The solution seems so easy to me, just tie the API access to reddit gold. It wouldn’t be completely smooth sailing, there would still would be many upset users. At the same time though, they would pick up a massive number of subs.
The users that don’t value the apollo experience enough to pay for it would switch to the reddit app, driving more ad revenue. The users that do value the Apollo experience would still keep providing their content to the platform, in addition to becoming paid, direct, subscribers.
This really seems like an amateur strategy.. kill the massively popular apps with ridiculous pricing and unfair timelines, and hope that the users of this massive community organizing tool you control don’t use it against you? Cut off (and piss off) a big segment of your power users in hopes that a sizable chunk of them move to the native app?
There are multiple solutions. This is one of them. Another one would be a bring your own api key and as a user be able to pay. Another one would be to serve ads in the feed for 3rd party api access and make it mandatory thay ads are not filtered out. Yet another one is to work w/ the 3rd party apps to display the ads.
Reddit is just a dumb corpo that doesn't understand why they have success. Watch them backpedal hard.
Overall very underwhelming... not even really a compelling landing page. Can't write code, can't understand any language besides english. If this is really the best they've got after months of 'code red' to catch up, they really seem to be in trouble.
I imagine it's due to the urgency with which they been trying to push Bard out the door. RLHF seems to be a key component in making an actually useful chat AI, so they want to start collecting that data ASAP. Management is probably breathing down that team's necks too, which also doesn't help.
This is the prompt I use, for which ChatGPT outputs a perfect program [0] on the first try (and can even convert it to use material-ui, with some prodding):
> Write a react program that I can paste into a codesandbox. It should display a counter that increases by one every second. Below the counter is a button, and when I press it, it decreases the counter by two. If pressing the button would result in the counter decreasing below zero, the button should be disabled.
Bard refuses to attempt it. It did attempt it for one variation of the prompt, but it used class components with an invalid render function and no timer; it looked like it was just pasting something from the first StackOverflow answer from a Google search of the prompt.
To be fair, Google isn't claiming to support this yet. But I'm just replying to the comment saying it can write code.
In the FAQ[1] it explicitly states that it cannot help you code. Maybe that's a limit of the prompt or just a limit of Google using a smaller/cheaper model? We'll have to test it out to see.
Bard faq says writing is not supported and that bard is still learning. I left with the impression that it can but maybe not well enough to have them even attempt to promote it as beta.
Don't forget bicycles are much safer as well. The tiny wheels on scooters make it incredibly easy to fall if you hit a good sized crack in the road, whereas a larger bicycle wheel can just roll right over it.
Bicycles are also safer, because the first-time or casual rider doesn't zoom at 15 mph on walk paths. And I really doubt scooters can break as well as a decent bike.
Electrically powered vehicles (scooters, hoverboards, gyropods, monowheels) must use the bike path when it exists. Bicycles must travel on the roadway or on the bike path (if there is one).
The user must ride on the bike paths. If there are no bicycle lanes, the user may ride :
- on roads with a speed limit of 50 km/h,
- on pedestrian areas, provided that they ride at a moderate speed (6 km/h) and do not obstruct pedestrians.
Sidewalks may not be used unless authorized by the Mayor. In this case, the user must ride at a moderate speed (6 km/h) and not interfere with pedestrians.
Parking on sidewalks is permitted, provided it does not interfere with pedestrians. However, the mayor may decide to prohibit it.
Use is also permitted on private roads (e.g., private property road, driveway of a private residence).
The user must be at least 12 years old and must not exceed a maximum speed of 25 km/h. The vehicle cannot carry more than one person at a time.
The driver must wear retro-reflective equipment when
- traffic at night
- or insufficient visibility during the day.
As of July 1, 2020, your vehicle must be equipped with a braking system, a horn, lights (front and rear) and rear and side reflectors.
- steady 25 km/h on a bike path is still too much, where you can see 8 year olds on their first bikes, runners, animals, roller bladers and all kinds of distracted and inexperienced bikers; 25 km/h is faster than most other users on the already narrow path; 25 km/h + 25 km/h = 50 km/h in possible frontal collisions
- headlights of the e-scooters are pointed straight ahead and cannot turn away; on a straight path they basically blind other users
- in the dark, it's very difficult to estimate the distance to an approaching e-scooter, because it has only one (blinding) headlight, and it moves at a constant speed, and it moves _fast_
You can’t really do multiple layers of tunnels with cut and cover. It also requires the tunnels to follow where you can actually cut instead of the most direct route.
> You can’t really do multiple layers of tunnels with cut and cover.
Sure you can. I ride such a line every week day.
True about the most direct route, but having uncontested right of way tends to make a big difference in travel times. And in grid layouts, where you want to go is closely aligned to how you get there.
Giving the founders 20-to-1 voting rights over regular investors while losing as much money as they are is insane. This really looks like a lemon, and the VC's want to cash out and leave public investors holding the bag.
I don't see how self driving cars are going to help them. Waymo and Tesla are both the farthest along, and they're going to run their own networks - competing against Lyft and Uber.
