timewarrior

I see many comments wondering why does Tesla not recognize stationary objects when it has a forward looking radar.

The radar can't distinguish between overhead signs, trees and things on the side of road. It only track moving things, otherwise there would be too many false positives and the car will keep braking every few minutes. Other Level 2 autonomous systems have the same limitation.

You might recall, Tesla hit a stationary fire truck in Jan [0]. This was caused due to a similar limitation in their software. Tesla was following a car at a slow speed. When that car suddenly moved out of the way to avoid hitting the Fire truck, Tesla AutoPilot accelerated to match the set speed and hit the fire truck. To summarize, if a moving car suddenly moves out of the way, Tesla will almost always hit whatever is stopped further down the lane. They even call this out in their User Manual.

Tesla is hoping that it would be able to get it's Vision capabilities to the point where it would be able figure out obstructions using cameras working in conjunction with the radar. But there is no committed timeline for this.

LIDAR is able to avoid this by using directed LASERs to build a 3d outline of all the surrounding objects. A system relying on LIDAR would have challenges in RAIN and falling snow. However under natural conditions (where most Tesla deaths with Autopilot have occurred - hitting a SEMI from side, hitting the barrier recently), LIDAR would be able to completely avoid the accidents.

However, if and when, Tesla vision capabilities are able to detect obstructions, they might be much cheaper and more effective than LIDAR in all kind of weather condition. Hoping they can get there soon. And I also wish they would fix their marketing to more accurately reflect the capabilities of the driver assist system and stop misleading the public.

[0] https://www.mercurynews.com/2018/01/22/tesla-on-autopilot-sl...

Exactly. I used Cruise control for the longest time. However having used Autonomous features on Tesla and Mercedes, I can't use Cruise Control anymore. It's just too risky.

Tesla Autonomous features have the brains of a 2 year old and consistency of a machine. Which is why there is some value to it.

I own a Model X. I love the car in many ways. However, I also think the automation is stupider than a 2 year old. Would you trust a 2 year old to drive your car or to decide when to close your Garage door?

I use the Auto-Steer, only when road is empty, is in a reasonably straight line and I am in the middle lane of at least 3 lane road. Otherwise I don't use it. It might also be fine to use it in stop and go on a divided highway because there is on way you will die at those speeds even if you are in an accident.

But sadly, some people trust the car automation with much more than what it is capable of. This leads to accidents like these. And I would blame Tesla here. Tesla has clearly been misleading in marketing this feature. Not everyone is very savvy and they sometimes take what a company says at face value. I have really smart friends who have seen the AutoPilot link [https://www.tesla.com/autopilot] and think that Tesla already have these advanced features.

In the end Tesla is just slightly more advanced than autonomous features of other manufacturers. However, Tesla claims to be much more advanced than others. Because it claims a lot more credit than it deserves, it's only fair that they also face criticism proportionately when the system fails.

Normally, I am against unnecessary regulation. However in this case, I think it is important to bring regulations around consistent messaging and 3rd party rating for autonomous features from all manufactures. We can't have "Move fast and break things" kill people. It's not the same thing as a social app losing a few posts.

Tesla should rename it's feature to "Drive Assist+" whatever, because that is all it is.

On an ending note - I feel sad that the deceased continued to use Autopilot at this spot, when it was known to them that is has failed at the same spot multiple times. If someone well educated, tech savvy and already aware of the problem can fall for this, I worry that a lot more people might do worse.

Yea. But if giving up 20% of SpaceX gives him 100% of Tesla, he might go for it. Hope it doesn't get to this!

I think is it more likely that Elon will use his equity from SpaceX to take it private.

Isn't this an oxymoron. Open borders completely, but against H1B program.

There are a few reasons why AirBnb is succeeding and many other gig economies are struggling.

1. AirBnb is a global market place. It’s much more difficult to build a global market place and its much easier to defend your moat once you are a global marketplace. Travelers from all over the world want to visit all over the world. There is a chicken and egg problem when you start. This benefit applies to Uber to much lesser degree because they are a city level market place. Most riders ride in their home city majority of their time.

2. Repeat business between travelers and hosts are much lesser compared to other gig economies. While people might get tempted to cut a direct deal with their cleaner or dog walker after a few good experiences, it’s much less likely to happen with AirBnb. Same benefits applies to Uber.

I can't edit my previous comment anymore.

The point 2 should have been: >> 2. Behind Silicon Valley on latest technology and product building practices by at least 2 years.

I instead used the word "best" which doesn't really capture what I had meant. This thread also helped me develop a better understanding of my learnings.

