3 ms·

Even though I think the market for programming languages is saturated and it will be for the next decade or more, I think it's always nice when someone tinkers

Even though I think the market for programming languages is saturated and it will be for the next decade or more, I think it's always nice when someone tinkers and creates something new, so, my congrats to the authors!

mumblemumble 6y ago
The market for programming languages has been saturated at least since I got into programming. Which was not all that recently. The market for music has been saturated for even longer. But nobody really complains when someone wants to start a band, because we recognize that creative work has intrinsic value, even when it produces little economic return, even if only by enriching the life of its producer. Experimenting with programming languages is only very superficially different from that.
- neutronicus 6y ago
  No reason to complain until people start bugging you to go see their band or use their programming language.
dgb23 6y ago
I think this is a weird way of looking at it. PL are evolving continuously on some level or another. New languages pop up with new concepts or ways to do things, old languages can then incorporate some of these things. It works with all things similarly, there is a conventional main-stream that most adhere to, for efficiency gains but the creative risk takers are the ones who drive things forward, or fail to do so.
- qalmakka 6y ago
  Yes, I mostly agree with you, but I think that if we look closely to how programming languages have evolved over the years, we can more or less define them into "generations" i.e. periods where certain concepts have more or less dominated the interest of the community and have dictated how software and languages in general are designed. These are usually linked with one or several languages springing out and becoming popular, either by embracing new paradigms or by strongly rejecting a mainstream concept whose usefulness have been cast into doubt either by the industry or the community. Outside these periods, the overall interest tends to stagnate while the newer languages become stable and gain mainstream acceptance. In such a situation, a new contender must either find itself a niche, by covering very specific needs in very effective ways, or it must compete in mindshare in order to justify itself among the newest big names (i.e. any new compiled, strong typed and memory safe language nowadays is almost forced to compare with Rust and I dare say Zig). That's what I meant with "market saturation": I think the stagnant landscape of the 00s, dominated by classic, OOP languages, gave birth to a huge wave of "safe" languages in the 10s, many of which share similar concepts and ideas; I think we're now in a moment where some of these have made a name for themselves and have been established as mainstream, serious solutions (see Rust or Go), requiring newer contendents to be really innovative in order to capture mindshare and not be considered "me too" languages. These are, obviously, just my two cents; I really love compiler development and language design, and I think we should always support new languages and the work people do towards the goal of designing better languages and improving existing ones.
mikewarot 6y ago
There is plenty of idea space to explore in programming languages. ColorForth, for example... lets you do some of the work a parser would do, at edit time, by adding additional attributes to the source. Verilog, Ladder Logic, and Register Transfer Language all allow the programmer to specify a flow of logic and expressions that is always running, instead of a flow of instructions that we're all used to. Mixing this paradigm with the traditional code we're used to offers powerful possibilities. Transforming spreadsheets to a set of statements that are always executed, and vice versa could be fruitful as well.
zomglings 6y ago
It's important not to think of it as a market. It is pretty valuable (and not in terms of dollars right now) to create things like this.

mumblemumble 6y ago

The market for programming languages has been saturated at least since I got into programming. Which was not all that recently. The market for music has been saturated for even longer. But nobody really complains when someone wants to start a band, because we recognize that creative work has intrinsic value, even when it produces little economic return, even if only by enriching the life of its producer. Experimenting with programming languages is only very superficially different from that.

neutronicus 6y ago
No reason to complain until people start bugging you to go see their band or use their programming language.

dgb23 6y ago

I think this is a weird way of looking at it. PL are evolving continuously on some level or another. New languages pop up with new concepts or ways to do things, old languages can then incorporate some of these things. It works with all things similarly, there is a conventional main-stream that most adhere to, for efficiency gains but the creative risk takers are the ones who drive things forward, or fail to do so.

qalmakka 6y ago
Yes, I mostly agree with you, but I think that if we look closely to how programming languages have evolved over the years, we can more or less define them into "generations" i.e. periods where certain concepts have more or less dominated the interest of the community and have dictated how software and languages in general are designed. These are usually linked with one or several languages springing out and becoming popular, either by embracing new paradigms or by strongly rejecting a mainstream concept whose usefulness have been cast into doubt either by the industry or the community. Outside these periods, the overall interest tends to stagnate while the newer languages become stable and gain mainstream acceptance. In such a situation, a new contender must either find itself a niche, by covering very specific needs in very effective ways, or it must compete in mindshare in order to justify itself among the newest big names (i.e. any new compiled, strong typed and memory safe language nowadays is almost forced to compare with Rust and I dare say Zig). That's what I meant with "market saturation": I think the stagnant landscape of the 00s, dominated by classic, OOP languages, gave birth to a huge wave of "safe" languages in the 10s, many of which share similar concepts and ideas; I think we're now in a moment where some of these have made a name for themselves and have been established as mainstream, serious solutions (see Rust or Go), requiring newer contendents to be really innovative in order to capture mindshare and not be considered "me too" languages. These are, obviously, just my two cents; I really love compiler development and language design, and I think we should always support new languages and the work people do towards the goal of designing better languages and improving existing ones.

mikewarot 6y ago

There is plenty of idea space to explore in programming languages. ColorForth, for example... lets you do some of the work a parser would do, at edit time, by adding additional attributes to the source. Verilog, Ladder Logic, and Register Transfer Language all allow the programmer to specify a flow of logic and expressions that is always running, instead of a flow of instructions that we're all used to. Mixing this paradigm with the traditional code we're used to offers powerful possibilities. Transforming spreadsheets to a set of statements that are always executed, and vice versa could be fruitful as well.

zomglings 6y ago

It's important not to think of it as a market. It is pretty valuable (and not in terms of dollars right now) to create things like this.