Beta phase!

Development OS
1

I would suggest that Haiku should enter beta phase now and all new features applying to R1 be delayed to R2 if possible. The longer we wait for R1 the harder it is going to be for Haiku to be compatible with the computing requirements of its users. Once R1 has been released then work should begin on R2 by making the default GNU Compiler Collection version 4.X.

While I understand that being BE OS 5.1 compatible is important, I also believe that stagnation is a real danger here. Be OS was and still is a brilliant and wonderful OS! It had features that truly were innovative and ahead of its time. While I believe Haiku should be a reimplementation of Be OS, it should not be a mirror image of a dead OS forever!

Again I am forced to put away my love and admiration of Haiku here temporarily and look at the state of the matter! In the beginning it was and still is an admirable thing to try to preserve Be OS and it’s legacy. But it has been eleven years since Be OS died and there has not yet been a R1. Here we are on the verge of a fourth Alpha release and still many bugs are yet to be fixed and missing features have yet to be developed!

My opinion is to drop all features that have been planned for R1 that have not been done and wait to do them in R2. This way we start working on a Beta 1 and hopefully within a year to two years we will have a R1! Then work can begin on R2. Depending on an antiquated gcc 2.95 is already causing problems and will continue to cause problems! There are many programs that will not compile on gcc 2.95 and having two C compilers adds extra complexity!

That is my two cents! Spend it wisely.

2

Now that document viewer is out, I don’t think that gcc2 is needed, unless there are more closed source apps that don’t yet have a replacement.
As far as beta phase goes, we shouldn’t go into beta until we are almost feature complete with the basics of an OS. Feature complete to me means at least wifi and probably package management. I hope these two things are what the majority of the donations go towards.

3

I also agree, there is nothing worse than nothing happening. It’s stable enough to go into Beta. And really, I just can’t wait!

4

It would be great if webpositive gets html 5 and css3…

this would be a visible and a surprise feature for most ordinary PC user.

5

I’m not too familiar what alpha and beta technically mean in software development, what separates them, but what I know is that it’s been too long from last official release. My secondary desktop PC has Haiku on it. It’s alpha 2, since alpha 3 refuses to even run so long that I could test it out, even from CD. Random nightly builds I have tried all end up in KDL on boot from CD. So it’s alpha 2 for me, for way too long. I boot it maybe once in 2-3 months now, I tweak around and that’s that for another 2-3 months. So I have nothing to give back to community and I also feel like my connection with Haiku is dying. So please, whatever it is - alpha, beta, R1, 2, 3 - please put something out that can be tested, talked about, compare to and keeps the community alive.

PS. Big thanks to You who You are developing Haiku. I truly appreciate Your work. I wrote about how I feel, and I know nothing about developing OS. It’s just to give You a layman’s view.

6

Have you reported a ticket for that?

7

Its been way too long, I hardly have any interest anymore
Ive waited 11 years, and lately, it just doesn’t seem worth the wait anymore

I don’t understand why you cant have something right here, on the Haiku website, to tell people what and when something may actually happen

I tried A1, 2 and 3, and each time, I had to find out about the download from OSnews, because Haiku.org doesn’t tell you anything until the day it happens

I dont care if R1 is 10 years away, as long as something happens every few months. If its not R1, I dont care, just have something new every now and then

Knightly builds are too hard to understand or use for the average person, they need a A3 . 5, or .6, .8 etc

Just give us something, to show us something is happening

When A1 came out, I told everyone I know about the good news, and that they would be able to try it soon. Now they dont care anymore, and I dont either. Its just been too long

A3, was almost useable, but not ready for 24/7 usage
Any little update or fix, could have made it useable 24/7
But waiting another year, for something with lots of updates or fixes
Just aint keeping people interested anymore

8

Have you reported a ticket for that?[/quote]

Hi, it was long time ago but I think I posted about it on forum. IIRC I had problems with TRAC back then, like now. But the possible bug has a ticket now.

9

I kind of agree, taking too long, not keeping Haiku in the press with some new release info, no hardware that we know Haiku will run on so it is buy and hope for the best kind of approach, and it just seems like a stalled project with too little developers doing anything anymore.

Don’t get me wrong, I understand it is a project that developers make Haiku in spare time it seems free of charge, but at the same time, it finally seems like our beloved BeOS that had become Haiku has so little contributors and news and updates it has become a project not many care about anymore.

