Categories
flickr ixd web

IxD / Dear everyone, I hope you can find my albums

Flickr navigation hack

What we have here is both a fail­ure to com­mu­ni­cate and an inge­nious workaround. To Kris­ten & Rob: Kudos.

Categories
ixd lit

National nightmares / Restoring a modicum of utility to the Complete New Yorker

I was one of the suck­ers who pre-ordered The Com­plete New York­er mag­a­zine. I am a long-long-time New York­er read­er, and the entice­ment was just too pow­er­ful — 8 DVDs filled with 60+ years of cul­tur­al com­men­tary, quirky car­toons and cool cov­er art, all in a dis­tinct high­brow-yet-prac­ti­cal-mind­ed voice and scanned in at super-high-res? For a few extreme dorks, this was intense­ly excit­ing. Expec­ta­tion-wise, it was like the release of a smar­ty­pants Playsta­tion 3.Upon arrival, it also resem­bled Playsta­tion 3, in that it sucked, big-time. My expe­ri­ence improved slight­ly after The Occa­sion­al Scriven­er post­ed a hack that allows you to copy issues from the 8 inde­pen­dent DVDs onto your hard dri­ve. An extreme dork after my own heart. Many thanks.The real­ly big, un-hack­able prob­lem: The search tool is a house of hor­rors. Imag­ine that you've final­ly been intro­duced to a long-time idol, let's say Bob Dylan, and he agrees to come home with you and sit in your liv­ing room and tell you any­thing you want to know. But then when you ask him to tell you the com­plete sto­ry of the "Judas!" show, you real­ize that he doesn't speak Eng­lish; he just sits there silent­ly, impas­sive. That's how this thing makes me feel. The whole point of get­ting Com­plete New York­er is to have your mind blown by the wealth of cool stuff in the old­er issues. There­fore, the chal­lenge faced by the inter­ac­tion design­ers is to facil­i­tate get­ting at that stuff, i.e. MAKE IT EASY TO SEARCH for what you want. The shot below rep­re­sents the Pro­crustean bed on which each searcher must lie.

The real­ly egre­gious crimes have been doc­u­ment­ed else­where, but I would just like to add: 

  • Per­for­mance that reminds me of the 90's. If this had been released in 1998, I could eas­i­ly for­give the lag every­time a but­ton is pressed or a search is exe­cut­ed. But real­ly, when I type "white" into the gen­er­al search field, and it churns for near­ly 20 sec­onds, I don't know, it makes me homi­ci­dal­ly mad. Anger at slow per­for­mance is like road rage — once you've got it, you can't get rid of it, no mat­ter how much you avoid being in a car.
  • Why the cru­el and unusu­al search com­plex­i­ty? Search­ing is nev­er made eas­i­er by sur­fac­ing every pos­si­ble method of doing so right off the bat. Google — the world's most pop­u­lar search inter­face — seems like an effec­tive guide here. Start sim­ple, and reveal sophis­ti­ca­tion when nec­es­sary. There aren't real­ly even that many ways I could con­ceive of search­ing the Com­plete New York­er — author, date arti­cle title, date range … That's about it.
  • Wast­ed ver­ti­cal real estate. Near­ly 33% of the ver­ti­cal space is con­sumed by tool chrome, those thick gray bars seg­ment­ing the screen. Com­bined with the often biz­zare and most­ly use­less "Abstract" below, this leaves 11 rows for search results, the place where users (I) make deci­sions on what to launch in the view­er. Unforgiveable.
  • What the heck is this thing called?. The fact that the search results do not con­tain a high­ly valu­able piece of infor­ma­tion — umm, the title of the piece — makes it a pain in the butt to scan (for instance) the sto­ries of JD Salinger, the assort­ed work of EB White. Actu­al­ly, pret­ty much every search is com­pli­cat­ed by this.

I could go on and on, but I won't. Here's my sug­ges­tion for CNY 2.0: Con­sol­i­date the exist­ing wid­gets into one wid­get with mod­est dynam­ic behav­iors. The wid­get would have one sim­ple ini­tial menu that deter­mines how you want to search — key­word, author, issue, depart­ment. This selec­tion then deter­mines the fil­ters you'll need — if you choose "key­word," maybe you get "depart­ment" and "date" as fil­ters. In doing this, you buy back all of that chrome real estate, allow­ing more results to be dis­played. Win, win, win. Of course none of this mat­ters much if data­base per­for­mance isn't improved, but here it is anyway:A modest proposal