cpprototypes

Lyft and Uber have wasted so much time and money. This is the problem with VC money focused on growth. There's no pressure to build anything sustainable, no pressure to build competitive advantages. Arbitrary point to point level 5 self driving is decades away. And even if it came tomorrow, what competitive advantage do these apps have? They are a single Google Maps update from being wiped out. They should've spent past few years using network effects in the same way as Facebook to protect themselves. They should've built loyalty programs like frequent miles. But they completely wasted their opportunity.

This is a lot lower than I expected. I wish governments would just add this as a tax directly. But seems almost impossible in the current political climate.

A lot of gas stations here ask "Do you want a car wash for $x.xx" directly on the gas pump after you slide your card. If gas stations just added "Do you want carbon neutral for $x.xx" I think many would push yes in the right markets and a reasonable $x.xx value.

This was true years ago, but I think times are different now. Theres a feeling of near panic among many as the effects of climate change become more obvious and the new Trump environmental policies. I think theres strong desire to do something and this could tap into that.

Also by being high end, theres opportunity to be very different. These stations shouldn't even look like traditional stations. They could provide services that appeal to affluent markets and provide good profit margin. This will be necessary at first because raising gas price alone won't be enough to cover costs.

I often wonder what is the natural market price of these services? Basically, without massive VC money distorting everything, how much would a Lyft or Uber ride really cost? I have a feeling it would be much higher than now. Despite all the hype, true level 5 self driving is decades away. When the VC money stops, what are these companies going to do?

I think the only viable option for them is Lyft Line/Uber Pool. It benefits from network effects (more using the service, better it gets) which also prevents newcomers. It could become a monopoly like Facebook. They should be pushing this harder than unrealistic dreams of level 5 self driving.

I think organics food shows this isn't true. These gas stations would first target the affluent liberal markets. The marketing could be, don't feel bad about buying that luxury BMW gas guzzler, come to Neutral gas station. The station should feel high end, like the Apple store of gas stations.

This is how organic food started with Whole Foods. And the same strategy could work here.

I wish some billionaire would start a chain of carbon neutral gas stations. I wouldn't mind paying more for gas if I knew it was carbon neutral. There are several ways the gas could be carbon neutral, either through production (solar/wind into fuel creation) or offsets (plant trees or carbon sequestration). Instead of slowly trying to change our transportation infrastructure to EV, which will take decades, this could have immediate impact now.

The "smartwatch" from Apple and Android is not a good idea. Fitbit and Pebble are closer to the right direction. Apple/Android keep trying to make watch-sized smartphones. But they will never compete with actual smartphones, they will just always be annoying inferior smartphones on your wrist.

The key factor is to consider which senses are optimal UI for each device. Smartphones dominate the visual and audio senses. Smartwatch simply can't compete there. So smartwatch should have a very simple very low power display like the Fitbit and Pebble and maybe a simple speaker/mic.

So what sense is the smartwatch good for? It's touch. There is a lot of potential of using vibrations/touch as a UI. For example, I think the Apple watch vibrates for turns when you do a Maps route. That's a good start. They need to focus more on things like this. There's a lot you can communicate with vibration, kind of like morse code (such as two quick vibrates, a long vibrate, etc.)

And now Apple has a very advanced touch sensor (the new one in iphone 7). Get rid of that silly big screen on Apple watch. Put the touch sensor there. Why not just allow thumbprint scan, wave my watch at a Apple Pay terminal, and buy things? Way less effort than taking out my phone.

These kind of pure touch and vibration UI is a huge open field. And I think Apple somewhat understands this. But they need to make the big step of dropping that screen. Because that's not where the smartwatch is strong. And it would also allow long battery life which is more important for smartwatch. People are used to daily charge for phones. But much less tolerant of that for watches.

The software field really needs to stop associating "top engineer" with Google interview process. The Google interview process has nothing to do with "top engineer". The only thing it measures is how much time you spent studying DS and algorithm books. It's just like the SATs for college, study and you do well, but has nothing to do with actual college work. There is only one correlation and it's as a measure of work ethic. If you study hard for SAT, you will probably study hard in college. Same thing for the Google interview.

