Beads: Computer language and toolchain • Hacker News Board

Re: Beads: Computer language and toolchain

#111

Post by morelisp » Thu, May 27, 2021, 6:57 PM UTC

discardable_dan wrote: post #77

kortex wrote: post #48

- promises far fewer bugs - promises to replace entire stack - ...and Excel - "declarative languages have almost no bugs" - built-in database Yeah color me skeptical. I love experiments, but I prefer those that under-promise and over-deliver. People have been promising unification languages since there were only two languages. People have been promising cross-platform since there were two platforms. And people have b…

The thing that really bothered me was the FizzBuzz example [0]. The core logic is: cell // this routine will be called 100 times, and the implied block variable // b will hold values, like the sequence number b.cell_seq var ss : str case mod(b.cell_seq, 15) | 0 ss = "FizzBuzz" draw_rect(fill:LIGHT_SKY_BLUE) | 3, 6, 9, 12 ss = "Fizz" draw_rect(fill:LIGHT_GREEN) | 5, 10 ss = "Buzz" draw_rect(fill:YELLOW) else ss = to_s…

> The entire point of FizzBuzz, as a programming practice, is to ensure the programmer does not try to write this exact code

No, the point of FizzBuzz, which is an interview practice is to make sure that the candidate has literally the minimum ability to translate a natural-language problem description into code.

There is no point using it for anything else. Not demos of languages or tools, not looking for a particular shape of solution, you don't even really need to let them finish writing a bug-free version of it. It is only to filter out the people who "learned a language" but cannot actually write a program. Which is unfortunately a ton of people.

https://imranontech.com/2007/01/24/using-fizzbuzz-to-find-de...

Re: Beads: Computer language and toolchain

#112

Post by magicmouse » Thu, May 27, 2021, 7:04 PM UTC

vogon_laureate wrote: post #90

"Some people, when confronted with a problem, think 'I know, I'll u̵s̵e̵ ̵r̵e̵g̵u̵l̵a̵r̵ ̵e̵x̵p̵r̵e̵s̵s̵i̵o̵n̵s̵ invent a new programming language that will unify everything.' Now they have two problems."

Funny you should mention Regular Expressions. In Beads there is a complete rewrite of the syntax for regular expressions, replacing the meta characters with a more readable vertical format that facilitates comments, and offers subroutines. For example, compare the IPv4 address regular expression as done in JS:

``` (\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5])\.(\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5]){3}| ```

with the Beads notation:

``` pattern octet group or digit // matches 0..9 set:'1-9' digit // matches 10 .. 99 '1' digit digit // matches 100 .. 199 '2' set:'0-4' digit // matches 200 .. 249 '25' set:'0-5' // matches 250 ..255

pattern IPv4 octet '.' octet '.' octet '.' octet ```

The Javascript version is extremely hard to read. I can't change the underlying engine that supports regular expressions as that is built in to the Browser runtime, but i can at least make notation friendlier.

Unified languages in the past like VB6 and Borland Delphi were beloved, and many people were productive in those environments, and clinged to them long after MS for example wanted to migrate people into .NET, because they were so much simpler.

The current complexity level that people have to endure is unnecessarily high. Never before in history did people write in 3 different languages in the same source code file (as they routinely do in JS + CSS + HTML), where they don't even agree on how comments are notated.

The current situation benefits the large incumbents like Google, Facebook and Amazon.

Re: Beads: Computer language and toolchain

#113

Post by jasonshaev » Thu, May 27, 2021, 7:09 PM UTC

I would seriously consider toning down the marketing speak. This project is making very bold claims, without citations, while trying to appeal to an often skeptical crowd (developers).

"Graph databases are considered more powerful and modern than relational databases," ... okay? According to what benchmark and for what workloads? A statement like that is an immediate turnoff because it is dismissive of 30 years of database development.

"In many languages the slightest error in input data can cause a program to seriously malfunction. Beads has special rules of arithmetic and a robust mathematical model, that makes it extremely difficult to have a serious malfunction." I don't see how "special rules of arithmetic and a robust mathematical model" makes it "extremely difficult to have a serious malfunction."

Replacing "Unix, SQL, Rails, Ruby, RSpec, Templates, HTTP, HTML, CSS, DOM, jQuery and Javascript" in one language? How does this replace Unix? HTTP? It's not an operating system or a transport layer.

