Category Archives: writing

Betteridge’s Law, or: Are Questions in Headlines a Good Idea?

Pick up any tabloid news­pa­per today and take note of how many art­icle head­lines are phrased as a ques­tion. I under­stand that these head­lines are an attempt to piqué our interest (or the res­ult of lazy copy editors/writers), but are they a good idea? What is the end res­ult of using a ques­tion as a head­line or art­icle title?

Now we know, thanks to Betteridge’s Law of Head­lines:

Any head­line which ends in a ques­tion mark can be answered by the word ‘no’.

Named for Ian Bet­ter­idge, this simple max­im was first expli­citly found in journ­al­ist Andrew Marr’s 2004 book, My Trade. This is why the law tends to be “uni­ver­sally true”:

Because of a simple prin­ciple of head­line writ­ing: if a story has enough sources to have a high chance of accur­acy, a head­line will be assert­ive (e.g. “Microsoft to release OS update on Fri­day”). If sources are weak, or only a single source is found, head­line writers will hedge their bets by pos­ing the head­line as a ques­tion (e.g. “Will Microsoft release an OS update on Fri­day?”).

The Wadsworth Constant: Ignore 30% of Everything

I’ll start with a story.

Last year my girl­friend and I watched the pilot epis­ode of a new TV show and were imme­di­ately hooked. The pilot epis­ode was refresh­ingly com­plex and forced us to guess miss­ing plot details con­tinu­ously: it’s adven­tur­ous to make your audi­ence work so hard dur­ing a pilot, we sur­mised.

We later dis­covered that, due to a tech­nic­al glitch, we actu­ally missed the first fif­teen minutes of the show (about 30%). The ‘com­plete’ ver­sion of the epis­ode was less sat­is­fy­ing.


Last year Steve Yegge wrote about life at Amazon.com and what it’s like work­ing under Jeff Bezos. On the top­ic of present­ing to Bezos, Yegge gave this tip: delete every third para­graph.  Why?

Bezos is so god­damned smart that you have to turn it into a game for him or he’ll be bored and annoyed with you. That was my first real­iz­a­tion about him. […]

So you have to start tear­ing out whole para­graphs, or even pages, to make it inter­est­ing for him. He will fill in the gaps him­self without miss­ing a beat. And his brain will have less time to get annoyed with the slow pace of your brain.


Around the same time as Yegge’s post­ing, a Red­dit user known as Wadsworth poin­ted out that the first 30% of “nearly every video in the uni­verse” can safely be skipped. As such things go, this soon became a You­Tube URL para­met­er: just add &wadsworth=1 to skip the first third of the video.

This ‘law’ soon became known as the Wadsworth Con­stant. It works.

Personal Pronouns as Relationship and Company Indicators

The per­son­al pro­nouns used by couples dur­ing “con­flict­ive mar­it­al inter­ac­tions” are reli­able indic­at­ors of rela­tion­ship qual­ity and mar­it­al sat­is­fac­tion, accord­ing to a study track­ing 154 couples over 23 years. The study showed that We-words’ (our, we, etc.) were indic­at­ive of a more pos­it­ive rela­tion­ship than ‘Me- and You-words’ (I, you, etc.) (doi).

Using We-ness lan­guage implies a shared iden­ti­fic­a­tion between spouses, even when the con­ver­sa­tion is focused on an area of con­flict. Con­sist­ent with this, We-ness was asso­ci­ated with more pos­it­ive and less neg­at­ive emo­tion beha­vi­ors and with lower car­di­ovas­cu­lar arous­al. In con­trast, Sep­ar­ate­ness lan­guage implies a great­er sense of inde­pend­ence and dis­tance in the rela­tion­ship. Com­pared with We-ness, Sep­ar­ate­ness was asso­ci­ated with a very dif­fer­ent set of mar­it­al qual­it­ies includ­ing more neg­at­ive emo­tion­al beha­vi­or and great­er mar­it­al dis­sat­is­fac­tion.

