Haven't studies shown that it's not the physical act of interacting with a phone that causes distraction, but rather focusing on conversing with someone not physically present?
"Using a cell phone while driving - whether it's hand-held or hands-free delays a driver's reactions as much as having a blood alcohol concentration at the legal limit of .08 percent. (University of Utah)"
Quickly reading though that, it doesn't strike me as convincing.
People drinking to 0.08 drove slower, braked faster (!), had normal reaction times, and got in 0 accidents. Yet the study says they are more aggressive because they followed the car in front of them by 9.5 meters instead of 10.3.
This study doesn't make me think hands-free sets should be illegal, but that the blood alcohol limit level is too low.
I dunno about studies, but for me it's definitely the eyes-not-on-the-road. I'm so thankful I haven't gotten into a crash despite the few times I've been stupid enough to text while driving.
"While many people know texting while driving increases crash risk, the lack of understanding about the risks of phone conversation while driving remains a challenge. Talking on hands-free or handheld cell phones requires the brain to multitask – a process it cannot do safely while driving."
Part of the problem with continuous partial attention, like the problem with alcohol, is that the practitioner/victim loses the ability to judge his/her own performance accurately.
Intuitively I would expect composing a text via voice to be more distraction than holding a conversation: for me, the very act of composing necessitates visualizing the composition.
Perhaps, but I'm less concerned about getting in an accident than I am getting busted by the police for using my hand held while driving (for those of us in such places). This, hopefully, solves that!
I was watching PG's interview on Bloomberg yesterday on he said that most YC startups choose to develop on iOS first then port it to Android. (I can't remember if he said he advises startups to do iOS first before Android).
We did choose to start with Android because it has more powerful APIs. We don't want to cut out half of the smartphone market though, so we'll get once we're happy with Android we'll move onto iOS.
The upside to Android is you can push fixes to market in a matter of hours - so you can iterate faster and worry less about accidentally pushing crippling bugs.
This is absolutely correct and I had forgotten about this huge advantage until you just mentioned it. We've been on a very aggressive release schedule, pushing major overhauls to the app every 2 weeks since we started in August. It definitely wouldn't have been possible to move as fast on iOS.
I can assure you that the process of iteration is horrible on iOS. It's awful to be in the situation having a team that is able to learn fast and develop improvements within days, but then having to wait a 7 day cycle for every new improvement or test. Plus from time to time Apple will just reject you - last weekend our update (a bug fix update!) was rejected because of a method name we were carrying for months. This is really not helping anybody, neither app developers nor users.
It's my understanding that some of the key features are not possible under iOS due to restrictions in the SDK, namely:
- Reception of SMS by the app and subsequent features (like automatically reading the SMS aloud) because iOS doesn't allow the app to either be notified of an SMS or to intercept SMSs before the other SMS-handling apps (like Android does)
- Sending an SMS directly from the phone programatically - as iOS demands that the SMS send form is presented to the user and then requires a button to be tapped. So you'd be reduced to sending via an SMS gateway using a data connection.
I'd really like to be wrong here as we're currently dealing with these limitations too, so if anyone has any counter experience, I'd love to know about it
Thanks! Curtis definitely deserves more credit than I do.
We haven't done iOS before and are aware of these kind of restrictions on SMS apps. We think we can come up with something that will get around them and offer just as good or almost as good of an experience on iOS, although we'll find out for sure when we get to iOS.
"Amazingly, the trick even works when the phone is just sitting there, its screen black, and sleeping. Say “text by voice” and it wakes up"
I'm pretty sure this is impossible without aggressive battery draw. Altho if you're driving you can be using a car charger. The GPS already makes this neccessary unless you want to risk killing your phone before it guides you home.
It would be nice if they could hook into the 'car dock' app or google maps nav - so that their background service auto launches when those launch. I believe that's possible with some egregious permissions.
Also maybe get the 'charger in' permission and actively listen whenever the phone is plugged in.
All the credit for getting background listening working seamlessly goes to my cofounder Curtis. He's a beast.
We already launch automatically on the phone docking with a car dock, and have other triggers you can specify like on Bluetooth connect and when the GPS detects you're moving more than 10mph.
Right now the battery draw is significant if you keep the app on all the time (about 5-7% an hour in our tests). This is unacceptable to us, so we've been working on reducing it, and still have a good number of technical tricks up our sleeve.
>I'm pretty sure this is impossible without battery rape.
I'm starting to sound like a broken record on HN about this, but this kind of comment trivializes actual, no-kidding rape by comparing it to a low battery, and comes across as inherently misogynistic even if you didn't intend it that way.
