4 ms·

This seems like the brain child of one ambitious smart person that still needs to receive feedback from the outside world.

jb1991 5y ago
It also feels like at least as much effort went into the marketing (site, videos) and PDF resources (ref cards) as went into the language itself.
sz4kerto 5y ago
We need these kind of experiments. If you learn too much about contemporary software engineering, then it is almost certain that you'll end up with a React front-end talking to a NodeJS/Java back-end hosted on Kubernetes. Which is obviously the right thing most of the time :) but the world is richer because of these super-ambitious, super-naive experiments. Also: imagine how much has someone learned from implementing all this.
- zitterbewegung 5y ago
  I've done a super naive and ambitious startup idea. It was a big waste of time and I should have listened to more feedback. The world isn't richer if you listen to feedback quite the opposite it makes you improve what you are making. I agree about the learning aspect but overhyping something doesn't really pan out that well.
  - galangalalgol 5y ago
    You weren't richer but the worl might be. Sometimes clean sheets and first principles discover assumptions that have changed. Either in how people use software, how the domain itself works, or the constraints of computing.
    - mikepurvis 5y ago
      I think that depends a lot on whether the failed startup produced any code, papers, conference talks, even podcast episodes, etc. If not, and everything was just buried in the person's head, then no, the world is not richer for it. This kind of thing is why I'm so grateful to the founders of startups like Starsky, who are obviously still immensely passionate about the domain they were in. That even in the wake of the company failing, they've continued to advance the state of the art by sharing what they learned, especially where it's wisdom that goes against the current dogma (eg, that teleop is a hugely important part of a self-driving vehicle strategy, but that investors of the late 2010s were turned off by that focus because they wanted to see companies trying to shoot the moon on L4/L5 autonomy): https://medium.com/starsky-robotics-blog/the-end-of-starsky-robotics-acb8a6a8a5f5 https://medium.com/starsky-robotics-blog/the-end-of-starsky-...
      - zitterbewegung 5y ago
        My startup idea was to recognize objects by their shape. I got it to the point where I was able to setup an elasticsearch index that looked at point clouds uploaded and performed feature extraction on what was ingested. I haven't open sourced by code because I probably had committed API keys (but I should make it public now since all those keys stopped existing). The biggest issue was a large amount of debt and lost opportunity cost. But it gave me a new perspective and at times it was really fun learning so much. I think it has made me a better developer because I made so many complicated mistakes and now I can simplify things.
- hackerfromthefu 5y ago
  that sounds more like a beginner to mid level stack
- jrm4 5y ago
  Exactly. This particular iteration is going to fail miserably, but the world will be better because people are at least thinking like this. I hope that this idea happens, somehow.
- catblast01 5y ago
  Matter of perspective I admit but it also could just be a weird pastiche of elements of C, COBOL, M, Ada, and JavaScript. Nothing here seems particularly ambitious really.. idiosyncratic however. A good counterpoint might be Go. The author used to at least actively solicit feedback, I never once noticed acknowledgement of any of it though.
  - magicmouse 5y ago
    We have a Discord group, and am very eager to hear from users. https://discord.gg/pTAdsSW https://discord.gg/pTAdsSW User suggestions and bug reports are promptly followed up on. There are almost no known errors at this point, so it is very good shape. You are correct that it has elements of JavaScript, but that is inevitable given that the primary output of the transpiler is JS code for use on the web, so some JS functions have to be present for that to work, and to inter-operate with JS, one has to use their string system. The attempt to offer an integrated programming system, where we return to the simplicity of VB6, or Borland Delphi, is actually quite ambitious, as in order to offer such an integrated product one has to build in a database (graph in this case), a layout engine (novel, Renaissance proportion based layout model), a drawing system, and a way of coordinating between client/server (a novel subscription to a single source of truth system, with remote procedure calls). The language is primarily based on Modula-2, which the author of Beads used for 20 years quite successfully in large commercial products. There is nothing from COBOL or Ada that i can think of (although Prof. Wirth did contribute to Ada I believe), not sure what the M language is. There are some elements of PROLOG built in that are not readily apparent, as the core programming pattern is State-Action-Model (see sam.js.org), and if you change a state variable, then any drawing code that used that variable is scheduled for refresh.
    - bawolff 5y ago
      > but that is inevitable given that the primary output of the transpiler is JS code for use on the web, so some JS functions have to be present for that to work, and to inter-operate with JS, one has to use their string system. I don't see why this is true (i also don't see it as a bad thing. Take what works and all that).
daenz 5y ago
Probably, but I think that's also a little dismissive. Maybe they have received plenty of feedback but still think their ideas take precedence over that feedback. I don't think we should confuse outside-the-box thinking with thinking that hasn't received feedback.
- potsandpans 5y ago
  they've received plenty of feedback from the PL community. Beads is a fun an interesting project, but the over the top marketing language and hype is a bit silly. The author really is a rare mix of driven PL engineer and overly grandiose personally.