I only keep an eye here on the forums because I still pimp a BeBox and have been hoping and dreaming one day Haiku would have a HaikuBox that is known to run Haiku 100% and could be used also to run other OSes while Haiku was being worked on.

Today it is still buy and hope Haiku works, too little, seeming too late.

I guess we can still dream of the BeOS glory days and tinker with Haiku but was hoping Haiku would be more than tinkering OS.

I still thank the people that made Haiku happen and for their hard work but I guess like anything, time and years puts things to rest, and things out of site become out of mind, and maybe that is what we are experiencing today with just a few soles still working on Haiku but just not enough to make something happen.

tj

10

Also, from where I am coming from, compare Haiku with other OSes and platforms of old like Atari and Amiga. Both Atari and Amiga have a vibrant following, new hardware and software still being made to this day.

Why is that? Is it because Atari and Amiga have know hardware that supports their favorite OS? On Atari you have individuals that have made the new Firebee computer that is actually getting in some folks hands these days and with Amiga, systems like SAM 460 and AmigaOne X1000.

Is it that these platforms have more “crazy ones” like the Apple commercials of old that really LOVE these platforms and Haiku/BeOS just had too little?

Or do these other platforms make it because they have unique hardware that the OS is made to work on, even if outdated to some degree, but keeps these platforms alive.

Would a HaikuBox that developers can develope Haiku for that at the same time runs other modern OSes be what it takes to keep Haiku moving forward?

I know I would buy a system that is made to run my specific OS of choice like Haiku yet can run some Linux distro to get us though when we need something today not yet available for Haiku.

Give something tangible Haiku can sell to us and money can be made to help make Haiku better. Without something to sell us, I think it has just stalled project with not enough people to do anything anymore.

TJ

11

Get the fact: chrome is version 19 nowadays, firefox 12, MacOSX 10.x.bla.bla… in almost ten year Haiku is still R1 alpha 3. GCC is 4.7.0 and Haiku uses a version 20 years old to keep compatibility with a piece of software written for BeOS 20 years ago… And perhaps it is a text editor or stuff like this…
When I find Haiku I thought it could have helped me get rid of GNU/Linux, in a year we are where we were. I appreciate the work of developers, what is missing is a plan. A plan can help new devs to get involved. A plan can help Haiku to get attention. Let’s say, a release every year, every year a task: R1, internet, reduce number of critical bug and improve BFS stability. R2, from GCC2 to GCC4 or LLVM/Clang or whatever and package manager. R3, office and a complete browser. R4, video and audio drivers. R5, bluetooth, DVB and a project of a media center. In five years there would be a new OS. Today is: for R1 we need stability, but also internet is lacking, and what about a pkgmanager? Yes, we need a browser and office, and also video drivers.
R1 tasks changed three times since I’m following Haiku… it doesn’t help.

12

For me at least the alpha releases should happen more frequent. One month ago, there was the intuition/feeling that the alpha release would happen any next days, and now a month past and it’s still not here. I was awaiting it, and not continuing work on the documentviewer in the idea, that i better wait till alpha4 is released, so that i dont need to setup my machine twice.

The problem is, that there are also no information about the delays, so that one could understand it, and to be able to know approximately how long the delay will be.
Somehow it’s the feeling of waiting in the darkness.
the communication could be really better…
and at least on the google+ haiku-page one could give more informations, if it is not desired to put such kind of informations on the front page of haiku.

13

I agree with the general sentiment here that the project does seem dead. It may not be the case, but that’s what it looks like to an outsider. It’s bad enough that ALPHA versions are being released only once a YEAR, but it’s been almost 14 months now with no real news on the next alpha or beta stage. And let’s be honest, as cool as Haiku is, it’s been in development for what, 12 years? And it still is not in beta.

With no news on the web site or anywhere and no communication on what’s wrong, what is good, any plans, additions, etc… there’s just dead silence.

What incentive is there is there for developers to waste their time developing for a seemingly dead OS? There isn’t. You’re getting yourselves into a vicious cycle of abandonment by not offering any communication with the userbase

14

I guess I don’t quite understand what more can be done.

