Hey, y’all. Good morning.

I real­ly real­ly love it when a con­fer­ence has a one-word theme, because that’s right in my wheel­house as a dic­tio­nary edi­tor. I feel very com­fort­able with a sin­gle word at a time. I can work with this. Did any­one else read the logo as rebel lion?” I was like, That is an awe­some name for a band.”

Obviously lex­i­cog­ra­phers are the first peo­ple you think of when you think of rebels, right? Dictionary edi­tors. You can see that we’re always the first on the bar­ri­cades when the time comes. And of course this is what we think of when we think of rebel­lion. Guns, shout­ing, fire in the streets. But in my pro­fes­sion­al opin­ion, I think that we can broad­en the def­i­n­i­tion of rebel­lion. Really, rebel­lion is just try­ing to make the world match your under­stand­ing of it. You believe things should be one way, the world believes things should be anoth­er way, and you’re going to change the world to match your under­stand­ing. Why force some­thing that does­n’t fit? Why not just make a square hole?


So if we think that the rebel­lion changes the world, does it real­ly mat­ter what the time scale is? It does­n’t have to be an overnight over­throw. It can be the steady remak­ing of the world through pure force of con­vic­tion, like water wear­ing away stone. We have slow food, we have slow fash­ion, why can’t we have slow rebel­lion? Many things that we think of as being rev­o­lu­tions, like sci­en­tif­ic rev­o­lu­tions, were not the kind of sud­den overthrow/lots of shout­ing rebellions.

This is what Max Planck said very famous­ly about sci­en­tif­ic revolutions: 

A new sci­en­tif­ic truth does not tri­umph by con­vinc­ing its oppo­nents and mak­ing them see the light, but rather because its oppo­nents even­tu­al­ly die, and a new gen­er­a­tion grows up that is famil­iar with it.
Max Planck

They don’t tri­umph. You just have to out­live the bastards.

So espe­cial­ly when you’re think­ing about the rev­o­lu­tion, the rebel­lion, that comes with a new idea, some­times a slow rebel­lion is the way to go. And of course when you want to change the world, it helps that the way that I’ve want­ed to change the world affects one very very small part of the world. I want­ed to change how dic­tio­nar­ies were made. This is how peo­ple real­ly think dic­tio­nar­ies are made:

Illustration of a hand-cranked meat grinder

You take the lan­guage, you grind it up, and then you have this kind of con­sis­tent sludge of dic­tio­nary. That’s not very enter­tain­ing to me, and I don’t think it’s very help­ful to peo­ple, so here’s the way that I think the dic­tio­nary should be made. They should rep­re­sent every sin­gle word in the lan­guage, no mat­ter what. And that usage is what deter­mines mean­ing; mean­ing is deter­mined by how peo­ple use a word. And that yes, that means every word, even the ones you don’t like. If you want to argue with this, I’m real­ly hap­py to argue with you at the par­ty tonight, but I should warn you that I don’t drink, and if when we argue about this you’re drunk and I’m sober, I’m going to win. But it’s okay because if you’re drunk enough you won’t remem­ber los­ing. So that’ll be fine.

If you were here at PopTech in 2008, you might be hav­ing a moment of déjà vu right now because I was right here on this stage when we announced word​nik​.com, which is the dic­tio­nary that I have been mak­ing. And I should prob­a­bly make some kind of Inception joke right here. I thought about wear­ing that dress again. But I decid­ed to just show you our throw­back logo.

You can hard­ly tell that it was done in a style I like to call half-assed Illustrator.” 

Wordnik has been going on for six years, since 2008. The peo­ple at Poptech were the first peo­ple to see our alpha ver­sion, the first peo­ple to get to try it out. I think that if one human year is like sev­en dog years, then six human years is like forty-two Internet years, right? Six years on the Internet is a real­ly long time. The advan­tage of a slow rebel­lion is that it gives you time to learn things while you’re chang­ing the world. If you have a sud­den over­throw rebel­lion, there’s not real­ly time to reflect. But six years is a long time to reflect.

