The List of N Things

http://www.paulgraham.com/nthings.html

by djm • 17 years ago

146 64 17 years ago

djm• 17 years ago

You seem to be leaning towards the idea that these list type articles are inferior to the essay format. When reading an article I consider the clarity with which the author has expressed himself to be most important - if he can express ideas more clearly in a list than otherwise then I am all for it.

I do tend to think this format is abused when people wish to write quick and dirty articles online though. For me, an article titled 'N things...' tends to be reason enough to ignore it in most cases.

Also, I will admit to preferring the essay format overall. You cannot lose yourself in a list like you can in a good essay regardless of how good the content is. An essay is more like a story which sucks you in.

cperciva• 17 years ago

I think PG is missing something in his comparison of "list of N things" to "beginner" 5-paragraph essays: Cohesion matters.

Even if they cover the same points, a good essay is far more informative than a "list of N things" article / powerpoint presenation: An essay has to tie the points together, pointing out the connections and drawing conclusions. When I was in school, my English teachers often returned essays (but not mine!) to students with "a list is not an essay" (or similar words) written in red ink.

Sure, poorly written essays, in the limiting case, end up being simply a list of disconnected statements -- but students are asked to write essays in order to develop their ability to write cohesive essays, not in order to demonstrate their knowledge of the underlying facts, so PG's suggestion to 'let them write lists of n things like the pros, with numbers and no transitions or "conclusion"' utterly misses the point.

sachinag2• 17 years ago

Essays are planned; papers are stream-of-consciousness. Readers can't follow the internal machinations of a particular writer's psyche; while you can have a revelatory essay, you still need to plan it.

My best example: Woolf reads like stream of consciousness, but she was a meticulous planner: she obsessed over every word - her daily output was a mere 75-150 words.

tptacek3• 17 years ago

I think it's the opposite, isn't it? Essays are "tentative"; they're a written document of the writer's grappling with their subject, which is why you get so many ideas while writing them.

ankeshk• 17 years ago

I remember Guy Kawasaki saying he tries to follow this list of N things tactic for all his speeches so that people don't have a problem of following him. They won't doze off.

Two other quick formats of writing essays:

"The secrets of telling a story well are three:

i. How to end

ii. Where to begin

iii. What to leave out."

- Roy H. Williams.

In that order. Determine the end first. And you can take your readers on to a journey. Surprise and delight them by connecting seemingly unrelated stories. Malcolm Gladwell uses this format very well.

Another format is answering 3 questions:

i. What is the problem?

ii. Why is it the problem?

iii. How to solve it?

Makes writing easier, quicker and more comprehensive.

jamesbritt• 17 years ago

"I remember Guy Kawasaki saying he tries to follow this list of N things tactic for all his speeches"

It helps the audience judge about how long something will be going on for. With print, you don't need that, since you can see the length of the piece in advance.

pingswept• 17 years ago

I don't know-- I've eaten some pretty bad cheeseburgers at good restaurants.

tptacek3• 17 years ago

The simplest dishes are usually the hardest to make; they're least forgiving of errors or compromises. I read that in the essay and wondered whether that was really the case with top-N-list essays as well.

baha_man• 17 years ago

How can you call a restaurant good if they can't even serve a decent cheeseburger?

jokull2• 17 years ago

I live in London and these Brits cannot make an honest, decent hamburger. The good ones are ones in fancy dress. The simple ones are dry meatballs of a pale sickly color. They've got a nice sandwich chain called Pret a Manger that make consistently good food however. And a very strong café americano.

movix• 17 years ago

Timeline from live chicken to Pret a Manger Chicken Wrap? Live, dead, cooked, chopped, wrapped = 55 minutes

byrneseyeview5• 17 years ago

I've eaten some amazing cheeseburgers at really bad restaurants.

Maybe they serve bad cheeseburgers so you'll tell your friends -- that way, they'll avoid having any customers who regularly converse with someone willing to admit he orders cheeseburgers.

whatusername2• 17 years ago

Chicken Parmigiana is my "safe-bet" food if I'm eating somewhere I don't trust.

netsp• 17 years ago

Bad Aussie!

andreyf2• 17 years ago

Sorry for nitpicking, but first sentence should be I'll bet you, not I bet you.

abstractbill2• 17 years ago

I think the "I'll bet you..." form is more standard in the US, but saying "I bet you..." is definitely acceptable, and in my experience is actually much more common in some other English-speaking countries (e.g. England itself).

andreyf2• 17 years ago

That's strange. I can't find a single example online where "you" isn't the subject of the bet (for example, I bet you look good on the dance floor).

jokull2• 17 years ago

Paul Graham has managed to set himself up an army of nitpickers (free editors).

pg• 17 years ago

An alarming vision.

http://dictionary.reference.com/browse/nit

tptacek3• 17 years ago

Lists are just writing devices. There are lots of other writing devices, and they all serve the same purpose: as scaffolding for building up ideas and communicating effectively. Some of the popular ones:

* Socratic dialog (ie, the openings of G.E.B. chapters)

* Extended metaphor (ie, mechanical_fish's zoo comment)

* Anthopomorphism (ie, Gruber's adventures of the washed-up brushed metal theme)

* F.A.Q.

