The billionaire’s typewriter

A friend pointed me to a story on Medium called Death to Type­writ­ers,” by Medium de­signer Marcin Wichary. The story is about the in­flu­ence of the type­writer on dig­i­tal type­set­ting. It ref­er­ences myex­cel­lent list” of type­writer habits.

Thank you for the com­pli­ment, Mr. Wichary. I can’t quib­ble with the de­tails of your piece. It’s true that Medium and I are op­posed to cer­tain ty­po­graphic short­cuts im­ported from the typewriter.

But by the end, I re­al­ized I dis­agree deeply with Medium about the ethics of de­sign. And by ethics, I mean some­thing sim­ple: though Medium and I are both mak­ing tools for writ­ers, what I want for writ­ers and what Medium wants couldn’t be more dif­fer­ent. Medium may be avoid­ing what made the type­writer bad, but it’s also avoid­ing what made it good. Writ­ers who are tempted to use Medium—or sim­i­lar pub­lish­ing tools—should be con­scious of these tradeoffs.

So, a few words about that.

For those who don’t in­ces­santly fol­low In­ter­net star­tups, Medium is a blog­ging ser­vice run by one of the founders of Twit­ter, multi­bil­lion­aire Evan Williams. Though it owes much to blog­ging ser­vices of the past (in­clud­ing Blog­ger, also founded by Mr. Williams), Medium is ori­ented to­ward longer, less di­aris­tic stories.

Medium also dif­fers from ear­lier blog­ging ser­vices in a sig­nif­i­cant, con­trar­ian way: it of­fers you, the writer, nearly zero op­tions for the pre­sen­ta­tion of your sto­ries. No mat­ter what kind of story you write, or who your read­ers are, it gets pack­aged into a sin­gle, non-ne­go­tiable template.

Medium isn’t the only blog­ging ser­vice rid­ing this wave, though so far it seems to have the biggest surf­board. Oth­ers in­clude Svb­tle, Posta­gon, and Sil­vr­back. They all pro­mote a sim­i­larly con­strained ap­proach to de­sign, which is some­times called min­i­mal­ist.

As a fan of min­i­mal­ism, how­ever, I think that term is mis­ap­plied here. Min­i­mal­ism doesn’t fore­close ei­ther ex­pres­sive breadth or con­cep­tual depth. On the con­trary, the min­i­mal­ist pro­gram—as it ini­tially emerged in fine art of the 20th cen­tury—has been about di­vert­ing the viewer’s at­ten­tion from overt signs of au­thor­ship to the deeper pu­rity of the ingredients.

If that’s the case, we can’t say that Medium et al. are of­fer­ing min­i­mal­ist de­sign. Only the ve­neer is min­i­mal­ist. What they’re really of­fer­ing is a shift from de­sign as a choice to de­sign as a con­stant. In­stead of min­i­mal­ist de­sign, a bet­ter term might be ho­mo­ge­neous design.

On the one hand, Medium’s ho­mo­ge­neous de­sign works and reads well. Mem­bers of Medium’s de­sign team have cat­a­logued the many ty­po­graphic de­tails they’ve im­ple­mented. Good for them. If they some­times act as if they dis­cov­ered ty­pog­ra­phy like it was the Higgs bo­son, we can for­give their ex­cess of en­thu­si­asm. Bring­ing these de­tails to a wider au­di­ence, and rais­ing stan­dards for ty­pog­ra­phy on the web gen­er­ally, is a wor­thy project.

On the other hand, a nec­es­sary side ef­fect of Medium’s ho­mo­ge­neous de­sign is that every story looks the same. If you agree that the role of ty­pog­ra­phy is to en­hance the text for the ben­e­fit of the reader (as I con­tend in who is ty­pog­ra­phy for?), then it stands to rea­son that dif­fer­ent texts de­mand dis­tinct ty­pog­ra­phy. As I say in What is Good Ty­pog­ra­phy?, one size never fits all. Ty­pog­ra­phy wants to be heterogeneous.

