I like the fax machine idea. Reminds me of an idea I had. Get some receipt printers for my friends and we can print to each other's printers to send text messages
That brings back memories. I once built an XHTML page served over WAP 2.0 (on 2G networks) from my home server that could send messages directly to the printer in my mother's office. It was incredibly clunky, but a lot of fun. I forgot to tell her before I tried it, so random pages just started coming out of the printer one day.
I was thinking about this the other day. Now that software is within reach of most idea makers, it opens the door for a much deeper level of tinkering. With very affordable, if not slow, 3d printers, and an abundance of hardware interfaces, I think we are going to see some really great weekend projects that will turn into beautiful, "where has this been until now" utility for the world!
I'm excited to see software engineers and teams morph into the next stage of product builders!
My little brother is a beach lifeguard but in the last year he’s pumped out so many incredible projects. It feels like he’s been unleashed. Such a cool era!
> ...the focus of hackathons has completely shifted away from typing code...
> ...iterating on intricacies of implementation with radical refactors has become a trivial task...
The irony is unreal. Where's the hardware?
Since the advent of SBCs and microcontroller kits, software devs have felt the same way about hardware being trivial. Yet, a hardware engineer still makes a massive difference in the outcome of the project.
It's the rotary phone and the raspberry pi, of course. Don't gatekeep.
The fact that microcontrollers are so cheap now means for most (but, sure, not all) applications they're strictly superior in every way compared to e.g. 555 timers and LM386 amplifiers, or whatever. This is because, critically, you can debug and reprogram a micro. To do the equivalent with a 555 timer means, at minimum, de-soldering a bunch of components and probably poking around with a logic analyzer or an oscilloscope.
What's more, you can get a full tcp/ip stack in a surprisingly small and low-power package these days. No need to futz with analog telemetry, or even SPI/I2C unless you really need to.
The "hack" in TFHackathon is altering the function of a phone. Who cares if they used a ras pi to do it vs something else? In what possible way does that diminish their feat?
Someone was bragging to me about their new AI startup a few months ago. I went to look at their website. It was some AI slop. I checked out the code for their form to register interest for the launch… it wasn’t setup at all. It was just a form that went nowhere. They had an idea, told AI to make a site about it, didn’t bother replacing the boilerplate to make it work, hosted it, and then started bragging to their friends about how they were going to be rich.
there's an Ice cream shop in my city that has obviously generated all their signage and menus and everything else, even on their website the photos section has AI generated people smiling and happy next to the sign.
I get that it's a small business that has probably saved a couple of G's on design/code but it's all so sloppy and obvious.
I'm excited to see software engineers and teams morph into the next stage of product builders!
Its a fantastic deal for management if you can find people gullible enough. But a raw deal for the worker bees themselves
> ...the focus of hackathons has completely shifted away from typing code...
> ...iterating on intricacies of implementation with radical refactors has become a trivial task...
The irony is unreal. Where's the hardware?
Since the advent of SBCs and microcontroller kits, software devs have felt the same way about hardware being trivial. Yet, a hardware engineer still makes a massive difference in the outcome of the project.
It's the rotary phone and the raspberry pi, of course. Don't gatekeep.
The fact that microcontrollers are so cheap now means for most (but, sure, not all) applications they're strictly superior in every way compared to e.g. 555 timers and LM386 amplifiers, or whatever. This is because, critically, you can debug and reprogram a micro. To do the equivalent with a 555 timer means, at minimum, de-soldering a bunch of components and probably poking around with a logic analyzer or an oscilloscope.
What's more, you can get a full tcp/ip stack in a surprisingly small and low-power package these days. No need to futz with analog telemetry, or even SPI/I2C unless you really need to.
The "hack" in TFHackathon is altering the function of a phone. Who cares if they used a ras pi to do it vs something else? In what possible way does that diminish their feat?
"and then this happened, then this happened, then this feature, then this feature"
Wow that's crazy...
What a joke.