Bike lanes and right turns. Seriously, who designed this system?

A friend of mine recently gave up on Los Angeles after being mowed down on his bike by an uninsured SUV driver making a right turn. He’s now living in Seattle. We lost one of the good ones, guys.

I’ve always thought right turn lanes were a threat to people on bicycles, but this incident in particular got me thinking about it a bit more seriously, and in doing so, I’m becoming increasingly baffled as to how we came up with the system we have.

right hook

First things first, let me be clear: no matter what anyone does, right turns are always going to be a mess. If both the driver and the cyclist are turning, most intersections produce a situation in which a driver must drive in the bike lane in order to make a right turn. The only time that doesn’t happen is when the street being turned onto also has a bike lane, which means that the driver and the cyclist both have a lane and don’t need to merge, but in the much more common case where the road being turned onto does not have a bike lane, the driver must merge into the bike lane before making the turn in order to avoid crashing (via a “right hook”) into any cyclists that might also be making that turn. Like I said, it’s a mess, but even so, it’s nothing compared to what happens when a cyclist is going straight.

Criss cross crash cycling

When a cyclist goes straight, there’s this little dance that happens between her and the cars that are turning right. They essentially need to trade places; the cyclist must move to the left of the right turn lane so that she’s not blocking the driver’s turn. When this happens, the paths of cars and cyclists form an X. If you need a refresher on why that’s a problem, please enjoy the best commercial ever.

Needless to say, this is an extremely dangerous situation, and when you throw someone into one of those, at the very least, you should let them know that it’s about to happen. A lot of cities around the world figured that one out pretty easily, coming up with solutions like this:

Green paint in Chicago reminding drivers that they're crossing cyclists' path.

Green paint in Chicago reminding drivers that they’re crossing cyclists’ path.

Bike boxes put cyclists in front of cars so that they're not in a turning driver's blind spot.

Bike boxes put cyclists in front of cars so that they’re not in a turning driver’s blind spot.

Despite everyone else’s success though, Los Angeles has failed to rise to the challenge. The vast majority of bike lane-equipped intersections I’ve seen here look like this one on Los Angeles St, next to City Hall:


Seriously. That’s a solution someone came up with.

This may as well be the transit equivalent of that movie The Purge–for a brief moment, all rules are abandoned; What happens during that moment is anyone’s best guess. Cyclists have no idea where they’re supposed to be, and drivers have no idea they’re about to enter a bike lane. It’s perfect chaos.

The absolute worst thing about the above situation though is that the dotted line that used to be the right boundary of a vehicle lane turns into the right boundary of the bike lane! That gives the very clear message that the bike lane is on the left side of the current lane, and that there should not be any cyclists on the other side of that line, exactly the place where they’re going to be as they cross over to where the bike lane continues. If that paint has any effect on traffic, it’s to make cycling more dangerous, not safer.

Come on guys. This is a really simple fix. I need to know: What’s the deal? Why are we doing it this way? Is there some ill-conceived engineering guideline that no one has ever bothered to change? Is it a cost-cutting technique? Or has literally no one ever even given it a thought?

No matter what the answer, something needs to be done, because we’ve let this go on long enough. It’s a simple change that will make a huge difference and save a lot of lives.


Why neighborhood identity matters: Mapping sentiment across Los Angeles

tl;dr | I mined ~200k recent tweets in the Los Angeles area and used them to determine where in the city people are the most positive and where they’re the most negative. The results were interesting. If you just want to see the map, go to the map.

A few months ago, I was browsing Zillow for listings in the central city as I often do when I’m bored, when I happened to come across an old house that I particularly liked. It was on Solano Ave, which is a little-known street that creeps up the east side of Elysian Park, just north of Chinatown. Without giving away too much of the secret, the street is full of beautiful (though sadly decaying) old Victorian homes, and if you walk up it, you’ll feel like you’re in San Francisco. It’s crazy. It’s awesome. And I’ve always secretly dreamed of living there. I had to go look at the place.

The next day I biked over, and as I cranked up the hill, I suddenly realized why this house hadn’t already been taken: It was right next to the freeway. The 110, for some ridiculous reason, charges straight through the middle of Elysian Park. Why anyone thought cramming nature full of noisy tires and exhaust fumes was a good idea is beyond me, but they did, and the impact is very real. It feels like a war zone. The houses that butt up against the freeway are subject to the relentless noise of zooming cars, and the people that live there can’t just get up and leave.

