Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Dark emerges with ‘deployless’ software model

techcrunch.com

Post reply on HN
Print view View original

151–160 of 174 posts

Re: Dark emerges with ‘deployless’ software model

#151

Post by dkersten »

I don't get it. I tried to find out anything about what Dark is, how it works, or what it looks like and found almost nothing. They make claims like "holistic programming language" but don't show me what they mean. The What is Dark[1] blog post makes a lot of claims, but I've seen these claims made many times and they rarely live up to the hype. They don't show me how they plan to meet these claims or what it looks like. Finally, their marketing focus is on deployment, but honestly, deployment really has never been the biggest pain point for me. Sure, it can be a headache, but its nowhere near as much work as actually writing the applications themselves. I'm not convinced.

[1] https://medium.com/darklang/the-design-of-dark-59f5d38e52d2

Re: Dark emerges with ‘deployless’ software model

#152

Post by HereBeBeasties »

pbiggar wrote: post #142
danpalmer wrote: post #94

I'm excited by almost everything about this. I think it could be transformational in some categories of software engineering, but I do have a big concern... > I think the biggest downside of Dark is definitely that you’re learning a new language... But this is not it. Learning new languages isn't that hard, and a language so well designed for an ecosystem would likely be a pleasure to use. They are probably giving th…

Founder here. Totally fair concern - we're not thrilled about the lock-in either because we want people to try Dark and see that it solves a problem for them. We're looking at ways to improve this - would love ideas, and will be writing up some of our ideas soon. Around some of the specifics: we've no intention of pivoting. If we discovered that Dark had a massive opportunity to be the AirBnb for market analytics (or…

Your vision is extremely compelling and well articulated.

I'm not sure I buy the "coalescing multiple things into one makes things simpler" argument - the simplicity comes at the expense of expressiveness, flexibility and optionality. A large, highly-opinionated monolith, right the way across the stack, is bound to have made some design trade-offs and decisions that are either plain wrong (we all make mistakes) or don't fit with however I want to actually use it.

To do this at the AST graph/language level strikes me as both genius and absurd - not only is language design very hard, but you clearly have a massive uphill struggle to provide sufficient library and framework-level code on top of that to be able to start to compete with more mature options. Not to mention things like code review tooling (it's neither text-based, nor even in any form of conventional revision control system, so likely start from scratch). Security controls, auditability, vulnerability management for any library ecosystem that springs up around this, etc. etc.

By rejecting text you are having to reinvent and support a compiler, an IDE, github/gitlab (or equivalent, with their massive functionality set including protected branches and security controls and the like), package management maybe, debugger tooling (on the web), release/rollback systems/UIs, perhaps even monitoring and alerting (because you're promising the users they don't need to run infrastructure/services like Prometheus and the like and your "roll out" process looks so different).

Does Dark really have sufficient expertise around all of that to want to make this stuff totally monolithic? Is it even possible for a startup company to really compete in that sort of global (massively open source) scene?

Convincing people to bet their company on your company, both as an ongoing enterprise and as a place that can get all the inherent trade-offs here well-matched to their individual use-cases is going to be an almost impossible sell. Regardless of how much you assert that you believe in your mission and don't want to ever pivot. You can't blame people for being sceptical, especially when the opening paragraphs essentially say "this is a silver bullet".

I wish you luck.

Re: Dark emerges with ‘deployless’ software model

#153

Post by StyloW »

pbiggar wrote: post #148
StyloW wrote: post #17

It sounds pretty cool until the point where you read something like: "Deployment is risky because you’ve only tested on your own machine, and now you need to run the same code on many different machines, to serve (millions of) users." Which makes it seem like they don't even know about the existence of Docker; making me doubt most of what they'll try to sell us.

Founder of Dark here. Here's a thing I wrote about docker before, that led me to create Dark. https://circleci.com/blog/its-the-future/

Don't get me wrong, I understand where you're coming from and what your goals are; it'll be grand if Dark can deliver on its promises. I just don't understand why that particular statement on deployment had to be included, as it makes it sound like you're solving a problem with an existing solution.

Re: Dark emerges with ‘deployless’ software model

#154

Post by pbiggar »

StyloW wrote: post #153
pbiggar wrote: post #148

Earlier quoted context omitted.

Founder of Dark here. Here's a thing I wrote about docker before, that led me to create Dark. https://circleci.com/blog/its-the-future/

Don't get me wrong, I understand where you're coming from and what your goals are; it'll be grand if Dark can deliver on its promises. I just don't understand why that particular statement on deployment had to be included, as it makes it sound like you're solving a problem with an existing solution.

