also have contributed to the drop in “chocolate” from #2 last year to #17 this year. “Valentines” is #97.

“Horse meat” (#20) refers to the ongoing European scandal.

The only celebrity to make the list was British boy band One Direction, up substantially at #41.

I learned several new words this year: “twerking” (#34), a type of dance move, “selfies” (#46), or self-shot photos taken with a phone, “subtweeting” (#57), or tweeting about someone without mentioning them by name, “oomf” (#71), or “one of my [Twitter] followers,” and “Nando’s” (#76), a chicken restaurant.

This list draws from 263,000 tweets from February 10-15, 2013, and excludes most retweets.

Rank	Word	Count	Change from last year’s rank
1.	Being pope	5,654
2.	Swearing	4,944	+1
3.	Soda	2,648	+2
4.	Social networking	2,264	+19
5.	Alcohol	2,217	-1
6.	Chips	1,690	+8
7.	Virginity	933	+23
8.	Marijuana	784	+17
9.	Fast food	776	-2
10.	Instagram	755	+270
11.	Twitter	672	-10
12.	Cookies	643	+19
13.	Valentine's day	514
14.	Masturbation	510	+18
15.	Takeout	465	+59
16.	Sweets	444	-7
17.	Chocolate	417	-15
18.	The pope	394	+10,224
19.	Facebook	380	-13
20.	Horse meat	375
21.	Junk food	362	-8
22.	Smoking	355	-3
23.	My swag	331	+373
24.	Desserts	325	+21
25.	Life	325	+40
26.	New year's resolutions	313	+47
27.	My boyfriend	309	+99
28.	Catholicism	255	+11
29.	Straightening my hair	228	+89
30.	Fried food	225	+5
31.	Netflix	216	+255
32.	Work	216	-5
33.	Sobriety	213	+4
34.	Twerking	185	+698
35.	The playoffs	184	+3,556
36.	French fries	173	+19
37.	Coke	168	+1
38.	Feelings	168	+207
39.	Laziness	160	+28
40.	Meat	158	-30
41.	Onedirection	155	+103
42.	You	154	-24
43.	Procrastination	153	+1
44.	Makeup	150	+16
45.	Internet	149	+61
46.	Selfies	149	+2,328
47.	Exercise	144	+58
48.	School	141	-36
49.	My phone	135	+15
50.	Classes	129	+84
51.	Dip	127	+132
52.	Pinterest	125	+133
53.	Church	124	+33
54.	Emotions	122	+397
55.	Going to school	119	+163
56.	My girlfriend	111	+207
57.	Subtweeting	110	+253
58.	College	106	+5
59.	My face	106	+4,168
60.	Ice cream	106	-27
61.	McDonald's	102	-32
62.	Being ugly	101	+256
63.	Snacking	99	+19
64.	Spending	96	+89
65.	Dunkin Donuts	96	+475
66.	Chew	95	+418
67.	Eating out	94	+28
68.	Elevators	94	+99
69.	Food	93	-47
70.	Moaning	93	+123
71.	Oomf	93	+78
72.	Chick Fil A	90	+135
73.	Healthy food	88	+180
74.	Football	87	+145
75.	Swimming	87	+200
76.	Nando's	86	+72
77.	DVDs	84	+1,326
78.	Snapchat	84
79.	Broccoli	83	+206
80.	Ranch	81	+250
81.	The snooze button	80	+176
82.	Crystal meth	80	+219
83.	Dignity	79	+116
84.	Cake	77	-13
85.	Unhealthy food	77	+34
86.	Homework	76	-65
87.	Busyness	75
88.	Schoolwork	74	+88
89.	Chemistry	74	+34,949
90.	Frozen yogurt	72	+480
91.	iPhone	72	+100
92.	FIFA	71	+143
93.	Betting	70	+315
94.	Doing homework	69	+158
95.	Myself	68	+267
96.	Supermarkets	67	+1,797
97.	Valentines	66
98.	Domino's	63	+323
99.	Being negative	63	+212
100.	Hookah	63	+340

Categories

Rank	Category	Number of Tweets
1.	food	11,642
2.	habits	8,083
3.	religion	6,519
4.	technology	4,782
5.	smoking/drugs/alcohol	3,928
6.	sex	1,771
7.	relationship	1,399
8.	health/hygiene	1,270
9.	school work	1,095
10.	irony	792
11.	sports	648
12.	entertainment	392
13.	celebrity	246
14.	clothes	235
15.	money	133
16.	shopping	111

The image is a Wordle.

See the top 100 things people are giving up in 2013 for Lent on Twitter, continually updated until February 15, 2013.

