previous Show HN post submitted by the author (109 comments) - https://news.ycombinator.com/item?id=41224689

> You can shard your SQLite tables into multiple database files and query across all of them from a single connection.

You mean using ATTACH statement, right? If you use WAL mode, then you cannot get transaction safety / ACID with ATTACH [0]

> If the main database is ":memory:" or if the journal_mode is WAL, then transactions continue to be atomic within each individual database file. But if the host computer crashes in the middle of a COMMIT where two or more database files are updated, some of those files might get the changes where others might not.

Moreover, ATTACH do not support more than 125 databases, so that limits the shards to 125. [1]

ATTACH does not solve the concurrency problems. That's why SQLite also has a BEGIN CONCURRENT experimental branch

[0] - https://www.sqlite.org/lang_attach.html

[1] - https://www.sqlite.org/limits.html

I would argue that BEGIN CONCURRENT doesn't solve the concurrency problems either, which is why SQLite also has an HCTree experimental branch :P

how does collaboration works for SQLite, since the db is embedded?

When you upload a database to the site, it is stored in your browser's memory or uses OpFS — a local storage within your browser.

You can share a link to grant access to your database, with the connection handled via P2P through WebRTC.

Uhm, by allowing multiple people to connect to the same database through a webapp like this?

I have an old 2019 MBP, now I am tempted to try Pop on it. Does the external displays work?

PopOS (just as Ubuntu) comes with a live-linux for its install media -i.e. you can try with the very same installer usb-stick if the system works for you. In generall I'd say the older a system, the better the likelihood of linux compat (if it is not WAY to old).

On that 2014 MBP Retina, I have attached a 4K TV via HDMI. It works in dual-screen, even though I use it with the lid closed in single-screen mode (4k TV only), but only 30Hz are supported (I can run 1080@60). Limitation of the Intel onboard GPU I assume.

You probably have your reasons why you do not want macOS on that system anymore - for me the 2014 MBP fell long out of macOS support and while I had Sonoma with Opencore Legacy Patcher running on it, the OS was just unbearably slow, plus some audio issues (along the fact that with opencore legacy patcher your security is also at risk). So that was a no brainer, because macOS just wasn't an option anymore.

Another word of warning: I had the very same 2019 Intel MBP and it died just a couple of weeks after it fell out of Apple Care. Just turned of right while using, never came back. That series is notorious for having thermal issues, and a friend of mine had the same model dying the same way just a couple of weeks after. Maybe you want to sell it while it still has macOS support (higher prices on the 2nd hand market) and get a different laptop if you are after Linux.

You can try using DP adapter/cable: https://web.archive.org/web/20160708162656/https://support.a...

I just tried to put omarchy/arch on my 16” mbp. Everything worked (eventually… speakers and keyboard backlight needed special stuff) except for suspend or hibernate. After about a week, I gave up and put Monterey on it.

> I've been working on something similar... although with slightly larger scope (intended to be used within containers/sandboxes) https://github.com/andrewbaxter/passworth

> stored in encrypted sqlite3

you had me at encrypted sqlite3. it would be great if you mention in readme that it uses SQLCipher

any opinion on how does this compare with ente?

I find this interesting. This is probably the first time I am seeing a large-ish project ban AI outright

Does substack let you add canonical URLs for the post? Otherwise won't you penalised by search engines for duplicated content?

I decided I am willing to take that risk.

This is cool! The code base is small to follow easily and also TIL ggez.

In the single player mode, the computer is not so intelligent. So initially i just kept winning without understanding what was happening.

Is it possible to create a filter that can work over a complex join operation?

That's what IVM systems like Noria can do. With application + cache, the application stores the final result in the cache. So, with these new IVM systems, you get that precomputed data directly from the database.

Views in Postgres are not materialized right? so every small delta would require refresh of entire view.