My goal isn't to crush someones idea, but consider your audience: developers.

Re: Beads: Computer language and toolchain

#114

Post by magicmouse » Thu, May 27, 2021, 7:16 PM UTC

jackcviers3 wrote: post #109

vogon_laureate wrote: post #90

"Some people, when confronted with a problem, think 'I know, I'll u̵s̵e̵ ̵r̵e̵g̵u̵l̵a̵r̵ ̵e̵x̵p̵r̵e̵s̵s̵i̵o̵n̵s̵ invent a new programming language that will unify everything.' Now they have two problems."

Not to mention not researching your problem space seems to be an afterthought for these types of people. It's not a novel problem or solution. Haxe already exists [1] to unify everything, similar to beads' claims. It doesn't. Unifying has been done better before with it, C (it does run almost everywhere!), Java, Flash/Flex, and JS+HTML. Many have entered. Few win for a little while, and eventually everything loses. 1…

I consider the unified languages of the past to be VB6 and Borland Delphi as two successful environments that had many happy users who preferred to live inside that world and get things done without a lot of fuss. I know my friends when VB6 started to be disfavored by MS and they felt pressure to learn .NET refused to use .NET as it was just too hard compared to VB6. The reference manuals for .NET are 1000's of pages when you print them all out.

My measurement of complexity includes both the syntax of the language, and also how many library API's you have to learn.

Haxe, which has roots in an open source version of Adobe's ActionScript is an excellent product. I often consider switching to Haxe for the implementation of the transpiler as it would yield Linux support. I currently emit AS3 or JS code, but Haxe is a very strong possible direction in the future as Haxe has a lot of portability.

I don't consider C a unified language. It had no database, no drawing model, no event model, and Berkeley Sockets was a library added in to support the internet, that unfortunately did not achieve complete standardization. C is the most portable language, but it just a step above Assembler, and full of pitfalls.

Tools do wax and wane in popularity; they are all useful in certain circumstances, and the more generally useful the tool, the more it gets used. VB6 was abandoned by its owner MS, when Gates retired (Basic was his pet project, as it was the original foundation product of MS' fortune). People used it as long as they could (and some still do). VB6 didn't make the jump to the Internet, but it could have if MS hadn't tried to force everyone into .NET (which was very complex).

Re: Beads: Computer language and toolchain

#115

Post by kortex » Thu, May 27, 2021, 7:18 PM UTC

magicmouse wrote: post #106

discardable_dan wrote: post #77

Earlier quoted context omitted.

The thing that really bothered me was the FizzBuzz example [0]. The core logic is: cell // this routine will be called 100 times, and the implied block variable // b will hold values, like the sequence number b.cell_seq var ss : str case mod(b.cell_seq, 15) | 0 ss = "FizzBuzz" draw_rect(fill:LIGHT_SKY_BLUE) | 3, 6, 9, 12 ss = "Fizz" draw_rect(fill:LIGHT_GREEN) | 5, 10 ss = "Buzz" draw_rect(fill:YELLOW) else ss = to_s…

There are about 3 slightly different methods for doing FizzBuzz, in any language, and this was but one of them. You could express it also as an empty string, then concatenating as necessary: ''' // to those objecting to using a modulo-15 test, // we could have done the above code as follows: var ss = "" if mod(b.cell_seq, 3) == 0 "Fizz" &=> ss if mod(b.cell_seq, 5) == 0 "Buzz" &=> ss if ss == "" ss = to_str(b.cell_se…

I think that is a better FizzBuzz example. You are right that it isn't meant to showcase the whole language, but lacking a composable solution (mod N1 || mod N2... for N prime factors) makes me immediately think the language does not lend itself well to that kind of composition.

And as others mentioned, Fizzbuzz is merely a "can you program anything test." But I do prefer the "general" fizzbuzzjazz, which accepts a list of numbers and a list of words [fizz, buzz, jazz,...] and emits the strings for appropriate prime factors.

Re: Beads: Computer language and toolchain

#116

Post by kortex » Thu, May 27, 2021, 7:20 PM UTC

magicmouse wrote: post #108

ZiiS wrote: post #63

Earlier quoted context omitted.

Given the implementation is `cookie_write(KEY, my_state, duration:1 day)` I am fairly sure even this example has a bug.

