TiddlyInstall appears to be unrelated (and not by the same author?), and serving a very different function. The only similarity is the "single HTML file" part?
---
Previous discussions on TiddlyWiki for context:
Why every programmer should have a Tiddlywiki (2011)
I love tiddlywiki. It even has plugins for AES encryption. So you can save all your notes encrypted every time. It's really interesting to consider the design of tiddlywiki as a decentralized application system. I think the design could work well for a lot of things. As it is built around users managing their own data.
> TiddlyInstall is a single HTML file. This page carries the installer programs, the runtime catalogue, and all the code that makes, edits and signs them
The neat thing about TiddlyWiki was that you could edit the page (WYSIWYG, I think?) and it would save the modified HTML. At least, it did 20 years ago.
Author here: TiddlyInstall is based on the observation that programs written for runtimes like Python, Nodejs, and so on -- all follow the same basic requirements for installation. In that they need the runtime installed and then the software. Existing approaches like freezing apps and trying to port it to an exe directly usually break.
The simplest possible approach is to install the runtime on a users computer, isolated, with the package inside it. It's wasteful in terms of disk space but can be generalised to build reusable installers that work for just about any language.
Reusable installers
A reusable installer reads its own meta-data to decide on the runtime to install and the project. Linux and Mac are supported through a bash script while all windows versions use an NSIS program. The unusual thing about this is once you build the reusable installers since it's just meta-data everything can be edited inside the browser without compiling new binaries.
Like TiddlyWiki: You can actually download this entire system as a single HTML page to create installers without ever needing a server. Though Github project support only works online at the moment (not a limitation, haven't added this feature yet.)
Edit: thanks HNers for checking out my project. I am actually really excited about it. I know installers aren't exactly the most exciting topic but I thought you would all appreciate a different take on it.
Although I don't fully agree. With AI the focus now shifts a lot from 'being able to implement the thing' to 'being able to know/figure out exactly what you want the thing to look like', also against the siren song of the LLMs constantly having ideas of what you could _also_ build.
So, I agree that it's not a direct translation from 'pro pizza cook' is 'pro pizza orderer', but I'm quite sure the pro pizza cooks would generally have the better pizzas in the end, if there were infinite variations of pizza on the menu and the ordering agent was constantly trying to upsell you on slathering on another layer of sauce and cheese.
Which is not to say there aren't exceptions in both directions, because as a pro pizza connoisseur you could even be better positioned to 'know exactly what you want', without ever having made a pizza; and as a pro cook you might still be overwhelmed by the ordering process.
Though all of this still assumes of course that we are in the 'complexity region' where we can expect the LLM/the pizza agent to deliver _exactly_ what we wanted, without issue. As soon as we leave that region (though it expands every week, it feels like), the skills of the cook are still very much necessary to make sure things are as they should be, and fix them if not.
there are two completely different activities both called "writing software using AI".
At work, for important stuff, I go through every change, mine or LLM's, edit it, clean it up, rewrite, until I am satisfied with the code.
At home and for one-off unimportant stuff I go "computer make me sandwich".
The second is nothing like the first. No amount of design and pro I have can make sure stuff coming out of LLM actually implements it, as opposed to me be tricked into thinking it does. After all, that is exactly the training criteria – convince the "eval" that task is accomplished, isn't it?
And the second kind of activity takes skill that no amount of "I am pro dev" can compensate.
I don't really get the hate. Everyone here uses AI for just about everything yet they almost never credit the models. I think this misrepresents the work so I credited the tools. But I'm by no means just handing a blank script to a model and hoping for the best.
This project is inspired by TiddlyWiki, which is from 2004, and apparently still going!
https://tiddlywiki.com/
TiddlyInstall appears to be unrelated (and not by the same author?), and serving a very different function. The only similarity is the "single HTML file" part?
---
Previous discussions on TiddlyWiki for context:
Why every programmer should have a Tiddlywiki (2011)
https://news.ycombinator.com/item?id=2913670
TiddlyWiki: a local, single-page wiki (2017)
https://news.ycombinator.com/item?id=13523660
Tiddlywiki – A non-linear personal web notebook (2018)
https://news.ycombinator.com/item?id=18107271
I love tiddlywiki. It even has plugins for AES encryption. So you can save all your notes encrypted every time. It's really interesting to consider the design of tiddlywiki as a decentralized application system. I think the design could work well for a lot of things. As it is built around users managing their own data.
