You underestimate the task by multiple orders of magnitude. It would be a significant effort even to build it on libreoffice.

> And I mean starting from scratch, not using the messy Libreoffice code base.

Given the libreoffice codebase is in its dataset, is that really from scratch?

I guess that means none of what a professional software engineer does is from scratch either, given that they're also trained on existing code.

Kind of hard to write code when you haven't seen a lot of it.

The actual codebase of libreoffice is in the dataset, and a model is designed to replicate plausibility from its dataset

I think I need to write a full blog post on this.

While I think there are better introductions to "first principles thinking" than this post, the idea of "...When I step back and ask what we’re actually trying to do, why it matters, and how the pieces connect, I usually find more ways forward than I expected."

That's an important skill to have.

However, I think sometimes we overvalue thinking from first principles when it isn't warranted. I've seen times in my career where first principles thinking led to a solution that ignored key non-technical constraints. (For example, it would require a full re-architecture of the system and require deferring all feature work for a year. Another example: the proposed solution breaks Conways law in a way that would require a reorg that would break other organizational constraints.)

Sometimes, we need to recognize our constraints. Spend the time to question them when appropriate, but realize that there are other tools that are more appropriate in some cases -- such as anthropological thinking.

You apply first principles thinking, but also be mindful of your role in the team.

When you say something isn't warranted because it takes too much time, it isn't valuable: you're still thinking about what it is you're doing (working), what tools you have (your team, people), to create value (your product or service).

It's the measurement of value part that's tricky. Who determines value? Sometimes it's the customer, sometimes it's just you.

It's about perspective, and seeing where you fit in the picture. Apply the principles from there. Nobody says you have to be a jerk about it, that's up to you and how you wish to apply the output of the first principled thinking.

> However, I think sometimes we overvalue thinking from first principles

The problem isn't the thinking though, but the lack of grounding.

You don't have a problem with people coming up with a technically better solution. Your problem is with the people not realizing that they exist in reality and not in a vacuum.

Please don't question the act of thinking itself. It is the wrong target.

Thinking is great. I am arguing that “first principles “ is one tool - and it may not deserve the primacy that its advocates propose.

I mean that's always been the paradox that good devs need to live in, and I often go backwards and forwards on the scale between technically pure, and straight up problem solving.

Fundamentally software we create exists to solve problems, if bad code solves the problem is it bad code? Counter to that is we are engineers and it's our job to design systems that mean problems can be solved safely and effectively.

Like most things in life the truth lies in the middle

Zipcar started in 2000/2001 on Oracle and relied heavily on stored procedures, particularly in billing (as SaaS didn't exist back when it started and was a bad fit for a long time). It took an entire rewrite of the system to get off of Oracle.

I've heard the exact same thing about DB2, Sybase and Terradata over the years. It's not some uniquely Oracle issue.

I don't know about it being pretty nice - I just think of Nasubi and what happened to him on Susunu! Denpa Shōnen.

Or, alternatively, I was a lot more willing to put in the time to self host before I had more money and less time. If it's truly important it's not in public. If it's public, I don't have time to keep up with security patches if I'm not being paid.

For those on plus, are you seeing astra limited to medium? Considering the rate limits that probably makes sense, but I'm wondering what different groups have access to.

Consider that if this was ran from your machine - it would have all of your memories and might be relying on the past 2 months. I'd be curious what would happen if it was run independently.

(I am assuming they didn't train on your prompts, which is always a worry)

This isn't talking about bespoke software, though. This is talking about upstreaming contributions to others. This puts a low barrier to create work and a lot more work for unpaid volunteers.

They're also fantastic for small measures for baking and cooking when it matters.

Curious: What do you make where you need such accurate measurements?

I just eyeball everything when I cook. (For other things that I'm passionate about I'm much more careful.)

There's a few modernist cooking tricks that require small amounts - tenths of a gram - for big impacts. Sodium citrate can make any cheese melt beautifully into a smooth sauce, using something like 1% of the weight of the cheese.

Any juice can take the roll of lime or lemon (in cocktails or baking) by adding a bit of citric or malic acid. Orange juice is 1% citric acid, so add 3g citric and 2g malic per 100g juice and now you can make an orange daiquiri or "key" orange pie.

Emulsifiers get goopy if you use a lot, but can make a smooth syrup in small amounts. You can make a nice syrup from any nut milk with a small amount of xanthan and gum arabic.

There's a wide array of ingredients including xanthan and gum arabic used in tiny amounts in gluten free baking

I sometimes wish my scale was precise to below 1g when measuring yeast for long-rising (bread) dough. A recipe may ask for 1g, and adding 2g will make a huge difference.

Baking often rewards precision.

I'm also an eyeballer. I'm a decent cook but a truly lousy baker.

Somewhere there’s probably a coffee nerd who wants to measure small fractions of a bean.

It depends on what you're trying to do. For non-coding tasks luna is quite often enough. Flash models are more than enough for summarizing a text, for example, or whipping up a small script to save me fifteen minutes. If you're on a 200/month plan, I see your point. If you're on a dollar limit - or worse, paying per token out of your pocket - you look to be more efficient.