deepGem

I never fully understood why students would go to a bottom tier law school and take out 200K in loans, when you very well know the value of a bottom tier law school. I mean, today is not the lack of information age. You can pretty much find out all the data points needed about a school. For instance, bottom tier schools typically don't advertise the graduating salaries or employment data. Just this one data point is an indicator of future earnings potential.

That being said, top law schools or top schools for that matter are within reach for low income individuals as long as they clear the entry criteria.This entry criteria is a whole another debate but Lambda and other schools also have an entry criteria.

There are need based scholarships and I think Stanford/MIT/Berkeley offer free tuition if your parents earn less than 100K per annum or something like that. So opportunity is definitely there but then not everyone can make it to Stanford, yes but that doesn't mean you saddle yourself with a huge debt by going to a lower tier school, knowing that your loan repayment capacity will be very low in the initial years of post graduation employment.

You got to be kidding ! I would say Taxis are shit. It's not like taxi drivers are leading stellar lives in their mansions. Most of them are struggling to make ends meet, while benefiting the taxi union or whatever shit that runs the monopoly. How is this any different from Uber/Lyft. At least, here you have an option you can drive for multiple ride share without having to fork out your kidney for a deposit at the taxi union.

Local transportation is a basic service. Ideally, your taxpayer money goes into providing one. But we all know how inefficient that is, especially in areas with low population density. So, short of that, Uber and Lyft are the next best thing to happen in transportation. Will that increase binge drinking, yes, so be it. So be it, because people who binge drink will do it anyway. If they don't drink at a bar they will drink at home. The correlation of binge drinking to the availability of ride hailing needs to be questioned.

Reminds me of feature flags. They seem sub optimal at a code level but highly beneficial at the System level. But why force people one way or the other ? Let them make their choices. Why not give optionality and then optimize. I don’t know much about SV traffic but in Bangalore people were routinely spending 4-5 hours in traffic everyday pre covid. I don’t see how this can be beneficial for anyone at an individual level and for the company as a whole.

"Git for everything" is the first thought I had when I opened this article. I am trying to keep all my notes in Google docs and it's all good as long as you stick to their format. I can't edit text files in Google docs ! I mean, how absurd is this.

I feel like building a Google docs clone for simple text files. Just one feature - versioning (A Time machine like interface) and perhaps add collaborative editing later. I just want to write and store a text document without having to create multiple files. Automatic versioning of snapshots so I can go back in time and refer to any timestamp.

In addition, standardised tests do not have any bias about your past or about your education. Anyone can start from scratch, prepare and ace them. All other forms of admissions, GPA, essays etc have biases of other parties. Essays and recommendation letters especially. Also they are so easy to game.

They should clearly eliminate the test fees and make free prep material available for those who can't afford them. Conduct free preparation classes for anyone to attend etc. A lot can be done to make standardised tests more accessible. That would serve a greater purpose.

Am not from the US but this seems very risky. Is there someone at the government who can clarify such guarantees. Am also curious what are the high level logistics to carry out something like this. Do you seek an untraceable route to Ukarine from US ? Do you unleash a worm like Stuxnet that will eventually cripple all Russian cyber attack infra ?

Purely anecdotal again. I had absolutely no concentration to write code about 5 years back. I treaded mostly a solutions engineer path since it was the most comfortable option. Somehow stumbled into running and my career path changed. Now I wouldn't say I am a genius programmer but I am certainly much better at design and coding. Running and lifting weights, surprisingly have resulted in me aspiring for coding tasks that take me to mental exhaustion. Some of the best days I've had are when I am mentally exhausted trying to figure out a solution and just fall asleep at my desk.

" that URL often just points to some VPS running Apache somewhere. Anyone with access to that machine, anyone who buys that domain name in the future, or anyone who compromises that machine can change the image, title, description, etc for the NFT to whatever they’d like at any time (regardless of whether or not they “own” the token). "

This is how I felt when I first created a NFT. Man, the contract is so secure and all that but the raw asset - forget about it.

https://twitter.com/pbanavara/status/1457675453565599748?s=2...

