Survivorship Bias, Survivor Guilt, and Opportunity Cost, Oh My!

November 16th, 2011 § 6 comments § permalink

N.B. For best results, try to get Destiny’s Child’s “Sur­vivor” going in your head before proceeding.

Two recent blog posts by Larry Cebula and Hol­ger Syme high­light the deep divide that sep­a­rates the pes­simists from the opti­mists in acad­e­mia. Cebula explains why he steers his stu­dents away from pur­su­ing a career as a pro­fes­sor, essen­tially argu­ing that the odds are sim­ply too stacked against them even under the best of con­di­tions. Syme, in con­trast, sug­gests that Cebula’s dream-crushing advice is short-sighted and ulti­mately dan­ger­ous to the long-term via­bil­ity of the pro­fes­sion. While Cebula’s rea­son­ing will be famil­iar to many — he’s work­ing the same rich vein as has William Pan­na­packer under his nom de doom Thomas H. Ben­ton in the Chron­i­cle of Higher Edu­ca­tion — I sus­pect he’s also still greatly out­num­bered among the greater pop­u­la­tion of aca­d­e­mics, thanks in no small part to sur­vivor­ship bias and an unwill­ing­ness to grap­ple with the unfor­giv­ing cal­cu­lus of oppor­tu­nity cost.

“Sur­vivor­ship bias,” you ask? Let’s roll Wikipedia:

Sur­vivor­ship bias can lead to overly opti­mistic beliefs because fail­ures are ignored […] It can also lead to the false belief that the suc­cesses in a group have some spe­cial prop­erty, rather than being just lucky.

Sur­vivor­ship bias can lead to all sorts of hilar­i­ous sit­u­a­tions, like your grand­par­ents advis­ing you not to bother wear­ing a seat­belt because they never did, and look, they’re still around! Unfor­tu­nately it also can drive you to imag­ine that there’s a rea­son for your suc­cess. You cer­tainly deserved to win, and oth­ers who deserve it will suc­ceed, too!

In con­trast, Cebula, Pan­na­packer, et al. appear at least par­tially moti­vated by sur­vivor­ship bias’s evil twin, survivor’s guilt. They just know that they didn’t do any­thing par­tic­u­larly spe­cial, and they’re mor­ti­fied by the thought that oth­ers have failed and are fail­ing while they haven’t. They blame them­selves for the gross imbal­ance in the job mar­ket and believe that by dis­cour­ag­ing stu­dents from begin­ning the process, they can mit­i­gate some of its dele­te­ri­ous effects. I’ll admit that I find this posi­tion rather appeal­ing, in part because it’s self-deprecating, but mostly because in my case it’s actu­ally quite easy to cal­cu­late what’s at stake.

The two core ele­ments of the argu­ment against pur­su­ing a career in acad­e­mia are the poor prospects for employ­ment at the end of the road and the oppor­tu­nity cost along the way. Few pro­fes­sors would deny the exis­tence of the first chal­lenge: we’ve all run the gaunt­let of the job mar­ket, often mul­ti­ple times, and we’ve seen our friends and col­leagues and rivals do the same. Peri­odic reports on the state of the job mar­ket con­firm what we instinc­tively know: there are too many can­di­dates chas­ing too few jobs. And even among those avail­able posi­tions, fewer still con­form to the ideal of a 2–2 tenure-track post.

The point about oppor­tu­nity cost is far more con­tentious because it’s less eas­ily quan­tifi­able (and his­to­ri­ans at least are of course explic­itly trained to avoid indulging in such hypo­thet­i­cals), and pos­si­bly because so many of us are bad at math. Cebula was (rightly) called out by my col­league Zach Schrag for advanc­ing the fig­ure of “a mil­lion dol­lars” as an indi­ca­tion of the income for­gone by stu­dents pur­su­ing grad­u­ate study as opposed to some more remu­ner­a­tive career (and here every­one seems oddly obsessed with Hoot­ers). But the oppor­tu­nity costs can still be enor­mous, as my own expe­ri­ence shows.