It's actually not a bad system for fresh graduates. If they studied hard in college, all that DS and algorithm knowledge should still be fresh. And it's away to see who studied in college and who didn't. But it's only useful here because there's nothing else to look at for fresh graduates, they have no experience.

However, it's a very bad process for senior engineers. Google should have a different process for seniors, but they probably don't care enough to change.

The plan in the article is complementary, not against public transit. It has the potential to fix the "first and last mile" problem for trains. If successful, it would greatly increase use of rail public transit.

I looked at ebikes a few months ago and it's still too expensive. Even the kits for converting a regular bike was almost the price of an ebike. Is it this expensive in the rest of thr world? If not, why is it so expensive here, why isn't someone just importing it and selling it for much cheaper?

I don't understand why Toyota is not the leader in plugin hybrids. They had such a big lead in hybrid tech and the next natural step was plugin hybrid. Imagine a RAV4 or Highlander plugin hybrid with 50 mile range. The demand would be incredible. The marketing would be so easy, "Drive your SUV to get groceries with zero gas! Save money, help the environment, and drive a big car! win-win-win for everyone!" A SUV like this would fit so well with the suburban lifestyle in the US. Instead they are wasting time and money on the hydrogen fuel-cell nonsense. How can their management have enough vision to create the Prius and hybrid tech and yet so ignorant of this next obvious step?

And this is why software today is in such a sad state. Managers think it would be great to have an army of young engineers who don't question and just do what they're told. But any leader surrounded by sycophants will eventually fail. Management often doesn't realize that they really need the senior engineer who has the experience and confidence to say "NO". Because that engineer is the one who will prevent the project from collapsing under the weight of scope creep and tech debt.

Recently I got an email about a job that seemed interesting. It's a well known company that used to be a startup but is now a mature company. I went to the job description page: 8+ years experience, tech stack I'm familiar with, mostly backend work, seems ok so far.

Then I get to the last line. Open office, has video games, happy hours, other typical things designed to attract those in their 20s. It's all a hidden message screaming, if you're old, don't apply. And they want someone with 8+ years experience! It's ridiculous.

FTA:

LA Metro opened up the Expo Line, a light rail between downtown LA and Santa Monica, in May as part of its effort to wean people off car ownership. When it began running, Uber ran a promotion for $5 off Pool rides to or from Expo line stations. For ride-hail companies, partnering with public transportation agencies to market themselves as companion services can increase mutual ridership. Kan, Lyft’s LA general manager, said three of the top 10 destinations for Lyft rides are metro stations.

If this trend continues and becomes more common, it has potential to really change transportation in LA. LA Metro is working on expanding the rail lines but there's always the "last mile" problem (not literally 1 mile, but usually last few miles to get/from rail station to destination). UberPool/Lyft Line can be the solution for this and the more people use it, the better solution it becomes.

Does anyone know how strict is the Google two year update policy for Nexus? It's clear they won't do major releases (such as Android 7.0) for old phones. However, what about critical security updates? For example, if something as bad as Stagefright was discovered, would Google do nothing for all the older Nexus phones out there? Or would something like that be a special exception to the rule?

The word "remote" is so general it's basically useless. We need new common terms to describe the different types, such as:

- Long remote: When someone is offshore or many time zones away (such as India or Eastern Europe).

- TZ remote: Same timezone, but requires airplane flight to meet.

- Local remote: Within driving distance.

Local remote is the best of both worlds and all companies should be doing this. It can be a 3-2 balance (3 days remote, 2 days in office) or 4-1, whatever makes sense for the team and company. Employees get the benefit of remote (flexibility, no commute) and in person meetings (such as designing a new system on a whiteboard). It's sad that so many tech companies (such as Google and FB) don't do this. Considering that it's tech that allows this (with video conferencing, git, slack, etc.), it's ironic that tech companies are so resistant to doing local remote.

TZ remote can work, but it should be only for employees who are already established at the company (spent many years doing local remote). The connections developed during the local remote years will help make TZ remote work. And they can fly to the office a few times during the year to refresh those connections.

Long remote has a lot of issues. The big time zone difference, lack of any history of in-person meetings, and culture differences make it very difficult for this to work.