* Narrative (ie, Ivan Krstic's Porsche story --- which, again, who knew Ivan had that in him? Wow!)

And my point is just, all these devices can be abused, and all of them can serve a purpose. There's no profound reason Top-N is popular; the reason it's there is straightforward: every "how to have a popular blog" post lists it as a key technique for engaging A.D.D. Internet readers.

fjabre2• 17 years ago

The 3 Reasons I agree with tptacek's statement:

1. It puts everything in perspective. After all, essay is just another one of several devices to convey a point through writing.

2. He clearly states that any one of these devices can be abused. Surely anyone can write a crappy and spam-filled essay with a catchy title just as they can do so with a list.

3. Clearly many people surfing the Internet do have A.D.D. Perhaps the cause is the Internet itself but it doesn't matter, the end result is the same - namely, reading material for A.D.D. Internet users is in high demand.

EOL

10ren3• 17 years ago

* mechanical_fish's zoo comment http://news.ycombinator.com/item?id=469940

* Gruber's adventures of the washed-up brushed metal theme ????

* Ivan Krstic's Porsche story http://news.ycombinator.com/item?id=424506

ableal• 17 years ago

brushed metal theme ????

http://daringfireball.net/2005/09/anthropomorphized

http://daringfireball.net/2007/06/brushed_metal_leopard

mechanical_fish• 17 years ago

As PG points out, lists are an especially low-risk literary device.

Socratic dialogue requires you to write dialogue, which only some people have the ear for. It can also tend to be long-winded and indirect, like real dialogue.

Extended metaphors, like cantilevered structures, will eventually break. Usually sooner rather than later. When I started writing the zoo comment, I had no idea the damn thing was going to sustain itself across a whole essay; I figured it would reach a length of two sentences and then need to be edited down to one or zero, which is the fate of most newborn metaphors.

Anthropomorphism is another form of extended metaphor, except that it is especially likely to overreach badly. You can end up making big logical mistakes, because the human brain is very forgiving of anything that lets it exercise its built-in machinery for understanding humans. We love that. So history is rife with epic-level anthrocentric fallacies that lasted for hundreds or thousands of years. (How can the Earth be so much older than humans? How can life have been designed, if not by an independent consciousness that likes to build things? Of course the sun orbits us, and not the other way around!)

The typical FAQ is terribly disorganized and unbalanced, even more so than the list. (To achieve a FAQ that is actually readable, you must carefully compose your FAQ. Do not make the natural mistake of thinking that throwing all your FAQs into a pile will produce a readable FAQ. ;)

And, while narratives work well, only a tiny percentage of anecdotes are of any use as entertainment:

http://www.youtube.com/watch?v=n7KQ4vkiNUk&feature=relat...

[This link is to anecdote professional Ira Glass. I recommend watching the whole series on YouTube but do not miss segment three:

http://www.youtube.com/watch?v=-hidvElQ0xE&feature=relat...

which is not exactly on topic but is too excellent not to plug.]

RyanMcGreal2• 17 years ago

>Socratic dialogue requires you to write dialogue, which only some people have the ear for.

A great negative case study for this is Jane Jacobs' second-last book, The Nature of Economies, which was written as a socratic dialogue among, regrettably, a bunch of insufferable hip urban New Yorkers. The ideas she expresses are first-rate (as usual), but the presentation borders on the unreadable.

joez• 17 years ago

My favorite is when N is a nice round number like 25, 50, 100 or 101. What is the likelyhood that the exhaustive list ends up being such a marketable number? Must be very low. So I always wonder what was omitted or, more likely, what pork was added to get to the round number.

kabuks2• 17 years ago

I actually tend to trust non-round numbers. Like 9, or 17. It indicates to me that author didn't try to fit the ideas to a number, but actually thought things through and came up with the number after the essay was written

byrneseyeview5• 17 years ago

That's why, when I write an article like this, I pick a number like that.

riffer• 17 years ago

The insight that readers like knowing that writers have agreed to be constrained is a great one. Witness Twitter.

fjabre2• 17 years ago

While PG makes some good points it sounds a little bit like why to read Shakespeare instead of Poe.

I happen to like both. Although I might add the list of N things does get abused more often.

I think using Cosmopolitan as an example and/or relegating the N-List format to newbie writers takes it a bit far.

Dilpil• 17 years ago

I wonder if this article will be a huge hit on digg?

sayrer• 17 years ago

They're called listicles.