For three years between under­grad­u­ate and grad­u­ate study, I worked at IBM, first as a con­trac­tor and then as a direct hire employee. When I left in August 1999 for the Uni­ver­sity of Michi­gan, my salary was $72,000, exclud­ing bonuses, awards, etc. which were fre­quent and not par­tic­u­larly dif­fi­cult to obtain. I then spent the next six years as a grad­u­ate stu­dent, eking out a mod­est exis­tence on teach­ing fel­low­ships and research grants. So what did I give up?

At IBM, ten per­cent raises were the norm for rea­son­ably com­pe­tent indi­vid­u­als, and I had in the pre­vi­ous year received a 13% increase and far larger increases in pre­vi­ous years. To be very con­ser­v­a­tive, let’s assume 10% per year and no pro­mo­tions, bonuses, or awards.

Year IBM Michi­gan
2000 $72,000 $14,000
2001 $79,200 $14,000
2002 $87,120 $13,000
2003 $95,832 $27,000
2004 $105,415 $21,000
2005 $115,957 $6,000
Total $555,524 $95,000

In con­trast, dur­ing my six years of grad­u­ate study, I earned a total of around $95,000 in stipends and research grants, mak­ing for a net oppor­tu­nity cost approach­ing half a mil­lion dol­lars. Not Cebula’s full mil­lion, but still noth­ing to sneeze at. Had I remained in grad­u­ate school longer than six years, the gulf would of course have widened even more rapidly, since I would have traded the com­pound­ing effect of raises for the rapidly dimin­ish­ing hand­outs granted to lin­ger­ing ABDs.

What crit­ics often fail to remem­ber is that oppor­tu­nity cost doesn’t mag­i­cally van­ish even if one is lucky enough to land a good tenure-track posi­tion like I did. Even with the rel­a­tively minor infla­tion we’ve expe­ri­enced over the last decade, 1999’s $72,000 is equiv­a­lent to just over $98,000 today. Not exactly the aver­age assis­tant pro­fes­sor salary in the human­i­ties, is it? Indeed, the oppor­tu­nity cost asso­ci­ated with acad­e­mia effec­tively con­tin­ues to mount indefinitely.

When we tell our stu­dents that earn­ing a PhD is essen­tially a very costly way to pur­chase a lot­tery ticket (albeit with less astro­nom­i­cal odds but also a vastly dimin­ished pay­out), we’re already mak­ing a good case for what might not be there at the end of the road. But we also need to explain to them what absolutely won’t be there: oppor­tu­nity cost’s for­gone earn­ings. Syme oddly claims that Cebula is cyn­i­cal to address these finan­cial con­cerns, continuing:

Has any­one ever been under the illu­sion that work­ing as an aca­d­e­mic in the human­i­ties was a quick way to wealth, home­own­er­ship, and a sta­ble nuclear fam­ily existence?

Well no, prob­a­bly not. But unless I’m mis­taken, what ani­mates Cebula’s argu­ment isn’t regret that work­ing as an aca­d­e­mic in the human­i­ties is not a quick path to these goals — which, aside from “wealth,” are hardly grandiose aspi­ra­tions — it’s the under­stand­ing that work­ing as an aca­d­e­mic in the human­i­ties more often than not rules out these goals entirely.

On Usage Figures

May 21st, 2011 § 9 comments § permalink

Among the more eye-popping num­bers asso­ci­ated with LinkedIn’s recent ini­tial pub­lic offer­ing is the 100,000,000 mem­bers it claims. What do those hun­dred mil­lion peo­ple do with their LinkedIn accounts? If they’re like me, they qui­etly ignore the end­less spam but never quite moti­vate to unsub­scribe. Or maybe they occa­sion­ally click through a link returned by a Google search, only to dis­cover the limp résumé of some sad sack look­ing to escape the Enter­prise rent-a-car counter, not the super cool and attrac­tive “Sean Takats” that they went to high school with and are stalking.