What I deduced was that the chain will somehow reject any other contract referencing the duplicate asset and only preserve the original contract. Something similar happened on OpenSea with Moxie but isn't this centralisation ?

But doesn't this apply to all off chain assets ? Essentially any underlying data for a smart contract, because you know storing even a byte of data on chain costs a lot.

So the smart contract - essentially a set of instructions is distributed, decentralised but prone to security lapses ( a whole another story ) yet somehow the data for these smart contracts is centralised ?

I am so confused.

I chime with this sentiment. The biggest issue in non-fiction books that I have is this - lack of a progress indicator. If it's a biography or an autobiography for example - I intuitively know the progress and this builds some sort of a curiosity to complete the book. In fiction, this is simulated by the story telling. You are almost always eager to find out what next.

In non fiction however, the lack of progress indicator combined with fillers totally takes away the what next phenomenon. On top of this, most non-fiction books try to convey the core of the meaning in repetitive language. Probably there is some science to this. The more you find repeated words sprinkled across, the more they tend to stick in your head. Apparently something that advertisers and marketers use all the time. So may be that's what drives this repetition and filler materials. I've seen this pattern in quite a few non-fiction books. Hooked, Deep Work, Grit etc.

Which is why I think, it's incredibly hard to pick a reading habit if you start with non fiction books. Also I don't think developing a reading habit with fiction will lead to a non fiction reading habit. Personal anecdote, I used to devour fiction as a teenager and in my early 20s. I remember reading Partner by John Grisham in one sitting, without even getting up to eat. Now I can hardly muster the energy to finish a book. I mean, I can but it's no longer that easy.

I read this sentence as "No convincing scientific evidence that hangover, cures work". A great start to the new year I must say.

This is pretty sad. Emirates, for all their glory doesn't pay their pilots well. I mean, what's the point in spending on stuff like fancy first class seats and all that without paying pilots well. True irony. Fanciest and the most comfortable airline for the passengers, the exact opposite for the pilots.

Well, I wouldn't doubt one of the best airlines on the planet. However, the article claims that they hire pilots with lower flying hours as they cannot find experienced pilots like American or United does. I fail to understand this. Can't they just hire experienced pilots from other airlines instead of hiring the inexperienced ones ?

The only site that I visit every day multiple times without fail. I can't remember a single day that I have not been on HN in the last 5 years. There's a lot I have learned and it would be silly on my part to say that I haven't been entertained. So a huge Thank you to HN and particularly the moderation team for all that you do. Happy Holidays !

I would just build a simple web service using FastAPI and use React on the FE. FastAPI is a really nice framework for prototyping or even for building POCs. When the need arises to scale you don't have to to undo a lot of your work. Swagger first class support is a boon. Pydantic and all other database related support also comes at no cost.

You already know about React so just go for it.

DB - You won't go wrong with either MySQL or Postgres or you could save some setup time and go with one of the PaaS offerings from Google or AWS.

This is in the first episode of Seinfeld season 1

Reminds me of the ass-out hug remark in Seinfeld.

So I have done this a couple of times and both outcomes were not so great during the journey. In hindsight however, yes it feels great.

First job at IBM - I quit in a heartbeat. Had no idea what to do next. Attended a Startupweekend event and started working on a startup, didn't go anywhere. After a year or so got hired at MSFT as a contractor. Quit that for another startup. Reason - this was just turning into another cushy job, no real focus just some high level strategy etc.

Again quit that startup, without any plan. I was the engineering manager I wanted to try building a startup again. A long time friend had come down so we teamed up. No idea, nothing to go with. We were not in our 20s and had families to support and all that. Tried multiple ideas for a year and we split. I pursued startups for about 5 years got tired and am now a DE at big tech. Like I said, in hindsight all this feels great - but during the journey I felt quite miserable. None of my ideas were taking off, damn frustrating.