The goals for R1 are on the wiki and haven’t changed much as I understand it. I use the RSS feed on the homepage to recieve updates from the developers. New working versions of Haiku are released nightly, and I update my installation every few days. I follow the developer mailing list to read some interesting discussions and advancements. If I wanted more information, the homepage has a rolling feed of bug reports / fixes and source code changes.

Many things have advanced in Haiku over the past year, including wifi support, modern Radeon video card support, printer support, and web browsing support. There has been a lot of work on integrating many Qt applications, although that doesn’t interest me very much.

What more would you like the Haiku developers to do or talk about?

15

Problem is, to me, that there’s not a plan. Keep on adding things on R1, and R1 becoming more and more distant in time, that’s what I’m noticing. Still there’s going to be a Alpha4, whose date is not known. Then there will be a beta release, perhaps it will take a year or so. And then another beta maybe, and a RC, and so on. Haiku R1 will be release in 2015 maybe, and it will be again outdated. And devs number will be the same, if not fewer.
All I ask is this: http://techbase.kde.org/Schedules/KDE4/4.9_Release_Schedule
and this: http://techbase.kde.org/Schedules/KDE4/4.9_Feature_Plan
Proceed at little steps, development will be easier, because you don’t have to complete 25 task at a time, and Haiku will gain in popularity.
When I see Ubuntu releasing a new version, I always give it a try. Let’s say in six months you pass from R1 to R2, bloggers, journalists, hobbyist, students would say: “Hey, let’s see how’s it going”. If in 14 months you pass from alpha3 to alpha4, I don’t think everyone cares, the thoughts are “well, let’s wait for the stable release”…

16

I think for launching Haiku there is a first TODO list that should be satisfed:

  • all NVidia drivers & other drivers included
  • Firefox
  • a bootloader that support to switch between windows and Haiku (installed by the installer)
  • BEOS Partition creation on an empty space
17

[quote=oueshcousin]I think for launching Haiku there is a first TODO list that should be satisfed:

  • all NVidia drivers & other drivers included[/quote]
    No NVidia’s specification are not available so it would consume many developer resources for a small improvement since Haiku does not have accelerated 2D/3D.
    VESA is enough.

Porting Firefox is too much work and Webkit/Webpositive is quite good + Alexandre is going to work on it.

[quote=oueshcousin]- a bootloader that support to switch between windows and Haiku (installed by the installer)
[/quote]
Already done! You may install Haiku BootManager, boot on an USB drive or you can “chainload” Haiku from Grub2 or Nt_Loader.

Already done! Just use DriveSetup! (Linux fdisk also does the job)

18

Get the fact: chrome is version 19 nowadays, firefox 12, MacOSX 10.x.bla.bla… in almost ten year Haiku is still R1 alpha 3. GCC is 4.7.0 and Haiku uses a version 20 years old to keep compatibility with a piece of software written for BeOS 20 years ago… And perhaps it is a text editor or stuff like this…

I can’t even begin to explain how wrong this is.

Version numbers mean jack all. Look at Firefox 4-12 for example. There’s hardly any noticeable difference between consecutive versions, because of the rapid release cycle Firefox changed to. If you compare Firefox 3.0 with Firefox 4.0, you’ll see a world of difference. Let’s look at Windows now: Windows 95 has an internal version number of 4.0, Windows 2000 has an internal version number of 5.0, and Vista has an internal version number of 6.0. You can’t tell me that the progress between versions of Windows is comparable to versions of Firefox.

Onto the compatibility argument. For any OS to be successful, it needs applications. Like it or not, Haiku is a reimplementation of BeOS, and as such, is required to be compatible with BeOS applications. The downside to this requirement is that it needs to support a rather outdated version of GCC (by 12 years, much shorter than your 20 years, which actually predates the public release of BeOS by about 5 years anyway). If you don’t like it, use a GCC4 nightly, but don’t come moaning to the devs when none of your applications work. The official releases are GCC2/4 hybrids anyway, so apps can be written and executed for either GCC version anyway. (WebPositive being a good example of a GCC4 application).

19

[quote=The123king]
I can’t even begin to explain how wrong this is.

Version numbers mean jack all. Look at Firefox 4-12 for example. There’s hardly any noticeable difference between consecutive versions, because of the rapid release cycle Firefox changed to. If you compare Firefox 3.0 with Firefox 4.0, you’ll see a world of difference. Let’s look at Windows now: Windows 95 has an internal version number of 4.0, Windows 2000 has an internal version number of 5.0, and Vista has an internal version number of 6.0. You can’t tell me that the progress between versions of Windows is comparable to versions of Firefox.