So basically Nix, but more user-friendly except WAY more limited?
> One file, like TiddlyWiki
> TiddlyInstall is a single HTML file. This page carries the installer programs, the runtime catalogue, and all the code that makes, edits and signs them
Click save page, buddy.
The neat thing about TiddlyWiki was that you could edit the page (WYSIWYG, I think?) and it would save the modified HTML. At least, it did 20 years ago.
Author here: TiddlyInstall is based on the observation that programs written for runtimes like Python, Nodejs, and so on -- all follow the same basic requirements for installation. In that they need the runtime installed and then the software. Existing approaches like freezing apps and trying to port it to an exe directly usually break.
The simplest possible approach is to install the runtime on a users computer, isolated, with the package inside it. It's wasteful in terms of disk space but can be generalised to build reusable installers that work for just about any language.
Reusable installers
A reusable installer reads its own meta-data to decide on the runtime to install and the project. Linux and Mac are supported through a bash script while all windows versions use an NSIS program. The unusual thing about this is once you build the reusable installers since it's just meta-data everything can be edited inside the browser without compiling new binaries.
Like TiddlyWiki: You can actually download this entire system as a single HTML page to create installers without ever needing a server. Though Github project support only works online at the moment (not a limitation, haven't added this feature yet.)
Edit: thanks HNers for checking out my project. I am actually really excited about it. I know installers aren't exactly the most exciting topic but I thought you would all appreciate a different take on it.
> Designed by Matthew Roberts and implemented by Claude. If you like this software why not hire me?
Why would I hire you when I could pay $17 for a Claude subscription?
Yeah I agree with that sentiment. The readme is also just a run on mess as well, so it doesn't not inspire confidence.
If you can do x and a pro can do x better who do you think can do it better using AI?
so basically pro cooks are better at ordering pizza?
"pro"-ness does not transfer. I can be both good at reading english, and have AI hallucinate non-existing details from the documents it processes.
Ohh I'm gonna steal that one.
Although I don't fully agree. With AI the focus now shifts a lot from 'being able to implement the thing' to 'being able to know/figure out exactly what you want the thing to look like', also against the siren song of the LLMs constantly having ideas of what you could _also_ build.
So, I agree that it's not a direct translation from 'pro pizza cook' is 'pro pizza orderer', but I'm quite sure the pro pizza cooks would generally have the better pizzas in the end, if there were infinite variations of pizza on the menu and the ordering agent was constantly trying to upsell you on slathering on another layer of sauce and cheese.
Which is not to say there aren't exceptions in both directions, because as a pro pizza connoisseur you could even be better positioned to 'know exactly what you want', without ever having made a pizza; and as a pro cook you might still be overwhelmed by the ordering process.
Though all of this still assumes of course that we are in the 'complexity region' where we can expect the LLM/the pizza agent to deliver _exactly_ what we wanted, without issue. As soon as we leave that region (though it expands every week, it feels like), the skills of the cook are still very much necessary to make sure things are as they should be, and fix them if not.
there are two completely different activities both called "writing software using AI".
At work, for important stuff, I go through every change, mine or LLM's, edit it, clean it up, rewrite, until I am satisfied with the code.
At home and for one-off unimportant stuff I go "computer make me sandwich".
The second is nothing like the first. No amount of design and pro I have can make sure stuff coming out of LLM actually implements it, as opposed to me be tricked into thinking it does. After all, that is exactly the training criteria – convince the "eval" that task is accomplished, isn't it?
And the second kind of activity takes skill that no amount of "I am pro dev" can compensate.
Perhaps so you could spend more time with your loved ones, or doing literally anything else, instead of talking to a machine? Just saying.
By "design" I mean choosing how the software works and the engineering trade-offs. I didn't type "make install system" into claude. I designed how the software worked based on my past research and my engineering knowledge. This included the original NSIS code I wrote: https://github.com/robertsdotpm/win-auto-py3 and file-based meta-data data here: https://robertsdotpm.github.io/software_engineering/program_...
I don't really get the hate. Everyone here uses AI for just about everything yet they almost never credit the models. I think this misrepresents the work so I credited the tools. But I'm by no means just handing a blank script to a model and hoping for the best.