jb1991 5y ago

It also feels like at least as much effort went into the marketing (site, videos) and PDF resources (ref cards) as went into the language itself.

sz4kerto 5y ago

We need these kind of experiments. If you learn too much about contemporary software engineering, then it is almost certain that you'll end up with a React front-end talking to a NodeJS/Java back-end hosted on Kubernetes. Which is obviously the right thing most of the time :) but the world is richer because of these super-ambitious, super-naive experiments. Also: imagine how much has someone learned from implementing all this.

zitterbewegung 5y ago
I've done a super naive and ambitious startup idea. It was a big waste of time and I should have listened to more feedback. The world isn't richer if you listen to feedback quite the opposite it makes you improve what you are making. I agree about the learning aspect but overhyping something doesn't really pan out that well.
- galangalalgol 5y ago
  You weren't richer but the worl might be. Sometimes clean sheets and first principles discover assumptions that have changed. Either in how people use software, how the domain itself works, or the constraints of computing.
  - mikepurvis 5y ago
    I think that depends a lot on whether the failed startup produced any code, papers, conference talks, even podcast episodes, etc. If not, and everything was just buried in the person's head, then no, the world is not richer for it. This kind of thing is why I'm so grateful to the founders of startups like Starsky, who are obviously still immensely passionate about the domain they were in. That even in the wake of the company failing, they've continued to advance the state of the art by sharing what they learned, especially where it's wisdom that goes against the current dogma (eg, that teleop is a hugely important part of a self-driving vehicle strategy, but that investors of the late 2010s were turned off by that focus because they wanted to see companies trying to shoot the moon on L4/L5 autonomy): https://medium.com/starsky-robotics-blog/the-end-of-starsky-robotics-acb8a6a8a5f5 https://medium.com/starsky-robotics-blog/the-end-of-starsky-...
    - zitterbewegung 5y ago
      My startup idea was to recognize objects by their shape. I got it to the point where I was able to setup an elasticsearch index that looked at point clouds uploaded and performed feature extraction on what was ingested. I haven't open sourced by code because I probably had committed API keys (but I should make it public now since all those keys stopped existing). The biggest issue was a large amount of debt and lost opportunity cost. But it gave me a new perspective and at times it was really fun learning so much. I think it has made me a better developer because I made so many complicated mistakes and now I can simplify things.
hackerfromthefu 5y ago
that sounds more like a beginner to mid level stack
jrm4 5y ago
Exactly. This particular iteration is going to fail miserably, but the world will be better because people are at least thinking like this. I hope that this idea happens, somehow.
catblast01 5y ago
Matter of perspective I admit but it also could just be a weird pastiche of elements of C, COBOL, M, Ada, and JavaScript. Nothing here seems particularly ambitious really.. idiosyncratic however. A good counterpoint might be Go. The author used to at least actively solicit feedback, I never once noticed acknowledgement of any of it though.
- magicmouse 5y ago
  We have a Discord group, and am very eager to hear from users. https://discord.gg/pTAdsSW https://discord.gg/pTAdsSW User suggestions and bug reports are promptly followed up on. There are almost no known errors at this point, so it is very good shape. You are correct that it has elements of JavaScript, but that is inevitable given that the primary output of the transpiler is JS code for use on the web, so some JS functions have to be present for that to work, and to inter-operate with JS, one has to use their string system. The attempt to offer an integrated programming system, where we return to the simplicity of VB6, or Borland Delphi, is actually quite ambitious, as in order to offer such an integrated product one has to build in a database (graph in this case), a layout engine (novel, Renaissance proportion based layout model), a drawing system, and a way of coordinating between client/server (a novel subscription to a single source of truth system, with remote procedure calls). The language is primarily based on Modula-2, which the author of Beads used for 20 years quite successfully in large commercial products. There is nothing from COBOL or Ada that i can think of (although Prof. Wirth did contribute to Ada I believe), not sure what the M language is. There are some elements of PROLOG built in that are not readily apparent, as the core programming pattern is State-Action-Model (see sam.js.org), and if you change a state variable, then any drawing code that used that variable is scheduled for refresh.
  - bawolff 5y ago
    > but that is inevitable given that the primary output of the transpiler is JS code for use on the web, so some JS functions have to be present for that to work, and to inter-operate with JS, one has to use their string system. I don't see why this is true (i also don't see it as a bad thing. Take what works and all that).

daenz 5y ago

Probably, but I think that's also a little dismissive. Maybe they have received plenty of feedback but still think their ideas take precedence over that feedback. I don't think we should confuse outside-the-box thinking with thinking that hasn't received feedback.

potsandpans 5y ago
they've received plenty of feedback from the PL community. Beads is a fun an interesting project, but the over the top marketing language and hype is a bit silly. The author really is a rare mix of driven PL engineer and overly grandiose personally.