Happy Emergency

when you have to do something wonderful.

[happyemergency]

1 Flight on Frontier Airlines: 11 Breaches of Trust, 13 Delays totaling 49 hours

Summary

Frontier Airlines broke trust with the travelers in my flight 11 times, each completely avoidable and the product only of poor management by the company. The four people in my party, including a 6 year-old girl and a 87 year-old man, and the hundred-plus other travelers had our flight delayed 13 times totally 49 hours. As a User Experience Designer and elected official, I understand the importance of maintaining the trust of your users and constituents. Any company that is at all concerned with its long term viability will consider preventing breaching the trust of its clients as the very lowest bar for delivering its products. Frontier Airlines seemed oblivious to the existence of that bar.

Though I think it's entirely reasonable to, I don't blame Frontier for the original hardware failure. These things happen and I'm glad it happened when the plane was on the ground. What follows is a brief account of the repeated breaches of trust and failure to take responsibility that lead to the worst customer service I've experienced and stranded us for nearly a full 49 hours.

Below is the play-by-play of the events.

Timeline of Events

This is regarding Frontier flights 1021 (Orlando-> Madison) originally scheduled to depart at 3:05pm on 1/3/2015.

Day 1 - 1/3 Saturday - 2:20pm We boarded the plane set to depart at 3:05pm. The plane seemed unusually hot and no air was coming from the vents. While we were boarding desk-staff was talking with the pilots but nothing was announced.

3:10pm After being settled and sweating in our seats, the captain told us there was a hardware problem and we had to get off the plane while they tried to fix it. Though I would have preferred to have been told before boarding as the issue was clearly known earlier, I understand that that may have added undue complications if the problem was quickly fixed.

4:46pm After 90 minutes we were finally given an update. I don't consider this a breach of trust, but it's absolutely awful customer service. The update was that the air conditioning (which is actually critical for the plane to function) was broken. No estimate about how long it would take to fix.

5:30pm We're told the plane is fixed and board again. The plane still feels rather warm.

5:50pm The captain tells us the air conditioning is broken again, and we need to deplane. Without explanation, we're told that Zones 1 and 2 should go to the baggage claim, but that Zone 3 should stick around (later clarified as at the gate).

7:45pm After almost 2 hours with no new information, the airline issued food vouchers "if you think you are entitled to one." Not a breach of trust, but more egregious customer service.

8:20pm Still no new announced information, but the flight (now over 5 hours late) showed Cancelled online. When questioned by a woman staff said that the flight was canceled but they wouldn't announce it to the gate until after a 8:45pm flight at the same gate departed. Breach of trust #1.

9pm Staff finally announced that the flight was canceled, an hour after it was updated online. The remaining were told to get our luggage from the baggage claim then go to the Frontier ticket desk to get vouchers for a hotel room for the night. No information about when another flight would be.

9pm Hotel voucher "Here you are. Next!" Wait, what about flight information. "7am, come back here at 4am"

10:30pm Got to hotel (by taxi), the small shuttle they provided would have taken over an hour to get the entire flight there.

Day 2 - Sunday 1/4 - 6:35am After the plane should have been loading for it's 7:05am departure, they announced it was delayed for an inspector to check the new part, but didn't know how long it would take. The plane was dark and there was on flight staff around. No bags had been loaded and there was no activity around the plane. The crew had clearly been alerted in advance, possibly even being able to sleep in. That passengers weren't alerted earlier when it was clearly known was a breach of trust (#5) and terrible service. Further, the internet listed the flight as delayed 3 hours to 10am. This wasn't announced to passengers, who were told it would be minutes, not hours. This was another breach of trust (#6).

8:00am After it had been announced online for 1.5 hours, we were finally told the flight would be delayed until 10am. The broken part was installed and they needed to first test it.

8:08am We were told it would be delayed until 12:00am. And that the part actually hadn't been installed yet but they thought they could install it in two hours, then have at least an hour to test it. When asked about meal vouchers for lunch, we were told they would issue them at 10am.

Around 10:30am The online flight status tool now lists the flight as delayed until 5pm. No update given for the people living at the gate. Breach of trust #7.

