avrionov

The need for caution is one reason, but the other reason is the huge capital expenses which are needed to deploy a fleet of cars in a major city.

This is the price of training if nothing fails.

It is a feature:

https://en.cppreference.com/w/cpp/utility/launder

The current generation of generative AI models is quire far from perfect. They'll need more and more data. For bigger models, for models which are targeting specific areas, etc.

But is possible that social media as source is already poisoned by Gen AI postings.

Are you sure about this? Polycom is part of HP Inc. This article is about HPE - the other part of the split.

IRS already audits at least 10% of the returns. They should have the entire tax system implemented as a software. It is madness that this part was hidden from the wide audience.

As many software projects, the internal system may not be appropriate for external use, but still they should have components very long time ago.

These are numbers for 2022. This year they are supposed to be profitable.

He is probably right to some extent, but he didn't solve self driving. He left his own company claiming that the problem is solved and only the boring part was left.

He also claimed that he can fix the search in Twitter, but he left after 2 months and now he is working on something else.

The company is not acquired yet.

This is bad news because it happened before the acquisition. When Cisco buys them, there are going to be more rounds of layoffs.

Red Dwarf predicted this:

https://www.youtube.com/watch?v=JMIHNiR3CP8

Werner Herzog on chicken is the all time best: https://www.youtube.com/watch?v=QhMo4WlBmGM

Millions of dollars. More than they can make anywhere else.

As iPhone X owner I can confirm that this is one of the worst combination for mobile browsing: small screen, safari and internet full of adds.

The best option for browsing internet is large screen Android Phone.

It is much harder to measure muscle mass at home, or even at the doctors office. Different methods show very different results when people are measured at the same time.

We need a simple proxy method for estimating the healthy weight range.

It is too early to say, if it is important or not.

Snowflake is not cheap, but they had the right idea to separate the compute and storage.

I wonder if this segment is ready for disruption. Splunk is very expensive, ElasticSearch is still lacking many of the features of Splunk and when hosted on AWS is very expensive. SumoLogic was acquired by private equity, which means that it won't get cheaper. DataDog is also very expensive.

Solution like SnowFlake for logs / telemetry where compute and storage are separated might be the future.

Sinofski left after Windows 7.

Every website looks like TikTok now.

Thank you for the link. It is interesting read, but it didn't say what you are implying. The article in the link states that 90% of the predictions of 1 Medusa corespondent were wrong. This is very different than the entire publication being 90% wrong.

For people who are wandering like me, the big jump in Django searches was the release of Django Unchained by Quentin Tarantino on Dec 25, 2012.

From the article:

> Another example is figuring out the right tradeoffs between using local SSD disks and block-storage services (AWS EBS and others).

Local disks on AWS are not appropriate for long term storage, because when an instance reboot the data will be lost. AWS also doesn't offer huge amounts of local storage.

Not all meetings are e2e encrypted, because encryption disables tons of features, like cloud recordings, apps, etc.

If you record your video call, Zoom will store it for you. Otherwise, it shouldn't be stored. Also if video streaming protocol like HLS or MPEG-DASH is used, this will store the stream in video chunks, which are deleted later.

What they meant is that it is not possible to fix it without replacing the hardware.

It is an interesting theory, but it is not how it works.

Private Equity partners become board members of the squired companies, some time CEOs, but they don't manage day to day activities of the squired companies.

You can check a company like Francisco Partners which one of the biggest PE firms now. They have another company called Francisco Partners Consulting which helps with the management. The partners are too specialized to become middle managers in the squired companies.

Scott Alexander reviewed the book "San Fransicko" by Michael Shellenberger and as part of the review analyzed the top reasons for people to become homeless. [1]

Some takeaways from the review: - Number one predictor for homelessness are the housing prices. - 30% of the homeless in SF are from outside the city

[1] https://astralcodexten.substack.com/p/book-review-san-fransi...

According to this they own part of it:

https://www.gv.com/portfolio

All FAANG companies had dedicated money pools for share buybacks. Apple even pays dividends. The common belief is that is a company is doing share buybacks, it doesn't innovate anymore. These buybacks have been taking place for the past 10 years and they are still growing.