V Playground
vlang.io
vlang.io
...or is this supposed to be the future of programming languages, compiler-as-a-service where you get a playground but not an actual compiler?
Free: Limited to 1k lines of code
Hobbyist ($29.99/user): Unlimited lines of code, no warning messages
Pro ($99.99/user): Warning messages
Enterprise (Contact us): Link to librariesI thought the days when anyone seriously toyed with the idea that you could build a business selling a programming language (or perhaps compiler implementation?) were gone.
Oh well, some interest -> zero interest, but hey, best of luck to them~
At least officially, the idea announced was that the language will be available (and open source) later in the summer.
Hopefully this goes well, as the language makes some big promises that make it sound like vaporware, but at least we do have a playground accepting the syntax now.
I’m 100% serious; that’s a business model with no value in it, and they few vendors who have tried it in the past are all fading into obvilion.
Getting adoption of a language is hard enough as it is (see the slow uptake of say, kotlin despite large scale corporate backing) making it a pay-to-use is just a dead end.
How are you going to sustain the user growth for a paid product when better or equivalent free alternatives exist? How will you build an ecosystem when the core product is paid-to-use? If I’m paying to make something, I’m not giving it away for free. Not only that, it’s a proven failed business model see Borland for example. The few vendors that remain are for embedded devices.
There are so many thing wrong with the idea it’s hard to know where to even start enumerating them.
/shrug
> They are not, and they're not helpful either.
...but please, don’t try to put words in my mouth, that’s just offensive.
I am completely serious, and was in my previous comment too.
I started developing Volt/V in the middle of developing gitly.
That's my biggest drawback. I finish 90% of the project, and jump to a different thing.
I've grown a lot since then, and I'm slowly wrapping up everything.
Claims like C++ translation is definitely one of those 90% things that may look easy to some people until you actually try to do it.
There are literally dozens of good enough small languages, e.g.: Nim, Zig, Crystal, Pony, etc etc, and at least the first two were written by a single person for the most part.
There are many other single-author single-serving languages (used to make games for example GOOL in Crash Bandicoot etc).
Not saying this can't be vaporware, but it's not beyond comprehension.
Heck, not long ago, a guy (with some psychological issues too) made his own OS, high level language compiler (HolyC), drivers, UI, and a flight simulator in assembler and HolyC, all working too (TempleOS).
I'd be ok with V though, even if it didn't do C++ translation (or even C translation at all).
Just a fast building, Algol-style syntax language, with the basic language features mentioned.
Look, you don't owe anybody anything, and that's fine. Just don't make promises you can't keep. If you're not going to have something done in 2 days, don't say it'll be done in 2 days. Just say "you'll get it when you get it" and that's an absolutely fine response. But promising something in hours or days and missing that deadline will lead reasonable people to take your claims with a huge grain of salt.
All I'm saying is you've made some very big claims, so now you might want to master the fine art of setting expectations or risk making people lose interest in your project when you finally release a compiler.
[1]https://news.ycombinator.com/item?id=19403747
[2]https://old.reddit.com/r/vlang/comments/b2k8j7/playground/
I no longer name dates.
I'm not sure why it's being presented this way. I'm guessing it's probably structured this way for sake of Patreon, but it's actually a huge turn off.
I'm working on articles about C/C++ translation right now. They will be up in a couple of days.
What else is hurting you? I'll update the website.
So the claimed C/C++ translation looks misleading.
So, no, I don't think you have done even a tiny fraction of what C++ entails.
Maybe you are hardcoding detection of a set of known types by hand?
Hmmm I'm not sure if this is true but he says that "V can compile to (human readable) C, so you get the great platform support and optimization of gcc and Clang."[0]
If you need optimization, for now you have to emit C. Like the website says, in the future V will be able to emit optimized machine code.
V will be open sourced this June.
This is definitely not a toy language. Right now it can emit x64 machine code, more platforms will be available after the open-source release in early June.
Supporting more platforms will not make compilation slower.
> I suggest porting V to llvm IR
No thanks :) I'd like to keep 1.5 million loc per second.
And that's fine too. I don't want support for some BS platform slow down the platform 90% of people and servers use.
Perhaps add ARM to x86, but that's it.
Your point being? Even if he's misleading people, people can always JUDGE by themselves and not give him money.
(As if he gets anything -- even established projects with tons of supporters and huge communities hardly make $1000/month on donations).
Yeah, so? Do you think you have some unique insight, that you have to plaster all over this thread 20 times? Say it and be done with it.
We don't think it's the new production ready language from Microsoft or Google either...
>His benchmarks exaggerate by comparing the speed of print statements, and the language itself is essentially a transpilation (a direct rule based conversion without an AST) to x86 so no other platforms can be supported.
If a language has a good enough syntax and semantics, that's totally find with me.
I could not give less fucks for non x86 platforms, especially at such an early stage of a language's development.
That said, it also says it supports transpiring to C, in which case the whole point of "not supporting other platforms" is moot.
For example, V has a powerful graphics library that can be used for writing 2D/3D games and all kinds of hardware accelerated apps.
Volt for Linux uses it, so does the vid editor.
Developer here!
Happy to answer all questions.
Even if it is half done and it doesn't fill many of the claims you might have made, push it online to github when the day comes. Inevitably some people will throw stones at you for not meeting their expectations and even you might be harsh on yourself but still push it online. There will be some of us who will be supportive and try to help the project according to your vision regardless if it is launched full of bugs, incomplete and with problematic or missing features.
Also, if the language never really existed and it was a fluke, I personally forgive you in advance, hope you find peace and wish you the best in life.
I will release it in June without delays, that's the plan. Whatever state it is in.
I just realized I maybe have answered my own question but I don't think even Go should use `go` keyword for that.
Maybe give it a more technical name?
I'm not very confident about it :(
And if it does compile to machine code, is it really running as fast as C as you claim? Or did you just say that because it trans-compiles to C.
Clearly this advanced language is too much to handle for mere mortals like myself.
This is an important bug and will be fixed asap.