y1426i

We did the same for our app as well. I wrote a little library to make it as simple as FastAPI to generate swagger specs - you can try it out - https://github.com/sleeksky-dev/alt-swagger .

Over last 6 months, since their latest upgrade, I have been driving my Tesla mostly on FSD. And it is nothing short of amazing, out of this world. The car drives like a pro and exceptions have become very rare. You have to drive this to believe it. Unlike Waymo which is full of gizmos, this is just a regular car that drives itself. The bet on camera and simplified architecture shows here.

If you just extrapolate what has already been possible and bet on the pace of growth of AI algorithms and techniques, robotaxi is almost a certainity and humanoid robots a possibility.

It should at least be possible to ban YouTube shorts. I wish those were served from a separate domain to make it easier to block just those.

If I sell an apple for one gold coin, that gold coin will probably fetch me two apples in the future. Which happens for a resource whose need is growing while supply is limited. This also means my gold appreciated without me generating any value over time. It creates an incentive to save and not spend.

Therein lies the problem with gold, which is why a decoupled currency was required. A natural resource such as gold or bitcoin, appreciates as the economy grows, which in turn slows the economy down. With a printable currency, a country can control inflation or depreciate past value to create agony for people who have to continuously work to create more value. That is what forces nation-building and what capitalism helps with.

Dollar worked. And it won't be replaced with a fixed resource such as gold or bitcoins. But as the article mentions, it may not remain unique. That will be very interesting though, since the world has never experienced those dynamics before.

I worked for years helping procure government grants and saw how it was used. People who have lived in just one place (aka. California) and have nothing to compare don't realize the amount of waste happening here. The cities that collect the highest taxes have the infrastructure and facilities of a poor town. The prop monies go down the drain or are grossly misspent all the time. DOGE is necessary and needs this level of access and authority to make this scale of change in such a short time.

The comments here seem mostly against DOGE, but I have seen the waste in these organizations firsthand, and we all pay for it. Musk hopes to cut spending by 10%, but that is only because he is limited in what he can do. A Twitter-style cleanup would at least reduce it by 50%, but it is not feasible. Know that those 10% or 50% directly map to a percentage of your income and lifestyle directly (higher taxes) or indirectly (higher inflation).

I don't think this is a practical solution. And it does not solve the underlying issue, which is the attention economy.

Here's a better solution option that is easier to implement; even adults can benefit, and I think it solves some of the problems:

1. Have an easy option to turn off feeds and enforce for non-adults. This would apply not just to meta, twitter, but also to Youtube, LinkedIn, etc.

2. Disable like display. The like counts are what hooks people and gives the dopamine kick. Add the ability to hide it and not show for under 18 easily.

3. Social and news sites should not be allowed to send notifications, period. Not on phones or browsers, at least not for those under 18.

Something along these lines would improve social media for everyone, not just kids. Parents' mental health affects kids the same. So blocking it just for kids only goes so far.

Building a single tenant system is 10x simpler than building a multi-tenant SaaS. App development is so much faster these days and you can build a significant system with just couple of remote devs. Even easier when there is an app from which they can copy from.

Quant has been about finding secrets/patterns that no one knows. Secrets because once they are known, the benefits go away or are greatly reduced.

Rather than finding patterns in historical numbers, LLM can help quantify the current world in ways not possible before. This opens up a new world of finding new secrets.

I think this is a different case. The purpose of OpenAI could not have been achieved had it not been converted to a for-profit organization. Profits are necessary since they incentivize the innovation that AI calls for. Non-profits can never achieve these.

Today we all benefit from OpenAI, but its the for-profit Open AI that made it possible. How else would they spend billions on compute and take those large risks, on whose money?

Create social networks by verticals, and later cross integrate it. Too many problems with LinkedIn - cannot find a handyman on it, plumber on it, hard to get relevant recommendations, professional networks are a chaos, no proximity features, student profiles are not that good, not personalized to industries/professions etc.

I haven't seen any idea for which there isn't a successful startup, company or exit.

The problem is internal tools are less then 1% of what an actual multi-tenant product would be. So the skills to build an internal tool, don't always map to the rest of the 99%.