11:45am After being told lunch meal vouchers would be issued at 10am, they are still not forthcoming. Breach of trust #8.

12:20pm After 4 hours without an update (4 hours without an update, particularly when the internet has more info, is breach of trust #9) or even a staff member at the desk, an angry mob of fellow stranded travelers attracts so much attention police are called. Four police talk with the Frontier customers, sympathize, then go off to bring back a Frontier employee to talk with us.

12:30pm The flight is canceled. Rescheduled for 7:30am the next day. We begin to queue up for hotel vouchers.

Day 3 - Monday 1/5 - 5am The 7:30am flight is rescheduled to 9am. Those of us who hadn't left for the airport got an extra 45min of sleep.

6:30am 9am flight is rescheduled to 10:30am. We had just woken up our 6 year-old to rush to the airport.

~10am The flight is rescheduled to 1:30pm. The mechanics talking to the captain and gate staff said it would likely be at least 3pm, despite what "corporate" says. It has become abundantly clear and openly talked about with some Frontier staff that the company is playing games with the office time delays to intentionally make the flights look more likely to leave soon, with the presumed intention of preventing people from switching to flights with another carrier. When you consider all of the other flights in the chain that relied on this plane, there were a lot of travelers that would constitute a lot of lost revenue if they left. Breach of trust #10.

~1:15pm Everyone knew the 1:30 was unrealistic, Frontier officially changes the time to 2pm, which is still unrealistic.

2:10pm 2pm passes and the flight time still says departing at 2pm. Ironically the board says the flight departed, the people in the gate cheered "We finally left!" The flight is eventually delayed to 3:30pm.

2:30pm We get word that the tests are going well and will be done in about an hour.

3:20pm The plane passes the tests and we're told to board in 10 minutes! After the flights was delayed 12 times, I think giving people only 10 minutes to get to the plane was unreasonable, breach of trust #11.

3:30pm We hurriedly board. The mechanic is still in the plane updating paperwork and the logbook. The flight is rescheduled one last time (the 13th) to 3:40. We speculate that this adds to some metric they have that says they depart when they say they will.

3:40pm The plane takes off. Once we get some altitude and get set on our straight-shot course we accelerate two more times, as though the pilot is incrementally moving us to max cruising speed to test the plane. When I travel I pay attention to how my plane flies (and my dad is a pilot) so I feel pretty confident saying this was irregular behavior.

~6pm Central We land and are jubilent to be back in Madison. The sister group of travelers who were stranded in Madison cheer for us, and thank us for testing the plane for them.

[happyemergency]

UX Principles: Naming Features is Like Golf, Not Pool

Most UX, IxD, and Product Designers find themselves occasionally designing a novel interaction. In most cases it’s relatively easy to find an analogous interaction in the real world, or try your bag of tricks for explaining it at a different level. But sometimes you get stumped. When that happens I believe it’s best to think of naming features more like hitting a golf ball than a cue ball.

Names are vehicles for meaning that depend on shared cultural references. Even the most basic of words can have shades of meaning that change depending on what subcultures you have been a part of.

On top of that, cognitive load needs to be taken into account, particularly with analogies and metaphors; that one small additional step can be enough to cause people to halt, or not make the connection at all.

When operating within those limitations, it’s better to use a name that gets 70% of the meaning across to 100% of users, than 100% of the meaning to 70% of users. Yes, those 70% of users that fully understand the meaning are more likely to enjoy your product, but the 30% that didn’t understand will lose flow, connection, or trust in your product.

For example, let’s take the Lasso tool in Photoshop. I assume most people reading this are familiar with it, but for those who aren’t it is a selection tool that uses freehand clicking to define the selection boundary. You draw around what you want to select, you lasso it. It’s catchy (;p), it conveys the meaning well with the metaphor, and if you’re making a mass market app it might be a bad idea.

Unknown tools become blind spots in a user's vision, we tune out what we don’t understand. This is as true for inscrutable icons as it is for inscrutable feature names.

