sacheendra

The authors point out that the PL community values formal methods and hard implementation, but not human factor studies and alternative programming approached like spreadsheets. I'm not from the PL community, but my first instinct was other communities such as social science, software engineering, HCI can do those things. PL has to be about building languages.

However, the example that drove the problem home for me was socioPLT. If award winning work on why program languages succeed is not highly successful, then maybe the community should indeed look into incorporating more research approaches.

The points the authors make about lack of community engagement for non-mainstream problems, lack of female creators in books about PLs, and gatekeeping of the term programming language are also convincing.

No satellite was pointed at the path when MH370 was flying.

That's only for Gmail, outlook.com, and samsung browser. It is possible to argue that these are not gateways without mental gymnastics.

It is fairly easy to migrate out of Gmail and outlook to another email provider. Not self host, but that's a different issue.

Samsung browser... Don't know any sites which only work in Samsung browser.

Storage manufacturers now give the application more control over SSD FTL operations through ZNS. https://zonedstorage.io/docs/introduction/zns. Curious to see how it will be used

The project's other github repo is better to learn more about the project: https://github.com/xtofalex/naja

The linked repo doesn't have an informative Readme. An example showing showing Naja differs from existing tools would help people unfamiliar with Electronic Design Automation, like me.

It is not that surprising. Facebook is complex ecosystem, and therefore experiences lots of outages/problems.

Looking at their status page (https://developers.facebook.com/status/dashboard/), they seem to have a problems every week. These are only the public ones!

I guess they figured they are losing more money due to these problems than the additional 5% they have to spend on infrastructure.

This I highly unlikely. Hyperscalers like AWS have custom power delivery, rack organization, redundant networking, etc. which make their instances reliable.

Long running jobs often use checkpoints which require high speed networking and storage, which I don't see an option for. Eg, I cam get EC2 instances with 100gbps networking.

Great job starting the service! But, I think you have ways to go before reaching Hyperscalers level reliability.

yes

It seems like they plan to do hardware-software co-design. Devices (servers, switches?) and firmware designed specifically for their software stack. A software stack (drivers, OS) designed specifically for their hardware. Really interested in seeing where this goes.

That is not true. The synthesis and layout have to be done separately for the ASIC. That is a more time consuming process than writing verilog code.

< 1000 pounds for what is basically license for a livelihood seems cheap.

I don't know how to feel about this. On one side are the privacy implications.

On the other hand, 16 year old me would have loved this. This is the pervasive computing of sci-fi made real. All information is just a tap away. Control lights or run a compute job using a custom Alexa Skill.

Let me take a shot at answering this. The idea is that the internet is a social construct. The information it holds is shared knowledge accessible to everyone.

Contrast burning your photos with burning all copies of a book. On the internet, a lot of things exist in the gray area between them.

That makes it difficult to judge if the info that was is private or public.

A one dimensional comparison like good or bad is not how these decisions get made. Current ARM based chips are good enough for current applications. They come with massive app support and a mature ecosystem.

Given the risk of being kicked out of that ecosystem. The score tips in favor of this new technology.

The USA is one of the few places which confers citizenship based on place of birth. Most countries confer citizenship based on the citizenship of the child's parents.

The property taxes example doesn't seem right.

Cost of a small apartment in NYC is equal to the cost of several acres of farmland. Proportional property taxes would impact both demographics similarly.

I guess it comes to supply and demand. Other countries want US technology and services. They don't want US agricultural products as eluded to in the above thread. Thus, trying to sell stuff others don't want ends up making more noise compared to selling stuff they do want.

The US does this because: 1. National security (US farmers wouldn't grow stuff if nobody bought it, which is a problem if the US ever goes to war. Then, they might suddenly need it.) 2. Vote banks (Rural US happens to be the key in deciding elections and support for laws. Due to the electoral college and each state having 2 senators.) 3. Ideology (The reasons you mentioned. Free market ...)

Looking a the source code of their website, they seems to use react for at least part of it.

I agree that the regulation is not perfect. We can do better to accommodate the needs of enthusiasts. But, in the case of brute regulation, I prefer we err on the side of safety vs enthusiast features.

Here in the netherlands, really old cars that fail to meet emission standards have special dispensation to be used on the road for less than 500km a year. Maybe something similar for sports cars would work. As you said, nobody is using their 911 GT3 for 10,000km a year anyway.

Safety regulations, stricter emissions regulations, drive by wire tech, crash zones and everything else petrolheads don't like about cars make them better for commuters.

Even with all these features, the average increase in weight for cars of the same size has been a paltry 23kg. The weight increase of the "average car" comes from people buying SUVs, which people increasingly are.

For a track day car, a cheap modern sports car like a miata, or a 1980s/90s sports car has all the characteristics you want for the price of a modern commuter car. Its just that a commuter cares more about relaxation than road feel. :-)

For those use cases where retention is a priority, they maybe better served by bluray discs or tapes.

HDDs are being squeezed on both sides. By SSDs for daily use. By discs and tapes for long term storage.

I'm partial to plotnine (https://github.com/has2k1/plotnine). It is an implementation of grammar of graphics in python (like ggplot2).

It provides a stateless layer like interface which feels intuitive, and avoids bugs due to global state.

But, it is built on top of matplotlib. So, if plotnine cannot do something, you can go back to the matplotlib way of doing things and take advantage of the large number of blogs, example, tutorials, and stackoverflow answers.

Performance wise, it can handle more data than altair. But less than matplotlib, if you use the built in data manipulation features.

Stratos Idreis's team has been working on exactly this. Automatic design of data structures based on workload.

Their work has been on HN before: https://news.ycombinator.com/item?id=18314555

Funding their campaign for re-election in the form of ads.

The easiest way to raise "awareness" is making it difficult to do the wrong thing. Or making it very easy to do the right thing. Again, this depends on your environment. Techniques you could use are easy default classifications for the data, ease of browsing existing code so that it can be reused, etc.

Moving from your existing infrastructure to a new infrastructure incurs a cost in both design and implementation. There is also a cost in training existing people on the new infra.

Without much info to go as to how the infrastructure is complex, the first thing I would suggest is to get rid of operations overhead by going with a hosted service. AWS EMR, Databricks, Qubole, etc. offer a service with S3+Hive+Spark.

The reason for using the tools might not just be volume. It could also be due to vicissitude (large number of different data sources using spark for pre processing), or scaling, or something else. Try to understand why the existing solution is used before planning a migration.

If you are determined to go with a new pipeline, build a prototype for a small subset of data science tasks you have and carefully evaluate the pro and cons vs the existing approach for that subset.

I was speaking in the context of the cloud. If you are using S3, I presume you are using AWS EC2 or other services. Otherwise network egress costs dwarf everything else for most use cases. Except maybe as a repo of database backups.

Latency I can see. I find it hard to believe S3 has a lower availability than a sata drive. Given that S3 is replicated.

How is cheap labour where people commit suicide ethical in any way?

One of the reasons Chinese companies can manufacture so cheaply is the possibility of IP theft. The west want to pay their poor less to manufacture things cheaply. The Chinese want to pay the western rich less to use their technology.

Calling IP theft unethical in a sentence with cheap labour is oxymoronic.

If the person didn't want the political views to be public, they shouldn't have tried to buy ads about them.