Nowhere did I say that Swiss startups are hopeless, either due to this reason or overall. This was a conclusion driven by nextstep and few others. And I think it was largely due to my mistake. I should have used "latest" where as I used "best". What I really meant was:

>> 2. Behind Silicon Valley on latest technology and product building practices by at least 2 years.

In fact I said that if it weren't for immigration uncertainty, I would move there tomorrow. Last year, I moved to Seattle from SF to experience a change of pace. So it's not that I have blind love for all things SF.

SF has a lot more startups trying out things in a small geography. They are trying a lot more things and failing a lot more and consequently learning a lot more.

There are benefits and drawbacks of not being the first mover. Many a times you do not waste time, effort and money going after dead end ideas (remember HomeJoy). Sometime it leads to missing out on things like AirBnb (especially relevant to as tourist driven economy as Switzerland).

With lesser startups being formed, comparatively lesser capital available - Swiss ecosystem is trying to find it equilibrium. And as I originally mentioned - they are really smart people. They will figure out what is best for them.

This thread was also a good learning for me. Every word matters. Sometimes, you try to sneak a post in between other work and things slip!

Probably I used the wrong word. Maybe a better word is no definitive path after getting permanent residency. It depends on the whim and fancy of the local canton. Most countries have a definitive path after getting permanent residency.

Maybe bleeding edge is a better word.

They just try a lot more things and find out faster what is working and what isn't working. Most of them turn out to be useless but the ones that work give them an early start.

Bigger a part of your business software is, faster the growth happens.

Ebay like marketplaces still have a big physical world component present. Even if it isn't ebay, it might be some other company in Australia. Or maybe trademe becomes the company which tries to take over in Austrlia. It's inevitable and just a matter of time!

No, this isn't satire. Please check my other comment: https://news.ycombinator.com/item?id=16218591

There are always exceptions (like your containerization example), which can be attributed to serious tech players like Yahoo, Google and FB in South Bay. However holistically, SF has had much more unicorns and near unicorns. The contrast between SF and rest of the world has been more pronounced in recent year vs pre-2010 (except China of course - due to the growing economy and closed nature making it difficult for outside companies to operate).

Quite a few of these were produced out of Google and FB, so it makes sense they would be based out of MV/Meno Park.

However most of the unicorns or near unicorns since 2010 have been based in SF. Dropbox, AirBnb, Uber, Lyft, Stripe, Zenefits, SoFi, Credit Karma, CloudFlare, DocuSign, Wish, Slack, Github, Instacart, Prosper, Automattic, Eventbrite, Okta, Docker, Gusto, Nextdoor.

Let list could go on and on. I stand by my statement that more talent in concentrated in a smaller area vs South Bay and they are trying out a lot more things and also faster. Dynamics of an urban area also play some role where, where you can quickly walk to, use public transport or get an Uber, vs driving to another place in South Bay. All this leads to SF being ahead of South Bay in terms of learnings and trying out things.

And name calling doesn't get us anywhere!

Maybe choosing own network and use of own time wisely would help. Filter might be better vs ignore.

If would rather be in a position where I can get potentially useful and useless information so that I can filter vs not having much information.

Not just that. This happened when a Uber like model doesn't have any inherent global market place advantages. Most users in a city take a cab in their own city most of the time.

If even having this in their favor, small players are finding it difficult to survive - imagine if a bigger player comes in a space where there is a play for global level market place.

Not all global mega corps need to salt the earth. To some extent disruption will have side effects. Some like Uber are deliberately evil and others like Apple, Salesforce etc might not.

My original point was that people were attempting ideas which were in the realm of potential global mega corps. If you don't grow to be one, you get eaten up!

I used to be in SF and in South Bay. Recently moved to Seattle , closely associated to the India and Swiss tech ecosystem.

In my experience further away you are from SF, less cutting edge you are. In my experience even South Bay is 6 months behind SF.

SF is a big melting pot, where a huge amount of talent are trying out new things and they are open to sharing their experiences - be it about latest technologies or latest customer acquisition strategies. Because of this, you find out about the latest strategies and associate advantages and disadvantages much earlier. By the time these strategies turn in blog posts and capture popular mind share - it takes months and years.

Of course there are exceptions everywhere and some companies outside SF are great at being at cutting edge. However on an amortized basis, further away from SF you are, more likely you are to be behind the cutting edge.

Yes, that is one of the main factors. And the ground reality is that right now a new employment based Green Card petitioner from India would need to wait 70 years.

I agree. But my point was that there was a natural resistance to scaling to global levels. Which is detrimental if you are in a market which has a potential to be dominated by a global mega corp.

