Notices by Daniel Kochmański (jackdaniel@functional.cafe)

Sigh, it seems that there's more work for me. I want to implement an extension for #McCLIM that is part of older CLIM implementations (although not part of the specification), namely the ACCEPT-VALUES-PANE. While it shares many similarities with the object constructed with ACCEPTING-VALUES, it has a few nasty differences; most notably it is not called synchronously, so it must cooperate with standard frame control loop. On the other hand it seems to be useful enough to take the extra mile. #lisp

They have a point you know?

I've been configuring a new android phone for my aunt and buy, it is horrific:

- tons of preinstalled applications including wtf games, linkedin, netflix, spyonme-soft, duplicate-this-stock-functionality-but-worse, spyonme-hard and much more

- more apps that can't be uninstalled yet are clearly spyware (xiaomi cloud? mimusic?) plus standard google spyware

- gemini this gemini that, "yes please" vs "ask me later", "make your phone your wallet" with "yes please" vs "ask me later" and much more (again!)

Then I proceed to install the only application of her choosing -- whatsup, but whatever, so I open play store (there is also a mistore:) and type whatsapp

The first result is a green icon similar to "whatsapp", I install it, but no! it is actually "telephone" with overreaching permissions that wants to replace the standard application for calling.

Only the second result is the application with the name that exactly matches the query. Apparently "telephone" is sponsored, but it is only very subtly hinted.

I wouldn't call it a circus - it is an animal experiment facility.

RE: https://mastodon.social/@iiradned/115650316040761029

That would be comical if it weren't horrifying.

Slightly improved version. And now, just like that, making a wonky spreadsheet is not a distant future :-)

It just occured to me, but with multiple prompts we can do this

#McCLIM #lisp #wip

I wrote a blog post that summarizes the current progress on Web Embeddable Common Lisp:

https://turtleware.eu/posts/Using-Common-Lisp-from-inside-the-Browser.html

This work is possible thanks to funding from @nlnet via @NGIZero program.

#lisp #ecl #webassembly

GitHub CEO: "embrace AI or leave"
GitHub CEO: "leaves"

They should have embraced more AI!

Did I write an interactive #forth compiler to #ecl bytevm?

@amszmidt but vectorp.1 test for a different thing...

@amszmidt ansi-test uses rt, a system for testing developed at mit in 1990.

@amszmidt yes, it is my explanation for you. but yeah, reading the source code is the best way to confirm (I did, so I can only encourage it).

@amszmidt

(deftest vectorp.1
(vectorp 1)
nil)

-->

(pass-only-if
(equal (multiple-value-list (vectorp 1))
(list nil)))

--------------------

(deftest arrayp.1
(notnot-mv (arrayp #(a b c)))
t)

-->

(pass-only-if
(let ((result (multiple-value-list
(arrayp #(a b c)))))
(setf (car result)
(not (not (car result))))
(equal result (list t))))

@amszmidt

It is not broken, because it checks for generalized boolean false, and the only value that is generalized boolean false is NIL. If there are more returned values, then it will fail, as per the other comment.

@amszmidt Both tests are justified and correct.

notnot-mv transforms (values a b c) into (values (not (not a)) b c)

(deftest name form &rest expected-values)

for arrayp.1 we test that
a) it returns a generalized boolean true (hence not not)
b) it returns exactly one value

for vectorp.1 we test that
a) a number 1 is not a vector
b) it returns exactly one value

@PaniczGodek @weekend_editor @amszmidt

DO is a syntactic sugar for tail recursion. I've even made a short blog post about it:

https://turtleware.eu/posts/How-do-you-DO-when-you-do-DO.html

@amszmidt @PaniczGodek such disputes are silly and I like Common Lisp better than others, but "a lisp" refers to many languages and I'd wager an opinion that Common Lisp is closer semantically and functionally to r6rs than LISP 1.5

@amszmidt @PaniczGodek name gatekeeping did more harm than good to lisp community, just saying.

@sznowicki @Watchdog_Polska @icd

Niedługo trzeba będzie mieć konto GMail żeby głosować :3

I know that people snark at HN for many things, but I'm more concerned that it becomes boring. Rare pearls between valuation, ai and opinionated snubby blog pieces. I remember that there was way more hacker than news there at some point and that was fun.