My choice is very easily made: I support the hardware I already own. I will be happy if I manage that already.
My current laptop is 10 years old and I have not gotten the video card fully working yet. Clearly I cannot honestly accept a partnership where someone donates me hardware and I make Haiku work on it. I will just pile up the hardware here and never do anything with it.
How do I know? Very simple, here is a list of hardware that I have here that was donated to me or to Haiku inc in the hope that someone would write support for it:
A SATA blueray drive
4 Sun Ultrasparc machines
A VIA C-3 based embedded PC with a GPS receiver and a multi-UART board
A DVB-T TV tuner
With all this (some of these I own for more than 7 years now), all I had time to do was getting the bootloader running on SPARC.
So, how would a partnership go? “donate me some hardware now, and in 2028 I will be posting on a forum about how I did nothing with it” doesn’t look like a very promising thing.
Also, personally I have zero interest for 3D acceleration. I don’t play video games. So, someone else will have to do that work.
If we want to do partnerships, we need someone who can dedicate their time to that. Personally I can’t, because I’m already very busy with just keeping Haiku usable for my own needs on the hardware I already have. And that’s one case where someone being paid to do things may be helpful, since they could dedicate some time to it.
Perhaps, finding students who want to work in the domain.
Haiku could mentor them in the discover of working with an open source project and basics of drivers. Later the manufacturer could offer a stage on how to write a driver for specific hardware. Even if the said driver isn’t for Haiku, the student would acquire knowledge about the piece of hardware that would be helpful.
If all went well, at the end of this stage they could be offered to work for Haiku for a year or for a specific task.
I’ve read over the above, I have some thoughts of my own to add later and will keep updated on the thread, but just to pop back to the original topic for a second, and confirm I’ll continue working on my setup and then read the docs on Gerrit and take something from Trac and submit a few patches for review; we can then see where we are then so I’ll leave it at that until I have those patches for the moment. Oh, I also learnt that Github… sponsors(?) exists only just earlier today, no idea how it works as of yet but wonder if that’s in any way a useful revenue stream.
It would be very usefull for 3D apps like Blender and a lot of video editors would benefit.
Yes that money would be wise invested in helping you get more time for your interests and development of difficult tasks, also for doing small projects with more speed up! If you pay someone it is work to do not for pleasure.
Now that Beta 3 is out, I would like to resume the discussion about the developer contract offered by Haiku Inc. Based on the recent email convo in the Inc’s mailing list (https://www.freelists.org/post/haiku-inc/Employment-contract-proposal-general-longterm-development), I would like to know from @waddlesplash & @leavengood whether there are any active steps in this direction? I stopped donating to Haiku Inc. this year (though I continue contributing to individual Haiku devs) but plan to resume if the contract is finally announced.
Yes, hiring waddlesplash will be happening, we just need to work out the details. His current job ends this week and then he wanted to take a bit of time off for himself and would be available to start for Haiku later in August. I thought that was good as it gives Haiku, Inc some time to get a bit more organized as well. The details we need to work out are whether we can make use of a service to hire him as a normal employee or whether a contractor makes more sense. We need to set up some sort of employment agreement as well.
Haiku, Inc plans to have a meeting soon to discuss this and other issues (we haven’t had a meeting in too long, so either way we need to.) This is mainly a formality as over email everyone has agreed this is a great idea. Also I am putting finishing touches on the 2020 financial report and getting things in places so future reports will be easier. This way everyone is better aware of our financial situation.
I will try to start doing quarterly reports with this system in place as with an employee we will be using our funds much more quickly. We currently have a bit of a “runway” to pay waddlesplash but that will be depleted within a year or so and if we don’t get a steady income stream to support him obviously we won’t be able to do so. But I think things will work out as we have plenty of time to show the results of hiring him and also add on to our funding sources. Plenty of people and open source projects can survive on mainly personal donations so I think Haiku can too.
Regarding this thread I was talking with stellarpower about working for Haiku, Inc but for his needs it was better he found another position. Someone else here in the forums contacted me as well but we had not finished working that out yet. We had kind of given up hope on waddleplash but it was always the slam dunk hiring choice as we know he can be trusted to write good code and he has a good understanding of the project. Depending on how things go with him and our funding maybe we can hire more people. Starting with waddlesplash greatly reduces the risk to Haiku, Haiku, Inc and all our donors.
Well, if it is now public knowledge, guess it is time to write up an official announcement (and a “call for donations” indicating the new donations target ) I will look at doing that this weekend.
Luckily for you, all of these are on my list (and indeed I mentioned them in the mailing list post linked above.) USB audio will probably be first, and GPU support later down the road (there is other low-hanging fruit I want to get at first before that.) We shall see what course I wind up charting…
Aside from the mentioned hardware support (and related R1 tickets about those), I would personally very much love to see R1 milestone tickets down to zero, so that the project can have the release it deserves! A stable system will hopefully attract so many developers, and with @X512’s planned hardware acceleration efforts, I think it’s safe to say that at last a bright future awaits Haiku.
Yay!
(I’d also love to see proper sleep/hibernation/power management support before R1, so that Haiku will be much more usable on portable devices. I think this is not a part of the R1 milestone.)
I started a thread about the possibility of sleep-mode support a while ago, and the general response (by @PulkoMandy in particular) was that it’d require a lot of code to be rewritten. I believe that’s why it’s not in the list of goals for R1.
Besides, the startup and shutdown times are much quicker in Haiku than in most other OSes, so I’d probably be OK with just a shutdown or reboot instead of sleep mode or hibernation.
Yes, I can see how being able to save your computer state for quick retrieval and switching to a low power mode for power saving is such an unnecessary feature.
Some simple sleep mode can be implemented that don’t require special hardware and driver support. It will turn off screen, disable response from some input devices and pause most threads. When user press power button, simple sleep mode will be exited.
Now may I ask about HDMI audio support?
I bought a soundcard just because HDMI sound is not working atm but I think most no audio problems could be about non HDMI audio support!?
On Intel chipsets, the audio is all handled by HDA. Then this has some kind of digital audio output that goes the the graphics part of the chipset. On that side, the graphics driver needs to configure the HDMI port to route that audio out of the machine. In HDMI, the audio data is transmitted in the same data stream as the video, so we need to make “space” for it by including it in computations for HMDI clock rates.
Probably most of the work will be on the graphics driver side rather than the HDA driver.
Oh, I also learnt that Github… sponsors(?) exists only just earlier today, no idea how it works as of yet but wonder if that’s in any way a useful revenue stream.
) I will look at doing that this weekend.