As I write this post, with about 5,000 tweets analyzed, the new hot topics so far this year are: meowing, Valentine’s Day, and Snapchat.

Look for the usual post-mortem on February 16, 2013.

This year, Twitter continues to take top honors. Facebook drops a few places compared to last year–has it become less-central to people’s lives? This year’s hot new site, Pinterest, almost makes the list, showing up at #118. (Incidentally, Pinterest has a number of Lent-related boards.)

Chocolate comes in at #2–however, if you add up all the mentions of chocolate in its various forms (“chocs,” “chocolate chips,” etc.), it totals over 14,000 mentions, enough to put it at #1.

This year’s biggest gainers are “breathing” and “makeup,” both of which jumped up more than 30 places in the list.

No celebrities make the top 100 this year. Boy band One Direction (aka #1D) is at #144, followed by Justin Bieber at #194 and Tim Tebow at #221. Last year’s curiosity, Charlie Sheen, only got two mentions; he dropped to #10,000 or so.

Overall, food was by far the most popular thing given up.

This list draws from about 300,000 tweets from February 19-25, 2012, and excludes retweets.

Rank	Word	Count	Change from last year’s rank
1.	Twitter	13,937	0
2.	Chocolate	13,001	+1
3.	Swearing	11,737	+1
4.	Alcohol	9,998	+1
5.	Soda	9,942	+2
6.	Facebook	9,025	-4
7.	Fast food	6,529	+3
8.	Sex	6,146	-2
9.	Sweets	4,973	+2
10.	Meat	4,444	-1
11.	Lent	4,171	-3
12.	School	3,976	+1
13.	Junk food	3,388	+6
14.	Chips	3,150	+4
15.	Coffee	2,263	0
16.	Candy	2,217	+6
17.	Bread	2,124	+3
18.	You	2,056	-2
19.	Smoking	2,002	+2
20.	Giving up things	2,001	-8
21.	Homework	1,908	+11
22.	Food	1,800	+5
23.	Social networking	1,754	-6
24.	Religion	1,701	-10
25.	Marijuana	1,594	+4
26.	Beer	1,359	+4
27.	Work	1,331	-3
28.	Stuff	1,302	-3
29.	McDonald's	1,249	+21
30.	Virginity	1,152	+7
31.	Cookies	1,137	+3
32.	Masturbation	1,134	+4
33.	Ice cream	1,113	+15
34.	Shopping	1,068	-6
35.	Fried food	993	-4
36.	Boys	956	+6
37.	Sobriety	910	+7
38.	Coke	899	+3
39.	Catholicism	881	-13
40.	Cheese	858	-7
41.	Nothing	831	+5
42.	Carbs	818	+16
43.	Red meat	758	-8
44.	Procrastination	738	+1
45.	Desserts	733	+26
46.	Pizza	714	+15
47.	Pancakes	650	-9
48.	Sugar	645	-5
49.	Rice	633	-10
50.	Breathing	631	+34
51.	Me	628	+12
52.	Texting	627	+3
53.	Starbucks	623	+1
54.	Fizzy drinks	595	+12
55.	French fries	593	+7
56.	Diet Coke	572	+21
57.	Porn	562	+10
58.	Tumblr	548	+12
59.	Wine	546	-7
60.	Makeup	539	+31
61.	Liquor	534	-5
62.	Booze	530	-22
63.	College	524	+18
64.	My phone	508	+30
65.	Life	486
66.	Caffeine	466	-17
67.	Laziness	453	+11
68.	Chipotle	452	+30
69.	Tea	445	+6
70.	Chicken	442	+2
71.	Cake	440	+3
72.	Sarcasm	429	+4
73.	New Year's resolutions	422	+15
74.	Takeout	417	+11
75.	Men	412	-10
76.	Pork	394	-3
77.	Christianity	388	-18
78.	Sleep	386	+1
79.	People	384	+8
80.	Caring	377
81.	Juice	357	+11
82.	Snacking	345
83.	Lying	333
84.	TV	332	-31
85.	Complaining	331	-2
86.	Church	328	-35
87.	Him	327	+2
88.	Sweet tea	326
89.	Lint	326	-21
90.	Vegetables	324
91.	Talking	323
92.	Bacon	321
93.	Being mean	320
94.	Pasta	316
95.	Eating out	316	+5
96.	Negativity	314	-39
97.	Eating	298
98.	Biting my nails	294
99.	Nutella	291
100.	Being nice	258

Categories

Rank	Category	Number of Tweets
1.	food	79,977
2.	habits	21,836
3.	technology	19,190
4.	smoking/drugs/alcohol	19,073
5.	health/hygiene	11,101
6.	sex	9,948
7.	irony	9,352
8.	school/work	8,567
9.	relationship	6,919
10.	religion	4,157
11.	generic	2,841
12.	shopping	1,491
13.	entertainment	1,344
14.	money	526
15.	sports	512
16.	celebrity	461
17.	possessions	376
18.	clothes	111
19.	politics	105

The image is a Wordle.