It was a sad realization. Sad not only because I knew I could never own that beautiful house (which, by the way, was built in 1891, long before the freeway imposed itself on the neighborhood), but also because I knew that this home was but one of many that had had its viability sucked out by a freeway. This situation happens over and over again, along miles and miles of freeway throughout the city.

Then I had a thought: What if there was a way to measure the damage freeways do to people’s happiness?

How do freeways affect people?

There’s this thing a lot of companies do called sentiment analysis, where you feed a bunch of text into an analyzer and it determines whether or not it thinks the text has a positive or negative sentiment. Positive sentiment isn’t exactly the same thing as happiness–excitement, love, and humor among others could also be considered to be sentimentally positive–but for the purposes of this experiment, it was good enough. So I came up with this plan:

  1. Use Twitter’s API to download and save to a Postgres database a huge number of tweets within a certain radius of Downtown Los Angeles. I picked 15 miles.
  2. Analyze each of the tweets for sentiment, and give them a score.
  3. Use the score in conjunction with a map to see what kind of impact, if any, location has on that sentiment.

Analyzing sentiment

Given that people are doing this in practice, I figured there had to be some kind of open source tool out there. After a little browsing, I found Sentimental, a simple Ruby gem that appeared to be exactly what I needed. It wasn’t too complex–it generates scores by matching words against their corresponding weights in a word list–but since my sample size was going to be large, that was okay; I didn’t need accuracy at the individual level.

After setting that up, I started mining tweets. I hit Twitter’s rate limit a couple times, but over the course of 24 hours I was able to store just under 200k tweets in my database. I was ready to set up a map.

Setting up the map

I needed a way to plot the data such that the density of tweets in a given area didn’t affect its score. Otherwise dense areas may have unwittingly gotten different scores than less dense ones. I eventually decided on a simple grid system, dividing the city into a series of squares and compiling data for the Tweets that fall into each square. Then I tried summing all the scores in each bucket, plotting the above average scores in green, and below average ones in red, which produced this:

Sums of sentiment scores plotted on a map

Sums of sentiment scores. Green numbers are above the average, red are below.

Hazah! In the back of mind there was always the possibility that all that data might not have been significant in the slightest, so it was pretty cool to see some patterns emerging. But I quickly realized that I hadn’t yet solved the density problem I’d been hoping to avoid. That is, in denser areas, even if the tendency toward positive sentiment wasn’t more significant than in other grid locations, the sheer number of tweets in those areas caused the scores to be much higher, which was throwing off the average. Because of that, I decided to use a median instead.

The median sentiment map

At first this produced a really strange effect: the grid sections that were over park land like Topanga Canyon state park and Los Angeles National Forest were overwhelmingly more positive than neighboring areas. For a brief moment I tried to come to the conclusion that nature simply makes people happy, but after a little investigation, I found the problem: each of those grid sections only contained one or two tweets, and the content of those tweets was usually something like “omg this view is amazing!!”

You can probably see where this is going.

Since not very many people are spending their time in the middle of the forest, all it takes is a single happy hiker to skew the score positive for that grid, and who ever had a bad time on a hike? To fix the problem, I decided to eliminate any grid buckets with less than 50 tweets worth of data in them.

The final map

Dimensions 24 rows x 24 columns
Threshold at least 50 tweets per bucket

A map of sentiments across Los Angeles

Sentiments across Los Angeles. Green represents areas that are more positive than usual, red more negative.

When I saw the above map, my original goal of visualizing the negative impact of freeways went out the window. Not only is there no evidence of such a trend visible, this map is also showing us something unrelated, and much more powerful: the importance of neighborhood identity.

LA lacks identity in a lot of its communities

I don’t need to tell anyone that Los Angeles is a city known for its sprawl. The endless series of strip malls and stucco can easily make it feel like we’re made up of only a handful of notable attractions, stitched together by miles and miles of filler and fluff that do nothing to create a sense of place or identity. And though there has been significant pushback against that type of development in recent years, there are still many who argue that it’s a small price to pay for the luxury of a house, a backyard, and a place to put your car. Well, this map seems to indicate the price may be steeper than we expected.

