|
Over the years, I've seen my own projects get torn apart or receive praise, and it feels like (though I certainly can't prove it) there is a direct correlation to how much I test before releasing. Because we have such a wider audience on the Web, it's really hard to develop an experience that fits everyone's expectations. And everyone definitely has their own expectations, with almost no consideration that theirs isn't necessarily the majority position. I've had people on Internet Explorer for Windows Phone bitch me out for not supporting their platform. Unfortunately, their noise can cause an out-sized effect on other people's opinion on other platforms, so you can't easily say that it's not cost effective to support niche platforms. On the other hand, something like an iOS app has such a smaller market, such a smaller set of potential use cases, with much higher activation energy, that in comparison it's a cake walk. The hard thing is that, if you're just starting out, you don't know what you don't know. I think it's possible to release a JS project that gets received well. But there are just so many things you have to take into consideration before releasing a project. I had started to write a list of considerations, but I quickly realized it was getting unwieldy fast and I don't want to be here all day. I think it all comes down to "make the technology transparent". People shouldn't know what you used to build your project unless they go to your Github repo and see it immediately in your README. And I mean that in both the sense that there should be negative consequences to your technology choices that bleed through to the user, as well as you should make no mention of it in your marketing materials. Because marketing material should be 100% focused on selling, and the technology used to build a thing is so very rarely ever a good selling point. At best, it's a distraction. At worst, you'll alienate people for no good reason. Put another way, it's hard enough to get people to show up, don't hand them excuses to leave. We get told to "release early, release often" a lot. And I believe in that concept very wholeheartedly. But it's in regards to features, not to defects. Work your TODO list until there are no known defects. HN is definitely a lot better than Reddit, though. Reddit is random, as far as I can tell. Whomever is the first poster gets to set the tone for the thread, and then it's just piling on after that, usually devolving into inside-jokes repeating sound-bites (which are expressly forbidden here and/or will get you down-voted to hell) |