I believe it was a good move for Adobe. They had a very specific market and the digital vocabulary wasn’t as well defined. Today such a tool might have been called Poly Select, Freehand Select, Draw Select, or in a touch-centric world maybe just Select.

All of those names do a decent job getting the concept across, but they fully convey that this is some form of select tool. It might not get all of the meaning across to all users, but it gets enough across that the users can mentally categorize it and file it away in the “some kinda select” category, instead of the “no clue” bin.

Which brings me back to golf and pool. In golf, you generally want to shoot for incremental steps, “what’s the best I can do *in this situation*”. You won’t get a hole-in-one, but if you can get on the green well under par, you’re doing okay. With pool every shot needs to be a win, there is no second chance so there is no opportunity to set up a next shot.

Get the ball on the green for your users, if you can get them close enough to the meaning, they’ll come back to putt it in on their own.

[turntable-fm]

Turntable Live

For the last few months our team has been focused on building and launching Turntable Live, an interactive way to hold concerts online. We’ve done 11 events and so far everyone has really loved the experience. To see what it looks like, check out this video or learn more at

[happyemergency]

Really sad to see Turntable shuttering their DJ service. I had written a critique of it last year with some suggestions. I really hope the live music direction they're taking works out. It's inventive and hopefully not ahead of it's time.

[happyemergency]

I enjoy reading your comments. You seem very informed. One question. would you ever try a Political Networking site again, since now you know how "not to do it," Seems like a no brainer

Definitely will do another political social network. At the very least the side project I'm working on: It's changed a bit, but the bones of it are here: http://www.youtube.com/watch?v=4ZjNtPbf8FY

[happyemergency]

Contemporary business model thesis: Money is never far from value.

[happyemergency]

Google Advice: The Trillion Dollar Marketplace for Information

(clickable walkthrough pdf here)

Wouldn’t it be great if talking face to face with an expert was as easy as a Google search? What do you want to know?

Everyone needs on-demand expert advice. A major purchase, tax question, a plumbing problem, medical concern, proofreading. But there are barriers to finding experts (it’s time consuming to find the right one) and getting what you need isn’t as easy as it should be (they don’t bill in seconds!). And search engines and Q&A sites? The last time you used one for expert advice you probably left feeling unsatisfied, or it took you half a day to find something you felt confident about.

Google Advice connects you with experts on demand. Have your everyday questions answered by experts in seconds or minutes, and for peanuts.

Google Advice enables people with full time jobs to make money sharing their expertise and hobbies. And freelancers to find work helping those most in need of their services. It’s the Amazon.com of the knowledge economy.

Who Should Build It?

Google: Google’s mission is to organize the world’s information and make it universally accessible and useful. The world’s information is still mostly in people’s heads and Google is continually finding ways to get it out. One Google product with a similar purpose was Knol, a sort of expert-based wikipedia that condoned expert opinion. But Knol treated information as a product, where Google Advice treats information as a service. It’s the difference between a book and a teacher. The industry as a whole has been learning that merely presenting information isn’t sufficient; it must be taught. This is why demand for UX and design has boomed, presentation and process matter. Another great reasons el goog should build this is it ties in beautifully with: Hangouts, Wallet, G+, YouTube (for recorded sessions), search (not finding what you want? How about instant advice), I could go on. And of course (opt-in) capturing information from the sessions themselves would be a gold mine.

Microsoft: Skype and the business community. Plus, like the other contenders Microsoft has the clout to get early mainstream adoption that network-size-dependant products like this need.

LinkedIn: Another natural fit, LinkedIn has been quietly expanding the breadth and depth of professional expertise indexing. If they don’t build something like this, whoever does will use LinkedIn data to the fullest extent possible.

Why I Made the Wireframes I designed this years ago while working on my first startup (I needed lots of help) and never finished all of the wireframes. It’s one of those products that will inevitably exist, it’s just a matter of time, form, and execution. Though I like tackling important and hard projects, I wouldn’t want to take this on as a startup. I just want it built and will gladly help anyone that wants to tackle it.

This product will change peoples lives by giving them an instant advisor when they need an answer, a teacher when they want learn, and a student when they want to share.