I try to learn some­thing every day, not mat­ter how dumb, and I like to tweet it. TILT” means Thing I Learned Today, and this is some­thing I learned back in August that I have not yet had the oppor­tu­ni­ty to put into practice:


So I try and learn some­thing every day, but I think that the most valu­able kind of knowl­edge is that which you acquire through noe­ge­n­e­sis. Noegenesis is a fan­tas­tic word, and it means The acqui­si­tion of new knowl­edge from obser­va­tion and expe­ri­ence, and from infer­ring rela­tion­ships between known things.” So with six years of Wordnik on the Internet, a lot of what we’ve learned, what I’ve learned, are some assump­tions that got validated.

My favorite thing that we’ve learned is the assump­tion we went with going in is that mean­ing of a word equals usage of a word, and you can infer mean­ing from usage. So STEAM, which stands for Science, Technology, Engineering, Art, and Math.” You read that sen­tence, you know what STEAM means, you don’t need a tra­di­tion­al dic­tio­nary def­i­n­i­tion. You’re good. You can move on with your life. I want to point out the cita­tion for sec­ond exam­ple. John Maeda was say­ing, Oh, you need the A for Art to turn STEM (which is Science, Technology, Engineering, and Math) to STEAM.”

Another thing I’ve learned that made me real­ly hap­py is that peo­ple real­ly do love words. Sometimes when you real­ly love some­thing you just extrap­o­late and believe that the uni­verse real­ly loves some­thing. But it turns out lots of peo­ple love words. There have been 158-some odd thou­sand words that have been marked as a favorite on word​nik​.com. That’s rough­ly three new favorite words an hour for six years, which is pret­ty cool.

And peo­ple like to col­lect words. People have made more than 48,000 lists on Wordnik, and those lists com­prise 1.7 mil­lion words. 1.7 mil­lion words were con­sid­ered impor­tant enough for peo­ple to put them on a per­son­al list.

We also found out that peo­ple need to add words to oth­er things. So the Wordnik API has been called [1,805,761,975] times. In fact, just back­stage I was talk­ing with Brent, who helps run this thing, and he’s actu­al­ly made some­thing with the Wordnik API. He has this lit­tle cool tool that puts his word of the day on his own blog. Learning all this was great, and learn­ing all this was grat­i­fy­ing, but there were a lot of things that we did­n’t know going in. 

One of the things we did­n’t know is what things were going to be easy and what things were going to be hard. Larry Wall, the inven­tor of the pro­gram­ming lan­guage Perl, which is the first pro­gram­ming I did any kind of pro­fes­sion­al non-school cod­ing in, and I should real­ly embroi­der this on a sampler:

Easy things should be easy, and hard things should be possible.
Larry Wall

Another famous Perlism is that there’s more than one way to do it,” which I think is also a great mantra for slow rebels, right? If at first you don’t suc­ceed, try, try again. He also said some­thing that I used to have on a sign above my desk that I had pho­to­copied from one of the Perl pro­gram­ming books that said, We want you to acquire the three chief virtues of a pro­gram­mer, which are: lazi­ness, impa­tience, and hubris.”

So, we found that a lot of things that we thought would be easy were easy, we found that a lot of things we thought would be hard were in fact pos­si­ble. But the hard­est thing, the thing that I was the most uncer­tain of when we first start­ed Wordnik was this Quora ques­tion: How does Wordnik make mon­ey? Because when you have a start­up, that’s one of the things that you’re try­ing to find out. What’s the busi­ness mod­el. This is answer I post­ed to this question: 

A whiteboard showing a plan in three phases, phase 1 is "WORDS," phase 2 is "?", phrase 3 is "profit"

I don’t know how many of you are South Park fans, but they had an episode where there were these under­pants gnomes, and this was their busi­ness mod­el. This is my most upvot­ed answer on Quora, by the way. Only two of those peo­ple who upvot­ed were actu­al Wordnik employ­ees at the time. But it turns out our the­sis was that if you knew a lot about a lot of words, that that data would be valu­able. And it turns out with Reverb Technologies (we’ve made the Reverb iPad and iPhone read­er app) that you can use a word graph to build a con­tent dis­cov­ery plat­form that helps you dis­cov­er arti­cles based on your inter­ests rather than your demo­graph­ics. It’s been down­loaded near­ly half a mil­lion times, and peo­ple use it to read about five to eight times longer than they do in com­pet­ing apps.

