I’ll be writing about this shortly (I hope this week). But let me briefly give some slight context on this. The goal of opening the repo was to support existing Dark users in extending the language and libraries, so that they aren’t as reliant on our internal roadmap (this is especially true now that “us” is just “me”, but was also the case before that).
It isn’t a goal to have non-Dark users contribute (indeed, if you don’t use Dark you’re unlikely to be able to make helpful contributions that fit within the needs of the language). It’s also not a goal for others to run their own instances of Dark - one of the goals of the language/platform is to have a single ecosystem where we run the infra, migrate the language, etc.
I recognize those goals don’t resonate with everyone, but I didn’t want to do “fake” open source: where a company releases something that’s technically open source but where they really do not want anyone else to run it, or where they have other controls to prevent forking (such as trademarks). I wanted to be explicit in the goals and rights here.
Bummer. That’s a rough transition to lead through.
Seems that if they actually developed a significantly faster way to build basic websites, then the obvious play is to start an agency and make money by cranking them out. Dog food the framework for a bunch of small business and marketing websites. Use that revenue to finance further development of the framework.
That is the beauty of bankruptcy. The business has already failed, basically. So instead of limping along as a one-man shop, they could formally shut down, return the remaining capital to the investors, and liberate the infrastructure. Maybe the founder walks with the code, maybe they release it as open source. The goal is to separate the useful technology asset from the failed corporate entity and enable it to be reused.
It worked for Iridium, and that was WAY more expensive.
I got a private beta invite back in February and it is fun to play with but doesn’t really fit into my problem domain. I think I can invite others, so it anyone is looking to try it, DM me an email address.
The license is pretty awful: https://github.com/darklang/dark/blob/main/LICENSE.md.
Basically you’re only allowed to touch the source code if you’re sending them improvements. Don’t even think about using it to run your own websites.
I’ll be writing about this shortly (I hope this week). But let me briefly give some slight context on this. The goal of opening the repo was to support existing Dark users in extending the language and libraries, so that they aren’t as reliant on our internal roadmap (this is especially true now that “us” is just “me”, but was also the case before that).
It isn’t a goal to have non-Dark users contribute (indeed, if you don’t use Dark you’re unlikely to be able to make helpful contributions that fit within the needs of the language). It’s also not a goal for others to run their own instances of Dark - one of the goals of the language/platform is to have a single ecosystem where we run the infra, migrate the language, etc.
I recognize those goals don’t resonate with everyone, but I didn’t want to do “fake” open source: where a company releases something that’s technically open source but where they really do not want anyone else to run it, or where they have other controls to prevent forking (such as trademarks). I wanted to be explicit in the goals and rights here.
‘Awful’ is relative. It’s their code, their license, anyone who doesn’t accept can avoid it. No big deal.
For some more context, Dark hasn’t found a market for their product, recently cut their entire staff, and is now a solo project
Bummer. That’s a rough transition to lead through.
Seems that if they actually developed a significantly faster way to build basic websites, then the obvious play is to start an agency and make money by cranking them out. Dog food the framework for a bunch of small business and marketing websites. Use that revenue to finance further development of the framework.
Agencies do not make a whole lot of money. Certainly not enough to make up for several years of engineering efforts up front.
That is the beauty of bankruptcy. The business has already failed, basically. So instead of limping along as a one-man shop, they could formally shut down, return the remaining capital to the investors, and liberate the infrastructure. Maybe the founder walks with the code, maybe they release it as open source. The goal is to separate the useful technology asset from the failed corporate entity and enable it to be reused.
It worked for Iridium, and that was WAY more expensive.
Ok, but, what IS this? The github repo doesn’t have an explanation of what the program actually does, which is a pet peeve of mine.
It’s a proprietary language and editor for building small web apis. There were a couple of posts about it here previously: https://lobste.rs/search?utf8=%E2%9C%93&q=darklang&what=stories&order=relevance
I got a private beta invite back in February and it is fun to play with but doesn’t really fit into my problem domain. I think I can invite others, so it anyone is looking to try it, DM me an email address.
Good point! Just added more context to the README, thanks!
Yeah. How does Unison compare?
MIT
https://github.com/unisonweb/unison/blob/trunk/LICENSE
Why would I even touch the source code of this. I just cloned and I feel dirty already…
My first thought was DarkBasic and the follow on Java like language they had although I can’t find any reference to it now.
https://en.wikipedia.org/wiki/The_Game_Creators#DarkBASIC
Hey I happen to be wearing my Dark tshirt today. It’s really nice and soft.