Still, I wouldn’t say that Medium’s ho­mo­ge­neous de­sign is bad ex ante. Among web-pub­lish­ing tools, I see Medium as the equiv­a­lent of a frozen pizza: not as whole­some as a meal you could make your­self, but for those with­out the time or mo­ti­va­tion to cook, a po­ten­tially bet­ter op­tion than just eat­ing peanut but­ter straight from the jar.

The prob­lem, how­ever, is that Medium holds out its ho­mo­ge­neous de­sign as more than a frozen pizza. It has be­come, by the Jedi mind trick­ery fa­vored by to­day’s tech com­pa­nies, a Bel­la­gio buf­fet of de­li­cious nonsense:

  1. Evan Williams frames Medium as aplace for ideas” with anethos” ofopen­ness and democ­racy—like the In­ter­net it­self.” Fine, but ide­al­is­tic plat­i­tudes ex­plain noth­ing. How, specif­i­cally, does Medium im­prove the Internet?

  2. Mr. Williams claims that Medium isthe best writ­ing tool on the web.” Okay, that’s at least con­crete. But we’ve got a lot of good web-based writ­ing tools al­ready. Medium does more than those?

  3. Ac­tu­ally, no—Mr. Williams con­cedes that Medium hasstripped out a lot of the power that other ed­i­tors give you.” So how is it pos­si­ble to bethe best” while of­fer­ing less?

  4. Here, Mr. Williams par­ries—he claims that think­ing about the pre­sen­ta­tion of your work isa ter­ri­ble dis­trac­tion and a waste of time.” Why?

  5. Ap­par­ently be­cause he’sone of those peo­ple who will open up Word and spend half [his] time defin­ing styles and ad­just­ing the spac­ing be­tween para­graphs.” Hmm, not every­one has that prob­lem with Word.

  6. Now comes the hand-wav­ing, as Mr. Williams as­sures us that Medium’s ho­mo­ge­neous de­sign isn’t a lim­i­ta­tion—it’s in fact es­sen­tial to let yourbril­liance and cre­ativ­ity flow smoothly onto the screen.”

  7. More­over, any­one who dis­agrees is a Lud­dite—be­causeevery­thing [other than Medium] feels like step­ping back in time.”

Like all non­sense, it’s in­tended to be easy to swal­low. But Mr. Williams’s ar­gu­ment is flawed in at least three ways:

  1. It makes no sense in the con­text of to­day’s web. If Medium had launched in 2005, it would’ve been as­ton­ish­ing. But it didn’t. To­day, the costs of web pub­lish­ing—in­clud­ing de­sign—have de­clined to al­most zero. Rel­a­tive to to­day’s web, Medium is not cre­at­ing new pos­si­bil­i­ties, but in­stead clos­ing them off. To pre­vail, Medium needs to per­suade you that you don’t care about the broader ex­pres­sive pos­si­bil­i­ties of web publishing.

  2. It sets up a false di­chotomy about writ­ing tools. Mr. Williams de­picts the writer’s choice as Medium vs. com­pli­cated tools like Word. Not ac­cu­rate. First, dif­fer­ent tools ex­ist for dif­fer­ent needs. It would be silly to use Word to make a web page, but equally silly to use Medium to pre­pare a print-on-de­mand pa­per­back. Sec­ond, any­one who’s used cur­rent blog­ging tools ap­pre­ci­ates that web pub­lish­ing has be­come heav­ily au­to­mated. Much of the for­mat­ting can be han­dled au­to­mat­i­cally (e.g., via Word­Press themes) or man­u­ally, as you prefer.

  3. You’re giv­ing up far more than de­sign choice. Mr. Williams de­scribes Medium’s key ben­e­fit as res­cu­ing writ­ers from theter­ri­ble dis­trac­tion” of for­mat­ting chores. But con­sider the cost. Though he’s bait­ing the hook with de­sign, he’s also ask­ing you, the writer, to let him con­trol how you of­fer your work to read­ers. Mean­ing, to get the full ben­e­fit of Medium’s de­sign, you have to let your story live on Medium, send all your read­ers to Medium, have your work per­ma­nently en­tan­gled with other sto­ries on Medium, and so on—a sig­nif­i­cant concession.