Sim­il­arly, the per­son­al pro­nouns used by CEOs in their annu­al share­hold­er let­ters provide a use­ful way of pre­dict­ing future com­pany performance. No doubt gleaned from the Ritten­house Rank­ings Candor Sur­vey, this is from Geoff Colvin’s book, Tal­ent is Over­rated:

Laura Ritten­house, an unusu­al type of fin­an­cial ana­lyst, counts the num­ber of times the word “I” occurs in annu­al let­ters to share­hold­ers from cor­por­ate CEOs, con­tend­ing that this and oth­er evid­ence in the let­ters helps pre­dict com­pany per­form­ance (basic find­ing: Ego­ma­ni­acs are bad news).

via Bark­ing Up the Wrong Tree (1 2)

Common Misconceptions About Publishing and Writing

After real­ising that “many people don’t have the first clue about how the pub­lish­ing busi­ness works — or even what it is”, the some­what pro­lif­ic sci­ence fic­tion writer Charlie Stross decided to do some­thing about it. The res­ult was a series titled Com­mon Mis­con­cep­tions About Pub­lish­ing.

This is admit­tedly only one author’s view­point and set of opin­ions, but Stross’ series of some­times lengthy but always insight­ful essays expose the innards of pub­lish­ing (at least, it seems to). Posts in the series include:

Some­thing that par­tic­u­larly struck me was this look at author income inequal­ity:

Research­ers [cal­cu­lated the] Gini coef­fi­cient for authors’ incomes — a meas­ure of income inequal­ity, where 0.0 means every­one takes an identic­al slice of the com­bined cake, and 1.0 indic­ates that a single indi­vidu­al takes all the cake and every­one else starves. Let me provide a yard­stick: the UK had a Gini coef­fi­cient of 0.36 in 2009, the widest ever gap between rich and poor — while the USA, at 0.408, had the most unequal income dis­tri­bu­tion in the entire developed world. The Gini coef­fi­cient among writers in the UK in 2004-05 was a whop­ping great 0.74. As the research­ers note:

Writ­ing is shown to be a very risky pro­fes­sion with medi­an earn­ings of less than one quarter of the typ­ic­al wage of a UK employ­ee. There is sig­ni­fic­ant inequal­ity with­in the pro­fes­sion, as indic­ated by very high Gini Coef­fi­cients. The top 10% of authors earn more than 50% of total income, while the bot­tom 50% earn less than 10% of total income.

This is the same Gini coef­fi­cient as Nam­i­bia in 1993 (the worst in the world at the time, accord­ing to the World Bank).

via The Browser

Contextual Writing (Telescopic and Responsive Text)

How can a writer cater to an audi­ence with diverse pref­er­ences and needs (par­tic­u­larly, how much detail they want and how much time they have)? One way is to use tele­scop­ic or respons­ive text.

Tele­scop­ic text is a meth­od of iter­at­ively dis­play­ing more and more tex­tu­al detail on request (I sup­pose the read­er becomes the user). Joe Dav­is’ bril­liant example of tele­scop­ic text starts with the phrase “I made tea” before pro­gress­ing to a 198-word short story through 45-or-so iter­a­tions. Won­der­ful.

Respons­ive text is sim­il­ar in some regards and vastly dif­fer­ent in oth­ers. Like a respons­ive design, respons­ive text ‘scales’ in response to the user’s screen size in order to dis­play an appro­pri­ate amount of tex­tu­al detail. If viewed on a lar­ger screen, Frankie Roberto’s respons­ive text example points out:

It’s a bit of an exper­i­ment, and I’m not really sure how use­ful it really is, but I think it’s an inter­est­ing idea.

It could also per­haps be com­bined with some form of a user inter­face that allows you to con­trol how much text you want to read. This might be really use­ful for news art­icles, for instance – you could decide wheth­er to read full quotes and a detailed back­story, or just the gist.

I think mak­ing this beha­viour user-con­trol­lable is key and an inter­face variable/bookmarklet is an inter­est­ing concept to fol­low. One issue I envis­age is that adop­tion of this will come from authors and mak­ing this easy-to-imple­ment on the pro­du­cer-side will take some skill.

via @fooman­doo­n­i­an