i know a lot of people do this thing where they"have an idea" and then theyre like,"if someone can just code this it would be so cool!"

this is going to sound like one of those. the difference is that when i have an idea like this, its usually based in part on some things ive already worked on, its probably something i could code myself if i devoted a couple years to it. and i talk about it to inspire other ideas.

in the 1980s (as a kid, and a basic programmer) i was impressed with eeproms. i had two ideas about these things:

1. we should make these into hard drives. you could carry all your files on a chip! (obviously, this is basically what happened.)

2. they shoud make a"programmable cpu" that can be different kinds of cpu-- depending on how you program it, it could be an apple (motorola) or an ibm (intel) or even something else!

when they put an fpga on the cover of scentific american years later, i was thrilled. i never thought they would do it, i was sure that every cpu was too proprietary for it to ever be an actual product.

for whatever reason i thought that was a good idea, i also want fpga-like programming languages (i know, it sounds like i dont understand what programming is, because all programming languages are programmable) and fpga-like gnu/linux distros.

what i have is this thing i made called mkfigos. and what it is, is a pretty straightfoward remaster script. it can turn refracta into fig os, or librepup into fig os, or antix into fig os.

but what else i can do is, if you for example had puppy tahr, and you said"i want this to be more like antix" if you tell me What You Mean by"more like antix" perhaps some compromise can be reached. and mixing distros together is almost more fun than designing them.

i understand the caveats of binaries from different sources. when i run a debian system i never mix stable with other branches, but i will sometimes copy mtpaint or gdmap from somewhere if for some reason installing properly isnt convenient.

you can make a"pretty good" c64 core for an fpga, and have a"pretty good" c64 clone on a modern fpga chip.

my thought is, given that we know what makes antix"antix," you could make a pretty good antix clone that fit on a distro called... how about"fpga-os?" ...technically"fpos" would be more accurate, but thats like naming your boy sue, youre just asking for name-calling.

so fpga-os would be a sort of"core distro" but you could customise it using"plugin scripts" (kind of like mkfigos.)

and depending on what plugin script you used, you would get an antix clone or a puppy clone or a refracta clone or a fig os clone...

youd have a base distro and all these heavy modifications-in-a-script options.

but you wouldnt"choose 100 options," you would choose *a* script."um, i want the antix 16 clone. or the antix 17 clone"

and the script would make it as close to that distro/version/flavour as reasonably possible.

perhaps best of all, instead of mucking about with creating distros, you could change a distro just by editing the plugin script. these scripts could be a lot easier to work with than tweaking distros by hand all day.

thats my idea for an"fpga" distro. its doable. ive done enough actual usable stuff to tell you how doable it is.

but i didnt think there would ever be actual fpgas, i just thought they were a really, really cool idea.

do whatever you want with the idea-- feel free to copy/use/quote this in whole or in part.

in the link is a quick animated demo of my programming language, which translates into python. i use that same language to remix refracta, puppy and antix 16 * 17.

(...and a fair amount of bash, which i find very tedious to use by itself but otherwise priceless.)