So when you know all the words, you know a lot about about­ness, and you can help peo­ple find what they’re real­ly inter­est­ed in. So yay, busi­ness mod­el, right? This works great. So I like to say that when we start­ed Wordnik, we were try­ing to make pan­ning for gold, find­ing out infor­ma­tion about words, more effi­cient and more scal­able. And we thought that that data would be gold, that it would be very valuable. 

And we were right. But it turned out to be even big­ger than we thought. So we scaled up pan­ning for gold. That was great. But in the process what we basi­cal­ly got was hydro­elec­tric pow­er. It was­n’t so much the gold that was impor­tant, but the process. The process of ingest­ing and analysing and build­ing a word graph out of bil­lions of words was very powerful. 

So that’s great, right? Rebellion accom­plished, indus­try dis­rupt­ed, prod­uct mar­ket fit, busi­ness mod­el found, cus­tomers hap­py. Well… When this slide came up yes­ter­day, did any­body else kind of feel like this was a mes­sage to them personally?

Photo of a text slide reading "If you want to find your greatest love. Look inside the hardest thing you do. That's where it will be."

It was like, Yay, the uni­verse! It’s try­ing to tell us some­thing.” So I’m very hap­py with Reverb Technologies. I’m real­ly proud of what we’ve built. We built a fan­tas­tic tech­nol­o­gy that helps pub­lish­ers reach mil­lions of peo­ple with rel­e­vant, use­ful con­tent. It makes read­ers hap­py. It makes writ­ers hap­py. But that actu­al­ly turned out to be eas­i­er than I expect­ed, so it’s not the hard­est thing. The hard­est thing for me is how can I make a 120,000 Wordniks hap­py? And how could I make that num­ber ten times bigger?

Because real­ly, the English lan­guage belongs to every­body. The English lan­guage is some­thing that we all share togeth­er. The only rea­son it exists is because we all agree to under­stand each oth­er when we speak it. That’s what the English lan­guage is. So today I’m hap­py to announce that in order to fur­ther our mis­sion of mak­ing as much infor­ma­tion as pos­si­ble about as many words as pos­si­ble avail­able to as many peo­ple as pos­si­ble, that Wordnik is going to become a not-for-profit corporation.

And this feels hard and scary, but I’m so hap­py about it because it means that I can ask for help. I can ask peo­ple who love and want to be part of the English lan­guage to join me to make that infor­ma­tion avail­able to every­one. This URL is live right now. I thought peo­ple love words, there are so many favorite words on Wordnik. Wouldn’t it be nice to let peo­ple adopt a word, just like you can adopt a high­way? And when you adopt a word we can help spruce it up a lit­tle bit. The charis­mat­ic words like serendip­i­ty” and cal­lipy­gian” will help sup­port the words that nobody likes, like impact.”

This is an alpha alpha alpha. Once again PopTechers are going to be the first peo­ple to get to try out some­thing new with Wordnik. That’s live and you can reserve your word today. John [Maeda] already grabbed design,” so sor­ry, folks. And let me know what your ideas are. What you’re inter­est­ed in, what you’d like us to do, because again the English lan­guage belongs to everyone. 

And by mak­ing Wordnik open, non-profit, we can do all the things for which there real­ly is no busi­ness mod­el. If any­body knows an excel­lent busi­ness mod­el for writ­ing ety­molo­gies, please let me know. Everybody loves them, nobody wants to pay for them. 

So I hope to be back at PopTech in 2020 in anoth­er few years to give you anoth­er update. Thank you so much for your kind atten­tion. I also want to thank Joey specif­i­cal­ly for all his work with Creative Commons, because all of these images were Creative Commons-licensed from Flickr. This pre­sen­ta­tion itself is Creative Commons-licensed, so if you’d like to use all these same slides to give a com­plete­ly dif­fer­ent talk, please let me know. I will make them avail­able to you.

Thank you so much.

Help Support Open Transcripts

If you found this useful or interesting, please consider supporting the project monthly at Patreon or once via Cash App, or even just sharing the link. Thanks.