If we really want to be serious about opening up tech and getting more women involved, and more importantly if we want a world with less rape in it, we've got to cut this kind of thing out.
If he said it "kills the battery" would you say that it trivializes actual, no-kidding murder? I doubt it.
Also, men can get raped too. I don't just mean women raping men; men rape men too. (Would a "Don't drop the soap" joke be too inappropriate here?) It could be argued that assuming only women can be raped is the misogynistic position.
Last, saying that we've got to cut this out "if we want a world with less rape in it" is, frankly, ridiculous. Nobody is going to read the "battery rape" comment and subsequently be more likely to rape someone.
Using it in this context is perhaps strong but definitely appropriate. Also, as kdommeyer noted, "kill" is commonly used in similar contexts (to kill an engine fe).
Apparently the term "rape" in terms of depletion of a resource (as in "rape of the rainforest", "rape of the ocean", etc) isn't keeping women out of the environmental movement. Not so sure how it causes actual rapes to occur, either.
It's a shame that we get this reaction. Emergency workers, truck drivers, and couriers all use our app and find it really helpful because they have to spend all day driving and want a safe way to communicate. (My favorite is a group of patrol officers in Kentucky who have Android phones attached to docks in their squad cars and use our app.)
Somehow even though it's risky and illegal, one third of drivers choose to text and drive. It's hard to get solid numbers, but around a few thousand people die every year in the US because of this, as well as causing hundreds of thousands of injuries[1]. Any way we can make driving safer overall is a net win.
People that are going to send messages while driving are going to do it no matter what. This app makes it much more safer that's all. Its a step in the positive direction, no doubt about it.
People that are going to send messages while driving are going to do it no matter what. This app makes it much more safer that's all. Its a step in the positive direction, no doubt about it.
_Safe_, or just safer?
There can be problems with people thinking that risk has been reduced or eliminated; it can encourage more risky behavior.
"It seems counterintuitive to give drivers less information, by taking away street signs, stop lights and lane markings, to make them drive more safely."
"Specialists in risk analysis argue that something else is in play. They believe that the increased use of bike helmets may have had an unintended consequence: Riders may feel an inflated sense of security and take more risks."
Absent specific studies on this there's room for doubt about "much more safer".
We do use the built in speech recognition that's part of Android. We definitely want to do other languages (and luckily Google supports a lot of languages), but are focusing on English for the short term and will expand once we get that right.
Vlingo's interaction is mostly hands-free, but there are a lot of points where it really wants you to push a button or read the text on-screen. We've also found it's confusing for first time users who want to send text messages, while we've tried to make it almost impossible to mess up while using our app.
Small anecdote: I knew we were onto something when I got my mom a new Android phone for Christmas and she managed to send a text without a hiccup on her first try- she had never used voice tech or even a smartphone up to that point.
Voice commands already do most of this on Android pretty easily, and have for a long time. It looks like the real advantage here is setting it up before you start driving and being able to interact without long pressing the search button - but that seems pretty rough on the battery.
We agree a large part of the advantage is not touching the phone at all. I disagree about voice commands though- if you put that app in the hands of someone who's never used it before, most people will give up before they successfully send a text.
Wonder how many traffic accidents these guys will prevent? I am guessing a lot! Actually, they will also prevent tickets you get from texting while driving in California. I have gotten one and it sucked!
Definitely worth trying, especially for a hyper-connected road warrior like myself.
Wow, I am the worst offender in terms of driving and texting.
Pretty scary (and stupid on the part of those of us who do text and drive) since according to Wikipedia, "the risk of crashing while texting /driving increases by 23 times, because reading or sending a text diverts the driver's eyes from the road for an average of 4.6 seconds"
I'm from Germany which has really strict laws on this and all phone activity (texting, calling, emailing) is forbidden while the engine is on (leave it to the Germans to be this strict).
Somehow in spite of all the risk and PR campaigns, one third of drivers still choose to text and drive[1]. We're happy to play a small part in reducing the number of accidents by providing a safer alternative.
It won't be able to popup behind the locked screen because of the security issues you mention, but you can still use the app through voice. The worst an attacker that got hold of your phone could do is use the app to send text messages and read any messages that the phone received while they had control of it.
Sorry it's giving you so much trouble! We're using Google's default voice recognition engine. It does allow for different accents, but we haven't turned this ability on in the app. If it's getting names wrong, you can also try spelling the contact name one letter at a time, that usually works for people.