See the top 100 things people are giving up for Lent on Twitter, continually updated for the next few days.

Look for the usual post-mortem later this week.

If you’re storing people’s Bible annotations (notes, bookmarks, highlights, etc.) digitally, you want to be able to retrieve them later. Let’s look at some strategies for how to store and look up these annotations.

Know What You’re Modeling

First you need to understand the shape of the data. I don’t have access to a large repository of Bible annotations, but the Twitter and Facebook Bible citations from the Realtime Bible Search section of this website provide a good approximation of how people cite the Bible. (Quite a few Facebook posts appear to involve people responding to their daily devotions.) These tweets and posts are public, and private annotations may take on a slightly different form, but the general shape of the data should be similar: nearly all (99%) refer to a chapter or less.

Compare Bible Gateway reading habits, which are much heavier on chapter-level usage, but 98% of accesses still involve a chapter or less.

The Numbers

The data consists of about 35 million total references.

Percent of Total	Description	Example
73.5	Single verse	John 3:16
17.1	Verse range in a single chapter	John 3:16-17
8.4	Exactly one chapter	John 3
0.7	Two or more chapters (at chapter boundaries)	John 3-4
0.1	Verses spanning two chapters (not at chapter boundaries)	John 3:16-4:2
0.1	Verses spanning three or more chapters (not at chapter boundaries)	John 3:16-5:2

About 92.9% of posts or tweets cited only one verse or verse range; 7.1% mentioned more than one verse range. Of the latter, 77% cited exactly two verse ranges; the highest had 323 independent verse ranges. Of Facebook posts, 9.1% contained multiple verse ranges, compared to 4.2% of tweets. When there were multiple ranges, 43% of the time they referred to verses in different books from the other ranges; 39% referred to verses in the same book (but not in the same chapter); and 18% referred to verses in the same chapter. (This distribution is a unusual—normally close verses stick together.)

The data, oddly, doesn’t contain any references that span multiple books. Less than 0.01% of passage accesses span multiple books on Bible Gateway, which is probably a useful upper bound for this type of data.

Key Points

1. Nearly all citations involve verses in the same chapter; only 1% involve verses in multiple chapters.
2. Of the 1% spanning two or more chapters, most refer to exact chapter boundaries.
3. Multiple-book references are even more unusual (under 0.01%) but have outsize effects: an annotation that references Genesis 1 to Revelation 22 would be relevant for every verse in the Bible.
4. Around 7% of notes contained multiple independent ranges of verses—the more text you allow for an annotation, the more likely someone is to mention multiple verses.

Download

Download the raw social data (1.4 MB zip) under the usual CC-Attribution license.

Data Modeling

A Bible annotation consists of arbitrary content (a highlight might have one kind of content, while a proper note might have a title, body, attachments, etc., but modeling the content itself isn’t the point of this piece) tied to one or more Bible references:

1. A single verse (John 3:16).
2. A single range (John 3:16-17).
3. Multiple verses or ranges (John 3:16, John 3:18-19)

The Relational Model

One user can have many rows of annotations, and one annotation can have many rows of verses that it refers to. To model a Bible annotation relationally, we set up three tables that look something like this:

users

user_id	name
1	…

annotations

user_id	annotation_id	content
1	101	…
1	102	…
1	103	…

annotation_verses

The verse references here are integers to allow for easy range searches: 43 = John (the 43rd book in the typical Protestant Bible); 003 = the third chapter; the last three digits = the verse number.

I like using this approach over others (sequential integer or separate columns for book, chapter, and verse) because it limits the need for a lookup table. (You just need to know that 43 = John, and then you can find any verse or range of verses in that book.) It also lets you find all the annotations for a particular chapter without having to know how many verses are in the chapter. (The longest chapter in the Bible has 176 verses, so you know that all the verses in John 3, for example, fall between 43003001 and 43003176.) This main disadvantage is that you don’t necessarily know how many verses you’re selecting until after you’ve selected them. And using individual columns, unlike here, does allow you to run group by queries to get easy counts.

annotation_id	start_verse	end_verse
101	43003016	43003016
102	43003016	43003017
103	43003016	43003016
103	43003019	43003020

Querying

In a Bible application, the usual mode of accessing annotations is by passage: if you’re looking at John 3:16-18, you want to see all your annotations that apply to that passage.