If you want to start your own thing, it's best to not look for ideas. My approach is to get to a job where you'll get as close as humanly possible to interact with customers and find out where their pain points are. Are they happy with the current company, what are the issues etc. Try solving those pain points from within the company - and you'll know what it really feels like to run a startup. If you succeed, great the company will support you, you grow and all that. If you figure that the company is unsupportive, quit and start up. By then, you already know the pain points, and you have some idea of how to solve them. You just have to execute.

So I took a weekend off to visit another city just for the heck of it, to figure out how inflation and COVID have impacted the travel industry. I am surprised.

I flew to Mumbai on a low cost airline and stayed at a 5 star property. The rates and service levels are at per-COVID levels. The flight was packed, but expensive compared to pre-COVID levels. The hotel a totally different story. The crowds are marginally thinner. On a Monday morning what would otherwise be a bustling morning at one of the flagship hotels in SoBo, there are a handful of guests. Yet the service levels are top notch, food is fabulous.

So I guess companies that are hard pressed to show profits at any expense are skimping on stuff to show profits but there are others who are not sacrificing service and quality levels for profits. Though the latter are far and few.

My reading is that, sadly the skimping will continue for a long long time. Once companies discover a new path to profitability, it’s very hard for them to give up on that unless there are several market and demand forces which act against the skimping. For instance, volume. Once volume picks up, there will be no option to but get rid of the skimping in order to save time. But volume picking up is a huge if. I doubt the travel and leisure volumes will ever reach pre-pandemic levels.

My lesson has been to neglect or ignore any statement made with certainty without any underlying data. The most credible statements ever made always had an element of probability/odds associated with them.

That being said, some of the points this chap brings up don't add up to they hyperbole.

AirBnB income - being double the actual rent in a particular neighbourhood ? How is this even possible. If you average out AirBnB rental income over a year, you probably will end up with a slightly higher percentage than rent perhaps 10-30%. Then it's not guaranteed, rental income is. So almost every landlord will choose a guaranteed rental income over the fluctuating AirBnB income. Unless they own multiple properties and can afford the risk. Also, good luck managing an AirBnB rental in Miami without ground support sitting somewhere in Manhattan.

Housing as a means of profit and leverage - To some extent this needs to be stopped. Shelter is a basic need and unlike food, hoarding costs nothing. I am yet to understand the mentality of hoarding houses. Many of my friends are owning 2-3 flats/houses for their kids etc. While, it is more prudent to invest that money in liquid index funds. There is still this perceived notion of 'houses are guaranteed sources of income that will keep increasing in value every year' amongst the average family man. This certainly needs to change, I don't know how. So housing as a source of profit and leverage is bad but I don't see a way out. Space is limited, everyone wants a 5000 sft house.

That being said, it's by no means a doomsday scenario as depicted in this article. I don't know why people get into such hyperbole just to bring about such unrealistic doomsday scenarios. The world is still a better place. To hell with these naysayers.

Yes technically this is a branch, but like a frictionless branch. You don't have to name a branch for instance. You don't have to worry about branch deletion, even though now this is all automated.

https://docs.github.com/en/repositories/configuring-branches...

A fork is still a lot less friction. Otherwise, technically yes it is a branch.

So the key takeaway here is that the Plan, build(iterate), ship cycle works very well. Most successful teams stick to this. What is not clear to me is how an abrupt new feature with an urgent deadline fits into all this ? "We'll fit this into the next release cycle" is not an option, let's say.

So how do you fit in this new requirement ? You are already in the middle of a feature built on day 3 or 4. Devs are working on their forks and have uncommitted code. Now you have a new feature that has to be given priority, assuming there are no dependencies on the existing cycle. Do you just keep a copy of your old work, and start working on this new stuff ? Feels like a mess. Any better approaches ?

I worked on a Tensorflow feature in 2018. There was no branching involved. Fork the main repo, build the feature on your child repo, make sure all unit tests pass and then send a pull request to the main/master of the parent repo. Typically someone reviews the code, you incorporate code review comments, pass all unit tests, CI-CD pass, PR approved. That's it. Your code is in the next production cycle.

