DylanDmitri

I've been thinking about a programming language where parts of speech are flexible, so if you define a "square" function you could write both "square 3" and "3 squared".

I referred my friend to triplebyte and never received the promised $5000 referral bonus. My emails to support were ignored.

I've wondered if you can get a system where groups of 20-30 people (so extended family, maybe a group of friends, parents at a school) appoint one of their members to research issues and vote on their behalf. I already do this by default, deferring on most local decisions to my friend who's really into politics.

There's maybe ten woodblocks with the potential for enduring modern appeal. But only one had an initial popularity spike that let it break through and claim enough cultural real estate to become self sufficient.

Read that Amazon had a similar problem when they assigned deliveries based on proximity. People would crowd the gate, use GPS spoofers, etc.. Amazon changed their algorithm to weight everyone within 500m or so equally, and those problems went away.

Sounds like the delivery apps have an exploitable flaw, and are claiming that attempts to mitigate it are illegal.

Like other design-by committee projects C# has considerable sprawl. As a result, it's overwhelming and confusing for brand new programmers to start writing hobby projects in.

But C# is beginner friendly in that it's easy for inexperienced programmers to contribute business value to existing codebases. The everything-in-one ecosystem (IDE + autocomplete + database + deployment + source control + package management) removes many sharp edges that experienced programmers are used to working around and take for granted in other languages. I've known smart high school kids with limited programming experience become productive C# developers in weeks, and that doesn't happen the same for e.g. Java.

People who build housing create value. People who manage properties create value. People who speculate on properties do not create value.

This is what Roku is attempting -- to control the point of contact (physical TV in living room) and commoditize streaming services. They've been able to squeeze some concessions out of streaming companies in exchange for customer access.

If all engineers worked at early-stage startups, this would be the case. However most engineers (upwards of 95% outside of Bay Area) work at established companies.

I think the real lesson is that under enough pressure every large system leaks. Anything that gatekeeps millions of real dollars (search engines, stock markets, Amazon reviews, insurance claims, etc) will constantly be exploited and patched by nature of the thing. Only "solution" is to decrease pressure, by say fragmenting market into 20+ search engines, so that SEO people can't realistically optimize for all of them at once.

Some smaller scope things can be made completely watertight, for example mathematically proven cryptography, but even that often leaks to government pressure.

Location: Seattle

Remote: Yes

Willing to relocate: No

Technologies: Python+Postgres for data plumbing+analytics, C# for building async web protocols

Résumé/CV: https://www.linkedin.com/in/dylan-dmitri-gray/

Email: d.dylan.g@gmail.com

Want to build out data analytics at a smaller company that's making a positive impact on the world. Fan of boring technologies and maintainable code.

To be fair, I go look at Stripe's homepage every couple months just because it's pretty.

Github is deeply subsidized by Azure's marketing budget and benefits from larger scale. Gitlab can't compete head-to-head; they'll have to pivot or partner soon.

Yeah it's a backwards compatibility issue, wouldn't be surprised if they make this change in Python4

*speculation warning*

We're still not comfortable checking if rationals are equal in any major language... there's a long way to go.

I could see a future high-level programming language (think Python design goals) using the rigorous foundation of surreals as the underlying numeric system. Sure 95% of the time it'll look the same to the programmer, and the performance will be worse, but you get inherent stability around infinities, irrationals, NaNs, floating-point comparisons, etc

It's reasonable to think that ~200 years from now, highschoolers will learn that numbers are all "really" from this structure of surreal numbers, and that most computers and physics processing etc uses them behind the scenes.

He probably does; he's got more tech credentials than every other major US politician. He founded a CRM company in the 90s, sold to Microsoft for $1billion, then worked as an executive there for multiple years (and was Satya Nadella's direct manager/mentor).

Factorio is the direct inspiration for both those games.

async/await came out of Microsoft research; each year they build and test dozens of concepts. Struck gold with async/await and within a couple years that pattern got retrofitted into most major languages (JS, Python, Rust, C++, C#)

I was wrong about Golang and Typescript, I did some reading and found most of their signature features in older languages.

WASM is some sort of political consortium building triumph, but I think has more potential than everything else on the list. The performance is shockingly good (see Google Blob Opera, Mozilla Pyiodide, etc). Over the next decade everything will become a fast Electron app.

I think PL is progressing just fine. If we look at major advancements in the past decade: LLVM, WASM, async/await, Typescript, Golang, Rust

The more interesting question is "what's happened to IDEs?" -- they're basically unchanged over the same time period. I think they feel "unglamorous" relative to PL and thus attract less passion projects.

I want to see an IDE focused on hybrid code/data systems (say Postgres/Python). Keep your schema, your data, and you glue language in one spot, edit it all together. Set up unit tests by specifying an initial "mocked" Postgres state by writing test data in columns, see it percolate through your code, etc.

If my understanding is right, if you purchase a house and you (or your kids) have been living in it continuously you pay taxes on the original purchase price, rather than current value.

These tax savings pass to homeowner families who stay in one location for multiple decades, especially as land values rise.

Homeowners are very opposed to being displaced, since moving means their taxes would reset to a higher level.

Renters are excluded from these savings.

They could do more part verification, but it might take twice the time and 3 times the money for the same insights. Just launch some prototypes and see what happens.

At least for C# ASP.NET it's basically what you described. There's a few optimizations, for example the threadpool has pretty smart autoscaling to maximize throughput.

I wrote an "anything-regressor" in SymPy, where you provide an equation (like y=mx+b) and a dataset of x's and y's, and it solves for constants to get a line of best fit. Super simple to implement and worked for any equation (quartic, logistic, etc) and any dimension of input variables.

Eventual goal was trying to implement something like XGBoost, but by applying successive regression equations to the residuals instead of successive decision trees. E.g. figure out that a sin wave best fits the initial data, and then a linear best explains the remaining difference.

Worked pretty well on toy data sets, but was much too slow to scale to anything real. Still think the idea has promise for producing human interpretable ML models.

Reddit does a version of this with shadow-banning and fuzzing vote numbers.

Ooh that's really good, thank you for sharing!

I'm hoping for a Jupyter/Excel hybrid, but that's mostly wishful thinking.

My brother met him in an informal setting about two years ago, and expressed concern then over "wow he seemed really old / out of it". But Biden's current speeches and interviews seem lucid and coherent so maybe he was just having a bad day back then.

The presidency, like most other executive jobs, ends up being run by a team of people working together behind the public figurehead. I have faith in Kamala, in the White House advisors, in the picks for cabinet positions, etc. Biden will probably invite back many Obama-era advisors, as opposed to white nationalists like Stephen Miller. Decent people tend to build decent organizations.

To me, the most interesting part of Dark is their dedication to tightening the feedback loop in website development.

It caches recent requests to your site, and as you make changes it constantly re-runs those requests showing you the resulting live values in code. This is close to something like Excel where code changes are instantly and visually applied to the input data.

There's incredible promise here, and I believe similar fast-feedback paradigms will become the mainstream programming experience within 30-40 years. We've already had a taste of this with Jupyter Notebook's ubiquity across data science; despite its many flaws it lets people iterate just a little faster.

As for Dark itself, it feels they put too much emphasis on language and not enough on ecosystem. I wish they had focused on a fresh new IDE experience for Django or Rails rather than additionally building a entire language and syntax from scratch.

They are indirectly if not legally liable. If they tried something as damaging as Nano Defender here, it would be massive bad PR and they would be forced to backpedal or lose industry trust.

Corel painter has terrible UX, to the point where I deleted it from my computer after three hours attempting to follow tutorials. On Mac, I strongly recommend Pixelmator.