News for hackers. Stuff that matters.

talos_

Re:Litestar Is Worth a Look on Thursday August 07, 2025 @07:02AM (#44822968)
Just want to comment that Litestar is awesome. Docs are great and the built-in caching feature is very convenient!
Re:Objects should shut the fuck up on Monday August 04, 2025 @11:28AM (#44787187)
I find it infuriating that navigation apps throw ads when I'm stopped at a red light. This is THE moment where I should glance away from the road and plan my routing
This analogy doesn't map to the actual problem here.

Perplexity is not visiting a website everytime a user asks about it. It's frequently crawling and indexing the web, thus redirecting traffic away from websites.

This crawling reduces costs and improves latency for Perplexity and its users. But it's a major threat to crawled websites

Re:Keep Pydantic out of your Domain Layer on Saturday July 26, 2025 @07:22PM (#44697704)
You should checkout the Python framework Litestar. It's an alternative to FastAPI that implements these ideas via their "Data Transfer Object" concept
Re:A Typology of Canadianisms on Thursday July 10, 2025 @02:56PM (#44524286)
> That seems unavoidable given almost all English words related to government/law/administration (including "state") derive from French!

Interesting. I always thought that Britain adopted parliamentary system earlier than France. I'm guessing this has to do with the period Normandie (i.e., the French king) ruled over England

Re:A Typology of Canadianisms on Thursday July 10, 2025 @02:50PM (#44524239)
I like to joke that "Dominion of Canada" is actually a railroad company
Re:A Typology of Canadianisms on Thursday July 10, 2025 @01:26PM (#44523318)
From the link you shared:

> The English word province is attested since about 1330 and derives from the 13th-century Old French province, which itself comes from the Latin word provincia, which referred to the sphere of authority of a magistrate, in particular, to a foreign territory.

The fact that British authorities picked a French word that the conquered would understand is significant.

> I wonder if the word choice was influenced by the US civil war ending only a couple of years previously

Interesting interpretation! I would agree given Canadians were given the opportunity to ally with the 13 colonies at the time (but didn't). British loyalists also fled the United States. "Province" made allegiance to the crown oversea clear

Re:A Typology of Canadianisms on Thursday July 10, 2025 @12:56PM (#44523045)
The reality in Montréal is that most people are bilingual. Outside older folks, unilingual French speakers are much rarer than unilingual English speakers, which are structurally preserved via the institutions you described. For instance, English-only schooling from first grade to university is available to them, but not to French-speaking households or immigrants. IMHO, it's a disservice to this population. I've had colleague born in Québec deciding to leave because they felt insecure about their professional abilities in French

English is associated with money (historically from colonial forces, and now foreign capital). Montréal, the metropolis, is an island that was unified as a city. Rich English-speaking borough lobbied in 2006 to become independent entities to control their regulations, policies and taxes. This includes the West Island (Dorval, Pointe-Claire, Beaconsfield), and even the very central Westmount near McGill. Nowadays, poor neighborhoods and their french names are erased by condo promoters: Mercier-Hochelaga-Maisonneuve is HOMA, Notre-Dame-de-Grâce is NDG, Ville-Mont-Royal is TMR, Pointe-Saint-Charles/ Le Sud-Ouest is Griffin Town

ref: https://en.wikipedia.org/wiki/Boroughs_of_Montreal

Re:A Typology of Canadianisms on Thursday July 10, 2025 @12:44PM (#44522910)
The original post was focused on history and language, and I added some political spice. Not discussing the politics of language (as in OP) is a bit outrageous.

You're right that French-Canadians are not guilt-free from discrimination et al. Québec only ever had French as an official language, but the last decades we've seen a series of dubious policies

Re:A Typology of Canadianisms on Thursday July 10, 2025 @11:15AM (#44521986)
Office Québécois de la Langue Française (OQLF) promotes the french language and adapt new English words (whereas France typically integrate English words in their vocab).

The website Banque de dépannage linguistique (BDL) will have a lot of useful resources if you're interested! For instance, how to write a professional sounding email, names of official documents, invoice templates.

Highlights (good and bad):

* emails -> courriels (courrier + iels; mail + similar sounding syllable)*

* spam mail -> pourriels (pourri + iels; rotten + similar lexem as courriels)

* to spoil (as in spoilers) -> divulgâcher (divulguer + gâcher; to reveal + to ruin)

* to mansplain -> mecspliquer (mec + expliquer; man + to explain); This one is outrageous (and uncommon) because it's an homonym to "m'expliquer" (explain to me)

* to browse (the web) -> naviguer (as in "to navigate"; browser -> "navigateur")

Re:A Typology of Canadianisms on Thursday July 10, 2025 @11:05AM (#44521906)
déjeuner is a literal translation though. "Breaking fast" -> "dé-jeuner" (undo fasting).