So yes, you work locally on a fork of the repo but not the main/master repo. I'd surprised if all devs have access to push to main/master.

I think having branches leads to a lot of merging issues. Because technically, the branch will then be used by multiple devs. So it's like dealing with multiple masters, kind of.

Personally, I prefer the TF approach. Oh I forgot to mention, the release tags are branches I think, perhaps read only branches.

I do miss it. It served one and only one purpose. Emails on the go. It was super fast to type and had an incredible battery life. I could type on the device without looking at the keyboard or the screen. In fact, that’s one of the reasons I didn’t adopt an iPhone early on. It’s hard to break out of muscle habits.

Let me predict Chris Lattner's return to Apple via the Sifive acquisition https://www.linkedin.com/in/chris-lattner-5664498a/

When I first moved from India to US, this is the first noticeable difference I found. Huge ass veggies, devoid of taste. I just couldn't believe tomatoes. They looked so pretty yet had literally no tart. So yeah I'm cool with all of these RNA fiddling etc but for the love of God, do not turn vegetables into tasteless elements that just look pretty in super market shelves.

My real shocker was onion. I mean, man I could just eat the red onion without a drop of tear in my eyes. Try doing this in India.

So those in US, if at all you ever feel like eating real vegetables hop on a 15 hour flight to Delhi or Bangalore :)

Your working theory is spot on. Coasting is almost impossible unless you constantly keep showing that you are valuable to the company. So yeah I did 100 meetings is one easy way of quantifying your presence.

What I am really surprised at - companies have been blind to these bad managers and are letting them run the show. Just blindly fire them. Life becomes simpler for everyone, including the company.

Remote is a win-win for everyone except the bad managers. Why keep them then. Is this so hard to realise ?

It's funny you put it this way. I actually did a couple of hard level Leetcode problems and I thought they would help me immensely in my day to day life in addition to helping me get better at interviews.

No such avail. In fact, unless these algorithms and problem solving methodologies are baked into your memory there's no way you are white boarding a Leetcode hard level problem in an interview.

What I was impressed at an Uber interview was their system design interview process - which basically boiled down to 'how do I abstract retrying a 429 - rate limit exceeded.

What I take is that - the interviewer is expecting a very specific solution even in an open ended system design question. It's like throwing a needle in a haystack at you and expect you to get to the needle in like an hour :).

Well, transparency works both ways doesn't it? Governments are largely not transparent. If they were, you wouldn't have something like the CIA, Guantanamo bay and the entire surveillance enterprise. I am just quoting the US government as an example but most governments are more or less the same.

Expecting others to be transparent, while governments go on about their private secret activities is very unfair.

Case in point - why hasn't any government adopted any open blockchain till date ? I mean, there are gazillions of operations that the government can use a blockchain for. OK, forget about a blockchain, at least an open distributed ledger ? For instance, all public RFPs and contracts can go on a queryable open distributed ledger. Why hasn't this happened yet ?

Let the governments show a spend log of every tax penny collected and then we'll talk about transparency.

I think the olden days of photography with the limits on the rolls etc, created some great photographers. You didn't have an infinite supply of film at your disposal so every shot had to be thought through.

Action and sports photography is an exception to this since you just click as many photos as you can to capture the right moment.

So I wonder if the explosion of digital cameras and smartphones have helped in creating better photographers. One argument could be that now there are lot more people who are taking up photography and some of them eventually will figure out the intricacies of the craft and become better photographers. Kind of like the 80-20 rule. On the flip side there is also a lot of noise (people just clicking photos), which is fine I think.

As a current Apple watch owner I was tempted to get a Garmin, just for the battery life. It is super annoying to remember to charge the Apple watch every 20 hours or so. I wish Apple got around to offering better battery life than adding features.

I finally caved and bought an automatic. I just want a watch to show me the time accurately, last long and offer good water resistance as I wear the watch all the time.