Unfortunately when you do press, you're at the mercy of whatever gets quoted and sometimes things that are kinda irrelevant get picked up. That's definitely not the thing we wanted as the lead. Our blog post on it was different: https://medium.com/darklang/dark-announces-3-5m-in-seed-fina...

Re: Dark emerges with ‘deployless’ software model

#155

Post by StyloW »

pbiggar wrote: post #154
StyloW wrote: post #153

Earlier quoted context omitted.

Don't get me wrong, I understand where you're coming from and what your goals are; it'll be grand if Dark can deliver on its promises. I just don't understand why that particular statement on deployment had to be included, as it makes it sound like you're solving a problem with an existing solution.

Unfortunately when you do press, you're at the mercy of whatever gets quoted and sometimes things that are kinda irrelevant get picked up. That's definitely not the thing we wanted as the lead. Our blog post on it was different: https://medium.com/darklang/dark-announces-3-5m-in-seed-fina...

That's a fair point. Regardless of my current impression, I'll have to try it out first hand in order to form a proper opinion, so looking forward to the public release. Good luck!

Re: Dark emerges with ‘deployless’ software model

#156

Post by crystaln »

pbiggar wrote: post #142
danpalmer wrote: post #94

I'm excited by almost everything about this. I think it could be transformational in some categories of software engineering, but I do have a big concern... > I think the biggest downside of Dark is definitely that you’re learning a new language... But this is not it. Learning new languages isn't that hard, and a language so well designed for an ecosystem would likely be a pleasure to use. They are probably giving th…

Founder here. Totally fair concern - we're not thrilled about the lock-in either because we want people to try Dark and see that it solves a problem for them. We're looking at ways to improve this - would love ideas, and will be writing up some of our ideas soon. Around some of the specifics: we've no intention of pivoting. If we discovered that Dark had a massive opportunity to be the AirBnb for market analytics (or…

Regardless of what you say as a founder, if the growth is not there and a massive opportunity arises, your investors will push you to pursue it. You won't have much choice, and this is an empty promise you can't keep.

Personally I don't see why anyone would choose such a closed ecosystem in this era when so many great open source options exist, where we can read and edit source code, substitute components, and pivot technically in so many ways impossible with a black box unironically named Dark. There is just way too much risk around hitting the limitations of the platform or corporate shifts. Why bet your biggest investment on something you have no control over? That would be incredibly foolish.

If you want this project to be successful, open source it and maintain your competitive advantage by adding value not monopoly power. If you are actually concerned about lock-in as you say, don't lock customers in.

Re: Dark emerges with ‘deployless’ software model

#157

Post by inlined »

twic wrote: post #64

I have a hunch that Dark won't generalise well. It might be good for making simple data-backed web applications. But a business of any nontrivial size has more than just simple data-backed web applications. Do you use Dark for part of your business, and traditional tech for the rest, in which case you now have the problem of integrating your traditional stuff with Dark's black-box infrastructure? Or do you build an M…

Or do you use micro service architectures where some of your services (eg ML training) runs on dedicated/specialized hardware and traditional event/request based business logic can be written more quickly and only moved to Dark incrementally?

Re: Dark emerges with ‘deployless’ software model

#158

Post by pbiggar »

HereBeBeasties wrote: post #152
pbiggar wrote: post #142

Earlier quoted context omitted.

Founder here. Totally fair concern - we're not thrilled about the lock-in either because we want people to try Dark and see that it solves a problem for them. We're looking at ways to improve this - would love ideas, and will be writing up some of our ideas soon. Around some of the specifics: we've no intention of pivoting. If we discovered that Dark had a massive opportunity to be the AirBnb for market analytics (or…

Your vision is extremely compelling and well articulated. I'm not sure I buy the "coalescing multiple things into one makes things simpler" argument - the simplicity comes at the expense of expressiveness, flexibility and optionality. A large, highly-opinionated monolith, right the way across the stack, is bound to have made some design trade-offs and decisions that are either plain wrong (we all make mistakes) or do…

Thanks!

If you want to use all that stuff, you can do that today. You can use GitHub and Go and Phabricator and Prometheus and whatever you want. People who want to a take a conservative approach have a million different options. Honestly, just run your app on AWS, it's got everything you need.

What we're trying to do is something different, removing a ton of stuff that we think we can allow you not need. And maybe it won't work for a whole lot of people - that's fine. Over time we'll fill out more of the product and support more use cases.

Re: Dark emerges with ‘deployless’ software model

#160

Post by inlined »

One question/unsolicited advice: the articles frequently mention “feature flags,” but these will cause simultaneous rollout in prod without any real testing or canary.

It seems like an experiment framework (eg whitelisted customers or IP addresses for alpha testing and percent rollouts) is a min bar for any reasonable implementation in Dark.

Post reply on HN