rpedela

A schema is basically a namespace within a database.

Yeah I'm one of the happy ones. I always loathed Python 2 and text handling, Python 3 was a huge improvement from my perspective.

This tool uses COPY which is highly optimized and supports stdout and stdin. The combination makes it relatively easy to copy tables around. I don't believe MySQL INFILE/OUTFILE is as flexible.

You can use spacy for english tokenization or you can use their neural model. The neural model will generally do better, especially sentence segmentation, but will be slower.

Face-to-face interaction is very important to both roles. You can certainly offload some, maybe most, academic instruction to online, but not the tutoring part.

Researchers always talk about what could be but not necessarily what should be. Don't worry too much, in practice what you suggest will most likely happen.

Should != Must

No one is saying Musk owes people or has to do it. No one.

It is not recommended to run two node ES clusters. The AWS service shouldn't allow that configuration.

There are cases where "dropping into your algorithm" would count as fair use such as a search engine of copyrighted content.

What about a known address when you are lost and don't have signal?

> I'm also confused as to how using google maps affects your ability to become familiar with an area - I've found that by the 3rd time I've gone somewhere I no longer need maps at all, but I was using it up until then.

Many people are naturally good at creating a mental map regardless of how they navigate initially. It sounds like you are one of them. However there are plenty of people who don't seem to have that innate skill. Wilderness orienteering experience would help those people. In my experience, it also helps everyone navigate in unfamiliar surroundings (urban or not) if they don't have signal. It is a form of transfer learning to use an ML analogy.

> Do you know people who Google Maps their way camping?

Yes. Google Maps has pretty good coverage of popular trails and topography and can be used if you have signal.

> it doesn't help you navigate to 9134 Grocer Place

It absolutely can if you have some familiarity with the area and no signal. If from previous experience, you generally know where Grocer Place is then finding 9134 just takes patience. You don't even need to have necessarily been to that specific street before if you have a mental map of the area.

> a topo map would be less than useless

For finding a street address, true. But knowing how to read a topo map helps someone learn how to read and navigate natural landmarks which can be helpful in a natural disaster scenario or in a city if they are lost and no signal. I invite you to spend some time orienteering, I think you'll start to understand where I am coming from.

I know people who are unable to navigate without their phones. If they spent some time in the wilderness, they would start to learn, intuitively, which direction is north, south, etc. They would learn how to read a map, including topo maps. They would start to learn how to create a mental map of their surroundings or at least remember where they had been. They would start learning what do when lost. All very helpful if the internet is down, their phone dies, or they get caught up in a natural disaster back in civilization.

The thing I dislike most about attributing success to mostly luck is that luck is almost always defined as "good luck", but there is bad luck too. Often talented, hard working people are better at maximizing the effects of good luck and minimizing the effects of bad luck. So it is not luck (good or bad), it is not talent, it is not hard work, it is all of the above.

It is closer to $3 trillion because children wouldn't receive the UBI as far as I know. Still a big number for sure.

Yes exactly, thanks.

There are plenty of places in the US with rents in the $100s and you could start a business whether that business is a hotdog stand or a software business.

As a product person, I want to hear someone's proposed solutions even if their ideas are crap. The reason is that sometimes their proposed solutions help me understand the problem better, and sometimes there is an aspect of their proposed solution that makes it into the final solution too.

They probably didn't realize they were in the high risk group until the diagnosis since the transfusion was so long ago. Even just talking to a doctor about screening tests can still be helpful since they will likely ask "have you ever received a blood transfusion" if you are asking about Hep C.

Number #4 is the law. However it is possible to convert an ISO to a NSO which doesn't have the 90-day restriction but has different tax implications. Some companies have started doing this, but I don't know how common it is.

If Peter's company hadn't worked out then your decision would have been correct in hindsight. It is very hard to tell with foresight which startups will succeed or fail. Given the statistics, at the time your decision would have been the most probable one to produce more personal wealth. Even if you had chosen differently, you may not have made the $200 million anyway. If you were given stock options, decided to quit before a liquidation event, then the rational choice would be to not exercise the options because you can get hammered on taxes.

I think your choice was perfectly reasonable and was not a mistake at all.

If Uber/Lyft didn't exist, public transportation in US would still be horrible for a myriad of reasons. I don't know if Uber/Lyft make things worse, but they certainly aren't the root cause.

It is, at least on HN, but it is also one of the best overall.

Doesn't Go have a lower memory overhead than any JVM language?

Product people, regardless of primary background, have to be competent in a variety of skills and experts at one or two to be extremely effective. The job simply requires it. It is possible to be an expert in engineering and be competent in business, design, etc. The blog post is simply describing an engineer who probably could be a product manager/owner if they had the desire.

The phrase "you are wrong" is one of the fastest ways to get someone's back up against the wall. In your example, completely removing that phrase doesn't change the meaning and it eliminates negativity.

I can at least give you some examples from my recent experience. A combination of text classification and NER can solve many business problems if the accuracy is high enough or you can mitigate errors.

For e-commerce search, a common problem is how to give relevant results for misspelled queries or unique queries. Text classification and/or NER can often identify the product type, brand, etc and hopefully give a relevant result which leads to a sale.

Another example is data entry or business analysis roles. Often they will manually extract data from sources and put it into Excel or some other BI tool. NLP ML has gotten good enough in just the last couple years that much of the data extraction can be automated, but you still need a human to verify accuracy.

There were also third party games such as a 007 game.

Valve originally was a game engine company just like Epic and Id, and they still are but have broaden into a platform and publisher. All three companies built games to show off the capabilities of their engines. Valve is unique for also caring about storytelling and rich game mechanics compared to Epic and Id, at least at the time. HL1 was as much about being a tech demo as it was about being a great game.