the big problem for me is also the same problem BeOS (freakn’) 12 years ago. lack of software, lack of drivers, and lack of motivation from the people behind it. i loved BeOS, and i love Haiku a3, but if i can’t do anything with it (i do use it to watch videos on my dell mini 10, since win7 is too slow), it’s not worth anything.
priorities for drawing interest:
a more up to date set of common network and graphics drivers.
---- get chugging on WPA/WPA2. that’s a big detractor.
---- the more systems it’s able to run on without resorting to VESA mode the better
---- lots of people have an intel gma500 based netbook, and more are still being made, that should actually be an easy boost in usable systems. case in point, this one really tickles my sidekick fancy http://www.dynamism.com/top-notebooks/highton-mi15.shtml
---- [i don’t know if my mini10’s wifi card (dell wireless 1520 802.11n) uses a standard chipset or not, so i won’t use that as an example, and you already seem to support the realtek gig-e in it just fine.]
get people to port Gimp, Inkscape and either LibreOffice or OpenOffice
---- gobe productive was nice for it’s time, but seriously?
---- abiword is also nice, but it’s only a word processor, a full suite is definitely needed
I don’t want to be trying to run 10yr old stuff on Haiku, I want to be using new stuff. Yes you need software to get users but I can tell you I did not start using Haiku because of any of the current software(ok read 10yr old programs) I came because I see the potential of Haiku.
What is going to kill that potential is if the devs want to wait and wait till every thing is perfect before making another stable version. When I read a msg about someone trying Haiku because of the IEEE article like I did and they load the Alpha and their video card is not supported and luckily the person asked about what he could do and was told to load a nightly and tada his video card is supported in the nightly but now he has to load a browser and such, I am thinking that an updated Alpha is WAY PAST DUE.
I think there needs to be a plan laid out, I like what franz1789 put down. I wouldn’t call it R1 but B1, but there has to be some movement forward. I have been part of some major software/hardware implementations and we started with a basic plan and modified it as we went in some instances(hardware not available or software still be fixed) right up until we put the system live. There where a lot of problems to fix after it went live but we had to put something up so others could do their work and start using the system.
Since I am new here, who are the people/person that decide to ok a new version? Right now I don’t see anyone person I could chat/email with on why it is not going forward.
Just an idea – but if Haiku was only ‘supported’ on a very specific set of hardware, similar to what MorphOS does, and there was funding available to give every Haiku Developer that particular hardware, I’m sure it would smooth a lot of things out. However, Haiku is expected (by the community) to run on every piece of hardware out there and that is a lot of work.
Also, I’d deprecate GCC2 just for simplicity’s sake. Sure, it allows users to run older software. However, how much regression-testing is done to make sure every ancient-closed-source-commercial-application still runs properly? I can only imagine how much of a nightmare it is to keep everything kosher.
I do agree on the planning part. I imagine they have one. I also imagine (well, hope) they follow some sort of an iterative style of development if they work as a tight-knit team. Just because their nightly builds aren’t labeled with some fancy version number doesn’t mean nothing is getting done and they aren’t following a plan.
Also, I noticed that Haiku only expects to get around $33k for the year. I’m sure on that budget there aren’t many (if any?) full-time Haiku developers. Therefore, I imagine most developers working on Haiku are strictly volunteers and work in their spare time. If most are actually just volunteers, then I have to admit they are doing a fantastic job at pulling this together.
Since the Haiku developers are currently preparing for the next Haiku release (alpha 5 or beta 1), it might be worthwhile for the development team to consider some of the opinions expressed in this old thread, because it appears not much has changed in 2 years, except now we have even fewer users than before.
How many of the people who posted in this thread are still with us?
Having a couple of articles published about Haiku once every 14 months is hardly conducive to keeping Haiku in the public eye and attracting more potential users and developers.
I am hoping that the next release, whatever they decide to call it, is worthy of grabbing some much needed attention. They really need to include more polished applications for the release, and put some of the stuff that is clearly not ready for prime time out of sight, until they are ready.
The state it’s in is moving anyway. I’ve seen a lot of progress happen on haiku in the last couple of months alone. These guys are also working as a hobby in their spare time, with exception to the contracts. And apparently they are nice enough to make it free.
I fail to see how people can seriously want more for nothing and think to be taken seriously. Could be worse, they could charge for it?
What I guess I’m getting at here is, they started the project to make a compatible os. They want to hit that target that’s fine with me. They started it anyway and did the work. Which btw, is a hell of a lot of work to do exactly that. Outside of the occassional issues here and there, it’s running pretty well from what I can tell at least on my hardware. Outside of NVIDIA and that’s understandable. Wish they would just open some info up on that already like they said they would. Just not enough info to grok the final drivers being released though.
And I don’t mean any of this in an ill will, etc. I just find it strange. If you want more information look in on the dev mailing list. You can read the discussions, etc there too. It’s pretty informative.
Just now when in the last few months more work (package management, web kit and wifi) has happened than I have seen in a long time you declare the project dead.
Finally the project is getting ready for a beta release and you say it is going nowhere.
You want (like so many others) that the devs absolutely drop the BEOS compatibility. I don’t get that either. The BEOS apps is all we have! Once we drop that and we release the OS people will start looking for software to run on it.