There is great potential for something like Lyft to fix the rail "first mile" and "last mile" problem. If they can integrate systems with government so that the Lyft cost is part of the rail fare, then it's possible it could be covered by the commuter tax benefit. That would be a big win for everyone, lower cost for commuters, steady income for Lyft, increased ridership for rail.

EDIT: The more I think about it, this is such a good idea I hope Lyft is seriously working on this. With something like Lyft Line, the optimization potential is immense. Assuming each Lyft Line car can have 3 passengers and most workplaces would be within 2-3 miles of a rail station, rides could be very cheap and fast. For example, most coworkers going home would probably share the same Lyft back to the station. The short distance (assuming a 2-3 mile radius) combined with Lyft Line would allow very high throughput during rush hours. If the Lyft cost can benefit from the commuter tax deduction, that's just icing on the cake since the car sharing alone should drop costs by a lot. And there's also networking effect. Coworkers will want to encourage coworkers to do this so they can lower costs for everyone.

I had no issue interpreting the likely intended meaning of this: This person is saying they have been keeping up with new technologies the past 14 years (such as RoR, NodeJS now, but years ago Java held that spot, etc.) https://en.wikipedia.org/wiki/Principle_of_charity helps here especially these days when typing on a mobile phone can easily lead to typos and mistakes.

The message is loud and clear for senior engineers (10+ years experience): startups don't want you. They may say they do, but they really don't. And the message from SV big tech is also clear: we don't want you unless you're in the inner club (Google, FB, Amazon, etc.)

The solution is to avoid these companies. There's a huge world outside SV. These may be known as "boring" (banks, insurance, etc.) but they will not have the stupid ageist algorithm games known as "technical programming interviews". And they respect experience. And they won't have an ageist culture, it will be ok to go home early and be with family.

Something that annoys me about all this self-driving focus is that they are not targeting the obvious low-hanging fruit:

1) Focus on perfecting low speed bumper-to-bumper traffic fully autonomous driving on freeways.

2) Focus on passing legislation to allow zero driver liability and awareness when using the low speed bumper-to-bumper traffic self-driving mode.

The speeds are low and so the risk factor is reduced by orders of magnitude. Even if an accident happens, it won't be fatal because of the low speeds. And that should also help with the legislation issues. Also, this feature would be very popular. Imagine a commercial with typical bumper-to-bumper traffic. Then the camera zooms in and the driver is sleeping. When the car detects that bumper-to-bumper traffic is going to end soon, it triggers a loud alert which wakes the driver.

And yet all the car companies are ignoring this. I remember reading about early lane keeping and automatic speed cruise control and it would not activate under slow speeds. I was incredulous when I read it. They made this not work for exactly the most useful scenario? I don't care about self-driving for road trips, they don't happen often. And door-to-door self driving is the ultimate goal, but it's going to take a long time to get there. But helping to alleviate the horrible bumper-to-bumper driving by allowing drivers to do something else can be done with current technology. The whole self-driving effort feels like no one has done any market research to see how to make it help with real world situations.

Samsung is a typical asian electronics company (has a hardware focused history and very good at it, but doesn't understand or respect software). I'm so glad that node.js is not under Joyent control.

Does anyone know if the ECMAScript group is seriously considering adding something like TS? Maybe in ES8 or ES9? TS 2 will be impressive and it seems TS has a lot of momentum (such as Angular 2).

Just a little bit of programming can be powerful. If I were designing an elementary school course, it would focus on two things:

1) How to clean data. Real world data is always dirty. It mixes strings and ints, has commas and quotes where they shouldn't be, etc. Knowing how to think of data in terms of a standard format and having a plan for exceptions is very important.

2) How to manipulate data. Python, JS, or some other simple dynamic language. Teach how to do functional style programming such as list.forEach(<lambda>) to transform or analyze data.

It's easy to make homework and tests for something like this. And these are fundamental skills. It doesn't matter if Python or JS fades away. It doesn't matter if popular formats change from JSON to something else. It's about teaching how to think about data and how to get it in a form where it's easy and reliable to do analysis or transformations with it.