I’ve been think­ing a lot about these kinds of num­bers as the Zotero team pre­pares for a major sum­mit this sum­mer. In our first few years, we used to mea­sure Zotero’s growth in terms of down­loads, but we quit doing so well over a year ago, when that num­ber was north of four mil­lion, hav­ing dou­bled from two mil­lion just a few months ear­lier. We stopped because down­loads are never a very accu­rate mea­sure­ment of adop­tion, and they are espe­cially prob­lem­atic for Zotero, which is avail­able from a vari­ety of repos­i­to­ries. Most users get our soft­ware from either zotero.org or addons.mozilla.org, but Zotero has also popped up else­where, mainly because we don’t restrict its dis­tri­b­u­tion in any way. In the absence of any other met­ric, how­ever, down­loads are bet­ter than noth­ing, and Mende­ley for exam­ple still uses down­loads to arrive at its fig­ure of 900K+ “peo­ple,” accord­ing to Ian Mulvaney’s recent code4lib talk. And when it comes to com­mer­cial prod­ucts like End­Note, we of course have no idea at all.

A sec­ond way to mea­sure usage would be to tally user account reg­is­tra­tions. Cur­rently zotero.org hosts 620,000 accounts. Note that I say “accounts” and not “users.” Indeed there’s no rea­son to think that this fig­ure is any­thing more than very slightly more reli­able than down­loads. Zotero was around for years before we even had server accounts, and we have never aggres­sively pushed users of Zotero to reg­is­ter accounts by con­fronting them with a sign-up form before offer­ing the down­load. We think server accounts pro­vide incred­i­bly valu­able func­tion­al­ity, but we also feel that it’s a lit­tle sleezy to try to co-opt peo­ple into sign­ing up for some­thing they don’t want. So the “real” num­ber could be much higher! Among that mass of accounts, there are hun­dreds of thou­sands of real, active researchers but also, inevitably, count­less spam­mers wait­ing to be weeded and dor­mant accounts sit­ting idle. Or maybe it’s much lower! But even if we were to pre­tend that all 620,000 accounts were tended to by highly moti­vated schol­ars, we would still be faced with an order of mag­ni­tude drop when com­pared to down­loads. A quick look at Mendeley’s peo­ple direc­tory reveals a sim­i­lar dis­crep­ancy: it lists fewer than 70,000 user accounts, which is noth­ing to sneeze at but of course well south of the down­load fig­ure. How many accounts does Ref­Works have? Again, we can’t know.

A final way would be to count how many peo­ple are run­ning Zotero each day. Because Zotero auto­mat­i­cally checks for updated trans­la­tor code on a daily basis, we know that at least 275,000 instances of Zotero ran today. But wait a minute, what’s with this “instances” and “at least” busi­ness? Well, maybe some peo­ple are run­ning more than one copy of Zotero on a sin­gle machine. We could account for unique IP addresses, which moves the num­ber down slightly, but then we would ignore mul­ti­ple instances of Zotero shar­ing a sin­gle pub­lic IP address. And of course, this fig­ure only accounts for copies of Zotero that have auto­matic updates active, and that man­aged to con­nect to the inter­net. Other soft­ware ven­dors could pre­sum­ably track sync activ­ity or other met­rics to arrive at anal­o­gous figures.

The basic moral of the story, if you haven’t already guessed, is that these num­bers are all pure shit, though some are clearly worse than oth­ers. All we can do is pro­vide an hon­est expla­na­tion of how they’re derived.