Their share structure prevents them from ending up in ETFs, so one of the largest segments of investors literally won't be able to invest in them. Seems more like VC wanting to cash out and leave employees who can't sell their shares yet holding the bag.
There has been a lot of discussion about this issue in the last years. Dual-class setups (at least for new IPOs) are penalized now by some index providers.
Ahh, thank you. That makes much more sense. So it's not that ETFs are specifically prohibited from purchasing them. Rather, if index providers exclude issuers who have a dual-share structure from their indices then any funds that track to an index wouldn't be purchasing stock in those issuers.
I think it's important to note that many ETFs aren't index-tracking so even though index providers may exclude the issuers with several share classes that doesn't mean the stock will be shunned by all ETFs.
Dual-class share setups are typically prohibited by most ETFs. That's why companies that do want to have more than one share class will trade them under different tickers ($GOOG vs $GOOGL).
Edit: Prohibited was the wrong word, looked down upon was probably better.
Thanks, can you explain a bit further? (Nerdy curiosity here, I've done a lot of legal work with mutual funds & hedge funds but have no experience with ETFs so I'm both clueless and curious.) Prohibited by what? The prospectus? And what exactly is prohibited that assigning a different ticker gets around the prohibition? SNAP has a similar structure but I don't remember hearing the same buzz about that being an issue for ETFs when they went public.
Agreed, and now that I think about it a bit more the ticker symbol is largely meaningless. Institutional investors trade by CUSIP, not Ticker, and a unique CUSIP is assigned to each security (and each share class of a company's stock is a unique security) regardless of whether they share an issuer.
Different classes always trade with different tickers. When they trade, of course: Google has also a non-traded B class. (By the way, I don’t think the first statement is correct.)
Prohibited was the wrong word, you're right. Institutional investors that run ETFs have become averse to dual class stocks ($SNAP, $LYFT) because they don't have the same voting power per share that they do with other share class setups. Indexes like S&P 500 have also stopped including listings that are dual-class -- this is changing though, see the articles I've linked below.
I don't really disagree with you, but I think (or, at least, hope) that there will be room for more than 2-3 players in the "self driving taxi" market.
The thing that would cause that is if there is some requirement for massive amounts of difficult to obtain data that is required to build a competitive system.
I do think data is a competitive advantage right now. But I find it really hard to believe that 10 years from now it will be harder to build a self driving system than today. That's just not how tech is. Building YouTube was a herculean endeavor 15 years ago, but today you can hack a YouTube clone together in hours.
Not just will we have more off the shelf software and hardware, but maybe you can license the data too.
In the end, I don't see the moat around a self driving cab company. If someone has a good app and a single car that operates in my area, why not switch to them?
Sure there are wait times and availability in odd places. But that doesn't inhibit a startup getting early adopters. You could literally just do the same commutes every day and have a profitable business.
Lastly, I expect the variety in vehicles and "mobile spaces" will provide a huge landscape of opportunity that a single company will not be able to fill. Just like there is not one "housing" company, there won't be one mobile housing company either. Too much variation in taste and preference.
> Building YouTube was a herculean endeavor 15 years ago, but today you can hack a YouTube clone together in hours.
Ha, good one.
Would your clone have large scale spam, fraud, and abuse systems in place? Would it work on all browsers, mobile devices, TVs, set-top boxes, etc? Is your streaming tech cost-efficient and can it deliver reliability across all regions? Will you be able to respond to DMCAA takedown requests and comply with IP laws across states and countries?
I can go on and on.
Simply put, serving a video over the internet isn't rocket science. Building YouTube...is much more like rocket science.
Personally, I think that Tesla could own the entire market, regardless of operator, if it were to develop the defacto standard rideshare vehicle, and financed it as such a % of the rideshare market.
See every yellow cab in NYC, as an example, is the same make/model of car.
Make a tesla Y vehicle and have the rideshare payment subsidize the car. So X% of every single ride goes to tesla to finance the car. And guess what they get as defacto: the data.
Let people qualify as a driver and put down a deposit, and the car must meet a quota of rides per month, which is tracked and displayed as the car is simply a 'device' at that point... and must be used to pay it off. Tesla doesnt care where the rider bookings come from, lyft, uber, whatever.
If the car is used for private purposes - then the owner is charged some function of the car's time as his "car payment"
Part of the reason rideshare works is because it puts the capex and credit risk on the driver, not the ridesharing company. Taxis are nice and all but that model didn't grow out of the most densely populated markets for a reason.
This model would be better suited for car rental companies, not manufacturers, to utilized unused assets (which I think some have already piloted).
Does my model still not work? Whereby instead of a "car payment" -- a price per mile/slice of your rideshare cost is automatically garnished from your services? And tesla pushes that particular vehicle as a vehicle that must be used as a rideshare vehicle - regardless of if its uber or lyft or whomever - and they reap the data and 'users flocking to their mobile devices'.
Who is paying for the cost of the car up front? That is the person who is carrying the capex and credit risk of this operation.