French people typically say: - breakfast - petit-déjeuner (small breakfast) - lunch - déjeuner (breakfast) - diner - diner

Québecois people say: - breakfast - déjeuner - lunch - diner - diner - souper (eating soup; probably historical roots like "getting your big meal of the day" which is likely broth + potatoes)

Re:A Typology of Canadianisms on Thursday July 10, 2025 @06:54AM (#44519552)
This is a remnant of British colonization. French-speaking population didn't know any English, so you have a lot of these literal translations.

I've heard "flour" uttered with the French pronounciation (fl-oo-r, instead of homonym of "flower") in New-Brunswick. I was floored. Took me a while to figure out what they meant.

Clearly, this originates from non-English speakers reading "flour" on a sign and just running with it.

Also, consider that the British conquest happened before watermelon was highly prevalent in France or North-America. It's unsurprising to see terminology diverge in this case.

Re:A Typology of Canadianisms on Thursday July 10, 2025 @06:36AM (#44519433)
Well, why not include the word "Canadian", which significantly predates the country, as the prime example?

It's derived from Iroquois Nation words and used by French settlers to refer to Indigenous people. The word "Canada" was used by explorer Jacques Cartier to refer to the city now called "Québec". It broadly refered to the territory of a specific Indigenous tribe. (could be derogatory, but seemingly accurate / matter-of-fact)

After the British invasion, the British start using "Canadian" to describe both First Nations and French settlers (derogatory, "non-British)

Over time, "Canadian" generally refers to habitants of Canada.

Related: the hockey team "Les Canadiens" is from Montréal in the province of Québec in Canada. It's the oldest hockey team (1909, pre-NHL). The name is a reappropriation of the word Canadian at a time where it was used derogatively against "French-Canadians" (term that didn't exist at the time). Their chant "go, habs, go" refers to the "habitants", i.e., French settlers.

Related: "province" originates from latin used by Romans to described conquered territory. This is the term founders of Canada in 1867 decided to use instead of "state"

For anyone interested in Canadian history, always check-out the French version of a wikipedia page (and translate it). English pages have a lot of hand-waving and start history with their conquest. Also, ChatGPT makes outrageous historical mistakes all the time, such as suggesting that French-Canadians were a minority group in the 19th century

edit: format, typos

Re:Timescale Is Now TigerData on Wednesday July 02, 2025 @03:01AM (#44440913)
In French, the river and the animal are both "tigre"
Re:MCP: An (Accidentally) Universal Plugin System on Saturday June 28, 2025 @02:22PM (#44406914)
You're getting an arbirary string back though...
Re:MCP: An (Accidentally) Universal Plugin System on Saturday June 28, 2025 @02:14PM (#44406864)
The VSCode extension Continue provides similar capabilities and gives you full access to your interaction traces (local database and JSON traces)
Re:Timescale Is Now TigerData on Wednesday June 18, 2025 @01:05AM (#44306815)
IMHO, the bigger name conflict is with Tigris Data. Tigris means tiger and despite no tiger logo, they did have tiger stickers at events
Re:The curse of knowing how, or; fixing everything on Tuesday May 06, 2025 @03:54PM (#43908966)
Agreed on this view! Sharing some similar thoughts

Paraphrasing a virtuous music band reflecting on their discography: "the first album was about what we could; the second one was about what we should"

It also aligns with Gell's philosophy of art. Here's a wikipedia exerpt:

> Gell argues that art in general acts on its users, i.e. achieves agency, through a sort of technical virtuosity. Art can enchant the viewer, who is always a blind viewer, because "the technology of enchantment is founded on the enchantment of technology"

I guess you would be paying for pushing all of these tokens to the LLM. Also, too much irrelevant context can "confuse" the model about the task at hand
My main gripe with Cursor is that they put MCP usage behind a paywall and their support for the protocol is weaker than Continue's in meaningful ways[1]. Given the big community push on MCP development, I'm a bit annoyed that Cursor monetizes OSS work with a paywall...

[1]https://modelcontextprotocol.io/clients

Congrats on the launch HN!

I've been following the IDE + LLM space. What's Continue's differentiator vs GitHub Copilot, Cursor, Cline, Claude Desktop, etc. ?

What are you looking to build over the next year?

dbt needs to play catch up against SQLMesh feature-wise, so they bought their other competitor SDF. SQLMesh seems to have more development velocity, and dbt will need to execute a smooth transition and integration to catch up.

For context, the team behind SQLMesh also develops SQLGlot, which powers the features dbt attempts to implement

This looks like an interesting tool to log data pipeline runs. Is it closer to an Airflow ETL dashboard or a MLFlow experiment manager? Who's supposed to manage it?
That makes sense. I'll look into it more!

One project was to create a layer between external APIs (OpenAI, Gemini, Claude) and our team. This way we can manage the cost of API calls, try different providers, and log API usage to find out what works / what doesn't.

How does burr scale with an application and support multiple users?

There's a LOT of LLM frameworks out there. The "bring your own framework" adds flexibility instead of endless integrations (like in ML libraries and MLOps tooling).

If I'm getting started with LLMs or have 1-2 POC deployed at my company what's the benefit of adding burr to my stack?