Querying MySQL

In SQL terms:

select distinct(annotations.annotation_id)
from annotations, annotation_verses
where annotation_verses.start_verse <= 43003018 and
annotation_verses.end_verse >= 43003016 and
annotations.user_id = 1 and
annotations.annotation_id = annotation_verses.annotation_id
order by annotation_verses.start_verse asc, annotation_verses.end_verse desc

The quirkiest part of the SQL is the first part of the “where” clause, which at first glance looks backward: why is the last verse in the start_verse field and the first verse in the end_verse field? Because the start_verse and end_verse can span any range of verses, you need to make sure that you get any range that overlaps the verses you’re looking for: in other words, the start_verse is before the end of the range, and the end_verse is after the start.

Visually, you can think of each start_verse and end_verse pair as a line: if the line overlaps the shaded area you’re looking for, then it’s a relevant annotation. If not, it’s not relevant. There are six cases:

The other trick in the SQL is the sort order: you generally want to see annotations in canonical order, starting with the longest range first. In other words, you start with an annotation about John 3, then to a section inside John 3, then to individual verses. In this way, you move from the broadest annotations to the narrowest annotations. You may want to switch up this order, but it makes a good default.

The relational approach works pretty well. If you worry about the performance implications of the SQL join, you can always put the user_id in annotation_verses or use a view or something.

Querying CouchDB

CouchDB is one of the oldest entrants in the NoSQL space and distinguishes itself by being both a key-value store and queryable using map-reduce: the usual way to access more than one document in a single query is to write Javascript to output the data you want. It lets you create complex keys to query by, so you might think that you can generate a key like [start_verse,end_verse] and query it like this: ?startkey=[0,43003016]&endkey=[43003018,99999999]

But no. Views are one-dimensional, meaning that CouchDB doesn’t even look at the second element in the key if the first one matches the query. For example, an annotation with both a start and end verse of 19001001 matches the above query, which isn’t useful for this purpose.

I can think of two ways to get around this limitation, both of which have drawbacks.

GeoCouch

CouchDB has a plugin called GeoCouch that lets you query geographic data, which actually maps well to this data model. (I didn’t come up with this approach on my own: see Efficient Time-based Range Queries in CouchDB using GeoCouch for the background.)

The basic idea is to treat each start_verse,end_verse pair as a point on a two-dimensional grid. Here’s the above social data plotted this way:

The line bisects the grid diagonally since an end_verse never precedes a start_verse: the diagonal line where start_verse = end_verse indicates the lower bound of any reference. Here are some points indicating where ranges fall on the plot:

To find all the annotations relevant to John 3:16-18, we draw a region starting in the upper left and continuing to the point 43003018,43003016:

GeoCouch allows exactly this kind of bounding-box query: ?bbox=0,43003016,43003018,99999999

You can even support multiple users in this scheme: just give everyone their own, independent box. I might occupy 1x1 (with an annotation at 1.43003016,1.43003016), while you might occupy 2x2 (with an annotation at 2.43003016,2.43003016); queries for our annotations would never overlap. Each whole number to the left of the decimal acts as a namespace.

The drawbacks:

1. The results aren’t sorted in a useful way. You’ll need to do sorting on the client side or in a show function.
2. You don’t get pagination.

Repetition at Intervals

Given the shape of the data, which is overwhelmingly chapter-bound (and lookups, which at least on Bible Gateway are chapter-based), you could simply repeat chapter-spanning annotations at the beginning of every chapter. In the worst case annotation (Genesis 1-Revelation 22), you end up with about 1200 repetitions.

For example, in the Genesis-Revelation case, for John 3 you might create a key like [43000000.01001001,66022021] so that it sorts at the beginning of the chapter—and if you have multiple annotations with different start verses, they stay sorted properly.

To get annotations for John 3:16-18, you’d query for ?startkey=[43003000]&endkey=[43003018,{}]

The drawbacks:

1. You have to filter out all the irrelevant annotations: if you have a lot of annotations about John 3:14, you have to skip through them all before you get to the ones about John 3:16.
2. You have to filter out duplicates when the range you’re querying for spans multiple chapters.
3. You’re repeating yourself, though given how rarely a multi-chapter span (let alone a multi-book span) happens in the wild, it might not matter that much.