Even with leases, the lease holder is paying with debt up front and is on the hook to fulfill payments, otherwise they're legally liable and their credit suffers. A rideshare rev split (or affiliate model) doesn't ensure the driver will cover the cost of the car and the manufacturer is on the hook with the risk of a depreciating asset not making money. If you do tie this to credit: - i.e. the driver pays in debt and loses if they don't fulfill rideshare payments - then you're just creating a lease with more strings attached.
> Who is paying for the cost of the car up front? That is the person who is carrying the capex and credit risk of this operation.
Banks, ostensibly. That's what banks do, buy credit risk.
Or Tesla might just take the risk themselves, they continue to be in a reasonable position to raise cash. That's not typical for them though, they've traditionally relied on banks for consumer financing.
This is speculation now, but I get the vibe from Elon that he is interested in a revenue share in the ridesharing world, but that he doesn't think Tesla should be own its own fleet. That puts them into a much more airy fairy financial model, and I think Elon likes the economics of selling cars. Tesla is already a hard enough sell on Wall Street as it is, without a big fleet of cars depreciating on the books.
That is fundamentally not how banks work - they lend money to the manufacturer or the driver, meaning the latter two pay the cost and carry the risk. Banks are not paying the cost.
Tesla has issues with cash and liabilities at the moment, and I highly doubt that any type of scheme like you're describing would be floated by them in the near future.
There might be but the point is that Lyft and Uber valuation is deflated if you assume they are not the dominant forces in a self-driving car industry.
I have Google Wifi at home and a unifi mesh at my office. Unless you are covering a massive area or have more than 25 clients, I’d have a real hard time recommending unifi. It requires much more configuration and maintenance. Google WiFi you really just forget about it - handles patches automatically, finds you the best channel automatically, tests your uplink daily and logs results.
Unifi’s software is much better than any other enterprise solution, but it still has a lot of rough edges and bugs if you’re doing anything non-standard.
I'd probably buy one, but I'm pretty happy with my 2 year old Unifi AC router, after initial setup, I haven't done anything to it - I have it set to update firmware automatically.
Case in point: bonding non-contiguous ports in Unifi is not possible (though it's possible on the Edge products). I think there's also a 4 port maximum.
LegalNature | Los Angeles, CA | Rails & Vue.js Engineers | Full-Time
At LegalNature, we are in the business of automating legal documents and processes. Our mission is to make the legal system far easier to understand, far less time consuming, and far cheaper for individuals and businesses.
We are in the process of building an 'autopilot' for government forms and beureacratic processes. We want our users to fill out a simple and easy form once, and from that, we will be able to generate any documents needed, file them (snail mail, fax, or online), and handle the response. If you have experience automating manual workflows and creating API's around them, we'd love to speak with you.
We are a non-standard company by most tech standards. We’re bootstrapped and profitable. Instead of worrying about the next round or pitch deck, we are only worried about our customers and making the best product possible.
Our engineering team is currently small, and will be growing throughout the year.
> The fact that we've lost faith in humanity's ability to improve things by taking on large, ambitious projects, even if some, even if most, will fail, saddens me every day. It doesn't prevent me from trying to make the world a better place though, and I hope nobody else was discouraged by low-quality, knee-jerk thinking like this, either.
Exactly! The author pick and chose grand ideas from the past that didn’t work. We’re a nation that built the interstate highway system as well as the transcontinental railroad. Both of those ideas were considered impossible by many at the time but completely revolutionized transportation.
> We’re a nation that built the interstate highway system
I think you mean this as a positive example, but a lot of people would disagree. It was an enormous public subsidy for a lot of behaviors that now don't look so good, including increased energy dependence, destruction of natural habitat, and a significant increase in racial segregation.
History is what it is, but I hope we can learn some lessons about the downsides of uncritical techno-utopianism.
Those were both networks built off of scaling existing technologies (the highway is as old as people, and the railway was ~40 years old at that point).
It's a very different thing than entirely new tech outcompeting old and outscaling it, which is what Musk is attempting.
The question is why don't we have a national high speed rail network and if that hasn't or can't work, what's Hyperloop and Boring gonna do better?
We don't have a national (or contintent wide) high speed rail network because North Americans by and large don't believe in community efforts and social investment (except for military). There is absolutely incredible private wealth and an incredible poverty of public wealth.
And so we are beholden to private visionaries like Musk.
It's rather misleading to ignore the convenience of existing modes fo transit (cars in particular) in North America relative to other places in the world. People don't widely support rail investment because cars work well enough. There is just enough parking, inefficient but tolerable traffic, etc. As a result, a lot of people just don't see the problem with the car culture here. Given an immediate problem, they (as with anyone else) will support a solution.
National high speed rail network sounds like something that requires a political consensus (and public money). Probably not feasible to try to build something like that as private company.
For me Elon's ideas feel more like something private company could take.
The users that don’t value the apollo experience enough to pay for it would switch to the reddit app, driving more ad revenue. The users that do value the Apollo experience would still keep providing their content to the platform, in addition to becoming paid, direct, subscribers.
This really seems like an amateur strategy.. kill the massively popular apps with ridiculous pricing and unfair timelines, and hope that the users of this massive community organizing tool you control don’t use it against you? Cut off (and piss off) a big segment of your power users in hopes that a sizable chunk of them move to the native app?