As for that en­tan­gle­ment among sto­ries, Mr. Williams has con­ceded that it’scon­fus­ing.” But this am­bi­gu­ity isn’t a bug. It’s an es­sen­tial fea­ture of the busi­ness plan. The goal is to cre­ate the il­lu­sion that every­thing on Medium be­longs to one ed­i­to­r­ial ecosys­tem, as if it were the New York Times.

But un­like the Times, Medium pays for only a small frac­tion of its sto­ries. The rest are sub­mit­ted—for free—by writ­ers like you. Af­ter a long time be­ing elu­sive about its busi­ness model, Medium re­vealed that it plans to make money by—sur­prise!—sell­ing ad­ver­tis­ing. This means dis­play­ing ads, but also col­lect­ing and sell­ing data about read­ers and writ­ers. So Medium will ex­tract rev­enue from every story, whether it paid for that story or not. (By the way, will that rev­enue be shared with writ­ers? Um, no.)

And com­ing full cir­cle—what’s the in­dis­pens­able tool for cre­at­ing this il­lu­sion of an ed­i­to­r­ial ecosys­tem? The ho­mo­ge­neous de­sign. The but­ter­fly bal­lot of 2000 (de­picted in Why ty­pog­ra­phy mat­ters) proved that er­rors of ty­pog­ra­phy can have his­toric con­se­quences. Medium proves that ty­pog­ra­phy can be used as a tool of eco­nomic lever­age and control.

In truth, Medium’s main prod­uct is not a pub­lish­ing plat­form, but the pro­mo­tion of a pub­lish­ing plat­form. This pro­mo­tion brings read­ers and writ­ers onto the site. This, in turn, gen­er­ates the us­age data that’s valu­able to ad­ver­tis­ers. Boiled down, Medium is sim­ply mar­ket­ing in the ser­vice of more mar­ket­ing. It is not aplace for ideas.” It is a place for ad­ver­tis­ers. It is, there­fore, ut­terly superfluous.

But what about all the writ­ing on Medium?” The mea­sure of su­per­fluity is not the writ­ing on Medium. Rather, it’s what Medium adds to the writ­ing. Re­call the ques­tion from above: how does Medium im­prove the In­ter­net? I haven’t seen a sin­gle story on Medium that couldn’t ex­ist equally well else­where. Nor ev­i­dence that Medium’s edit­ing and pub­lish­ing tools are a man­i­fest im­prove­ment over what you can do with other tools.

In sum—still superfluous.

Let’s re­mem­ber two points that get lost among the torches and pitch­forks car­ried by Death to Typewriters.”

First, al­though the type­writer did im­pose ho­mo­ge­neous (and ugly) ty­pog­ra­phy, it had ex­cel­lent ethics. The type­writer made it pos­si­ble to write more quickly, leg­i­bly, and ac­cu­rately than ever be­fore, with low cost and high porta­bil­ity. In short, it of­fered free­dom. For that, ho­mo­ge­neous de­sign was a small price to pay.

Sec­ond, though type­writer ty­pog­ra­phy was ter­ri­ble, it wasn’t a choice made by type­writer man­u­fac­tur­ers out of lazi­ness or ig­no­rance. These com­pro­mises were ne­ces­si­tated by the me­chan­i­cal lim­i­ta­tions of the type­writer. Type­writ­ers were never ideal, but as cer­tain lim­i­ta­tions were over­come, they got better.

With to­day’s net­worked com­put­ers, we’re get­ting closer to the ideal. We en­joy the ben­e­fits of the type­writer with­out any of its lim­i­ta­tions. We get more ef­fi­ciency, speed, stor­age, de­sign op­tions, and free­dom. The com­puter is the most re­mark­able de­vice in the 500-year his­tory of print­ing (which al­ready in­cludes a lot of re­mark­able devices).

This leads back to why those type­writer habits are so aw­ful in the dig­i­tal age. Com­put­ers have none of the me­chan­i­cal lim­i­ta­tions of type­writ­ers. So the ty­po­graphic short­cuts that were a nec­es­sary evil with type­writ­ers are like­wise ob­so­lete. Why per­pet­u­ate them?

