Pretty clear these days that the bottlenecks in technology manufacturing are now weaponising their monopolies/duopolies / triopolies.
They’ve become the trolls under the bridge and will squeeze every passerby for every dollar they’ve got.
The days of cheap computing have been in decline and are now dead, replaced with giga profits for this companies who managed to the the indispensable links in a chain with no or minimal competition.
They allow me to do work I could never have done before.
But there’s no chance at all of an LLM one shotting anything that I aim to build.
Every single step in the process is an intensely human grind trying to understand the LLM and coax it to make the thing I have in mind.
The people who are panicking aren’t using this stuff in depth. If they were, then they would have no anxiety at all.
If only the LLM was smart enough to write the software. I wish it could. It can’t, nor even close.
As for web browsers built in a few hours. No. No LLM is coming anywhere new at building a web browser in a few hours. Unless your talking about some super simple super minimal toy with some of the surface appearance of a web browser.
This has been my experience. I tend to use chats, in a synchronous, single-threaded manner, as opposed to agents, in an asynchronous way. That’s because I think of the LLM as a “know-it-all smartass personal assistant”; not an “employee replacement.”
I just enjoy writing my own software. If I have a tool that will help me to lubricate the tight bits, I’ll use it.
Same. I hit Tab a lot because even though the system doesn't actually understand what it's doing, it's really good at following patterns. Takes off the mental load of checking syntax.
Occasionally of course it's way off, in which case I have to tell it to stfu ("snooze").
Also it's great at presenting someone else's knowledge, as it doesn't actually know facts - just what token should come after a sequence of others. The other day I just pasted an error message from a system that I wasn't familiar with and it explained in detail what the problem was and how to solve it - brilliant, just what I wanted.
> The other day I just pasted an error message from a system that I wasn't familiar with and it explained in detail what the problem was and how to solve it
That’s probably the single most valuable aspect, for me.
Let’s say you want to indicate some relationship between part of your code and a specific requirement number. And you want to be able to do that in any computer language. At the moment the only way you can do anything like that at all is to put your reference in a comment And LLMs are extremely well known for deleting comments. So we need some sort of standard for meta-data or comments that LLMs are guaranteed not to delete and that is consistent across all programming languages. How can we do? This is a new standard needed? Can we piggyback onto some existing mechanism?
Satya has succeeded greatly in transitioning Microsoft to the cloud and open source coexistence. These successes alone make him one of Microsoft’s great leaders.
Satya has also completely fucked Windows and has been unable/unwilling/uninterested in making it a product that people love. Satya has led to people hating Windows.
Both things can be true. Paul Graham (bless the maker and his passing) declared MSFT dead, but it came back strong.
The copilot fiasco is classic Microsoft Bob territory. They have so much entrenched tech in govt and big corps they can shamble on for years.
Typical Microsoft, not great, not the worst but the overpriced fast casual hamburger of the tech world. An ok meal, somewhat disappointing leaving you feeling greasy with a feeling you paid too much but you know you’ll be back.
>> “ we want to be clear about what this means for customers.”
Nope, not clear.
This is a clear message “ the heroku product is cancelled but will not be shut down, will continue to operate exactly as before but no new features will be added.”
Took photos of the fridge and pantry and asked the AI to identity all the food and create recipes.
reply