If het project is finally able to do an official release it better be working, which is currently not the case yet (have you actually tried to get HAIKU working on real hardware?) Yes this is a big project done by a small amount of people en therefore, yes it’s going slow.
I too am often frustrated by the slow pace of the project but the timing of your rant really bothers me. BTW alpha 4 has been released some time ago (time to get up to date with the status of the project I think).
If you take the time to read the FAQ you would see that they have a kind of plan. I call it ‘kind of’ since the milestones that they want to reach are clear but the timing is not.
I think this is perfectly reasonable. They have put themselves a goal: make a binary compatible version of BEOS. They now quite well what they need to do for this (look at the tickets in the bug tracker) but since the work is mainly done by volonteers it is hard to put a timing on it.
If there are some donations or using the GSOC applications sometimes a considerable amount of progress can be archieved (like e.g. when Java support was added!). But nothing is definite.
I, just like anybody else, am hoping the next release will be Beta 1 instead of alpha 5 but really you have to understand that if you release a beta it will attract some attention. So it better be worth it.
As for the constant moaning about the compatibility with BEOS (not only by you but by almost any poster on this thread). What is the big problem with BEOS compatibilty? I always hear user complaining about it but never developers. So that means that it doesn’t stand in the way of the current development. So what’s the problem then? Just like this GCC business. What do you care about the kind of compiler is used? Also I gather that they have a hybrid build system in which you can use the version you like most. And the system runs apps that are build with either. So what’s the trouble all about then?
The support is there. It’s done. The developers are now working on other stuff. So what could be gained by dropping support apart from diminishing the already very small software base?
Hey, it’s obvious that a “beta” release would attract some more attention than yet another alpha. The question is not there. The question is, are we ready for that? With our previous alpha releases, we have had some people review the OS, thest things around, and most of the time, these reviews end in the line of “these guys are on to something, but the lack of hardware support and/or apps make the OS not suitable for production use yet. Let’s see where they go, it’s just alpha after all”.
The last thing we want is sticking the "beta" tag (or even worse, R1) on a release that isn't actually beta quality. Yes, everyone has high expectations on beta and R1, but just sticking the name on it won't magically fix all the bugs. We are still just out of the PM and Scheduler merges, and there are a lot of things to clear out. HaikuDepot isn't really ready for prime time, do you suggest we drop it from the image? To make sure people continue going on Haikuware and end up saying "too bad, most of the software on that site doesn't work"? Or messing with the command-line pkgman to install packages? This is not the experience we want to provide. Same for the several causes of kernel panics, and the currently glitchy web browser, and some annoying bugs in Terminal, and I could go on and on. Yes, we could remove all these apps from the image, pack up the things that are "done", and call it beta. But then, people would say "these guys have no software", and leave. Instead, we want to actually fix all those annoying bugs, and get a very good release out. This is work we will have to get done anyway, and if we rush R1 out, you all know what will happen: after two months, we will be telling people to use the nightlies, or R2alpha1, or whatever experimental release.
This is not what we want: we want an R1 that will last for years, so the developers can focus on adding the new and shiny features to R2, instead of bugfixing R1 and releasing updates to it. With the package manager, we are moving some of the R1 maintenance efforts to the community: you will be able to replace packages with newer versions easily, so the system doesn't get too much outdated. There are ongoing efforts to remove some code from the main haiku repo and turn it into packages (mostly the command-line tools, but also some libraries).
[quote=SomeoneFromBelgium]I guess I just don’t get it.
Just now when in the last few months more work (package management, web kit and wifi) has happened than I have seen in a long time you declare the project dead.
Finally the project is getting ready for a beta release and you say it is going nowhere.
You want (like so many others) that the devs absolutely drop the BEOS compatibility. I don’t get that either. The BEOS apps is all we have! Once we drop that and we release the OS people will start looking for software to run on it.
If het project is finally able to do an official release it better be working, which is currently not the case yet (have you actually tried to get HAIKU working on real hardware?) Yes this is a big project done by a small amount of people en therefore, yes it’s going slow.
I too am often frustrated by the slow pace of the project but the timing of your rant really bothers me. BTW alpha 4 has been released some time ago (time to get up to date with the status of the project I think).[/quote]
I’m not sure who exactly is being addressed in some of these comments, especially since it is a long thread, and there are some strong sentiments being expressed. Also consider that the thread was started about six months prior to the alpha 4 release.
Myself, I have 5 different PC’s that I am using to test Haiku on real hardware. Recently my wife told me to try f****** my computer since that is where I seem to be spending most of my free time after work every day. I really don’t see anything here which could be called a “rant”.
“HaikuDepot isn’t really ready for prime time, do you suggest we drop it from the image?”
The package manager is not what I had in mind when I made the statement about “prime time”. It is already in a useful, but incomplete state and needs to be included for testing and further development. I was thinking more about some of the apps that have limited functionality or poor support i.e. codycam and TV.
Otherwise, I agree with your assessment.
“The last thing we want is sticking the “beta” tag (or even worse, R1) on a release that isn’t actually beta quality.”
“…we want to actually fix all those annoying bugs, and get a very good release out”.
These releases aren’t just test platforms, but are also a showcase of the operating system and its capabilities. With each new release, the reviews are going to be less forgiving than before. Until package management is mostly complete, and the rendering issues in Webpositive are sorted out, it would probably be unwise to call it a beta (just my opinion!).