To illustrate the point, let’s go through some of the neighborhoods with positive scores and match them up with the people and culturally significant things that come to mind when we think of them.

Santa Monica Startups, the pier, 3rd St, KCRW, Yuppies
 Venice  Hipsters, a certain kind of hippies, architects, boardwalk, street performers, interesting food, motorcycles, bikes, surfers, Abbot Kinney, David Duchovny
 Marina Del Rey  The marina, boats, sailing, rowing, etc
 Downtown Culver Artists, restaurants, shopping (Note that the rest of Culver wasn’t included on the positive list)
 Sawtelle  Little Osaka, Art, Japanese culture
 Westwood  UCLA, boutique shops, historic film industry stuff
 Playa del Rey Startups
 Beverly Hills Rich people in film and entertainment
 Manhattan / Hermosa Beach Beach parties, bon fires, volleyball, skate & surf, post-college frat life
 Watts Watts Towers
 Downtown Hipsters, finance, farm to table food, oh come on it’s downtown
Silverlake / Echo Park A different kind of hipsters (though not that different)
 Monterey Park Great Chinese culture and food
 Burbank The other film industry, Bob Hope
Pasadena Old Town, Rose Parade
 Whittier College town

With the exception of a couple sections like the undefinable one along the river east of Compton (any ideas?), these places all have strong identities. Their culture is such that you know exactly what defines each neighborhood, and you could tell a tourist what they might find if they went there. The places in red, however, are a different story. What happens in Montebello? Or how about the parts of Culver City that aren’t downtown? What defines those neighborhoods?

Given the reason I started this project, it’s ironic that the San Fernando Valley is the most positive in the areas that are closest to the 101 freeway. Perhaps this is because the freeway runs alongside Ventura Boulevard, which is arguably the most unique and identifiable street in the area.

For kicks, I did a Google search for “Map of interesting things in LA” and got this:

Map of attractions in Los Angeles

I don’t want to speculate too much, but it’s hard to deny the similarities.

So what does this all mean?

Well, it’s hard to say exactly. Drawing conclusions would require making some assumptions we’re not exactly at liberty to make. For example, there’s a bit of a chicken and egg situation going on in that we don’t know whether people with already-positive attitudes created great neighborhoods, or already-great neighborhoods birthed the positive-thinking people who live in them. We also can’t definitively say that tweets are a good representation of a neighborhood. It may be that the only people tweeting are the people who are there specifically to see whatever attraction it is the neighborhood is known for, and they’re all just tourists who don’t live there. One likes to think that tourists would only tweet about the things they’re excited to see, which could explain the unusually positive attitude in those touristy areas, although the very negative sentiment in the Hollywood / Highland area sort of proves that one false.

Despite all this, I believe there’s something significant in this map. It would have been really easy to dump two hundred thousand tweets into a machine and get nothing but garbage out of it, but that’s not what happened; we got a pattern back. We got something real, something obvious. And something that, though we might not yet be able to explain it, is nevertheless worth noting.

There’s one thing we definitely do know though, and that is that communities, above anything else in life, make people happy. Belonging to a community of people with whom you share common interests–common identity–has been shown to be even more important than money in making people feel safe, comfortable, and content. Given that, I don’t think it’s too much of a stretch to say that the neighborhoods with the best defined identities might also have the strongest communities, which in turn makes the people who live there happier and more positive-thinking.

If that’s the case, we might seriously need to reevaluate our approach to city planning, because if the mistakes we’re making aren’t just inconveniencing people–if they’re actually causing them to have a negative outlook on life–then they aren’t just mistakes anymore. They’re crimes.

What do you think the map means? Leave a comment and share your thoughts.


Why Gil Cedillo deserves a chance

I don’t believe that a politician–or anyone for that matter–would ever make a decision knowing that it was wrong. But then, decisions can’t actually be right or wrong in and of themselves, because they’re entirely dependent on their context. The question to ask is, “Who is this decision right for?” Just because someone has deemed a decision to be wrong doesn’t mean it’s universally so; it’s just wrong for that person.

