I remember some years ago, there was a “R1 Features Poll” (that poll was open to all the community), and the “Package Manager” was one of the “must have” in the results.
I remember that too, does it still exists. I believe “user democracy” in this case is a mistake, the decisions must be taken by one head developer, if I recall correctly it was open to forum members. It should be interesting to see those results again.
EDIT: here are the results
1 Like
About the forks: there are two known ones currently. One is “Poem”, which is available now, and based on Haiku Alpha 4.1 (so it’s Haiku in the state it was in 2012). If you can’t stand the idea of a package manager or the way we implemented it, maybe Poem is a better choice for you.
The other fork, or rather, fork attempt at this point, is Senryu. Senryu existed in the past as an Haiku distribution including a lot of extra software. It was made by the guy running Haikuware, a website hosting software for Haiku. So far, any visible activity he had is announcing the fork, raising some money to fund it, then announcing a kickstarter campaign to fund it even more. There may be more things going on in private, I don't know. Note that the man behind the project is not an Haiku developer. While he did a great work over the years by keeping haikuware and bebits.com running, he never contributed to Haiku source code, or, to my knowledge, 3rd-party applications. So the success of his fork relies on managing to find some developers to work on it. If my understanding is correct, he plans to attract developers with a "bounty" program, where users can fund a particular feature. He experimented this for a while on the haikuware website, and while it did bring some developments to Haiku, most of the developers in our team agree that it is not a very good way to fund the development.
So, for now there is not much to see about Senryu. We have to wait a bit more and see what they offer in their Kickstarter campaign, where i hope there will be more information about the plans.
Note that in this vote, you have results from both the devs and the users. The user vote was not forcing the devs on anything, but more a way to gauge interest from users for each of the proposed features.
[quote=PulkoMandy]About the forks: there are two known ones currently. One is “Poem”, which is available now, and based on Haiku Alpha 4.1 (so it’s Haiku in the state it was in 2012). If you can’t stand the idea of a package manager or the way we implemented it, maybe Poem is a better choice for you.
The other fork, or rather, fork attempt at this point, is Senryu. Senryu existed in the past as an Haiku distribution including a lot of extra software. It was made by the guy running Haikuware, a website hosting software for Haiku. So far, any visible activity he had is announcing the fork, raising some money to fund it, then announcing a kickstarter campaign to fund it even more. There may be more things going on in private, I don't know. Note that the man behind the project is not an Haiku developer. While he did a great work over the years by keeping haikuware and bebits.com running, he never contributed to Haiku source code, or, to my knowledge, 3rd-party applications. So the success of his fork relies on managing to find some developers to work on it. If my understanding is correct, he plans to attract developers with a "bounty" program, where users can fund a particular feature. He experimented this for a while on the haikuware website, and while it did bring some developments to Haiku, most of the developers in our team agree that it is not a very good way to fund the development.
So, for now there is not much to see about Senryu. We have to wait a bit more and see what they offer in their Kickstarter campaign, where i hope there will be more information about the plans.[/quote]
and does not forget “discover haiku” from tts. this stick is just like a distro without name change ;-). a haiku version with package management und various software (qt apps too). good solution to take look and work with haiku. and on top a trial version of the tunetracker radio station software.
i have tried to download poem, but this does not match with bezilla,qzilla,webpositive.
I tried the VM image in VirtualBox (after fiddling around to get VirtualBox to add the image…). It’s about what I expected: Haiku alpha4 with (to me) horrible icons.
No insult intended, but I kinda doubt the forker’s ability to provide a “‘hybrid’ Gnu/Linux edition”. Other than forking ZevenOS.
Regards,
Humdinger
Prop #28 Driver issues (gfx, audio)
•All known bugs that render supported hardware useless to be resolved. (eg, distorted display)
must have 74%
Currently there are a lot of outstanding issues with Intel HD graphics which are still awaiting resolution - Ticket #8952,#9007,#9651,#9949,#11151,#11180,#11248,#11254,#11268
Currently the “fix” is to buy another video card and use that instead of the on-board graphics.
http://fatelk.com/faceplant.htm
It has been about two years since I wrote this. Not much has changed except that now there are fewer users and fewer application developers. Haiku looks very much like an OS in its death throws.