Zotero Versus

May 6th, 2011 § 11 comments § permalink

Brian Crox­all recently lit up the com­ment feed at the Chron­i­cle with his ProfHacker com­par­i­son of “Zotero vs. End­note,” where the debate cen­tered mostly around issues of cita­tion fidelity. As Fred Gibbs notes, how­ever, “while cita­tion for­mat­ting is one major rea­son to use bib­li­o­graphic soft­ware, it isn’t nec­es­sar­ily the only or even pri­mary rea­son, espe­cially in the human­i­ties.” Zotero’s cita­tion func­tion­al­ity was always imag­ined merely as bait: by pro­vid­ing this labor-saving func­tion­al­ity, Zotero would encour­age each user to move her research into what amounted to a fully search­able and share­able rela­tional data­base that could be sub­jected to text min­ing and other analy­sis. Here researchers could begin to do truly remark­able and new things with their evidence.

A few com­menters, as well as Fred, tried to shift the dis­cus­sion toward the issue of cost and open­ness, and in par­tic­u­lar to Zotero’s sta­tus as free/libre open source soft­ware (FLOSS). Many of Zotero’s most ded­i­cated users have cham­pi­oned the soft­ware in the name of FLOSS, but this line of argu­ment fre­quently falls on deaf ears, or even ears that are con­di­tioned to reject FLOSS as some­how anti-market or anti-capitalist. From my per­spec­tive, FLOSS in and of itself is a fairly unper­sua­sive argu­ment for using Zotero, akin to knee-jerk calls to “Buy Amer­i­can!” in the 1980s, when the USA still did some man­u­fac­tur­ing. Buy­ing Amer­i­can and using FLOSS might make one feel some sense of moral supe­ri­or­ity, but at the end of the day can those feel­ings still paper over our sense of exis­ten­tial dread when faced with dri­ving to work in our crum­bling K-cars or cob­bling together a dis­ser­ta­tion with shitty research software?

Just as Zotero’s cita­tion man­age­ment func­tion­al­ity is a means to an end, so is licens­ing and devel­op­ing the soft­ware as FLOSS. Far from just ide­ol­ogy, FLOSS has allowed Zotero to lever­age rel­a­tively lim­ited finan­cial resources to out­per­form vastly larger and bet­ter funded com­peti­tors, old and new. Zotero’s annual oper­at­ing over­head is only in the low six fig­ures. This amount cov­ers in-house devel­op­ment, out­reach, and infra­struc­ture costs. In com­par­i­son, End­Note and Mende­ley each have oper­at­ing costs that are an order of mag­ni­tude greater (or even more). And of course there’s an even higher, hid­den cost for these plat­forms: the expec­ta­tion of sub­stan­tial profit, which nec­es­sar­ily impinges on sustainability.

Why should any researcher care about these issues? Defend­ers of Zotero have often voiced con­cerns about “lock-in” with pro­pri­etary, for-profit soft­ware. Users might find them­selves unable to migrate their data out of one of these com­mer­cial solu­tions at some later date. But even if this worry were valid — and I don’t know that it is — lock-in in and of itself isn’t nec­es­sar­ily a bad thing. Who would com­plain about being locked-in to the very best solu­tion, par­tic­u­larly if that solu­tion also didn’t cost any money?

Unfor­tu­nately, the closed, for-profit soft­ware option has never been the very best solu­tion, and there’s no sign that that sit­u­a­tion is chang­ing. This isn’t ide­ol­ogy speak­ing; it’s his­tory. End­Note has been derided for well over a decade for its 1990s inter­face and preda­tory “upgrade” cycles. New fea­tures come late or never, and the soft­ware has yet to embrace online research and col­lab­o­ra­tion. Mende­ley, while far newer and the­o­ret­i­cally nim­bler, has like­wise only slowly moved to pro­vide the basic, core func­tion­al­ity that active, pub­lish­ing researchers require. It’s entirely likely that “fea­tures” like jour­nal abbre­vi­a­tions, cita­tion page num­bers, and sub­col­lec­tions will even­tu­ally make their way into Mende­ley, or that End­Note will one day dis­cover the inter­net, but the mere fact that these things haven’t yet tran­spired speaks vol­umes about the pri­or­i­ties of their par­ent corporations.