(edit 07/25/13: Turns out Google had the same idea! http://techcrunch.com/2013/07/24/meet-helpouts-googles-secret-plan-to-bring-live-video-commerce-to-local-businesses/

How cool is that!)

[happyemergency]

Pixel-Perfect UI: How to Design Upvote Arrows

Here are the behavioral and visual factors I used to inform my design of up/downvote arrows for MyMaryland.net. This is pretty extreme UI nerdiness so if you’re not in our tribe here are some kittens. Start with UX What are the users’ goals as they relate to your functionality? What actions does the user want to accomplish, how should the features allow them, what moods should they engender? What actions were users taking prior to voting, and what will they do after? Short answer: In our case we wanted users to be able to voice general support or opposition to a post or comment. Pretty standard. However we also wanted users to feel as though they needed to vote on every post. We wanted unvoted posts to feel unresolved, and post-vote to feel like an accomplishment akin to crossing an item off a to-do list. Use Cases

Users can vote up or down

Users can change their vote

(intentional omission: we do not want to allow users to “unvote”, although that is often desirable)

What is an “Upvote”? What an upvote is isn’t as clear-cut on MyMaryland.net as it is on most sites with content voting. Generally, an upvote is seen as a recommendation for that content. On MyMaryland.net, however, the noteworthiness of the content is different than the primary reason most users will vote: to express their support or opposition to an opinion. Initially, we thought that the vote arrows on the list view page would express whether or not the content was recommended and that the content specific voting--whether or not the user supported or opposed it--would be done on the separate content page. I thought the distinction between vote-types would be difficult for most users to understand. For example, I thought it would be difficult for a user that has a strong opposition to an issue to both vote against it AND recommend that content, even though the expression of their opinion is a clear indication that they found the content valuable. Instead we went with one vote per post/comment. Arrows versus Thumbs The question of whether to use arrows or thumbs wasn’t a clear one, and the choice is likely one of the biggest decisions you’ll make when designing vote arrows because it’s the biggest variable in defining the act of voting for your users. Arrows can be cleaner visually but there can be some amount of ambiguity for users. Thumbs-up and down are a clear indication of favor, at least in the western world, but there is also a degree of casualness to them that we felt would detract from the image of serious deliberation. You give a thumbs up to a movie, not a law. Precedent Study As a frame, let’s look at 3 of the most well known voting arrows in use today.

Disqus, with their minimalist primary-edge arrow pair. Tailless. Grey both when unvoted and when selected.

Reddit, with possibly the most used voting arrows on the web. Tailed, grey when unvoted, orange and teal when voted.

Stack Overflow. Tailless, grey when unvoted and voted. Simple, unexciting. Arrow Shape - To Tail or Not to Tail A lot of vote arrows are tailless, some even bodiless, just existing as their slicing edge, such as the Disqus arrows. For younger and more internet-acculturated users the tailless geometries are often the best choice because their modern abstractions allow for the greatest degree of minimalism, but to be most clear for our older users (some in their 80’s) I felt tails would be helpful. Arrow Angle The angle of the arrowhead is also a consideration. Though the three samples above are all obtuse, you probably have another arrow on this screen with an acute angle: your mouse pointer. Near-right-angle arrows have a geometric stability that gives them weight and presence without being overly dominating. Acute, sharp arrows will stand out from clean layouts, having a greater degree of object-ness than arrows that blend into the rectilinear page elements.

Plus, arrow head angle is somewhat of a secondary variable since it relies on height and width for variation, if you wanted an acute arrow you’d need to sacrifice tail or width. Arrowhead Barbs Similarly with the angle of the arrowhead, the barbs are best kept horizontal to integrate the arrow into the page. If you want a little more objectness for the arrow then barb angle is an effective tweak because the non-parallel lines will stand out strongly. Small Digression on the Design of Mouse Pointer Arrows

Mouse pointer arrows have a very particular job, and their dynamic form accomplishes it well. Why?

