Haiku Final Release

Feedback Ideas
#2

It maybe good to start looking and thinking towards moving to R1 after Beta 4 so this doesn’t take another 10+ years

In all fairness, it takes however long it takes :slight_smile:
Haiku is an Open source project driven by a small community of users and developers. Labels like Beta/R1 are indicative of what that community wants from the project in its final form now. If those expectations change, so will the labels.

I hope the following happens otherwise Haiku may stay in development forever

Every major OS in use is also still under development, and I hope Haiku will be too :slight_smile: That’s how open source projects succeed.

And if one wants to see things happen, patches are always welcome :slight_smile:

#3

Last thing I want is a final release… For me, this sounds like, “Job done! Let’s get rid off that!” As much I’d like a release, I sincerely hope Haiku will survive to it and that it will only be the first of long line.

2 Likes
#4

I think “final release” is misnamed. There will be room for glass-elevator releases after R1.

3 Likes
#5

Yup. You understood me but some others may have not. :slightly_smiling_face:

Should push to finish off R1 release instead of just coming out with non-stop betas and RC releases for the next 10 years.

Development will still continue for release version 2 (R2). R2 would mostly add more functionality, features and improved performance.

#6

What is the value of R1 release? Making release don’t make bugs magically fixed.

3 Likes
#7

Considering Haiku is written in object-oriented C++, it should be modular enough that it can be maintained better than C-based OS attempts (including successful ones). Unfortunately C++ authors made more promises of maintainability than can easily be kept using C++.

In reality, RC1 should be a locked-in ABI for all core features. Any future release-candidates should focus on bug fixing and polishing to a full release status. Good progress has been made. It’s a shame to have to hide it behind an unstable beta release.

Rabbit Trails

In the early to mid 2000’s decade I was fixated on writing a spinoff of Amos BASIC that could work on both 68k and PPC models of Amiga computers. Under the monaker of Mattathias BASIC (named after the son of the Biblical prophet Amos), my programming partner and I purseued rabbit-trail after rabbit-trail.

Ultimately some of the paradigms needed to be pursued by others using other projects:

  • AmosPro had never been object-oriented so that needed to be relegated to the AmigaE family of languages.

  • AmosPro’s tokenization algorithm was never intended to implement a portable bytecode so that needed to be relegated to WebAssembly.

  • The reusable nature of the code we tried to write lacked a competitive optimization framework, relagating that function to GCC and/or LLVM (whose documentation has been improved remarkably since then).

  • The original AmosPro worked on the Amiga multimedia chipsets so getting old source codes to recompile on modern graphics cores using shaders and 3d acceleration has been relagated to AOZ studio (written by one of the original principle authors of AmosPro).

  • This leaves the only remaining priciple of design: transpiling BASIC to another language like C++ which has been relagated to my YAB2C++ transpiler.

In closing, Mattathias was never as great as the sum of its parts because it went too many directions at once.

QA and stem-to-stern support
#8

I argue that all the betas could have been already labelled as release (of course taken the less buggy it is, the better it is) . The value of a “release” label would be mainly on a PR and psychological front to announce that haiku is “mature” enough to “support” some “normal users” and that is the conservative approach not to have an “unstable” system. Following something else than “release” is at your own peril, and therefore be prepared for things breaking.
Currently not having the “release” label means that haiku is not “stable” enough for “normal” users to use. Again, it is mostly psychological because if one has been following haiku long enough, one should already be considered a “power user” and know those things. Moreover, it helps developers (again psychologically) be less tempted to introduce new exciting features for the release and more willing to iron things out (debugging and testing is relatively boring).

3 Likes
#9

Agree with you quentinsk. R1 would get lot more users coming towards, trying and using Haiku and maybe add some additional developers too. Maybe even get noticed more by technology companies, which may lead to more donations or help with developing apps or getting patches, etc.

Staying alpha, beta or RC tells the public not ready yet. Just keep on waiting. So only devs and enthusiasts will be using Haiku until R1 comes out.

2 Likes
#10

If an R1 is released, an extra branch must be maintained, as with the official betas. This must be regularly maintained and updated. Unlike the betas, no bugs should be built here, so everything needs to be tested more thoroughly before release.

After all, an r1 does not mean that development has stopped. Bugs are still being fixed and new drivers are added.