[quote=bbjimmy]http://fatelk.com/faceplant.htm
It has been about two years since I wrote this. Not much has changed except that now there are fewer users and fewer application developers. Haiku looks very much like an OS in its death throws.[/quote]
It has been like this for ages, some developer leave and some new emerge 
I think that the pike was back in 2001-2005 :). But I think that today we have more developers than 3 years ago. (Active developers)
We have numbers to back our claims of still being alive: http://pulkomandy.tk/stats-201508/activity.html
The peak was actually around 2008-2010 (look at the "commits by year" graph)
http://pulkomandy.tk/stats-201508/authors.html has a "number of authors" in the "commiter of the month" graph. It was more than 20 for the first time in 2009, more than 30 for the first time in 2012, and currently in the range of 17 to 26.
I suppose that the peak you’re thinking of was of “developers” who responded to the initial OpenBeOS announcement by Michael Phipps. It’s actually kind of fun to look back at that because Michael had absolutely no prior experience, this wasn’t just his first operating system, or his first Free Software project, as far as I can tell it was the first time he’d organised anything bigger than a birthday party.
But while there’s plenty of enthusiasm showing in those early days there weren’t actually many developers. Michael found it very easy to attract people who had opinions, people who wanted to lead a group, be beta testers, or who had blue sky ideas for how the OS should improve upon BeOS, but much harder to get people who actually write code. This somewhat mirrored Be’s experiences with the BeBox, subsidised hardware intended to bootstrap a developer community for their product. Some people bought it to do exactly what Be intended, they built custom hardware, they wrote interesting software, or at least they ported big components from the outside world and gave useful feedback on the OS. But mostly people added it to their pile of curiosities they meant to get around to playing with later. Thousands of dollars of Be’s investor capital were effectively transferred to random hobbyists who liked to be different for the sake of being different. And meanwhile cheap consumer hardware that was just as good was coming onto the market. Not an efficient use of limited resources.
The period about ten years later is much more a peak of actual developers of Haiku. There was a perception, false as it turns out, that Haiku was “almost finished”. You could still make a real difference to a brand new OS if you got in quick. The OS was self-hosting, Haiku R1 was said to be 12-18 months away. And then another of Be’s old problems surfaced, shipping is hard. Releasing a product requires a different skillset, orthogonal to the skills needed for development. So there was the joke of “alpha” releases spaced a year apart, and then despite a post mortem identifying the need to do more releases instead of getting better it got worse.
Anyway, bbjimmy specifically said application developers. The OS exists to run applications. One choice you could make is to be very good at running other people’s applications, that’s why Windows is a huge success, and it’s why Linux is a huge success, indeed it’s even why the SCO lawsuits started, replacing a SCO system with Linux was a much faster and cheaper way to run the same application software. Haiku is not, to be frank, very good at running other people’s existing applications. So that leaves the other option, having your own compelling applications so that people are obliged to use your OS to get those apps. That can’t happen without an army of application developers.
I’m an application developer. My day job is working on some Qt/C++ applications that run on Linux and Windows and by evening I’ve been writing some Haiku applications. For fun. It’s also another valid motivator.
If I just wanted to run a lot of applications, I’d be running Windows or Linux.
I suppose that the peak you’re thinking of was of “developers” who responded to the initial OpenBeOS announcement by Michael Phipps. It’s actually kind of fun to look back at that because Michael had absolutely no prior experience, this wasn’t just his first operating system, or his first Free Software project, as far as I can tell it was the first time he’d organised anything bigger than a birthday party.
But while there’s plenty of enthusiasm showing in those early days there weren’t actually many developers. Michael found it very easy to attract people who had opinions, people who wanted to lead a group, be beta testers, or who had blue sky ideas for how the OS should improve upon BeOS, but much harder to get people who actually write code. This somewhat mirrored Be’s experiences with the BeBox, subsidised hardware intended to bootstrap a developer community for their product. Some people bought it to do exactly what Be intended, they built custom hardware, they wrote interesting software, or at least they ported big components from the outside world and gave useful feedback on the OS. But mostly people added it to their pile of curiosities they meant to get around to playing with later. Thousands of dollars of Be’s investor capital were effectively transferred to random hobbyists who liked to be different for the sake of being different. And meanwhile cheap consumer hardware that was just as good was coming onto the market. Not an efficient use of limited resources.
The period about ten years later is much more a peak of actual developers of Haiku. There was a perception, false as it turns out, that Haiku was “almost finished”. You could still make a real difference to a brand new OS if you got in quick. The OS was self-hosting, Haiku R1 was said to be 12-18 months away. And then another of Be’s old problems surfaced, shipping is hard. Releasing a product requires a different skillset, orthogonal to the skills needed for development. So there was the joke of “alpha” releases spaced a year apart, and then despite a post mortem identifying the need to do more releases instead of getting better it got worse.