Imagine you build an AirBnb like solution limited to Switzerland market. AirBnb will eventually come and eat your lunch.

In today's globalized world - if you are not one of the bigger players and do not have moat (govt contract, some other kind of exclusivity) - other bigger players will come after you.

I have been involved with Swissnex and been advising Swiss startups for some time. I also go to Switzerland every other year if not every year and meet with startups there.

Things I noticed:

1. Incredibly smart people.

2. Behind Silicon Valley on best technology and product building practices by at least 2 years.

3. Great work ethic and social responsibility.

4. Do not have world conquering ambitions - which kinda works against them. One of the founders was talking about an idea and was being very conservative with market size estimates. So I suggested to him - Think about how you can conquer the world. And his reply was - We are Swiss, we don't conquer the world.

I think one of the biggest challenge for Switzerland tech scene is that there is no viable immigration path for a non-EU citizen. Even after decades, you might not get a citizenship [1]. It makes it difficult for them to attract talent from outside, especially those with Silicon Valley experience.

Not that US is any better for Indians. Based on current estimates, if someone from India applies for an employment based Green Card now, they might get it in 70 years. And life is a pain because you have to do a lot of paperwork throughout.

If there was a viable immigration path in Switzerland, I would move there tomorrow!

1. https://www.swissinfo.ch/eng/naturalisation-controversy_amer...

In terms of my time, it has been very valuable.

I had to spend quite some time working around Heroku's limitations. We have had 100% uptime (based on external monitoring) since we moved to Elastic Beanstalk.

Health check based rolling deploys are awesome. Caught production bugs 3-4 times.

I agree that it was easier for me to do, because I have extensive DevOps experience. However if someone is getting to anything more than a hobby project, consider investing in someone to build a basic CloudFormation script for ElasticBeanstalk. I will open source ours in a couple of months (things really hectic right now).

While I agree that a script can't match what Heroku does, in the end Heroku shouldn't be used for anything serious.

I love Heroku for it's simplicity. They give you very few levers, making it less likely for you to screw up.

When we had a production add we faced following issues with Heroku:

1. Frequent downtimes. Heroku would have a 3-4 hour downtime every 2 months. In some cases, site would be down. In most cases you couldn't deploy and add machines.

2. There is an end user impact in terms of performance. After moving to Elastic Beanstalk, we had a 20% decrease in response time and 5% increase in revenue.

3. In the end Heroku is very simplistic. They don't have features like rolling deployment or Blue-green deployment. They had some beta version of rolling deployment, however that was buggy and caused version issues with deployment. With Elastic Beanstalk we have a health check based rolling deploy, which has caught production issues many times.

4. To do custom things, you need to deal with buildpacks etc, which is not very convenient.

5. Thought it wasn't an issue for us, Heroku can be expensive. Our server costs went down 80% while giving better performance.

So, in summary - Heroku is great to get you started quickly. However if you are doing anything serious but do not want to get deep into DevOps, consider something like Elastic Beanstalk.

Eventually we managed our environments using CloudFormation. That allows us to create repeatable Elastic Beanstalk instances. Once we got the CloudFormation template right, it was easier to maintain Elastic Beanstalk vs Heroku.

Medical system in India is a mess of kickbacks. A Doctor gave my cousin a prescription of Calcitriol because it was much more expensive. Led to major side effects which went away as soon as it was stopped.

Supplements don't have it. However that warning was for someone who goes out of way to acquire Calcitriol.

One point of caution:

Anyone who takes Vitamin D supplement, make sure you take 25-Hydroxy (Calcifediol) and not the active form - 1,25 dihydroxy (Calcitriol).

Active form has severe side effects and is not supposed to be used for supplementation unless in you have kidney problem which prevent metabolizing 25-Hydroxy.

Yep, they are and also the best form. Extremely unstable in capsules and oral supplements though.

Methylcobalamin isn't very effective orally. However someone with a gene mutation doesn't have any choices.

It's difficult to determine whether there is a selection bias. Many people who smoke, take Vitamin B12 supplements.

Anyways it's a good idea to not overdo these supplements.

Another big challenge in winter months is that people are so covered with warm clothes, that even going out in Sun doesn't help much.

Thanks for this added information. I discovered the same thing, with another family member where Methylcobalamin worked wonders and Cyanocobalamin didn't.

One thing I would add is that Methylcobalamin needs to be injected to work best. If people do not have your gene mutation, they are better off taking Cyanocobalamin.

PS: Whenever someone in family has major health issue which no one can get handle on - they reach out to me. My lifelong dream has been to cure Cancer using AI. Just started working on that.