Angled orientation, pointing to 11 o’clock: causes it to stand out strongly from the rectilinearity of most programs/pages; the body of the arrow angled at 11/5 o’clock sweeps it out of view for more precise clicking of text for us left-to-right reading westerners; the angle matches basic western eye direction of top left to bottom right, giving it a sense of pointing forward.

Subtle barbs create even greater contrast with content while maintaining its strength.

Acute arrowhead angle balances just the right degree of accuracy (sharpness) with presence (width).

Black and white fill and stroke guarantee it contrasts with content regardless of white or dark theme or content.

Drop shadow to add a little additional contrast and further strengthen its objectness.

I’m using OSX at the moment so I can’t say how universal this is, but the perfectly vertical left cutting edge of the arrow is clever for two reasons. First, having that vertical edge anchors our precision for the arrow point, making it much easier for users to get the point on the exact pixel they choose. Second, though antialiasing is becoming less of an issue as PPI gets absurdly large, having one non-AA edge gives it solidity; the non-fuzzy primary edge makes the pointer much stronger.

Gradient

A great way to add to the directionality of the arrow is to use a gradient fade in the direction of the arrow. I kept it subtle to stay at an unconscious level. Often the best design moves are those the user isn’t aware of even though they are under its influence. Colors Colors are powerful tools for conveying meaning. Given that the up/downvote arrows aren’t the strongest indication of support/opposition, I felt we needed to bring big guns: stoplight green and red. I picked hues that complimented the punchy color palette of the site, then saturated to the ideal level of impact. For the contra-down I went grey with a hint of red/green, returning to the unvoted color on hover. Once you vote the color conflict is resolved, red or green wins, but if you want to switch the color returns. Depth

Depth is a thorny issue; it tends to veer toward skeuo or look dated. I kept these arrows flat except in the down state, which are recessed. This sort of glowing void arrow clearly doesn’t look like it can be pressed; there is no button there. But the imprint of the arrow shape clearly signals it was there. And in its place the glowing light shows the effect of an action; something was activated. Our brains are hardwired that a glow is an effect caused by something. Final

The arrows in their three states. Weaknesses and Shortcomings These arrows were built in somewhat of a rush. I didn’t have the time I wanted to spend on the design(always the case), but it was even more hurried than “normal”. Consequently there are some things I’d like to address in the 2nd draft.

The feeling of resolution isn’t strong enough with the down/contra-down arrows. Maybe the colors could change hue and saturation. Maybe instead of grey the contra-down arrow could go with a subtle neutral color such as a warm grey/sea-green.

The vote behavior isn’t instant. The superscript % and button state should use ajax to change instantly.

The colors aren’t perfect, I’m not happy with the red and greens. They don’t have enough character.

The arrows shouldn’t expand on hover. Yes lots of people do it but its gives a sense of cheapness. It makes icons look like decals instead of fully integrated components.

[happyemergency]

TechCrunch: How To Build A Political Social Network That Actually Works

Here's another article I had published in TechCrunch recently about political social networking. Getting closer to making big change with small bytes.

http://techcrunch.com/2013/04/19/strategies-for-building-political-social-networks/

[happyemergency]

Inception Branding

Your brand doesn’t exist in your marketing, it exists in the minds of your customers. Simple, obvious, but it does raise a possibility: if your customer unlocks your brand meaning, that process of discovery will add a strong sense of ownership and desire.

To illustrate this idea let’s assume we have a clothing company launching a new product-line for which we’ve already developed a brand persona (a recently popular exercise involving the creation of richly developed fictional characters that embody the personality and philosophy of your brand.) Brand personas are useful visualization tools that give people permission to imagine the world through the eyes of their hero-product. Just as children who are asked “how would a smart kid answer this question?” will themselves produce better answers, so brand personas give employees the nudge to role-play beyond the professional personalities in which they’ve cast themselves and produce branding more in alignment with their brand’s values. For our example, our clothing company’s brand persona for their new product-line is targeted at young, edgy urbanites.

How do we derive the final branding from the brand persona, is it just a further derivation of the brand persona to media such as graphics and marketing materials? No, the final step is to add the customer persona, the fictional-character representation of the key traits of your customers, then find a vehicle to create your brand identity in the mind of your customer persona. Brand is not your product’s identity, it is the prism through which your customer creates that identity in their own mind.