Anyway, bbjimmy specifically said application developers. The OS exists to run applications. One choice you could make is to be very good at running other people’s applications, that’s why Windows is a huge success, and it’s why Linux is a huge success, indeed it’s even why the SCO lawsuits started, replacing a SCO system with Linux was a much faster and cheaper way to run the same application software. Haiku is not, to be frank, very good at running other people’s existing applications. So that leaves the other option, having your own compelling applications so that people are obliged to use your OS to get those apps. That can’t happen without an army of application developers.[/quote]
So port docker to Haiku?
Because, lets be frank, Docker is the ELF of the future, and right now other OSes are gaining capabilities that will most probably wipe Linux out in the near future, i.e. SmartOS, FreeBSD etc.
I would love to see this happening, but I am not very optimistic 
I suppose that the peak you’re thinking of was of “developers” who responded to the initial OpenBeOS announcement by Michael Phipps. It’s actually kind of fun to look back at that because Michael had absolutely no prior experience, this wasn’t just his first operating system, or his first Free Software project, as far as I can tell it was the first time he’d organised anything bigger than a birthday party.
But while there’s plenty of enthusiasm showing in those early days there weren’t actually many developers. Michael found it very easy to attract people who had opinions, people who wanted to lead a group, be beta testers, or who had blue sky ideas for how the OS should improve upon BeOS, but much harder to get people who actually write code. This somewhat mirrored Be’s experiences with the BeBox, subsidised hardware intended to bootstrap a developer community for their product. Some people bought it to do exactly what Be intended, they built custom hardware, they wrote interesting software, or at least they ported big components from the outside world and gave useful feedback on the OS. But mostly people added it to their pile of curiosities they meant to get around to playing with later. Thousands of dollars of Be’s investor capital were effectively transferred to random hobbyists who liked to be different for the sake of being different. And meanwhile cheap consumer hardware that was just as good was coming onto the market. Not an efficient use of limited resources.
The period about ten years later is much more a peak of actual developers of Haiku. There was a perception, false as it turns out, that Haiku was “almost finished”. You could still make a real difference to a brand new OS if you got in quick. The OS was self-hosting, Haiku R1 was said to be 12-18 months away. And then another of Be’s old problems surfaced, shipping is hard. Releasing a product requires a different skillset, orthogonal to the skills needed for development. So there was the joke of “alpha” releases spaced a year apart, and then despite a post mortem identifying the need to do more releases instead of getting better it got worse.
Anyway, bbjimmy specifically said application developers. The OS exists to run applications. One choice you could make is to be very good at running other people’s applications, that’s why Windows is a huge success, and it’s why Linux is a huge success, indeed it’s even why the SCO lawsuits started, replacing a SCO system with Linux was a much faster and cheaper way to run the same application software. Haiku is not, to be frank, very good at running other people’s existing applications. So that leaves the other option, having your own compelling applications so that people are obliged to use your OS to get those apps. That can’t happen without an army of application developers.[/quote]
True. But missig application developer on a non beta OS? Don’t think tah will happend soon :)… But I would not call it dead… Remember IBM’s OS? or Palm OS? those are dead…
I beleve that with PM we can replace BeBits introduce a Android Play type of set. I think thats the plan in the long run, if not it should 
Replying to NoHaikuForMe:
Be chose to create a new OS, and wanted to attract developers. Sure, they could (and did) have a MacOS emulator, but the point of an OS is to have native applications, not run applications from another OS.
I really struggle to see the reasoning behind your opinion that “there is no point to have BeOS/Aros whatever, because it does not have any applications”.
There is no OS that had applications before release.
Or, stated differently, “Since $OS does not run some already existing application in a better/faster/whatever way, there is no reason for $OS to exist”.
If this was true, then we would still run ZX Spectrum emulators, so that we can run the same software, since the Spectrum had orders of magnitude more software than any other computer back in the 80’s.
Your reply will certainly be, “But haiku does not run any application better/faster/whatever, than already existing OSes, so what is the point?”
If you do not see the benefit of designing and creating an entire OS, including the GUI, using the open source technique of creating software, then you have completely missed the point of Haiku.
Are you saying I am not a real user ?