I am not aware of any bug in this code. It works, and the selection persists. Of course it is a silly project. Classic FizzBuzz is just too simple a task to show off any language features, so the task was souped up so that it draws a 10x10 grid of the results, and lets you pick one of the 100 cells to highlight. It persists this selection for a day to show how one can save up to 1kb of state information trivially ins…

That's neat, but imho start with the plain fizzbuzz first. Everyone has it in their lexicon so it acts as a first-pass Rosetta stone for what the language syntax is like. Adding extra bits implies you need those to get the job done.

Re: Beads: Computer language and toolchain

#117

Post by magicmouse » Thu, May 27, 2021, 7:22 PM UTC

danpalmer wrote: post #16

The marketing effort here feels so unbalanced. On the plus side, there's a marketing video with a professional sounding voiceover. But then there's a website titled "Beads Language Home Site", with a barebones design, and which talks repeatedly about "Macintosh" and "Windows OS".

I apologize that the intro video is so much nicer than my own homemade videos. After paying a pro do to make that first video of a series of a dozen videos explaining the theory behind it, i realized it was going to cost too much, and the money would be better spent on the product itself.

So that video is just part of 1 of 15 parts, and with some support and enthusiasm from people, we can continue the series, because it is o much nicer to watch a professional video with proper British narration. As the great George Bernard Shaw once said, "The United States and Great Britain are two countries separated by a common language."

I hope that people will overlook the homwbrew quality level of the videos, but the language for a spin, because you might like the language a lot. It has a lot of simplifications, and particularly for client/server web apps, where it is much easier to build reliable products.

Re: Beads: Computer language and toolchain

#118

Post by magicmouse » Thu, May 27, 2021, 7:25 PM UTC

toboche wrote: post #69

Not even looking at the way the language and tool chain works, the home page is very confusing, with each following paragraph making almost no connection to others. What's worse, the YouTube video makes is mostly based on a fairly convoluted argumentation that doesn't seem to be logical to me. If the page is supposed to draw attention of any fairly technical folks, it definitely needs some more thorough reasoning and…

I apologize for the bad writing. I am not a good writer at all, but my programming is quite solid, and i hope you will take it for a spin. It has many nice simplifications, from the reformed Regular Expression syntax, to the automatic dependency checking in layout and refresh (if the model changes values, any part of the visualization that is affected is automatically refreshed).

Re: Beads: Computer language and toolchain

#119

Post by magicmouse » Thu, May 27, 2021, 7:32 PM UTC

tromp wrote: post #28

Reminds me a bit of Rebol [1] and its close cousin Red [2]. [1] https://en.wikipedia.org/wiki/Rebol [2] https://en.wikipedia.org/wiki/Red_(programming_language)

Rebol and is rebirth Red, is a very concise and powerful language. They are well funded, and will be a force to reckon with going forward. I would say the languages are very different in the sense that Beads is clearly aimed at graphical interactive software, with a layout/drawing/event model built into the language, while Red exists as a systems programming language, a variant customized for building crypto contracts, and then as a general purpose language.

They are so different that it is hard to compare. Red being a concatenative language has more in common with FORTH than Algol.

The closest thing to Beads is Elm, or visual javascript (was called Yazz) (https://yazz.com/visifile/index.html), which is another integrated product.

Re: Beads: Computer language and toolchain

#120

Post by ElFitz » Thu, May 27, 2021, 7:37 PM UTC

daenz wrote: post #88

ElFitz wrote: post #61

Earlier quoted context omitted.

Funnily enough; the last time I tried something like that I was 25, had barely 6 months of experience, and somehow convinced a friend I had introduced to programming not more than two months earlier. Waaay in over our heads. Since then, I’ve realised it’s been tried and is being tried an inordinate number of times and seems to usually fail for some reason or another. Kind of a wild goose chase for some (supposed) hol…

How did you convince your friend?

We had had a brush with AWS, it’s Lambdas, Step Functions, etc, as well as Zapier and Stamplay’s flow on our first project together.

We both came out of it convinced that there had to be a way to make it much easier yet still as powerful to orchestrate logic & services…

The goal was to make something we could use to drastically simplify the development of all our future endeavours, thinking we might also be able to sell it to others.

Little did we know how naive we were ^^’