When I was younger, I’d hear stories about the decisions politicians were making, and they would often make me furious. “How can they live with themselves, making decisions that are so obviously wrong?” I’d think. But I was missing a major point, which is that though those decisions may have been wrong for a lot of people, they weren’t so for the politician himself. Those politicians were just doing the same thing we all do–weighing the information, balancing all outcomes, making sure that the positives of the eventual choice outweigh the negatives. But the problem is that no one, no matter how admirable, assigns equal weight to the interests of all the parties affected by a decision. And thanks to lobbying, many politicians can’t resist assigning a lot more weight to their personal interests than those of the people they were elected to serve. But that said, we still can’t stand up and say that such decisions are universally wrong or bad. They’re just wrong or bad for those who don’t have the same set of interests and values.

So here’s the thing: when Cedillo decided to block the bike lanes on North Fig, he believed it was the right decision. He’s not a batman villain; he doesn’t do evil just for the sake of doing evil. But in spite of that, we remain in the dark as to who the decision is right for, because he never offered up an explanation. Is it right for drivers? Cyclists? North Fig residents? Local business? Perhaps the environment? Or is it simply right for Cedillo himself?

I think we should ask him. Click the tweet button below to let Gil know you’d like an explanation.

Who knows, maybe he’ll come back with something reasonable and fact-based. And if he does, even if the decision isn’t the right one for me, I’ll be able to rest easy knowing it was at least right for some people. Because that’s what’s important.


The problem with parking: My note to the Arts District HCNC

Last night when we were discussing the Arts District Development Plan, Mark said something that really stood out to me. He said that if we do this right, which we absolutely should, the Arts District could serve as a model for development in the rest of the city. I completely agree, and that’s everything I ever wanted for this neighborhood. But if being a role model for the future of Los Angeles is indeed our goal, rewriting that plan is not going to be enough, because at every single one of these meetings I’ve attended, we’re entirely overlooked the one thing that is going to rip our neighborhood apart at the seams: the parking.

Without sounding like a broken record, the way we’re treating parking in this neighborhood is naive. Every developer that rolls through proposes building a parking garage that’s essentially the same size as the development itself, and though we do push back a little on height and location, it’s mainly on the grounds that such structures are ugly. I’m happy that there’s any pushback at all, but in this day and age, there is so much available research documenting the detrimental effects of parking on communities that the aesthetics of a garage are the least of my concerns.

I was blown away when at the first ADC meeting I attended several months ago, our residents were asking developers for assurance that their new developments would provide enough parking so as not to impact the current parking situation in the neighborhood. Parking does not work that way. Providing additional parking does nothing more than encourage people to drive into the neighborhood, because the easier it is to park, the more likely people are to drive. Then, when everyone chooses to drive, the parking fills up, and it’s once-again a problem. Then we’re stuck with this situation where we have twice as much parking disrupting the fabric of the neighborhood, but just as much of a parking problem as before. Is that what we want?

This phenomenon has a name, and I recently wrote an article about it. Counterproductive as it may seem, the best thing you can do to keep the parking problem from getting worse is to allow for as little parking as possible when these new developments come in. That ensures that that the people moving into them are people who are planning on walking, biking, and using transit, not relying on their car. Smart cities have recognized this, and have abolished the concept of parking minimums for developers. Of course, developers aren’t going to be happy with that idea since it interferes with their ability to shower potential tenants in amenities, but these developers also have no incentive to maintain the integrity of the neighborhood, so why are we cutting them so much slack? Remember, the people who are moving into the neighborhood in 2014 are the people who can afford it, and often times, those people just aren’t a good fit. They drive in and drive out, contributing nothing to the sense of community here. There’s one guy who lives in the Savoy building whom I’ve never met, nor have I even seen his face, but I know him because he’s nearly run me over in his Porsche on several occasions blindly entering and leaving his parking garage. That guy takes from the community, and he offers nothing in return. If we keep building all this parking, those are the kinds of people we will continue to get. But we can fix that problem. If we were to build buildings that instead of having 150% parking capacity had only 50% capacity, that would not only enable the developers to dramatically cut the costs of the project itself, allowing for lower rents that artists can afford, but it would also keep the toxic people out by refusing to grant them access to the amenities they can’t live without. Artists are far more likely to live without a car than than others, especially in Los Angeles.