BRadmin• 17 years ago

"...the 5 most interesting startup founders of the last 30 years."

http://www.paulgraham.com/5founders.html

byrneseyeview5• 17 years ago

The beginning of that sentence is: "Inc recently asked me who I thought were..."

randallsquared• 17 years ago

> You think of n/2 of them in the first 5 minutes.

The word that leaps to mind here is "half". :)

izaidi• 17 years ago

Why use four characters when you can use three?

abstractbill2• 17 years ago

Every equation included in an essay will halve its readership? ;-)

kylec• 17 years ago

To generalize: why use n characters when you can use n-1?

caffeine2• 17 years ago

I think it's sort of a textual fist-bump to people who smile when they see n followed by n/2, because it means something's about to get logarithmic, and we like that.

zackattack• 17 years ago

He also recently used "orthogonal" in an essay when it really wasn't the best word choice. When given the option to maximize eloquence, the author favors impressing the reader with his technical know-how, sacrificing being articulate for upping his nerd cred. Possibly a smart move for communicating with the HN community and securing his status as leader. A political thing. I often find myself talking more nerdy than I need to in #startups, for example.

Tichy• 17 years ago

If your brain is in algorithm mode, n/2 might actually be easier to process.

gizmo• 17 years ago

Because the moment I read n/2 I wondered which kind of distribution n has, for t.

It could be (cumulative) exponential. Ideas would then be independent and random. That doesn't feel right, because ideas are very much related, but a better model doesn't immediately spring to mind.

Had he said half, I would never consider the distribution of ideas over time. And I bet pg considered the distribution while writing the essay, and he probably also realized it wasn't really all that interesting, and consequently left it out.

caffeine2• 17 years ago

You can play the Zeno game with this.

The first half of the items is actually an article of n/2 items - so you think of n/4 in the first 5 minutes.

So you never actually think of an item.

yan• 17 years ago

That protocol is cracked.com's livelihood.

billswift2• 17 years ago

"The list of n things similarly limits the damage that can be done by a bad writer."

I have more things I want to do than I have time to do them; why in the world would I waste time reading something if I expected it to be by a bad writer? I have quit reading writers before when I came across strong evidence of bias or incompetence in their writings.

whatusername2• 17 years ago

A bad writer does not bad ideas make!

billswift2• 17 years ago

Not necessarily, but the converse IS true, bad ideas make a bad writer no matter how easy to read or entertaining he may be.

wglb• 17 years ago

A nice way of suggesting what I have felt, and that such articles are quite often hastily prepared. Most importantly, how do we as readers have any confidence that an article titled "The 7 ..." is really inclusive or even instructive?

bcl• 17 years ago

This article may have held my attention longer if it had been a list. It lost me after the 7th paragraph when I realized he was going to continue to ramble on about lists. So I skipped to the final paragraph just to be sure.

caa09hh• 17 years ago

I really find that to be true all the time. I would much rather click on a list of n things rather than an article about the same topic.

kwamenum86• 17 years ago

Regarding making explicitly numbered lists of n things in school, I am pretty sure thy do that in the lower grades, although not in essay form. For example, children have to write "5 things I learned about x" or "5 things I like about y" and are graded on relevancy and sometimes persuasiveness...if I am remembering correctly anyway.

anigbrowl• 17 years ago

Irrelevant, but too good to pass up with Paul's title: http://webspace.webring.com/people/fc/churud_geo/cyberiadn.h... ...a short bit of sci-fi for your reading pleasure.

10ren3• 17 years ago

This is a charming little essay. But why?

- The "exoskeleton" metaphor for an N-list having its skeletal structure on the outside made me smile.

- I like the explanation of N-lists having a flat structure. Clean modules, with no dependencies. If your code can naturally fit this structure, it will be simpler. Alan Kay says that by inventing new representations, we make their users more intelligent (e.g. multiplication with Roman v.s. Arabic numerals). If your new representation enables its user to think in flat terms about what previously required nested or recursive structures, then you have done something great.

- These are playful and novel ways of looking at commonplace things. They evoke my conviction that gold is all around us; just a step or two off the path is something incredible. All you need do is look.

But underlying all this is the lack of any discernible ulterior motive: it's just for fun. :-)

10ren3• 17 years ago

I just noticed that I have an implicit conviction that pg's essays are entirely written for us, that the comments here are their beginning and their end - their entire purpose.

I don't know if it's true; not that pg intends it, nor that that is how they are received. I have a strong sense that it is, without evidence.

JBiserkov• 17 years ago

How to make a list of n elements?

  (defun list-of (n elt)
    (if (zerop n)
        nil
        (cons elt (list-of (- n 1) elt))))

ANSI Common Lisp, Paul Graham, page 38

1996 by Prentice Hall

The List of N Things

Related Stories