Because it’s FLOSS, Zotero has been able to add and refine fea­tures thanks to the con­tri­bu­tions of hun­dreds of vol­un­teer devel­op­ers and the feed­back of hun­dreds of thou­sands of users. The tech­no­log­i­cal suc­cess of this model is unde­ni­able: Zotero’s open-source cita­tion engine, entirely rewrit­ten by Zotero user Frank Ben­nett, and the thou­sands of user-contributed style files the engine uses have already been adopted by Mende­ley and Papers, and a rep­re­sen­ta­tive from XXXXX has expressed inter­est in doing the same.1 (Update: The indi­vid­ual who wrote regard­ing XXXXX and CSL clar­i­fies that the com­mu­ni­ca­tion was made in a per­sonal capac­ity, not as a rep­re­sen­ta­tive, and so I’ve removed the software’s name.)

Wiki­me­dia Com­mons Credit: Stan Zurek

And of course, there is no rea­son to think that any of these par­ties is act­ing in the inter­est of serv­ing ide­o­log­i­cal inter­ests. Indeed, if we look at how they pub­licly address FLOSS, we find ambiva­lence and dis­dain. Mende­ley only admit­ted its use of Zotero code when con­fronted, and avoided any men­tion of the prove­nance of its cita­tion styles for years. Frank’s cita­tion pro­cess­ing engine, despite sav­ing count­less hours of devel­op­ment and sup­port, earns faint praise. Papers like­wise ini­tially only con­fessed its planned use of the cita­tion styles when probed on Twitter.

Lib­er­at­ing researchers from the con­straints of com­mer­cial soft­ware devel­op­ment has been good for research, not for ide­o­log­i­cal rea­sons but for tech­ni­cal ones. It has also been extremely good for com­mer­cial com­peti­tors, who rec­og­nize the value in openly devel­oped soft­ware. What’s not at all clear is that attempt­ing to put the genie back into this par­tic­u­lar pro­pri­etary soft­ware bot­tle will sus­tain any of the remark­able momen­tum gained over the past few years, or whether inno­va­tion will con­tinue to be stunted or sti­fled in pur­suit of illu­sory finan­cial gain.

Wiki­me­dia Com­mons Credit: Finn Rindahl

As com­menters on Brian’s post noted, there is a real cost asso­ci­ated with mov­ing between research soft­ware, and it’s inevitably in the inter­est of for-profit enti­ties to keep those costs as high as pos­si­ble. Right now the mar­ket won’t bear very high costs, but that’s largely thanks to Zotero, not because it’s free but because it’s FLOSS. End­Note, Mende­ley, and the rest sim­ply aren’t equiv­a­lent play­ers, because the mar­ket that they’re squab­bling over is checked in growth and in all like­li­hood doomed to decline so long as there is a strong FLOSS competitor.

  1. To my knowl­edge, not a sin­gle pub­li­ca­tion style file has ever been con­tributed by a non-Zotero user. [↩]

Time Shifting and Historical Research

April 20th, 2011 § 20 comments § permalink

About ten thou­sand years ago, we were intro­duced to the phrase “time shift­ing” by a decade-long law­suit over the right to use VCRs to tape TV shows for later view­ing. Today’s DVR has of course made this process far eas­ier and prob­a­bly more wide­spread, but the idea remains the same: rather than watch some­thing right now, with no snack breaks, we instead put it off until some later time. Other than the occa­sion­ally self-serving gripe about hav­ing “a lot of TiVo to catch up on,” time shift­ing is a set­tled and dead issue, a non-story. Or it would be, if it were not for the trou­bling case of his­tor­i­cal research.

