Polls and statistics are of course always to be taken with a grain of salt. But Miro’s work is at least as scientific as it gets. It shows that the impression one can get if you plunge into this web forum may not represent the “community” at large. Few people are able to sway your perception.
Often the web forums are more casual and prone to unnecessary rudeness or, say, bluntness.
The mailing lists are more serious and the main channel of communication of the devs. Therefore shannanigans ignored in these forums are quickly frowned upon on the list. I recommend joining, even if should you just want to lurk a bit.
I personally do not think Haiku users are anti-Linux, they are however very anti-(Haiku==Linux).
Too many time linux fans posting in these forums have clearly acted like Haiku-OS is just another Linux Distro, they then make assumtions about how Haiku works, or what type of software should be added to it.
The problem with that is doing direct ports of Linux programs and their supporting libraries results in code that does not support Haiku-OS features like DataTypes, MimeTypes, MultiThreading from day-one. Much less the features that BFS and other services provide to a Haiku-OS user.
And when talking GPL, which one? Version 1 was easy to understand, and as a person who would like to port older BeOS code to Haiku-OS I would love to have source code included in as many programs as possible. But Version 2 starts to have questionable additions and of-course Version 3 seem to forget the purpose of open code as it gets more and more complex - ALL OF MY BeOS/Haiku CODE is in the public domain, I have never seen any harm to me not using GPL-3 as it tries to imply. The demand by many Linux users that I have meet personally that my code MUST BE GPL2/3 is why you find I will not touch Linux with a ten foot pole today.
There is one more thing as a programmer.
Linux code in general is a MESS. Years ago during my late Amiga/early BeOS stage I tried looking at Linux code to get an understanding on how to program for it. Coding standards were all over the map (assuming you could even figure out what standard they were following) with each program/function/library using a diffirent one.
And no, it is not better today! Last year I expressed interest in making a compressing RAMDRIVE, so I popped in to some Linux forums to see the discussions and code. No thanks.
Haiku uses a standard that I don’t like, but once I understand a part of the standard it stays the same thru-out all the code.
My code does not follow Haiku’s standard (a reason you probably will never see my code inside Haiku’s base image) but I try to document it clearly (checkout Haikuware), I don’t want the mess that I see in Linux coding coming to Haiku-OS.
So to repeat, we don’t hate Linux, we just don’t want Haiku to become another Linux.
Yes, native applications are of course preferable but if Haiku is going to wait for native applications emerging which covers the needs solved by already existing and widely used FOSS applications then we might aswell give up right now. Have you looked at Haikuware? How many native applications have you seen posted there in the past 3 years?
As for multithreading, I assume you mean the GUI? There is no magical multithreading inherent to Haiku/Beos applications that I know of, so if you want to make efficient use of multiple cores in Haiku applications you will have to do the work. And FOSS programs which benefit from multithreading already make use of it with standard posix threads which Haiku supports.
Also there’s nothing stopping someone porting FOSS applications to make the port appear/feel very native, like adding a native GUI and perhaps also making use of some of Haiku/Beos more exotic features where applicable.
[quote=Earl Colby Pottinger]
But Version 2 starts to have questionable additions and of-course Version 3 seem to forget the purpose of open code as it gets more and more complex[/quote]
Curious as to what these questionable additions are in you opinion.
Compared to what? Also can you point at something specific as an example?
With coding standard you mean what exactly? Naming convention? Formatting?
[quote=Earl Colby Pottinger]
So to repeat, we don’t hate Linux, we just don’t want Haiku to become another Linux.[/quote]
Well I can’t say you sell that notion very well as I personally find your comments here dripping with anti-Linux sentiment. As for Haiku becoming Linux, how could it? Unlike Linux which is just a kernel around which you add components depending on your needs, Haiku is a complete operating system purposely aimed at one type of use: the desktop. It all boils down to some irrational fear in my opinion, like when the Haiku devs declared that they were making a package management system and we had some people throwing a fit in a pure knee-jerk reaction.
