rotating the idea of Feature Completeness. when you're not targeting infinite growth, you will reach a point where you're Done, where everything that needs to be there is there, where you can safely go into "maintenance mode" and do something else. we're unfamiliar with this in web services because targeting infinite growth is the norm; you always have to ship something new because if you don't keep up with the Fads then you'll be left behind. stability is death. but if you don't give a shit about adding Stories or Live Streaming or Generative AI Chat Bots or whatever then all the big parts will be done eventually. what happens then? skeleton crew for maintenance/operations/moderation? 3 day work week? pivot to being a game publisher? who knows! i sure don't. but it'll be a while before we have to figure that out.
- they/she
- jkap.io/
in reply to @jkap's post:
thank you for understanding the exact thing that galvanized patreon, instagram, and truly just every platform that started out pretty okay. incentive to get MORE money instead of just a consistent ROI, venture capital is a scourge on the sharing of information and art
you can reach that point, but there's still upkeep to do, and even if you're feature complete, the world may introduce new requirements that you aren't ready for.
my forum (jul.rustedlogic.net) is something of a legacy codebase; it was originally written for like php 3 or something in 2000~2001, reached "maturity" around 2006, and stopped getting major new work around 2007-8 and, outside of emergency fixes or manic bursts of inspiration, has been dormant
even with a "finished" codebase, there's still things like "the platform it runs on is gone" (mysql_ deprecation, then several incompatible syntax changes in php 8), or the world's insistence for https at all times (tcrf ran on debian 6 until about 4 or 5 years ago, and letsencrypt's certbot stopped supporting it, so i very suddenly had to figure out how to "fix" that without doing a full system upgrade - thx to whoever wrote a bash-compatible letsencrypt client you are a lifesaver). even if you're done, the rest of the world isn't, and unfortunately the players are increasingly bending it towards "follow what we say or perish" (forced upgrade from tls 1.0, and the like)
in the case of a commercial product, there's also the whole, you know, money thing. even if you aren't targeting infinite growth, you have to at least reach stable (and ideally rather comfortable), whether that's working on new projects or refining/improving the features you have whatever
the real fun is in making "being feature complete" a problem you have to consider at all.
a website that just, like, maintains itself and a high quality of service feels incredibly powerful, akin to actually fixing potholes with taxpayer money
i think it's fascinating to think about this re cohost because there is an end in sight. like, i can easily imagine a day where we all go "yep, cohost has enough features." we're pretty happy with it as-is! i'd like inline images, video uploads (premium $$ feature) and like... 20 nitpicks. it's like 500 person-hours of work and then the site would simply be finished. and you know i'm not saying it's trivial, i'm just saying that unlike all other websites, at that point we would all simply stop expecting anything of you other than for the place to continue existing. and this is the only website i've seen in 20 years where it might just fucking be possible.
Having spent the last 14 years mostly not adding features to already mature codebases but still working on them full time, this concept feels weird to me ๐
Here's hoping you find an answer someday. Long live Cohost, and long live the post-capitalist world.
I think about this constantly and I'm not even a web developer, nor do I develop anything with that sort of content cycle. I use Discord on a regular basis (read: for many hours every day) and I'm constantly floored by the number of things being added that just... Don't need to be there. Maybe I'm the weird one, but I feel like tech companies could by and large get away with just shipping a successful product and then never changing anything again. Pepsi is never going to go out of business by selling the same sodas for eternity, so what makes a website all that different?
One thing that will probably continue being a chalenge then, would be balancing upkeep costs and subscriptions. Server costs may change for the worse and suddenly this is a problem to solve.
But honestly, it's way easier to manage than having a greedy hole that ALWAYS hungers for more income no matter what. Upkeep costs are solvable, as opposed to investors' greed.
Heh, having worked on more than one 10+ year online service... people def. do not stop asking for stuff and you def. do not stop having ideas of things to do. But hey if you don't have an unfeasibly huge staff to start with, it's a pretty gentle transition into "everyone has more side projects".