Commentary on the first few paragraphs of
his server's main README file:
Quote:
This website must have a structure such that it's ready to be compiled into our Operating System, and such that it's easy to include as little code as possible. We will need to create ordinal index text files and index source code files to create from the tiniest to the fullest system.
Yes, let's compile a website into an operating system. What does that even mean? Increasing the size of the operating system by including a local mirror of a disorganised pile of meaningless crap? Sure it's easy to include as little code as possible - let's just... not include any code! (That was really easy, and that is the absolute smallest amount of code possible.) I don't even know what the last sentence means.
Quote:
It must be able to compile even if it's fully casual in terms of the style of the code written, the documentation, and the overall directory structure. It would be like an OS distribution from which it's extremely easy to learn from.
I'm going to assume that "casual in terms of the style of the code written" refers to whitespacing, not to the actual code, in which case that's what preprocessors are for (if the latter, then by basic laws of computing that's not a suitable way to write code because it's impossible for a machine like a computer to understand). What the second sentence means, I have no idea.
Quote:
It will be like a set of full education courses, real world programs to compile and use as portable programs, a generic history of all of our activities, user help files and redundant pointers but preferably inclusions of full external resources containing only what we have exactly used in practice, all in one same place.
I think this is trying to be a description of his server. I'm a bit concerned about the redundant pointers though...
Quote:
We shouldn't normally include index.html files but _index.html to show the full directory listing always.
At least now I know why his server is just a messy tree of directory listings. Nevertheless, index.html is a standard for a reason and breaking that standard because you can't figure out the proper way to do something is a Bad Idea. While we're at it, let's write a replacement for HTTP that's portable, casual in terms of the style of the response returned, and can be compiled into our Operating System.
Quote:
If we are using or improving a given program, let's include it fully in its portable format to download and to inspect completely online, with source code and with reverse engineered versions available, no matter if it's open source. We need it to reach the full scope of computing.
I sincerely hope a DCMA takedown request is enough to get this tragic mess of a server off the internet. I'm sure that will help us to "reach the full scope of computing".
Quote:
Include a download index for that too.
index.html? That's typically how you include an index. What is a "download index" anyway?
Quote:
It all will be searchable. We will probably need to prefer direct decompression of PDFs to search text on them, or create full TXT versions.
Except that there's actually a standard to include real text in PDFs, which can be searched with any PDF reader and converted to plain text easily using a utility such as pdf2txt if you really have to use grep on a PDF file.
I wonder what his server configuration file looks like...