Hello creator of the project. This entire project was motivated by another HN post where someone was asking where we can get information about architectures and system designs across our industry. Seeing there wasn't anything available, I decided to start it; glad to see its resonated with the community.<p>I hope to create high quality posts from engineers that work on these systems and the challenges they are trying to solve. Really dig into the problems and the technologies and strategies they used to solve it.<p>What they have learned? Where they have failed?<p>I also plan to write up deep technical dives on technologies we all rely on and use everyday.<p>If there is any feedback to improve or you have something you want to write up with me please reach out.
This is the first issue of the new Architecture Notes publication (website and newsletter) run by Mahdi Yusuf.<p>Mahdi interviewed me about the architecture of <a href="https://datasette.io/" rel="nofollow">https://datasette.io/</a> - topics we covered include:<p>- Building a modern Python app using ASGI<p>- Benefits of SQLite<p>- Designing plugin hooks<p>- Safely allowing SQL injection<p>- Using SQLite from asyncio<p>- The Baked Data architectural pattern<p>- Bundling a Python web application in Electron<p>- Packaging a Python for WebAssembly
Here is the sketch note of the entire post as well.<p><a href="https://architecturenotes.co/content/images/size/w1600/2022/05/Datasette-FIN-01-2.jpg" rel="nofollow">https://architecturenotes.co/content/images/size/w1600/2022/...</a>
Datasette is pretty cool.<p>But AFAICT, it just doesn’t scale whatsoever. That SQLite db is both the dataset index and the dataset content combined, right? So you're limited by how big that SQLite db can realistically be. The docs say "share data of any shape or any size", but AFAICT it can't handle large datasets containing large unstructured data like images and video and multi-billion data point datasets are hard to store in a single machine/file.<p>Not really a criticism, but more wondering if there are scale optimizations in Datasette I'm not aware of since the docs do say any shape or size.
You can load your Stripe account into a Datasette instance using this:<p><a href="https://github.com/tabledog/datasette-stripe" rel="nofollow">https://github.com/tabledog/datasette-stripe</a>
I really like the design of the website, you earned my sub!<p>I just have one small suggestion:<p>I think it would make it better if the subscribe button disappears after you scroll down for a whole page instead of having it sticky.<p>Other than that, great work in terms of the article and the website! looking forward to read more
Is it possible to add an Excel-like spreadsheet data <i>entry</i> mechanism for ad-hoc data analysis (rather than receiving datasets from elsewhere)? If not, how should I create datasets?
I have a hard time believing people feel comfortable using Python applications in production for anything other than prototyping. I have seen some shit over my career, across many kinds of interpreted languages, that will never let me approve of that.<p>Unless one simply doesn't care about runtime quality.