I rely on a broader ver­sion of this prin­ci­ple in my own work. Tech­nol­ogy keeps im­prov­ing, thereby ex­pand­ing pos­si­bil­i­ties for us. So we have a choice. We can ei­ther ig­nore those pos­si­bil­i­ties, and merely ac­cept what tech­nol­ogy of­fers, which will ul­ti­mately make us lazy. Or we can ex­plore those new pos­si­bil­i­ties. But to do that, we need to ex­pect more of ourselves.

We also need bet­ter tools. I’d char­ac­ter­ize most of my work as tool­smithing—whether the project is de­sign­ing a font, writ­ing a book, or cre­at­ing pub­lish­ing soft­ware. I don’t con­trol how oth­ers use these tools. I don’t want to, ei­ther. For me, it’s far more in­ter­est­ing to share these tools and then be sur­prised by how oth­ers use them.

To that end, I de­lib­er­ately avoid cre­at­ing tools that do too much. Some as­sem­bly is al­ways re­quired. For in­stance, I’ll tell you the qual­i­ties of good web­site ty­pog­ra­phy, but I’m not go­ing to sell you a tem­plate. I want the cus­tomers for my tools to be re­spon­si­ble for some of the heavy lift­ing. That way, they dis­cover that what they get out of the tool has a con­nec­tion to what they put in.

So even though I op­pose the type­writer habits, I still ap­pre­ci­ate that core ethic of the type­writer—re­mov­ing lim­i­ta­tions when you can, do­ing your best with them when you can’t. That’s a great idea. Yes, let’s ex­plore all the pos­si­bil­i­ties of the tech­nol­ogy avail­able to us. Let’s hack the hell out of every­thing and see what hap­pens. In the type­writer era, the tech­no­log­i­cal lim­i­ta­tions were mostly hard­ware. To­day, mostly soft­ware. But if we treat these lim­i­ta­tions as some­thing to obey—not over­come—we’ll just be­come in­den­tured to who­ever con­trols that technology.

In Death to Type­writ­ers,” Medium in­sists that the type­writer is itssworn en­emy.” In cer­tain ty­po­graphic de­tails, maybe so. But as a de­vice that im­poses ho­mo­ge­neous de­sign, Medium still has a lot in com­mon with the typewriter.

In fact, its ethics are ac­tu­ally worse than the tra­di­tional type­writer. Why? Be­cause Medium’s ho­mo­ge­neous de­sign has noth­ing to do with lim­i­ta­tions of the un­der­ly­ing tech­nol­ogy (in this case, the web). As dis­cussed above, it’s a de­lib­er­ate choice that lets Medium ex­tract value from the tal­ent and la­bor of others.

Medium is a new kind of type­writer—the bil­lion­aire’s type­writer. It’s not the only bil­lion­aire’s type­writer. So is the Kin­dle. So is iBooks. So is Twit­ter. What dis­tin­guishes these new type­writ­ers is not the pos­si­bil­i­ties they make avail­able to writ­ers, but what they take away.

Whereas the tra­di­tional type­writer of­fered free­dom at the cost of de­sign, the bil­lion­aire’s type­writer of­fers con­ve­nience at the cost of freedom.

As a writer and tool­smith, I’ve found the rush to em­brace these sys­tems per­plex­ing. Not be­cause I’m cur­mud­geonly. Not be­cause I fail to un­der­stand that peo­ple, in­clud­ing writ­ers, en­joy things that are free and convenient.

Rather, be­cause gen­tle scrutiny re­veals that these sys­tems are pow­ered by a form of hu­man frack­ing. To get his frack­ing per­mit on your ter­ri­tory, Mr. Williams (the multi­bil­lion­aire) needs to per­suade you (the writer) that a key con­sid­er­a­tion in your work (namely, how & where you of­fer it to read­ers) is awaste of time.”

If you really be­lieve that, then by all means, keep us­ing the bil­lion­aire’s typewriter.

But if you have doubts, here’s a counterproposal.

As a writer, the biggest po­ten­tial waste of your time is not ty­pog­ra­phy chores, but Medium it­self. Be­cause in re­turn for that snazzy de­sign, Medium needs you to re­lin­quish con­trol of how your work gets to readers.