In a recent post I fret­ted about how shift­ing research prac­tices might affect the sig­nif­i­cance and allure of his­tor­i­cal fields. Here I want to exam­ine those shift­ing prac­tices in a bit more detail. The ben­e­fits of com­press­ing a research agenda or of greatly expand­ing the amount of mate­ri­als that can be gath­ered (or both) has encour­aged a whole­sale trans­for­ma­tion in the way that researchers now use archives. The point of vis­it­ing the archives hasn’t changed — peo­ple still go there to gather evi­dence — but before the wide­spread use of dig­i­tal pho­tog­ra­phy the col­lec­tion of evi­dence was lim­ited by what could be read, and then sum­ma­rized in notes or tran­scribed. All of this activ­ity nec­es­sar­ily had to occur on-site, dur­ing the lim­ited hours and days of oper­a­tion, fur­ther con­strained of course by strikes, hol­i­days, and hangovers.

With dig­i­tal pho­tog­ra­phy, a far greater num­ber of doc­u­ments can now be processed in a much shorter period. This isn’t really news to any­one who has vis­ited an archive in the last five years. And here, Robert Darnton’s recent defense of the ana­log rings espe­cially hol­low.1 In dis­pelling “5 Myths About the ‘Infor­ma­tion Age’”, Dar­ton claims,

“All infor­ma­tion is now avail­able online.” The absur­dity of this claim is obvi­ous to any­one who has ever done research in archives. Only a tiny frac­tion of archival mate­r­ial has ever been read, much less digitized.

This is cer­tainly an accu­rate state­ment, so long as we only look back­ward. What Dar­ton is ignor­ing of course is that essen­tially all archival mate­r­ial con­sulted today is being dig­i­tized, whether in the form of tran­scrip­tion or pho­tog­ra­phy. What’s miss­ing is the abil­ity to access and mine these innu­mer­able rich indi­vid­ual silos of data. Zotero is one step toward real­iz­ing this vast meta-archive, but how­ever out­ra­geously ambi­tious such a project might seem, it is triv­ial when com­pared to the mas­sive amount of labor that has already been deployed to dig­i­tize at the indi­vid­ual, cottage-industry level.

What’s espe­cially inter­est­ing, I think, is how this new prac­tice might qual­i­ta­tively affect research. In par­tic­u­lar, I won­der how the cre­ation and pop­u­la­tion of indi­vid­ual research queues, time-shifted for later con­sul­ta­tion, will influ­ence how schol­ars approach the gath­er­ing and analy­sis of evi­dence. Take, for exam­ple, the remark­able trans­for­ma­tion in the area of pre-1923 printed mate­ri­als. When­ever I encounter any ref­er­ence to any printed source, the first thing that I now do is to con­sult Google Books or Gal­lica to see whether there is a dig­i­tized edi­tion avail­able. If I find a dig­i­tal ver­sion — and most times I do — I add a copy to my Zotero library to be read later (and if it’s small for­mat, octavo or duodec­imo, usu­ally on my Kindle).

This work­flow has dra­mat­i­cally reduced the amount of time that I spend on-site in research libraries like the Bib­lio­thèque nationale de France, which is increas­ingly becom­ing just another nice quiet place to do work, a pur­pose much like that served by my neigh­bor­hood branch library when I was in grade school, only with RFID cards and a smok­ing lounge. But it has also hugely increased the amount of time that I now spend read­ing and “doing research” at home, at night, and on week­ends. More­over, it’s incred­i­bly easy to amass a mas­sive queue of dig­i­tized doc­u­ments and feel like one has “per­formed research” even though a good per­cent­age (most?) of those mate­ri­als might prove use­less. So in a sense, we’re not just talk­ing about time shift­ing an amount of research equiv­a­lent to say, 1998 lev­els, but rather that we’re simul­ta­ne­ously esca­lat­ing the evi­den­tiary basis for any research project.