For all the same reasons as above, it’s extremely frustrating to hear talk of “parking relief.” There is no such thing as parking relief. There will never be such a thing. We need to face the facts: the Arts District has become a desirable neighborhood, and all the development we’re doing in it is only making it more desirable. There is no desirable neighborhood in the world that has ever solved the parking problem, because solving the parking problem would mean that every single person that wants to park in the neighborhood has a spot when they want it. Can you imagine what that would look like? That’s an insane number of cars. How could we fix that many cars in one place without it looking like LAX? And yes, there are places out there that do not have parking problems, but they’re places like Burbank, where parking is provided at the expense of any sense of community or authenticity. Burbank is not desirable. It’s not a destination. That’s why they don’t have a parking problem. It’s also the reason I don’t live in Burbank. 

The Arts District of 15 years ago wasn’t a desirable location for most people, and that is in fact the only reason there was no parking problem. Those days are gone. We have to stop talking about them as if they’re coming back, because we sound like those people who talk about the original days of the freeway in Los Angeles, where the was never a traffic problem. There will never be parking relief in the Arts District, and if you think otherwise, you’re fooling yourself. It’s like a drug addition; The only relief we’ll get will eventually just be responsible for making the problem even worse. Let’s move past it. Let’s build the best community we can in acceptance of that fact.

The thing is though, all of the above pales in comparison to the most important point: We talk about being a roll model for the future of LA, but that vision can only come to fruition if we actually plan for the future. We’re not currently doing that. Look at the revolution we’ve seen in transportation over the last couple years. Über and Lyft have gone from nothing to commonplace in our lives. Google’s self driving cars are so close to ready that we can expect to see them carrying passengers within five years. There are excellent case studies out there documenting the impact that such technology will have on the future of transportation, and all of them point to a future where car ownership is in massive decline. One such study estimated that self driving cars would reduce the number of cars on the street by a factor of eleven. This is due to the fact that instead of being parked when you’re not using it, the car you just got out of will then head somewhere else to pick up the next guy. That being the case, what is our neighborhood going to look like years down the line when we’ve sacrificed so much land that could have been used for people to instead cater to the cars that are no longer there? It’s not like a parking garage is a temporary structure—these things are going to be around 50 years from now, even if the cars aren’t. That needs to be a consideration in making these decisions.

Guys, the thing that makes the Arts District so wonderful is the community. I know we all believe in that community. It’s why we do what we do. But we’re not being smart enough about preserving it. We’re putting personal interests ahead of our collective interests. We’re making decisions based on gut instead of facts, and we’re making mistakes that are really going to hurt us down the line. All the space we’re devoting to cars is space that we’re not using for people. It’s space that can’t be used to roast coffee. It’s space that can’t be used to make art. It’s space where you can’t walk your dog. It’s space that 20 years from now might be considered nothing more than blight.

I’m not proposing that we do away with parking altogether. Obviously cars are a necessary part of live for a lot of people and will be for the foreseeable future. But while I do see a need for them, I don’t see the need to encourage them. We should be smart about the way we offer infrastructure to cars, and free parking for all is not the way to do it.

In Tokyo, the interiors all their city blocks have parks, restaurants, and public art. Every block there is a community unto itself, open to both the residents of that block and the people simply wandering past. It’s amazing. We can have that here. We can be a role model for livability. We can turn Los Angeles on its head. We can do it all, if only we set our priorities straight. 

Please, let’s stop the parking.

Thanks for listening,



AI will probably make your entire existence pointless–just saying.

40,000 years ago, the last known Neanderthals died out due to our own species being completely incapable of coexisting with another similar species. Given that, is it not ironic that we now spend such a huge amount of time and effort making computers as powerful and humanlike as possible?

I just watched this fascinating TED talk which shows just how far the field of AI, specifically deep learning, has advanced in recent years. Given the pace of progress, it’s entirely likely that in the coming years, the problem solving ability of machines will start to outperform humans in a majority of fields . When that happens, basically every job in the service industry disappears. The medical field? Gone. Computers make more accurate diagnosis than humans do already, and combine that with self-driving car technology and computers will end up doing surgery, too. Writing and documentation will be left to machines as well. Economics, accounting, business, and stats– computers are better at all those things. Even creative writing and art could eventually be taken over by machines when computers learn so much that they can predict, based on the nuances of the human psyche, the type of work our brains find to be thought-provoking and aesthetically pleasing.

