> The ecosystem is vibrant, healthy, and will keep growing. > > in my mind there will eventually be a tipping point,

Why do so many people insist on Growth and Largest, as metrics for 'success'?

My local pub is a success. They are not worlds' largest pub. My Mom and Granny don't go there. They don't have parties with millions of people crammed in the place. Etc.

Why can't Linux be called 'a success', the very moment that it helps one person? Why don't we call Mastodon or Peertube a success the very moment a tiny group of people uses and likes it? Why do we insist that Peertube is only successful when my mom and my Granny use it? Why do we insist on huge market shares to call open source successful?

Your local pub might be trying to offer an alternative to local people staying on their homes watching tv instead. Linux and peertube are trying to offer an alternative to something of a much bigger scale, and most likely with more pressing ethical concerns (they often want to drive people away from the current mainstream option for ethical reasons). Not everyone might share that view, but they are definitely different situations. Not saying you don't have a point, but we can't ignore the differences in context.

> Linux and peertube are trying to offer an alternative to something of a much bigger scale, and most likely with more pressing ethical concerns

True.

But my point is that you could call both "a success" the moment that one singular person uses it as alternative. For that person it is a success!

With Peertube you need some network-effect, because one single person uploading and then watching those videos, can't be called "a success". But if a tiny niche, say a small group of specialist beekeepers, uses it to share instructional videos and no-one else in the entire world uses it, it can still be called "a success". Because for them it offers a solution.

I used to think differently, but I am starting to agree. Part of the issues with current generation of software is that it is effectively intended for masses, who barely understand what is happening under the hood or what is possible resulting in software that looks pretty, has severely limited options and constantly calls home for 'diagnostic' purposes.

That said, commercially, it is hard to argue with adoption as a metric. Problem is, focusing on that metric results in the same bs as in the real world: consumer ( crap tier ) and business ( good tier that tends to have good options if you are willing to spend money ).

Cost of physical distance enables niches that can be served by your local pub. If we could teleport (well, and cohabit the exact same space and tune out conversations of others at will so crowdiness is not an issue) they would have a much much harder time selling something unique. Scale aids in bringing people in, because it becomes the default choice.

You contradict, and invalidate yourself, in your own argument.

> they (the local pub) would have a much harder time selling something unique

> Scale ...becomes the default choice

Scale and monopoly removes choice and uniqueness.

There are people that love Starbucks, there are also people that love their local coffee shop.

I don’t those two thing are contradictory, it’s just that you are assuming you don’t get uniqueness inside a monopolistic platform, which is only partially true. The platform is fixed, but you get uniqueness inside the platform, with the content and contacts you follow. Facebook, reddit, etc. are not the same experience for all users. Comparing a network, where each vantage point is unique with Starbucks, which has maybe 50 products, is not an apt analogy.

I tend to avoid larger pubs because the quality of experience in smaller ones tends to be better.

Quantity has a quality of its own, and it's not always a good thing.

Aren't local pubs in the UK and certain other countries now mainly owned by corporate chains, not independent? In that case, there is definitely a concern with growth, though it might not be readily visible to patrons.

> The ecosystem is vibrant, healthy, and will keep growing.

I'm in favour of it, and hope it does well. But at the same time, constant growth itself doesn't mean that much.

There's a big difference between slow constant growth and explosive constant growth. Something can constantly grow yet still remain something most people have never heard of. I think Linux on the desktop is a reasonable example of this.

The alternatives will likely be constantly growing as well.

> in my mind there will eventually be a tipping point

it's going to be difficult to compete against YouTube... there has to be some reason why people use it instead of YouTube.

> there has to be some reason why people use it instead of YouTube.

I had non-technical people in my family asking me Youtube alternatives because they were afraid of censorship. I think the reasons are already there[0] but there are no "peertube marketing department" taking care of conversion.

0: https://en.wikipedia.org/wiki/YouTube_suspensions

YouTubers have been complaining about YouTube for nearly a decade now :

http://angryjoeshow.com/ajsa/forums/topic/10325-youtube-beco...

http://angryjoeshow.com/ajsa/forums/topic/10553-youtube-copy...

https://web.archive.org/web/20200806101856/http://www.joueur... (fr)

But in 2013 there were no real alternatives. This has started to change in the recent years.

That has me concerned that nutjob right wing conspiracy theorists are going to flock to PeerTube just as they did voat and parler. So long as they maintain their own stuff which is what I understand as to how this all works, thats fine. I just wouldn't want their stuff to pop up in other instances so that they're filtered out by default.

You know, Linux in general is probably the best example of this. Its success doesn't need to be defined by the desktop.

When people bring up desktop Linux, I always bring up that Linux impacts 100% of internet users because of its presence running web applications, and it impacts however many people who use Android or other Linux devices (embedded/IoT). Automotive is even a huge sector these days.

Depends on how one pats their back and the actual reality.

Yes, Android runs on Linux, yet that is 100% meaningless for app developers targeting Android.

Web developers don't care about what OS the browser is running on.

Native Android developers have a Java Framework and Kotlin libraries to work with, which are completely unrelated to Linux.

However they also have the NDK, one would say. Yes, indeed.

So what are the public APIs?

https://developer.android.com/ndk/guides/stable_apis

And recent versions will kill apps that try to be clever and access private APIs,

https://android-developers.googleblog.com/2016/06/android-ch...

https://android-developers.googleblog.com/2017/07/seccomp-fi...

As for automotive, while Automotive Grade Linux is getting some love, it is still not an option for real time requirements.

RT_PREEMPT is very much capable of real-time applications. Outside of control loops faster than a couple kHz, it's even suitable for hard real-time applications.

But those issues aren't fixed with Linux, but rather with CPUs design for hard real-time usage.

Industry seems to think otherwise.

That still isn't a match for something like INTEGRITY OS, vxWorks, QNX when human lives are at risk.

Or why despite using an heavily modified Linux kernel as basis for Azure Sphere, Microsoft rather advocates Azure RTOS for real time tasks.

https://techcommunity.microsoft.com/t5/internet-of-things/co...

Or why the Linux Foundation, ironically pushes Zephyr for such purposes, completely unrelated to Linux source code.

https://www.zephyrproject.org/

If Linux with PREEMPT_RT is unsuitable, all the modern x86 CPUs with system management mode interrupts are even worse. Because those are the worst and primary cause for that Linux being insufficient for control loops in the 10's of kHz.

Why does it need to compete with YouTube? Is it really a zerosum situation?

It doesn’t need to. But obviously there are people who want it to be at least fairly well used.

The value of such a thing has a lot to do with how many people use it and how much content there is. You need users to give incentive for people to upload content. And you need content to incentivize new users.

I disagree. Value here is a matter of quality, not quantity, and impact on humanity's welfare. Most "network-effect successful" networks have a low and decreasing signal-noise ratio, they merely produce unqualified "content" for the most part.

Tech diversity in itself is a value, even if the niche players never leave the niche for the most part. Diversity allows for adaptability in a changing environment/crisis.

I believe the mass social network epoche is about to end, as the advertisement/influence ecosystem is eating up its island's last trees. The dream of the global internet failed, for reasons we can't fix. ActivityPub is much better fitting a fragmented network.

Yup, same with whats going on in defi with stable coins… as the existing system crowds people out and benefits fewer and fewer people more, more people will be incentivized to seek out alternatives