Mike O’Malley and I have writ­ten about the chang­ing land­scape of his­tor­i­cal research in the face of abun­dant evi­dence.2 We agree that find­ing, as part of the research process, will inevitably decline as a val­ued skill as asso­ci­ated costs con­tinue to fall. In con­trast, syn­the­sis and con­tex­tu­al­iza­tion, always valu­able, will become even more impor­tant dif­fer­en­ti­at­ing qual­i­ties. Yet I won­der whether time shift­ing, and the risks it nec­es­sar­ily intro­duces, won’t so over­bur­den researchers that they fail to advance to the stage of the research cycle where they can begin to per­form mean­ing­ful analy­sis. How is time shift­ing affect­ing your research?

  1. Robert Darn­ton, “5 Myths About the ‘Infor­ma­tion Age’,” The Chron­i­cle of Higher Edu­ca­tion, April 17, 2011, sec. The Chron­i­cle Review, http://chronicle.com/article/5-Myths-About-the-Information/127105/. [↩]
  2. Michael O’Malley, “Evi­dence and Scarcity,” The Aporetic, Octo­ber 2, 2010, http://theaporetic.com/?p=176 and Sean Takats, “Evi­dence and Abun­dance,” The Quin­tes­sence of Ham, Octo­ber 18, 2010, http://quintessenceofham.org/2010/10/18/evidence-and-abundance/. [↩]

The End of (French) History

April 8th, 2011 § 3 comments § permalink

As a his­to­rian I like to think that I’m com­fort­able with the idea of fields dying. Maybe they’re reborn and live to fight another day (like diplo­matic his­tory, or so we all keep hear­ing), and maybe they never really die at all (like quan­ti­ta­tive his­tory, safely entrenched at Paris 1-Sorbonne).

But what about the old work­horse geo­graphic fields like, say, French his­tory?1 Unless you are Natalie Davis, being a French his­to­rian over the last five decades or more prob­a­bly meant that you, well, went to France to do your research. And why did you go? You told your depart­ment chair and your grant fun­ders and your col­leagues that you had to get to the archives and the libraries, of course. But let’s be hon­est here: French his­to­ri­ans don’t go to France to get to the doc­u­ments. Instead we wed our­selves to doc­u­ments that just so hap­pen to be found only in France, “forc­ing” us to go there, usu­ally when the weather there is super great, or at least super crappy at home. That’s why we became French his­to­ri­ans in the first place. There are a few excep­tions to this rule: the hand­ful of self-loathing sad sacks we all know who hate wine and cheese and cig­a­rettes, but this is no time to be cruel. » Read the rest of this entry «

  1. I’m focus­ing on France, of course, because it’s what I know. That said, I don’t see any com­pelling rea­son why the argu­ment that fol­lows couldn’t apply to any num­ber of national or transna­tional areas of inquiry. [↩]

FATCamp Florence

March 27th, 2011 § 3 comments § permalink

This past week’s THAT­Camp Firenze was a huge suc­cess, offer­ing plenty of oppor­tu­ni­ties to learn about new projects and the var­i­ous national and inter­na­tional fla­vors of dig­i­tal human­i­ties that are flour­ish­ing in Europe. But seri­ously, on a blog with this title, how can I ignore the most spec­tac­u­lar part of the trip?

So much pork fat. Note my plate's single sad leaf of radicchio (it was actually delicious).
Serge shows Sharon, Amanda, and me how it's done.

Zotero and AWS, or How We Learned to Stop Worrying and Love the Cloud (Part 1 of 2)

March 4th, 2011 § 23 comments § permalink

Zotero’s server infra­struc­ture has evolved in count­less ways since the project’s 2006 launch, but most of those changes are super bor­ing and not worth remem­ber­ing. Over the past two months, how­ever, we moved the bulk of Zotero’s back end to Ama­zon Web Ser­vices, a step that I believe is uniquely note­wor­thy in the con­text of dig­i­tal human­i­ties projects and their long-term sus­tain­abil­ity. In this post I describe the recent changes to Zotero’s archi­tec­ture. In the next post I’ll dis­cuss why these changes are impor­tant for the dig­i­tal human­i­ties. This story is long, but it has a moral, and also a van. » Read the rest of this entry «

