Hacker News Reader

Comment by forgotmypw17 | Hacker News Reader

This illustrates how in proprietary and centrally-developed software, once the owner decides to integrate or replace something, all alternatives are eliminated, usually quickly, in this case in several years.

Compare with GNU/POSIX/Linux, where you can still make a comfortable environment for yourself without Wayland or systemd or whatever it is you don't like, and replacements still continue to be maintained and developed.

View on HN

This is actually the core failure of the Linux approach and why they year of the Linux desktop never occurred. Because there is no 1-true approach, developers must support too fragmented of an ecosystem, thus Linux becomes too big a lift to target for software titles.

Similarly the mass market of users care less about pimping their ride (so to speak), as getting their work done. Thus little is gained by rearranging your display or initialization subsystem.

One of the key things I find attractive about MacOS is that design is more than skin deep or an ad-hoc assemblage loosely related libraries. Meaning that visual design, architectural design, APIs, and even silicon co-designed makes for better systems.

I've built several tools like this over time. One was rdocul.us. The point of these tools was not to make me money, it was so that I would work myself out of a job. I was at Rails Conf '08 and and some guys in the back room were walking by and saying "it'll be like rdocul.us but BETTER" and I was like cool I don't have to care about that site anymore. It was that or go way too deep into the documentation engine for ruby than I wanted to at the time, crap didn't even build. The other open versions were so tightly coupled to the html output I was going to have to write somthing else. Given I built the site after I 3rd degree separated my shoulder on day 3 of a 12 day ski trip and I was just in the bar self medicating out the rest of the trip till my S/O showed up for the last 2 days, I wasn't really interested in doing more work. So good for them. Didn't even want to say "yo I built that" as I wanted them to feel like they were the underdog. doc sites are so much better these days and I like to think I helped in just a little bit by showing what an online and searchable site would look like with essentially push hooks from github (it polled repos every hour as there were no webhooks then, and anyone could register any repo to be documented).

I've had other OSS projects I've worked on where someone came to us about all the issues with our system and we just said "yea the tech debt we have with this is too high, it'd be a lot easier to re-write and also dump all the old features, but people would be mad if we did that." Which had us put it on life support and suggest the new tool that didn't have those issues.

That's what I as a software dev want out these types of projects. Show what's possible, have the platform holder realize it and do a much better job with full time paid devs. Stuff like FLUX are just better if they're part of the OS. I can't even imagine the hackery that goes into building them when you aren't part of the OS. And how little personal benefit (monetary) you'd get out of them. As opposed to how easy it is for apple to just add it and keep it working as a tested part of the os.

Personally I also get bored with maintenance after a time, esp once a project really solidifies and it is truly just ops work. In fact I know I'm bad at it as I'll go do something else and my previous project will suffer. I take this into consideration highly when deciding what to take on.

Before OS X I used slackware. At the time it did most of what I needed. I found the community around it lacking.

When I found OS X open source folks they were beyond welcoming. This is one of the bigger reasons Growl was on os x and not Linux.

This seems to be more of an illustration of the “platform owner copied and obviated your idea, so what’s even left to do?” problem.

That's exactly my point. There is a platform owner.

Reply on news.ycombinator.com