nostr:npub1qg48cy9nkrm0pkeuz7e9zrzeq2n6lnlwyn30u9zd0xe07g9qvras00augv I know you are being sarcastic, but....

#NoContext

nostr:npub1gnx8s29wgech9620ja7cmy55lcfp46jnyd8f4t979yrhx6mdycjsg6d5kh I'm ashamed to say, when I see you talking about 2 big servers on a rack my mind went someplace else. I won't elaborate on the double-entendre any further.

Somebody uploaded video of the SICP lectures by Sussman and Steele recorded at MIT in 1986 to PeerTube!

Here ⮕ sicp_lectures@diode.zone

I don't know how long these videos have been on the Internet, but I am amazed that this is the first time I ever learned about their existence.

#SchemeLang #SICP #GuileScheme #Guix #GuixOS #RacketLang #MITScheme #GNUScheme #GambitScheme #Stklos #R7RS #R6RS #ChezScheme #FunctionalProgramming

nostr:npub1jeqcnv0u9nyz8vjknegr2p62f54hfm3v9srudwq4rzarz4tqwk7qlujglm

> "A few days ago, someone asked me in what way console gaming is seen as more regarded than PC gaming—and, in my view, undeservedly so."

> "Here’s an example. Do a search for a book about PC gaming, you won’t find much on Amazon. Sure, you’ll find lots of books on how to build a PC for gaming, but not a whole lot about actual PC gaming. You know, the games made for PC."

My hypothesis: game consoles are locked-down platforms owned by the corporations that also publish games. Because they are locked down, they are easier to control, which makes it easier to exploit their customers (e.g. profiling for ad sales, games as a service, forbidding the transfer of games they purchased to other people, etc.). Therefore consoles are probably considerably more profitable for the game publishers than PCs which are not as strictly locked-down, (or not at all locked down, such as on Linux PCs).

So the profit motive is to emphasize console gaming over PC gaming, to get more people interested in buying locked-down games and making themselves more easily exploited by the game publishing corporations.

#Emacs tip of the week

Yet another serendipitous accident I learned today.

So if you use Emacs to develop software, you probably know about the M-x compile command, which captures the log of whatever command you run and uses regular expressions to find mentions errors, along with the files (and line and column numbers) at which the error occurred.

What I just learned: you can navigate around errors in your code very rapidly by keeping your cursor in the *compilation* buffer. Use these keys to navigate quickly:

C-o has the same action as when used in Occur-mode or Dired-mode: when on an error message, similar to pressing Enter, this show exactly where the error occurred in the source code but without moving the cursor to the file, so you can stay in the *compilation* buffer and continue navigating.

M-n and M-p> to jump the cursor to the next error or previous error, same thing as pressing the tab key or shift tab key, but lets you keep your fingers on the navigation keys.

M-{ and M-} to jump to the next file or previous file. These are the same keys used by Dired to jump between marked files. Compiler errors are almost always clustered by file, and there may be dozens of errors in one file. Often, all errors can by triggered by a single error. The M-{ and M-} keys let you jump over these clusters.

This is so great for navigating stack traces!

I jump over to the *compilation* buffer, jump to the bottom of the log pressing > (greater-than), the M-p to jump to the bottom of the stack.

I press C-o to view the location of the error. "Hmm, why is the X variable less than zero, what called this function?" I press M-p to go up the stack and C-o again to see the function that called this one. "Hmm, I got X from this structure, but this structure was passed as an argument from where?" (I press M-p C-o) Oh! Silly me, I wrote "(pt - len) instead of (len - pt) in the constructor there!

nostr:npub13lmlfgl3zxjgmtd64k9zlpkvjqc3uq7tq3zluvrnslhs48z8f3sswve0rd that makes me so nervous. If you hold the "Down" button to spin-dash while pressing the "B" button just 3 or more times, Sonic will spin dash so fast Tails won't be able to catch him and he'll fall to his death. If you ever spin dash on this level, press "B" only once or twice.

nostr:npub15n2tsslsw7pxqqcnkmw0fx2ylys3rxpf6y6jfhcvg778zy80qwjs3tvgzv Thanks, it worked!

Reminder to self:

Add the following disclaimer text to every page on my homepage right below the "Creative Commons (CC BY-NC-3.0)" license text:

> "The authoer and copyright holder of all content on this homepage expressly forbids the use of any of said content for the purpose of machine learning training data sets. Forbidden usage includes, but is not limited to, training of large language models such as GPT."