Many of the responses are about the relationship and how that turned out. Instead, the decision to join or not join that startup should come, assuming you both are strangers. Is the opportunity good, does it progress your career, pays you well, and is the equity fair?

Given the answer to that is yes, the job may or may not work out well, that risk remains the same. The decision to leave that startup again should not be based on friendship but on personal situations and aspirations.

I personally feel its better to be working with friends than with strangers. The hierarchy equation is simply a matter of circumstances.

Open offices exist for the same reason WFH doesn't work, at least where it won't work.

I have worked in older organizations, and the culture there is that the most productive workers spend relatively more time on their work chairs. That is the only way managers have traditionally known to get work done from their teams.

An open office is a natural way to ensure everyone is working, at least in how those organizations measure productivity.

If these companies move to a closed office, they will have to change how they measure productivity, and their culture may not allow that.

True. But emergency only insurance can be much cheaper. Local labs can still send scan/blood results globally wherever the patients want. The bulk of lab costs is a medical doctor attesting/writing the report. That can be solved.

There was a post recently about a diabetic patient not able to afford the daily medicines. Most chronic problems require chronic medicines and regular consultation which can be solved.

One issue someone might raise is the quality of doctors. I have heard tons of cases from friend I know where the doctors have messed up here. It is mostly about reviews and trust, and applies globally, and can be an individual preference.

Allow everyone the freedom to own their health, see whichever doctor they want, get prescription from whomever they want and get their medicines shipped from wherever they want from around the world. Let health care compete globally. That will automatically fix this problem.

There are many reasons to not use foreign keys, but it also depends on the kind of application. For small databases, foreign keys do make things simpler from a validation standpoint.

When building systems for scale where the databases may grow large, foreign keys can cause many issues -

- ORM features around foreign keys can easily bring your system down when joining large tables with incorrect/missing indexes during heavy loads

- As the table grows, not having foreign keys makes it simple in taking out large tables into big-data solutions in the future

- The schemas and relations are sometimes hard to understand during the initial phases of application development. Not having those relations makes changing schemas simpler and faster.

- Sharding tables is much simpler when there are no foreign keys

- It helps to add some of the reference logic in the application rather than the database. Databases are the bottlenecks when it comes to IOPS and scaling. The more processing you move to your application server, the better scalability you can achieve.

Hire someone. Really. Spending a bit of money and having someone who depends on you for work can be a great motivator.

Sorry for being a bit vague here. And yes that's 200 vs 50 for the same level, say a Senior Software Engineer, one in San Francisco, the other in say Brazil.

I am trying to get a perspective on the trend of remote here. I mean how is this going to affect Silicon Valley since this is different than the dot-com boom/bust? The story for the bay area (or tech hubs) in the past was that the VCs, executives, and talent all were here and innovation/creativity required being in the room together. All that seems to have changed post-COVID, especially for Silicon Valley.

This is inevitable in a value based economy. And yes, there can be cost based economies. Something is terribly wrong when greed can legally come in the way of a sick person and their cure. There are too many solutions to this problem if you think about it. Government could sponsor all medical research and share the research to companies to manufacture based on cost based pricing. Companies can be restricted from increasing price of their drugs beyond what they were launched with. People should be allowed to freely import life saving drugs and equipments from other countries. The government has the solution, what is needed is a will to solve.

Not surprising. It's quality has deteriorated and not dependable. It's high SEO ranking has led to it being exclusively used for pushing an opinion or agenda. Articles on popular topics eventually end up being biased towards a viewpoint rather than being factual and chronological.

We use pivotal tracker which is awesome. We use labeling conventions heavily to customize for our workflows.

I find CLI a great interface. Its just stuck in pre web days! Here are some thoughts on its design / UX -

- contextual help. if i write a command incorrect, it should suggest correction and link to more help.

- cloud storage of all my command history

- be more understanding. talk a bit in natural language. Eg. show my ip address -> translated to the right shell command.

- learn and grow automatically as the community builds more tools. Like, hey used this command but check this one out!

It is not feasible to upgrade all existing scripts to adhere to this but a wrapper shell that can provide this functionality should be very well possible.