The reason we create these technologies is to make our lives easier. But what happens when our lives become too easy? People often speculate that in a world where all labor is taken care of by machines, humans are free to spend their time as they please, perhaps making art. But when even your own ability to make art is undermined by machines that do it far better and much more efficiently than you do, what’s the point?

Humans draw their satisfaction in life from the fact that they are needed. Currently, we’re the most intelligent (known) beings in existence, which means that we’re the driving force behind all progress. The reason humans can’t coexist with other similarly intelligent species is because the sheer fact that they might exist draws away from our own sense of necessity, making our own existence feel pointless. This is especially the case when the second species is more intelligent, as many believe Neanderthals were given their larger brains.

So then, we killed off the Neanderthals. And yet, here we are again, working to recreate a thing that we know is going to undermine our own sense of self-worth, a thing that we know we’re probably not going to be able to tolerate.

I’m not really sure how this one plays out. I’m not against AI by any stretch of the imagination. It’s just strange to think that our very purpose in life, to create value and progress in society, is in turn destroying our ability to create value and progress in society, and thus, eliminating our purpose in life.

I think that’s a palindrome.


The DTLA fire that may actually have done us a favor

Last night a huge fire broke out in Downtown LA near the intersection of the 101 and 110 freeways. I was fast asleep at the time, but after being awoken by a handful of text messages checking to make sure I was out of harm’s way, I hopped on the computer to check out what had actually happened. After mapping the location of the fire, I knew exactly which construction site they were talking about.

Palmer Da Vinci Burned down Palmer Da Vinci Burned down

Geoff Palmer building

The building affected was an under-construction apartment complex by developer Geoff Palmer. If you’ve ever driven in or around DTLA before, you’ve seen his buildings. They’re the unapologetically out of place, faux-Italian fortresses that decorate the freeways in the area, stretching from South Park along the 110 all the way to Chinatown on the north end. To be blunt, people hate them. Palmer has been accused time and again of having no respect for neighborhoods. His complexes destroy community, alienating neighbors with closed doors and barricaded facades in the name of exclusivity. A post on Curbed LA (from only 2 weeks ago!) fumed:

His squat, nearly-identical fortresses, with embarrassing names like the Visconti and the Medici, aren’t just ugly (although they are very ugly), they’re vacuums designed to suck the life out of a neighborhood that has worked so hard to become lively in the past decade.

Even the USC students who do succumb to the allure of cheesy amenities like swimming pools and gyms don’t stay convinced for long. Not a single one of Palmer’s downtown complexes breaks the 2.5 star mark on Yelp, and most of them are far below that.

Your maintenance will never get done.  Requests are rarely attended to.  Oh, and they’ve had both rats and fecal matter on my floor.

Geoff Palmer doesn’t care about people, and he doesn’t care about neighborhoods. His buildings are rotten cliches that pay more insult than homage to the Italian architecture they weakly attempt to emulate, and their foul presence in the neighborhood actively defies downtown’s authenticity. Palmer moved into the neighborhood because he saw the potential to make a lot of money on the growing demand for housing, and that’s exactly what he’s doing.

When I realized that the building burning down last night was Geoff Palmer’s latest development–yet another culture-void fortress doomed to endless 1-star Yelp reviews–I caught myself feeling almost excited. For years, I’ve wished there was something I could do to voice my opinion about those awful buildings–it’s actually one of the reasons I decided to start CityGrows–, but Palmer is wisely unavailable via social media, as he knows he’d get torn to shreds by the array of people negatively affected by his projects. His absence on the internet blocks any possibility for discussion, and it keeps people from rallying against him by simply eliminating the forum. There’s no way to push back against him at all.

No one knows what started the fire that burned down Geoff’s building last night. Maybe some construction equipment at the site malfunctioned, or maybe a nearby car caught on fire, I don’t know. But what I do know, is that it rained quite a lot last week. And what I do know is that that building was surrounded by a rather large swath of un-burnable material. And what I do know, is that when you make a lot of enemies and piss a lot of people off and deny those people an outlet to vent their frustration, they’re likely to find other ways to do so.

I’m not saying I think someone burned down Geoff Palmer’s building on purpose. All I’m saying is I wouldn’t be surprised.