crvdgc

> I sent a message to my friend, "Cracked it at last. It was the letter 'E'."

Enum types have saved me so many "typo"-related errors. Even in this case, it's not exactly a typo, but the encoding process should make you ask questions about the domain and the meaning of the strings you operate on.

I like how people give many reasons about why 32 hour work weeks is a bad idea, but are somehow settled on the magical 40 hour agreement. Even if all of those are valid, the very same reasons can be applied to an argument for 48 hour work weeks or 60 hours like in many Chinese tech companies.

> Also, it must be ensured that user data is accessible to all competitors.

This doesn't sound like a good idea. The collection of user data and the underlying advertising business model is a problem to be removed, not a playing field for more competition and "improvement".

Maybe slightly missing the main point here, but I find documenting your compromise is a good trade-off (TODO: BF later). Make a note in the comment or PR, create a ticket for it, but leave it for another day.

It takes little effort. Writing things down helps you reassess your choice. And it serves a good starting point for future revisiting. It also keeps you honest (certain culture promotes a pretence of perfection). Even if it becomes someone else's problem later, they'll be grateful for the explicitly.

The tech debt will occur anyway, so you might as well make it visible rather than invisible. Knowing what you owe is the first step of repayment.

Not sure about the post's article one, but yes on the ONS calculator. Basically one enters their expenditure by categories and it will calculate the overall inflation rate. Though personally I wasn't impacted by these two categories.

You can directly view by kind on ONS as well[1]. It states CPIH over 12 months rate in July 2023 is 5.4 for housing and 9.0 for health.

[1]: https://www.ons.gov.uk/economy/inflationandpriceindices/bull...

It's not as applicative in this case, at least for the UK.

As a software developer based in the UK, the inflation mentioned in the article is felt very deeply by myself. The article cited 7.2% as the price inflation rate, which is close to the rate of 7.6% calculated by the Office for National Statistics based on my household spending[1] and is consistent with my personal experience.

On top of that, the UK Treasury adjusted the personal tax threshold policy back in April this year, basically freezing the tax bands to no longer rise with the inflation, which caused a direct 3.3% hit on my income. This freeze is estimated to be the largest over 50 years according to BBC[2], but since we are discussing media sensationalism, so maybe that's not something that would convince you.

Anyway, maybe the situation is different elsewhere, but I do find this article more truthful than not from where I stand.

[1]: https://www.ons.gov.uk/economy/inflationandpriceindices/arti... [2]: https://www.bbc.co.uk/news/business-67031930

True to a certain extent. Being good at execution is both a blessing and a curse. Nowadays, it's just a matter of time before a product's anti-features catch up with its features. Then only implementing the features would get you ahead.

Tried it once. Its promise is to take the dataset's seasonal trend into account, which makes sense for Facebook's original use case.

We ran it on such a dataset and found out that directly using https://github.com/karpathy/minGPT consistently gives a better result. So we ended up using the output of Prophet as an input feature to a neural network, but the result was not improved in any significant way.

This sounds like a false dichotomy. It's possible to opt-in for a "dev" mode which gives the user more choice, but not implemented because of the potential profit loss.

Imagine arguing against press freedom because a relative might fall for some disinformation.

Don't know about ASP.NET, but one of the supported languages is assembly, so I'd presume C++ and Rust can easily add a wrapper around that.

> Binary-compiled languages in general: using the embedded libunit library.

From https://unit.nginx.org/keyfeatures/#supported-app-languages

There is a short one[1] from Functional Programming in Lean, which is itself an introduction to Lean 4 from the programming's perspective. Highly recommended if you're interested in dependent types.

[1]: https://leanprover.github.io/functional_programming_in_lean/...

> recursion is extremely rare in shipped code but extremely common in interviews

Though not quite important to the main theme of the article, I wonder what area the author is referring to.

It's an important problem and people in formal verification are aware of it. One example of tackling this is that the Lean LTE team accompanies the proof with "several files corresponding to the main players in the statement" and "[they] should be (approximately) readable by mathematicians who have minimal experience with Lean [...] to make it easy for non-experts to look through the examples folder, then look through the concise final statement in challenge.lean, and be reasonably confident that the challenge was accomplished".

Details in this blog post: https://leanprover-community.github.io/blog/posts/lte-exampl...

The implicit contract of GitHub and the developer community is that they host open source projects for free, provide a feature-rich and professional interface, in exchange for developers to push their enterprise subscription at the developers' organizations.

This shouldn't be spelled out, I'm sure that GitHub knows better, but they are just playing the dumb card now and hope to nudge the community to accept this round of good will burning before they start the next.

The current list of anti-developer behaviors:

1. Code search requires login: https://github.blog/changelog/2023-06-07-code-search-now-req...

2. Ads are shown in the UI

This topic among others was covered by a Computer Architecture class back in my university. That class is mandatory for all CS students. Though I have to add that it's never been brought up after I started to work.

Real World Haskell has this[1]. Every paragraph and code sample has a comment section that you can unfold.

[1]: https://book.realworldhaskell.org/read/

Kafkaesque

Yes, Tim-Berners Lee led the Solid project[1], which reverses the client/server identity and data model. The user will store their own data and the service provider can only access it under the policy set by the user.

The promise is that one can not only transfer the identity and all personal data across instances of a single service, but also across different services (imagine from mastodon to Lemmy).

Sadly, it never caught on.

[1]: https://solidproject.org/

Thanks for sharing the use case. However using it to defend the entire DMCA sounds like a false dichotomy. Surely there's a way to make rules only for those "awesome" uses, so that they don't have to be "clever" uses. C stands for Copyright and it's mainly used that way in practice.

In rare cases where I really need to remember something, I make an Anki card instead.

It's in the reference list.

> Do they have to pay more license fees to include English subtitles?

Yes. I asked the customer service in Japan and they told me that subtitles are licensed separately.

The Chinese footnote:

> 未经本人电邮及书面授权，不得复制中文译名

states that one is not allowed to copy the Chinese translated names without an email and written consent of the author.

It's the first time I see an attempt to copyright translated names, which would defeat the purpose of translation itself. Let alone most names are just translated by pronunciations. It's really weird to see the author being overly protective and the wording is much milder in English.

At least in Haskell this can be expressed as type classes. For each condition, you can create a (possibly empty) type class to guarantee that the condition is met. Then the call site type class constraints will be checked at compile time.

From the article:

> On Thursday, a university student named Marvin von Hagen independently confirmed that the list of prompts Liu obtained was not a hallucination by obtaining it through a different prompt injection method: by posing as a developer at OpenAI.

Although one cannot be 100% certain.

I used to write presentation slides in Markdown using pandoc+beamer. Basic contents are written in plain Markdown, but you can always resort to LaTeX beamer for advanced features. In the end, it compiles to a PDF file.

A blog post: https://ubikium.gitlab.io/portfolio/latex-beamer-flake.html Example code: https://github.com/crvdgc/latex-beamer-flake

Sounds like a racing condition. A lock (algorithmical or just through communication) should have been used.

I bought books from leanpub.com, and they could really improve their PDF generation.

In Markdown, there's no concept of pages. So you can get one row in a table get pushed to the next page and the rest of the page is blank, even there are contents after. Generating PDF with pandoc will usually take care of this through LaTeX, but I don't think leanpub is doing that.

... which is exactly the setting of We by Yevgeny Zamyatin, the novel that inspired 1984.

There's https://unhook.app/ to remove recommendation elements (front page feed, side bar recommendations, etc.). I use it to leave only the search bar and my subscription feed. YouTube feels so much better.