That is exactly what A-style did, the Italian clothing company whose logo you see at the top of this post. To many, their logo appears to be a sturdy “A” with a slumping umlaut. But to their young, sex-positive market their logo in fact shows two people having doggy style sex. This was a symbol whose meaning their target market was equipped to unlock. Their customers felt a strong connection with their brand because the customer’s minds recorded it as though they thought it up.

This is an easy point to miss, and many companies don’t trust their customers enough to make the connection sufficiently and thus opt for more prima facie branding.

As an example of a company that fell into this trap, French Connection was pursuing the same market at the same time. They wanted the same young, edgy, sex-positive customers of A-style, but as their central brand vehicle their new FCUK logo required no target-market specific decoding, let alone any decoding by anyone in the world. Everyone who even glanced at FCUK reads it as FUCK, there's no sense of personal accomplishment in recognizing it, it’s simply a here’s-our-brand pie in the face. It was edgy, it was sex-positive, but it didn’t create the same connection with its customers. It is an easy trap to create branding materials that give you the sense of your ideal brand, but it will never compete with brands that conjures the imagery in their customers’ minds.

To create a sticky brand your customers must create the brand themselves. You can bring them 90% of the way, but if they make that last 10% connection themselves they will make your brand their own.

[happyemergency]

Government will never be a proper organ of progress because it relies on the support of a population with disparate ideologies.

[happyemergency]

Shared Worldview Communities to Supercharge Activism - The House Foundation

I've been developing this concept for a few years on a back burner in the recesses of my mind. Over the last year or so it's been increasingly demanding, and about four months ago I started working with Lee Shanahan on the idea.

It's current iteration can be found here, applying for a grant to develop it with the Knight Foundation. I'd love to hear feedback, and if you like the idea please "Applaud" it.

Text/Submission: http://kng.ht/ZTZCkz

YouTube Video (2:11) : http://youtu.be/4ZjNtPbf8FY

[happyemergency]

Innovation: Connecting ideas in novel ways to create new solutions to old, previously intractable problems.

[happyemergency]

My Article in TechCrunch: How To Build A Revolutionary Political Social Network

I recently had this article published in TechCrunch, give it a read:

http://techcrunch.com/2012/11/24/blueprint-for-a-democracy-transforming-startup/

[happyemergency]

When life gives you trolls, make a fucking troll martini.

[happyemergency]

Poor writers try to write instead of trying to speak.

[happyemergency]

Google Play Store Loses Battle With Optical Illusions

Warning: If you aren't a designer you'll find this uninteresting. Move along. Move along.

I spent a nightmarish 1.5 hours getting the Nexus 4 yesterday--How could G blow that it in so many ways?-- and a few optical illusion UI issues jumped out at me. Let's see what you think.

I would recommend opening the two full-size screenshots linked below to see the true before and after. With this tumbr theme the size restriction forces me to use 500px wide vignettes which minimize the impact. Original Modified

Before then After

And since the changes are subtle (but #4, below) here are the differences called out and the modified css.

There are really two underlying components of environmental psychology that are operative in these examples:

We don't see with our eyes, we see with our mind.

Looking right is more important than being right.