Other CouchDB Approaches

Both these approaches assume that you want to make only one query to retrieve the data. If you’re willing to make multiple queries, you could create different list functions and query them in parallel: for example, you could have one for single-chapter annotations and one for multi-chapter annotations. See interval trees and geohashes for additional ideas. You could also introduce a separate query layer, such as elasticsearch, to sit on top of CouchDB.

Congratulations, I guess, go this year to Charlie Sheen, who came in at both #23 and, with “tiger blood,” at #90. Justin Bieber is up several spots this year, so he hasn’t quite crested yet. The next-highest celebrity, who didn’t make the top 100, is British boy band One Direction.

“Trophies,” at #69, refers to the English soccer club Arsenal’s recent defeat, or something.

The later start to Lent this year means that “snow” doesn’t appear on the list–last year, it was #48. Myspace hangs on at #99, dropping 48 places.

This list draws from 85,000 tweets from March 7-10, 2011, and excludes retweets.

Rank	Word	Count	Change from last year’s rank
1.	Twitter	4297	0
2.	Facebook	4060	0
3.	Chocolate	3185	0
4.	Swearing	2527	+1
5.	Alcohol	2347	-1
6.	Sex	2093	+3
7.	Soda	1959	-1
8.	Lent	1493	-1
9.	Meat	1352	-1
10.	Fast food	1303	0
11.	Sweets	1252	0
12.	Giving up things	778	+7
13.	School	768	+27
14.	Religion	745	+1
15.	Coffee	707	-3
16.	You	675	+6
17.	Social networking	665	+15
18.	Chips	664	+3
19.	Junk food	594	-1
20.	Bread	571	+6
21.	Smoking	555	-4
22.	Candy	541	-8
23.	Charlie Sheen	511
24.	Work	482	+4
25.	Stuff	467	-2
26.	Catholicism	436	-10
27.	Food	395	+3
28.	Shopping	363	+1
29.	Marijuana	358	+31
30.	Beer	346	-10
31.	Fried food	307	-7
32.	Homework	306	+27
33.	Cheese	297	+4
34.	Cookies	293	+11
35.	Red meat	285	-10
36.	Masturbation	285	+8
37.	Virginity	253	+26
38.	Pancakes	252	+20
39.	Rice	236	-5
40.	Booze	235	+2
41.	Coke	234	-3
42.	Boys	229	+24
43.	Sugar	229	-16
44.	Sobriety	226	+10
45.	Procrastination	226	-10
46.	Nothing	219	+21
47.	Winning	219
48.	Ice cream	211	-7
49.	Caffeine	203	-16
50.	McDonald's	195	+27
51.	Church	188	+28
52.	Wine	188	-3
53.	TV	184	-7
54.	Starbucks	183	-15
55.	Texting	182	-12
56.	Liquor	181	-1
57.	Negativity	180	+26
58.	Carbs	179	+10
59.	Christianity	177	-12
60.	Justin Bieber	176	+9
61.	Pizza	175	-11
62.	French fries	159	+2
63.	Me	157	+9
64.	Losing	155
65.	Men	152	-13
66.	Fizzy drinks	151
67.	Porn	147	+4
68.	Lint	147	-11
69.	Trophies	144
70.	Tumblr	144
71.	Desserts	142	-15
72.	Chicken	140	+15
73.	Pork	139	-3
74.	Cake	132	+8
75.	Tea	127	+19
76.	Sarcasm	127	+14
77.	Diet Coke	119	-16
78.	Laziness	118	-13
79.	Sleep	117	-6
80.	Jesus	115	-4
81.	College	111
82.	Internet	110	-46
83.	Complaining	108	-9
84.	Breathing	103
85.	Takeout	98
86.	Beef	98	-8
87.	People	96	+11
88.	New Year's resolutions	96	+1
89.	Him	94	-5
90.	Tiger blood	92
91.	Makeup	91
92.	Juice	90	-7
93.	Clothes	89
94.	My phone	88
95.	God	87	-15
96.	Abstinence	85	-15
97.	Stress	84
98.	Chipotle	82
99.	Myspace	81	-48
100.	Eating out	81	-25

Image created using Wordle.

Here’s a presentation I just gave at the BibleTech 2010 conference about how people tweet the Bible:

Also: PowerPoint, PDF.

I distributed the following handout at the presentation, showing the popularity of Bible chapters and verses cited on Twitter. It displays a lot of data: darker chapters are more popular, the number in the middle of each box is the most popular verse in the chapter, and sparklines in each box show the distribution of the popularity in each chapter. (Genesis 1:1 is by far the most popular verse in Genesis 1, while Genesis 3:15 is only a little more popular than other verses in the chapter.)