Tempt­ing per­haps. But where does it lead? I fear that writ­ers who limit them­selves to pro­vid­ingcon­tent” for some­one else’sbranded plat­form” are go­ing to end up with as much lever­age as cows on a dairy farm. (A prob­lem at the core of the re­cent Ha­chette–Ama­zon dis­pute.)

If you want to be part of some­thing open and de­mo­c­ra­tic, use open-source soft­ware. If you want to have your writ­ing look great, learn some­thing about ty­pog­ra­phy (or hire a de­signer). If you need a plat­form for writ­ing, try Pollen (the sys­tem I made for this site), or Word­Press, or a sub­scrip­tion ser­vice like Atavist. I pre­fer web pub­lish­ing de­spite its short­com­ings, but if you don’t, then make an e-book or PDF and dis­trib­ute it yourself.

As writ­ers, we don’t need com­pa­nies like Medium to tell us how to use the web. Or de­fine open­ness and democ­racy. Or tell us what’s awaste of [our] time” and what’s not. Or de­ter­mine how and where read­ers ex­pe­ri­ence our work. We need to de­cide those things for ourselves.

—Matthew But­t­er­ick
17 Feb 2015

Per­haps my com­par­i­son of Medium tofrozen pizza”—a fan fa­vorite—even­tu­ally reached the desk of Evan Williams. Be­cause in a re­cent in­ter­view with the Guardian, he de­rides to­day’s web as largely junk food.” This is ironic, of course, be­cause as the founder of Twit­ter, Mr. Williams is ba­si­cally the Ray Kroc of the web.

But never mind that. Medium, he ex­plains, is about serv­inghealthy, nour­ish­ing food.” Even the front page of the site says so: Medium isn’t just pub­lish­ing web pages—it’s go­ing to move think­ing for­ward” in ways that, ap­par­ently, the open web cannot.

Maybe it’s no sur­prise that Mr. Williams is reach­ing for the eth­i­cal high ground—Medium seems to be strug­gling. The past year has seen sev­eral reversals:

  1. Af­ter launch­ing Medium with­out a com­ment sys­tem—be­cause that was the junk food of 1999—last year Mr. Williams re­lented and added com­ments, mak­ing Medium more like every other blog (or more like Twit­ter with­out a char­ac­ter limit).

  2. Af­ter po­si­tion­ing Medium as a pub­lisher of orig­i­nal writ­ing by ac­quir­ing web pub­lisher Mat­ter, Mr. Williams re­cently spun it out of Medium to be­come his hobby in­vest­ment. Other Medium prop­er­ties that had fea­tured new writ­ing, like Re:form and Arch­i­pel­ago, have been shut down altogether.

  3. Af­ter try­ing to hatch a high­fa­lutin ad­ver­tis­ing busi­ness premised on Medium dri­ving traf­fic to the pub­lish­ers un­der its um­brella, Medium is now repo­si­tion­ing it­self as a high­fa­lutin Word­Press al­ter­na­tive. The new busi­ness model? Rather than Medium send­ing traf­fic to pub­lish­ers, pub­lish­ers will share their traf­fic—and rev­enue—with Medium.

Wait—what rev­enue?” Um, yeah. When a ven­ture-funded tech startup is look­ing to hitch its wagon to a team of starv­ing mares, it’s not a good sign. Es­pe­cially be­cause the myth that has been at­tached to Medium—pri­mar­ily by Mr. Williams—is that Medium is thefu­ture of pub­lish­ing” or sim­i­lar nonsense.

Why non­sense? Be­cause the ev­i­dence has been rolling in, and it doesn’t sup­port Medium’s grandiose the­ory of its own in­dis­pens­abil­ity. In­stead, the open web is still win­ning. That’s great news.

The con­tin­u­ing ab­sence of a rev­enue model, how­ever, is not great news. But that’s go­ing to be on us to solve, not Medium.