1. Weakened Edges The edges of white content background boxes are being eroded by the diagonal white line page background texture. That type of fine parallel hatching can cause tons of optical illusion problems, though because of their light color and relatively low contrast they're not overly problematic here. The solution is adding a mid-tone grey border at just the right opacity. You can't see the border, but the box is stronger and edges don't bleed. It's hard to tell in the vignettes, but on the full screen pngs linked above its a pretty striking difference. 2. Choppy Type Due to the same type of problem caused by the tight diagonal lines on the background, the "SOLD OUT" looks blocky, almost as though the font smoothing was disabled or there was a faint dark drop-shadow. Because of the higher contrast of the white diagonal on the dark background, as compared to the previous example, in this instance the diagonals cause the dark background to read as lines that erode the type. Compare the bold "SOLD OUT" to the bold "More options" below it. The original hardly looks bold at all (though this is also partly due to 13px vs 14px font size). To fix this I added 3 drop-shadows of varying tones and opacities. 3. Slumping Title The "Nexus 4" title should be vertically centered, but it isn't. Actually it is, it just doesn't look like it, instead it looks slightly lower. It's subtle but on some level people will have a sense that it's somehow "wrong". The cause is the thin dark blue bar on top of the main blue title bar. Even though the "Nexus 4", is perfectly centered in the original, the addition of that thin dark blue bar unbalances the lower bar. Since they are so similar in hue and shade their shared edge actually bleeds together, making the lower bar seem taller. You also detect the bleed in that the thin dark blue bar almost looks like a drop shadow from the menu bar, doesn't it? If you don't look directly at it that edge looks like a gradient. The dark bar is a nice touch though, so we don't want to get rid of it. Instead I just shifted the "Nexus 4" title up a few pixels, changing the balanced 15px-15px into 13px-17px. NOW it looks centered in the main blue bar, and doesn't look like it's weakly slipping down the page. 4. We Get it With the Lines Already Ok, this one isn't that much of an optical illusion issue. Yes, the white diagonals are eating away at the most hierarchically important part of the page which could be fixed by layering some text shadows like we did before, but in this case the better solution is to just lose the lines. Yes, we get that the gDevices are high tech, and edgy, and you're intentionally thumbing your nose at the skeuomorphic fruit-bois, but in this case just be simple and bold. You don't need that extra edginess, the slight bit of added interest is outweighed by the way it weakens the title of your flagship products. Androids UI is bold and edgy because of it's stark modernism. You're not selling the new Droid (yet). This is how Google makes a phone. Clean, simple, everything just works the way you want it.

[happyemergency]

Turntable.fm: Improve Your Algorithms or Songza Will Drink your Milkshake

The primary product of Turntable.fm (TT) is the curated music stream and that product is currently inferior to competitors such as Songza. The secondary product of providing a venue for aspiring DJs is often deleterious to the primary product by lowering the quality of the curated playlist. Nearly as bad, mediocre DJs prevent the quality of curated music from reaching its full potential. To be successful Turntable.fm can’t just be crowdsourced music curation, it needs to be functionally meritocratic. If TT can successfully provide a venue for aspiring DJs and keep a consistently high level of curation quality it could still meet its goals and beat its competitors. Ideally the UX/UI would be entirely redesigned integrally with a rework of the reputation system, however even with the UI almost entirely unchanged a few tweaks to the algorithms and system could help TT become viable again. Here are a few quick thoughts on algorithm design principles and some basic implementation suggestions: Consistent Quality: Minimize Troll and Amateur DJs

DJs next play is locked 60 seconds before it plays, which DJs on deck can see. If 2 of 4 DJs on deck downvote before it plays that song is skipped. If their 2nd choice is also voted down their play is skipped and they take a hit to their DJ score. When a play isn’t vetoed the voting DJs have their DJ score affected by the outcome of the play.

All rooms have minimum DJ scores and/or room-admin nominated DJs. 

Improved Identification of Pro DJs

Convert DJ score from an absolute accumulation of points to a dynamic score such as between 1 and 100.

DJ score unique to room.

Heart/Lame/Awesome/ ratio becomes primary DJ score determinant.

New/low-score DJs in the audience can virtually DJ: higher ranking DJs can see their next play selection and up/down vote it to affect their score.

Multilevel marketing! Seriously, hear me out: Higher level DJs that positively rate picks from a lower ranking DJs that then goes on to become a higher ranking DJ will also raise the score of the DJ that identified their talent. And have their scores hurt if they support a DJ that goes on to be a dud or troll.

These suggestions would change the nature of TT but if they were well balanced they could encourage vetting without enforcing conformity, improve the overall quality of the music, and make the social interaction even more fun and interesting. Ultimately TT DJs make the primary TT product and unless that product is solid TT will fail. In many ways these sorts of changes would transform TT from a crowdsource system to a collective intelligence system.