Zotero Storage Goes Global

February 15th, 2011 § 5 comments § permalink

Because I live under a rock (Viet­nam), I only recently dis­cov­ered the Google Charts API. When I saw that it sup­ported maps, I thought it might be fun to plot the sales data for Zotero File Stor­age pro­vided by the non­profit cor­po­ra­tion I started along with a bunch of other aca­d­e­mics. Bear in mind that these maps only reflect the billing addresses asso­ci­ated with pur­chasers of Zotero stor­age. Zotero’s gen­eral user base is even more glob­ally dis­trib­uted and sev­eral orders of mag­ni­tude larger than the sub­set depicted here. Nonethe­less the results are stun­ning, I think, and some­thing that pleas­antly reminds me of the last throes of a game of Risk. We have work to do in Africa and the Mid­dle East. Click the thumb­nails below for full-size, detailed images.

Test Your Digital Humanities Knowledge

November 16th, 2010 § 16 comments § permalink

I met up with my French col­league Marin Dacos today while he was in the mid­dle of giv­ing an exam on dig­i­tal pub­lish­ing to his human­i­ties master’s stu­dents. While many U.S. grad­u­ate stu­dents (and pro­fes­sors) would balk at such a “fac­tual” exam, I sus­pect that they would have a very tough time get­ting through it unscathed.1

How would you or your stu­dents fare with a test like this? I’ve trans­lated the ques­tions into Eng­lish. You have ninety min­utes. Go!

Basic ques­tions and def­i­n­i­tions (13 points)

  1. In Wikipedia, what do “Diff” and “Edit war” mean?
  2. Name the XML for­mats use­ful for elec­tronic pub­lish­ing and spec­ify their particularities.
  3. How does PageR­ank work? What are the advan­tages of this sys­tem? What are its drawbacks?
  4. Who is Tim Berners-Lee? What is the W3C?
  5. What is meta­data? What is Dublin Core?
  6. What is DRM? What are its advan­tages and disadvantages?
  7. What is sin­gle source publishing?
  8. What is the dif­fer­ence between the PDF and EPUB formats?
  9. What is Zotero? What does it do?
  10. What is a DOI? What is name resolution?
  11. What is inter­op­er­abil­ity? What is OAI-PMH? What are the main verbs of OAI-PMH and what do they do?
  12. What is the atten­tion economy?
  13. What is the Cre­ative Com­mons License? What is its purpose?

Syn­the­sis (7 points)

  1. Elec­tronic pub­lish­ing falls into three cat­e­gories. For each type of elec­tronic pub­lish­ing, pro­vide a def­i­n­i­tion, at least one rep­re­sen­ta­tive exam­ple, its main tech­ni­cal char­ac­ter­is­tics, its prin­ci­pal qual­i­ties, and its major faults.
  2. The pub­lish­ing indus­try is search­ing for an eco­nomic model of elec­tronic pub­li­ca­tion. Present the dif­fer­ent strate­gies cur­rently under devel­op­ment (name, basic descrip­tion, exam­ple, advan­tages, disadvantages)
  1. Except for #9. Every­one can answer that one. [↩]

Zotero and Open Access

November 12th, 2010 § 4 comments § permalink

Before the excite­ment sur­round­ing last month’s Open Access Week fades com­pletely — too late — I thought it might be appro­pri­ate to describe how and where Zotero inter­sects with OA. When peo­ple talk about Open Access, they typ­i­cally mean free access to pub­lished, usu­ally schol­arly, con­tent. It’s a con­cept that’s ide­o­log­i­cally easy for most researchers to get behind because few of us reap any direct finan­cial ben­e­fit from the major­ity of our pub­li­ca­tions, and we’re all very famil­iar with the annoy­ing fric­tions intro­duced by gat­ing access to con­tent. Cham­pi­oning Open Access is kind of like advo­cat­ing not club­bing baby seals: you’re unlikely to encounter much oppo­si­tion. » Read the rest of this entry «

« Older Entries