I was using beos since version 3, and still have beos running on an old P2 system, so yes I still use it, and no, haiku does not doing anything it cant. I’m not trying to put it down here, but yes, I also have Haiku running here, and no, I can not do much on it, at all, not as much as I can do with beos on my old system, it juts crashes too often, while good old beos never crashes… Again, that’s no disrespect, I think the world of you guys, and if I could do as much as you I would be helping for sure. I learned to write code on the Coleco Adam, when I was young, in the late 70’s early 80’s, but I never did it enough to be up to speed these days. I wish I did, and could help somehow
But back to the problem, Windows 10 is out now, and a million different versions of Linux are out there. all suffering from the same problems, they are all too big, too complicated, and too bloated. Anything that evolves that far will be bloated as all hell, with no real fix in sight, or possible. Just like virus’s and security, modern government, etc. It will never end, never be fixed, and never stop. Windows will never fully secure the OS, or be more then 1 step ahead of hackers. And hackers will always find a way around anything they do
I cant speak for the original writers of BEOS, but I did read about everything they ever wrote at the time, had the beos bible, and a stack of other books on it. And it seemed that the idea behind beos, was to keep it small and simple, strong and nimble, quick and easy, etc
And they even said, they didnt believe in the way Microsoft tried to do everything, and make every program people could want or use, they wanted people to use it to learn how to code themselves, and they wanted other companies to make a profit by making the big professional applications people would want and need, while they focused on keeping the core light and strong
Like the other hundred times Ive said it, Linux has a dozen package managers, its free, and strong, yet most people would never use it, because its just too complicated, its too messy, and just plain unorganized. I love it, and I can use it no problem, but I still usually don’t, just because I cant stand the see that bloated file system. it drives me nuts
That’s where Beos was different, the file system was small and simple, you didnt see the Linux like parts unless you really had to. the programs menu wasn’t cut or partitioned for games, internet, etc etc etc. The menus and clock, and everything else were simple and easy, and there wasn’t even a block on deleting files, at first, later it had the hold shift to delete a file etc. And that seemed weird, but I guess a little needed. Still, once you start trying to over think things, and stop customers from being human / stupid, etc. Your on the way to becoming just another bloated OS
the package manager, like Internet Explorer for Windows, should have been a free download for people that wanted it, not a main part of the OS, that’s so big and so complicated that it stopped Haiku from being released for 3 or more years
Microsoft’s problem was they wanted to be king, they wanted to rule everything, they wanted everyone dependent on them and them alone. If they never bought IE, and forced it into being a part of the OS, netscape and everything else would have ran better on a more simple / clean OS, that would have had less problems simply because they focused on the core, and not everything else
Remember the great Microsoft line, WE WILL NOT BE THE BIOS
How about a bar graph on the website showing a range of OS’s and how far they have progressed in relation to Haiku. e.g., Windows, MAC, and Android would have full bars since they are complete OS’s (in terms of functionality, consumer use, and app ecosystem). Then Linux would have a slightly less sized bar, then less again for BSD, less again for Unix, then maybe you would have HAIKU next. After Haiku the bars would be progressively smaller: Genode, Kolibri, Menuet, Vysopsis, and ReactOS. This could generate more interest from developers and users (which has ebbed and flowed before) since it would show that Haiku is getting up there with the more complete OS’s moreso than other OS’s.
Linux, the BSDs and the UNIXs already are complete OSs in terms of functionality, user base and app ecosystem (well, UNIXs are dying for lack of users, but they are still complete OSs).
[quote=tsteve]
I cant speak for the original writers of BEOS, but I did read about everything they ever wrote at the time, had the beos bible, and a stack of other books on it. And it seemed that the idea behind beos, was to keep it small and simple, strong and nimble, quick and easy, etc
And they even said, they didnt believe in the way Microsoft tried to do everything, and make every program people could want or use, they wanted people to use it to learn how to code themselves, and they wanted other companies to make a profit by making the big professional applications people would want and need, while they focused on keeping the core light and strong.[/quote]
This is called “making a virtue of a necessity”. Be Inc. had no choice about building this small, simple system, they had insufficient time and money to build anything more. In marketing, if you’re obliged to do X, and particularly if your competitors choose not to do X, you always tell the public you’re glad to do X, it’s the best choice and you would never choose anything else, even though that’s a direct lie.
And you proved my point - to generate interest you’ve got to have some discussion and controversy. So, even better than that graph, what about one showing OS innovative features. In this case, Haiku would have the full bar, followed by the lesser OS’s, with WIN, MAC, Android, and ReactOS at the end with the smaller bars - because they don’t innovate as much. That way Haiku is portrayed as the lead OS (which it should be anyway). This is like Donald Trump-style Presidential Marketing 101.