Claude argued with me about the quadratic equation the other day. It vehemently felt a -c was required whereas a c was the correct answer. I pointed this out showing step by step and it finally agreed. I tried Grok to see if it could get it right. Nope, the exact same response as Claude, but Grok never backed down; even after the step by step explanation of the maths.
My friend had a cute baby boy and mentioned difficulty in finding children's storybooks in Spanish.
Challenge accepted:
I built an AI generated multilingual storybook, just to see if it would work.
Tap or click the little monster to have it read to you.
Local LLM generated the story, stable-diffusion generated the images, AI converted text to speech in two languages: English and Spanish ( could easily do many more languages ).
I "filled the app out" by adding a simple landing page placeholder, login page and "library" page.
Not very phone friendly, was made for her iPad.
Just click login to move on, as it is currently not connected to a backend.
Only the second book currently has a story, the others are placeholder templates.
Cool. This concept is useful for adult language learners also. Depending on the language you are learning it can be very difficult to find reading material in the age 5 - 10 range to practice with.
Only if your third grade math teacher has claim to your algebraic code comparisons. Current trends are iterative training with positive first model, negative second model and let the AI 'scale' the learning from there.
The clear declarative: "Extract this to a separate function."
Everything thing else can be misinterpreted as a suggestion, dig, ect.
Actually an interview question to work with us. We need code, not egos. Write and be wrong, correct and continue. Do not sweat errors. None of us are perfect, and neither is this review.
I'm tired of people always jumping on the latest technological hype-train. O.K Typing is great but I used Speech successfully for decades before it was deprecated and I'm not going to jump on the latest trend and change my whole workflow and tooling just because young devs want to add the cool new tech like "Typing" to their resumes.
Words are semantic units, they are by definition and purpose: meaningful.
Speech is also:
* Synchronous
* A faster form of communication than the fastest typers
* A much much faster form of communication than the average typer
* Has incredibly good built-in features for communication like:
* The ability to see who's currently talking
* The ability to communicate a thought almost continuously rather than discretely
* The ability to be interrupted to provide new information when necessary
* Built-in styling features like tone, feeling, volume to emphasize important information
* Built-in synchronous next-speaker negotiation
* Privacy respecting and non-binding which is incredibly important in heavily regulated industries where everything recorded must be very carefully thought out because it might come up in discovery 10 years later
This is an absolutely horrendous claim. Every person is different, there is absolutely no replacement for face to face talk for me and my current co-workers. I need it, if my company decides it'll be replaced with written-only communication, I'll leave them in a second. Seeing people's facial expressions, hearing their tone, understanding how they react to spontaneous questions, what frustrates them, what things they feel confident about, what things they're puzzled about... There is no easy way to express these via text.
Please understand that I have no problem with remote work; I'm only responding to your implication that face to face meetings are "waste of time".
> Words are meaningless, easily misinterpreted and non binding.
I wish more people recognized this, instead of expecting everyone else to remember "that thing that we talked about, like, last week".
I imagine it should be possible to create an organization where all communication would be just specification and documentation. I wonder if there is anything that can't in principle be done that way.
It is nice where it works out. Some organizations have staff working manual labor who are not as literate or articulate with written communication. Some organizations are competing at a pace where a rival organization can beat them with a well run meeting over waiting for emails to be written, read, and replied to.
I think that could work in an environment where everything could be cleanly laid out and understood in advance, without room for error or confusion or misalignment. I think it is a lot easier to burn through misunderstandings with face to face or otherwise synchronous speech.
Detailed note taking is a good remedy for the whole "words only last as long as your throat cords vibrate" thing.
Although now that I think about it we're probably pretty close to agreeing, I'm really just pointing out the value of regular old talking as a good escape for when text based communications slow down or fail but implicitly I'm assuming a world where the default is text.. which is what it sounds like you'd prefer.
I mean in principle almost any form of communication is possible, but there's a reason you don't go to dinner with a friend and only communicate through post-it notes: because it is unnatural and inefficient.
I am a fan boy and did not find any of it offensive. Great write up, actually.
Windows simply does a better job at window management. I use full screen and three finger swipe between 5 separate desktops for work to overcome this weakness.
The flash drive did make me giggle... oh you poor windows laden idiot. Flash drives... like from 1999?
Most of your issues are just growing pains. I did it when I migrated about 10 years ago from my LeapFrog: no carry handle, screen too bright, no built in songs, ect.