With its var­i­ous piv­ots, Medium has es­sen­tially con­ceded that it doesn’t have a bet­ter idea how to make money from web pub­lish­ing. Itsnew” idea is as an­cient as hu­man com­merce: po­si­tion your­self in the mid­dle, where you can put your hands in other peo­ple’s pock­ets. At Medium, the fu­ture of pub­lish­ing looks in­creas­ingly like its past.

When I orig­i­nally wrote this piece, I didn’t un­der­stand the mo­ti­va­tions of Mr. Williams. It seemed to me that if you were a bil­lion­aire who really wanted tomove think­ing for­ward,” you’d have so many won­der­ful op­tions—fund schol­ar­ships, en­dow pro­fes­sor­ships, start re­search pro­grams, open non­profit or­ga­ni­za­tions. Yet Mr. Williams chose to side­step all those pos­si­bil­i­ties. In­stead, he cre­ated an­other tech startup. The bil­lion­aire’s typewriter.

But af­ter ten months of the Don­ald Trump pres­i­den­tial cam­paign, these mo­ti­va­tions are per­haps less mys­te­ri­ous. As a can­di­date, Mr. Trump has no in­her­ent in­ter­est in the de­tails of pub­lic pol­icy and coali­tion-build­ing that are cen­tral to po­lit­i­cal ac­tion. Rather, he’s in­se­cure about his pub­lic rep­u­ta­tion—in­ept in­vestor, re­al­ity-TV buf­foon, short-fin­gered vul­gar­ian—and sees his pres­i­den­tial cam­paign as his last, best chance to be taken se­ri­ously. How? By barg­ing into the con­ver­sa­tion about our na­tion’s most con­se­quen­tial issues.

I won’t spec­u­late about Mr. Williams in par­tic­u­lar—who dropped out of col­lege be­cause he con­sid­ered it a waste of time,” whose first suc­cess­ful com­pany nearly died, and whose sec­ond suc­cess­ful com­pany was the ac­ci­den­tal off­shoot of his failed startup. But any wealthy per­son who wants to or­ches­trate a con­ver­sa­tion about the fu­ture of pub­lish­ing—from a perch of sig­nif­i­cant self-in­ter­est—de­serves com­men­su­rate scrutiny and skepticism.

As some­one who had a good run in the tech world, I buy the the­ory that the main rea­son suc­cess­ful tech founders start an­other com­pany is to find out if they were smart or merely lucky the first time. Of course, the smart al­ready know they were also lucky, so fur­ther ev­i­dence is un­nec­es­sary. It’s only the lucky who want proof they were smart.

—Matthew But­t­er­ick
12 April 2016

by the way
  • Though I’ve been pok­ing holes in its rhetoric, I don’t have an­tipa­thy to­ward Medium any more than I do Google Fonts. I get it—it’s a com­pany set up to make money. It’s not a lit­er­ary foun­da­tion. I’m sure the peo­ple in­volved with it are tal­ented and sin­cere. And they cer­tainly don’t care what I think.

  • A cou­ple read­ers have pointed out that Medium doesn’t re­quire ex­clu­siv­ity—you own your sto­ries, and you can pub­lish them else­where. Fair enough. But this doesn’t change the core ar­gu­ment. Medium is def­i­nitely pitch­ing it­self to writ­ers as an all-in­clu­sive plat­form (Build your pub­li­ca­tion, blog, or writ­ing port­fo­lio”). As for those writ­ers who are us­ing it as a sec­ondary out­let, Medium is still ex­tract­ing rev­enue from their sto­ries that isn’t shared. The re­cent Medium for Pub­lish­ers ini­tia­tive seems in­tended, in part, to close this loophole.

  • I’m not the first to raise these is­sues. See also Rian van der Merwe (Medium seems to be more about Medium than about au­thors … The bar­rier to set­ting up your own site has never been lower”), Marco Ar­ment (con­sider whether it’s wise to in­vest your time and writ­ing in some­one else’s plat­form for free”), and Alexis Madri­gal (me­dia pro­duc­ers … have to de­cide whether Medium is a friend or a foe”).

  • Con­fi­den­tial to graphic de­sign­ers who are pub­lish­ing sto­ries on Medium: if you wouldn’t set your busi­ness cards in Times New Ro­man, then why would you … ah, for­get it.