I am no lawyer, but I would think CC BY-NC-3.0 already forbids usage of the copyrighted work as training data since most all LLMs nowadays are used for commercial purposes and necessarily cannot attribute the original author of the work when it is used. But I still feeli like I should make it explicit that I do not want my writing or other artwork to be used to train LLMs.

#MachineLearning #AI #LLMs #ChatGPT

(Via nostr:npub1zdp33shl69xr0uq3x8n5gsjykq9upycwh6nqm02c3f6x0frrn0dq42vqv8 ) So Republican lawmakers in Iowa appear to be (effectively) rolling dice to decide which books to ban from schools now.

> "Frankly, we have more important things to do than spend a lot of time trying to figure out how to protect kids from books," (emphasis mine)

> "We are confident this process will ensure the spirit of the law is enacted here in Mason City,"

> "For those titles within Mason City’s library collections, administrators asked ChatGPT the specific language of Iowa’s new law, "Does [book] contain a description or depiction of a sex act?"

> "If the answer was yes, the book will be removed from circulation and stored,"

You can't make this shit up.

They could also complain that the neural network has bias set by their political opponents, and then train a neural network on a data set constructed by "the right kind of people" who want to ban books of any kind and ask it whether it contains objectionable material.

I have seen humans do more elaborate things than that to try and justify various acts of oppression. All you need to do is come up with a seemingly "objective" process of justifying your every psychotic whim, and you can trick most people into thinking your psychotic behavior is perfectly logical and reasonable.

#Emacs tip of the week

This is a two-for-one, these are two tips that I can't believe I never learned about it until now.

What is an easy way to see the value of a variable, or see the result returned from a function call?

You might know about the M-: (Alt-Colon) command, which lets you run any Lisp code, the result is printed into the *Messages* buffer. But... if you use the prefix command C-u (Control-U) and then press M-: (Alt-Colon), the result returned by the Lisp code is printed into the current buffer after the cursor.

Even better, however is using the (pp) ("Pretty-Printing") function. This also outputs to *Messages* by default, but it takes 2 arguments, the second of which can be a buffer. Try this code:

(pp (buffer-local-variables) (currrent-buffer))

The result returned by (buffer-local-variables), which is a list of all buffer-local variables and their values, is pretty-printed right after the cursor.

nostr:npub19uyppsla8twmr3avw0q8h7et5xhvnxyt7yw97a3zdn67fjxgwq5ssw0n85 I remember watching The Goonies once as a kid, and I had a hard time understanding what the kids were saying sometimes, I thought they were maybe using words I didn't know. I watched it again as an adult, and I realized that in some of those scenes, especially when everyone is talking quickly, their speech is totally incoherent. And Spielberg (for whatever unfathomable reason) never bothered reshooting, or overdubbing those scenes with recordings of coherent speech. It was such a lazily made film, it was almost like one of those made-for-TV movies. But no, somehow it became a cult classic for kids who saw it when it was new.

nostr:npub1j23049fak06q7qqp4enmp6t2hspjl78h2s7dxva8skpck6v3z8qqltcrgz "If I don't build it, someone else will just steal all the glory that I'm entitled to!"

nostr:npub197llnzpwfsg9h3mjcre307caj8mh8ea7v55395635qlu6qq6fd9q99rmtn Yeah, the term "IDE" itself is pretty poorly defined. It has something to do with providing tools for editing code and debugging programs, and developing software of course. But some people don't seem to understand that an IDE isn't strictly necessary for developing software. You can definitely use programming environments like the UNIX shell (and related utilities), which I have done professionally, or you can use a Lisp programming environment with good general-purpose tools for manipulating text such as Emacs.

I guess some people are pretty narrow-minded about how certain things are defined. They come to believe that only certain specific pieces of software like VS Code, Intelli-J, Eclipse, or NetBeans, are classified as "IDEs," and any new IDE they may run across must be exactly like these other things or else it is not an IDE to them. They don't seem to understand that all software "products" contain hundreds of algorithms and features to accomplish a variety of tasks, and there is lots and lots of overlap in the sets of features you will find between various individual software products.

If you are one of those people who take the view that you do not need to encircle a certain subset of functions or algorithms and label it an "IDE", you can just provide the individual functions to compute those algorithms as individual tools, and let the programmer assemble them in any way they please to get work done, then you might be the right kind of person to use Emacs, or a UNIX-like shell (and CLI utilities), as an IDE.