Haiku was already usable in the alphas and especially in the betas and nightlys. For many this feels like a release as many have had little to no issues.

But when the status of the R1 has really been reached depends on the assessments of the core developers.

#11

I don’t see it that way at all, because a lot of what the “normal user” today regards as a “must have” and uses daily differs greatly from what haiku can offer. The old among us who haven’t arrived in the “mainstream” because they don’t know the world like that often don’t even know what young people care about.

Things like twitter, facebook, instagram, clouds, games, online games, steam, ms office compatibility, watching videos online (netflix, amazon…)… everything is online or network dependent these days.

Haiku cannot serve all of these wishes, there are drawbacks here.

If we have too many cutbacks, many articles will follow where people vent their frustration. Would that make sense?

#12

So do you prefer to keep Haiku as a permanent beta not to disappoint the average public (since Haiku will always be behind the major OSes due to its nature)?

4 Likes
#13

No i’m writing that it’s not time to release an r1 yet considering what the users of today need.

And no, I don’t mean that the required applications have to come from Haiku’s side, but the functions should be possible (as far as possible).

#14

since Haiku will always be behind the major OSes due to its nature

Considering pretty much all the major OSes are making decidedly anti-user moves, I wouldn’t be so sure. If things keep progressing as they are, Haiku has decent chance of catching up, even if it will take a while.

3 Likes
#15

In a perfect world, sure. In this one? Not likely. Companies only care as far as the can make a profit.

Linux only gets boosted because of the licensing. FreeBSD has been used by large companies like Sony but nothing ever is given back because they don’t have to.

Just need more devs is all. I myself just don’t have the time and most here are in the same boat. What they’ve accomplished is no small feat. It’ll get there in time.

#16

Citation needed.

#17

Look at the Playstation 4. They used freebsd 9, yet had accelerated radeon drivers that never made it into mainline. There are plenty more occurences of that kind of thing that goes unnoticed as well for both bsds and linux. Just Google search it and try to inform yourself.

#18

I think you have this backwards.

There are many bugs to fix and Haiku is not stable enough for users (normal or not?) to use. Therefore we don’t make a release that’s tagged final.

If we did it anyway, we would just set the wrong expectations. People would think the OS is stable and ready to use, while it is not. The result will be more users, yes, but these users will soon find that they have been tricked, and that the product isn’t as good as expected. So they will be angry at us and leave again.

We only have one chance to do this, so we’d rather get it right. Until then, Haiku is good for users who are not made afraid by the “beta” tag, and, of course, these users are the ones who don’t get the difference between a beta and a final release :slight_smile:

It’s great if you are happy with the current state, but many other people would not be, and so, we have to accept that we will keep a relatively small userbase for some time until we are ready to set more ambitious targets.

It is indeed psychological, but not in the way you think: the “release” tag sets higher expectations, and we must make sure we are ready to deliver on these. Otherwise it will just generate disappointment, people will react something like “seriously? 22 years of work for THIS? are you kidding me?”. Which is not the reaction we want.

The beta tag sets lower expectations which are maybe lower than our current state, so maybe we can consider switching to “developer release” or “release candidate” versions at some point. Are we there yet? I don’t know, and I don’t think it matters, we could make new names for release states as much as we want, no one would really get what they exactly mean (it is already hard to agree on a meaning for “beta”). So in the end we may as well stick with “beta” for one or two more releases.

5 Likes
#19

You don’t have proof they haven’t contributed anything, because they did.
You are just sour they haven’t contributed the stuff you want, which they probably can’t contribute at all because all the NDA jazz.

#20

OK, so Exo, I’ve been around before the community was here on this platform. What I do for work gives me knowledge you do not possess. So, first thing, stop assuming things like you do. There is a good reason I’ve stayed away after some newcomers have brought some toxicity. This is my 4th account now and well probably not much longer.

You have a habit of not reading as you’ve proven, again and in the past. I’m not bitter because I don’t care. What you’re referencing isn’t even connected to what I’m talking about. Plus what they do is up to them not me. I have no dog in that fight. I’m just a hired hand.

At this point stop engaging me at all. I’m not tolerant anymore and your education is you’re own responsibility, not mine.

#21

Stop bickering, guys. Take it to PM or I’ll just remove those off-topic, partly personal attack posts.

11 Likes