We don't need to make programming "fun". We need to treat it like algebra. A basic skill that all should learn. But that doesn't mean making web pages or mobile apps. And it also doesn't mean the theory focused CS topics like data structures and algorithms. Just start with learning how to think of data in a standard form like CSV or JSON. And how to use a programming language like JS/Python to do something with the data. After students master these skills, it can be integrated into science classes. Students will know how to write simple scripts to clean up the data from a lab test and analyze it.

What are some of these companies? A common advice I've seen here and other sites is that the best workplace for developers will be at companies where the software is the main revenue product. It's assumed that if software dev is seen as just a pure cost and not the main revenue generator, then workplace for dev will be horrible due to lack of influence and status in the company.

The reasoning for this seems intuitive and correct. But I'm wondering if there's evidence it's wrong. What are these great companies for developers and are they software revenue companies? If not, what are the forces that maintain a good workplace for developers when software is not the main revenue?

The energy efficiency of the process is bad, but the future economic conditions could change dramatically:

1) Fossil fuels costs will eventually start rising and never come down again. How much investment and attention will synthetic production receive when oil is $100/barrel? Or at $200/barrel, or $300/barrel? More investment could lead to significant process improvements.

2) Solar energy has improved tremendously and continues to improve more. But the great weakness of solar is that it's only effective at daytime. Anything produced in excess of daytime demand is basically waste energy. Since it's waste energy, the inefficiency of the oil synthesis process doesn't matter. The only competitor here is other energy storage methods such as batteries. The key question here will be, is it better to store this excess waste energy in batteries or use the synthetic process to make oil? From a pure efficiency standpoint, battery is better. But when considering the existing oil infrastructure, the answer is not as clear.

Ethanol is terrible, especially corn produced ethanol. The type of process I'm referring to is this:

https://en.wikipedia.org/wiki/E-diesel

The inputs to this are just energy + water + CO2 = oil. No corn or other plants required. Currently this type of technology is far from cost competitive. But that could potentially change if more was invested into this area.

When comparing just technology, EV cars are superior. Far greater engine efficiency, less complexity and moving parts, efficient transmission of energy from power plants to vehicle, etc.

But we live in an oil world. There are billions of cars out there. Cars are big investments that are handed down generation to generation like houses, especially in developing countries. EV car sales are still a drop in the bucket compared to ICE car sales.

If we want fast and rapid action on climate change, the quickest path is not pushing EV. It would be massive investment in reducing the cost to make synthetic oil. This may seem impossible, but that's what many said about solar competing with fossil fuels. EV car enthusiasts often talk about how EV cars get immediate environmental benefits from power plant upgrades such as burning coal to solar. But how much more orders of magnitude environment improvement would we get from carbon neutral oil creation? If a cost effective way was found that could compete with fossil fuels, billions of cars would immediately benefit environmentally.

Again, another computer analogy. Imagine someone invented a beautiful new elegant programming language that reduces CPU energy use by 50%. At the same time, someone found a way to reduce JVM CPU energy use by 20%. If we wanted the shortest path to worldwide energy reduction in CPU, what would be faster, just update the JVM for millions of servers or rewrite everything in the new programming language?

Oil is great for energy storage and we have over a hundred years of infrastructure built for it. And that oil technology is still being refined and advanced.

The big issues with oil are due to the source, not the technology itself. Right now the source is underground carbon sinks (fossil fuels). But what if we produced synthetic oil using excess solar energy during daytime? It would pull carbon from the air. And would burn cleaner than any fossil fuel source.

It seems many dream of a battery future. But is it really better to produce millions of batteries instead of just finding better ways to make oil?

It kind of reminds me of the classic desktop app vs web app debate. One big advantage of web app is that upgrages are single source, just update the server. But desktop apps, there may be many multiple old versions out there because users don't always update. In a similar way, why are we focusing on the very difficult and slow task of upgrading all cars to EV? Why not improve the source of oil itself?

I think LA has a lot of special advantages that other regions don't have:

1) It has the same strong CA protection against non-competes as Bay Area.

2) Starting to develop some big and significant "anchor" tech companies such as Snapchat.

3) Surrounded by good universities (UCLA, USC, Caltech)

4) Main tech hubs are Santa Monica and DTLA. These will soon be connected by a new rail line (opening this month)

5) Housing is expensive, but not as bad as Bay Area. And rail line should greatly expand cheaper housing options while being able to work at a tech hub.