Onto the compatibility argument. For any OS to be successful, it needs applications. Like it or not, Haiku is a reimplementation of BeOS, and as such, is required to be compatible with BeOS applications. The downside to this requirement is that it needs to support a rather outdated version of GCC (by 12 years, much shorter than your 20 years, which actually predates the public release of BeOS by about 5 years anyway). If you don’t like it, use a GCC4 nightly, but don’t come moaning to the devs when none of your applications work. The official releases are GCC2/4 hybrids anyway, so apps can be written and executed for either GCC version anyway. (WebPositive being a good example of a GCC4 application).[/quote]
I was thinking about an answer, all I have to answer on your first part is: and so what? Do you think that Firefox decided to begin crazy numbering cause they were tired? or because people is dumb and want to hear high version number? Above all Haiku needs attention: pubblic attention, dev’s attention and perhaps one day software house attention. We’re not talking about a lib, that can be version 0.0000018 and none cares. We’re talking about an OS…
Second part: I thought that Haiku was meant to take the “inheritance” of BeOS and taking it further. Do Haiku want to have a wider user base? Well, you cannot think that people using GNU/Linux, or Windows, or MacOSX will try apps that were written almost ten years ago.
Then, if Haiku wants only to become the clone of BeOS, and this will take three years more, sorry, I misunderstood everything in my life. If, let’s say, in one year Haiku devs will release a R1, how can I advice my friends to try an OS that has not a real plan for a better R2 in few times and how can I tell them to play with an almost incomplete browser, no office app, no drivers, some old audio tools, almost no recent vlc release?
Anyway, perhaps I’m not talking a good enough english for people to comprehend me, so, here’s what I’m asking:
R1: drivers, wpa, pkg manager, documentation. Release Jenuary (of an imaginary year 1).
R2: switch to GCC4, complete web browser, initial release of office. Release June (year 1)
R3: complete office, x64 initial support (don’t know what else). Release January (year 2)
And so on. Here’s what I’m seeing:
R1: complete web browser, complete drivers, initial office, stability, documentation, initial GCC4 switch, works on ARM, pkg manager, wpa, java… Release don’t know when.
Wouldn’t it be easier? Make a plan, take new and old devs, work on feature, release. I’m not asking the moon, I don’t care if there will be 14 release in 7 years to have a complete and stable OS, but at least when I’m following blogs, web sites, forums about Haiku I could always see where it is and where it is going.
Am I crazy or what? I follow loads of projects, everyone works like this, from text editor to KDE, from web browser to linux kernel.

EDIT: only Debian has a slow release cycle but, ok, they have a huge user base and a huge attention…

20

[quote=franz1789]I’m asking:
R1: drivers, wpa, pkg manager, documentation. Release Jenuary (of an imaginary year 1).
R2: switch to GCC4, complete web browser, initial release of office. Release June (year 1)
R3: complete office, x64 initial support (don’t know what else). Release January (year 2)
And so on. Here’s what I’m seeing:
R1: complete web browser, complete drivers, initial office, stability, documentation, initial GCC4 switch, works on ARM, pkg manager, wpa, java… Release don’t know when.
Wouldn’t it be easier? Make a plan, take new and old devs, work on feature, release. I’m not asking the moon, I don’t care if there will be 14 release in 7 years to have a complete and stable OS, but at least when I’m following blogs, web sites, forums about Haiku I could always see where it is and where it is going.
Am I crazy or what? I follow loads of projects, everyone works like this, from text editor to KDE, from web browser to linux kernel.

EDIT: only Debian has a slow release cycle but, ok, they have a huge user base and a huge attention…[/quote]

When you install Haiku on real hardware( sometimes impossible - http://dev.haiku-os.org/ticket/7665 ). You can see a cool, fast and classic operating system. With potential. But Haiku has serious lacks in every place. Many people are not able to connect to the Internet. How look the application to configure Internet connection?

All of this is needed to do.

Users do not come to us, just because you change a number.

For me it is a sick, that a small change gets a major version number( Chrome, Firefox).

Haiku development will accelerate after relase of planned R1. Because all of the important things, will be already.