Of course I also will assume that by ‘we’ you are actually only referring to yourself.
He refers to me also. We owe much to the GNU project, for some important software components, but for those who want Linux, you know where to find it. We don’t need to have a debate on the merits of Linux - both exist, you may choose for yourself.
The slow pace of native Haiku software releases doesn’t worry me, because it has never been released. I won’t argue with those who will distribute their new software built to an alpha release, but it’s also reasonable to wait and see before making that kind of investment. I can’t prevent anyone from porting software from other platforms, and in fact it’s important to get in on multi-platform software efforts like WebKit - but I think anyone who’s ever worked with something like that will appreciate how complicated it gets when you try to build software around several distinctly different operating systems.
I have a long standing attachment to NetBSD (also going back to the Amiga, when in the early '90s it was the alternative to Commodore’s Amiga/UX), and I think it addresses some of those criticisms, coding standards for example, but … who cares? We agree on the value of coding standards, and if Linux demonstrates their value by contrast, that serves us at least as well.
Who? I wasn’t referring to you, I was only aiming my post at Earl Pottinger.
I like Linux and I like Haiku, do I have to choose ‘side’??
I wasn’t arguing against ‘not wanting Linux’, obviously each and every one is the best judge of what he/she likes. He (Earl Pottinger) however went on to criticise Linux as being a ‘MESS’ in terms of code and GPL for having included ‘questionable additions’ and I want to know what he’s referring to.
[quote=donn]
I won’t argue with those who will distribute their new software built to an alpha release, but it’s also reasonable to wait and see before making that kind of investment. I can’t prevent anyone from porting software from other platforms, and in fact it’s important to get in on multi-platform software efforts like WebKit - but I think anyone who’s ever worked with something like that will appreciate how complicated it gets when you try to build software around several distinctly different operating systems.[/quote]
Let me assure you that for any non-trivial piece of software it’s generally a hell of alot easier to port an existing project than it is to write an equivalent from scratch. Ports will be a necessity for Haiku, unlike you I don’t think for a second that there’s some massive amount of devs sitting and waiting for R1 who will then start pouring out native apps. I also doubt Haiku will change alot from a user app development perspective between now and R1.
I sure wouldn’t mind being wrong though, the idea of lots of developers ready to jump onto Haiku with native apps in mind sure would be welcome, I’ve seen absolutely no indication of that though.
I’ve been thinking about that a lot, and I think if we can get R1 out sooner rather than later, and build a compelling “brand” for Haiku (whatever that may be), we will get some developers. It may not be like iOS, but I think there will be some people.
But we need to have a real market for Haiku, which could be as simple as a good, fast OS on cheaper and more environmentally friendly hardware (versus the monsters needed to run most other OSes well.) If we could provide some good “HaikuBox” hardware which fits that mold and works very well with Haiku, all the better. Sort of the anti-Mac, but in a good way.
But yeah, until there is a good market for them, there is little incentive for most developers to develop for Haiku. Fortunately because we aren’t Linux there will be less of a “all software must be free” stigma (which isn’t necessarily a fault of Linux, that is just the sort of community that has built up around it.) Though maybe the Ubuntu software center has changed things some, I don’t know the data.
[quote=leavengood]But we need to have a real market for Haiku, which could be as simple as a good, fast OS on cheaper and more environmentally friendly hardware (versus the monsters needed to run most other OSes well.) If we could provide some good “HaikuBox” hardware which fits that mold and works very well with Haiku, all the better. Sort of the anti-Mac, but in a good way.
[/quote]
Agreed, unfortunately with the general cpu/gpu/ram availability on people’s desktops today Haiku doesn’t get as much praise for it’s small memory footprint and generally being so lean on system resources as (imo) it should. It offers a full desktop environment using only ~90-100mb. Sadly alot of clueless people has bought into the ‘memory should be used…’ mantra, yes it should be used, but by your applications, not hogged by the OS (unless for pure caching purposes where it can be reclaimed in an instant).
Haiku would perform splendidly and thus shine on the new wave of low powered devices but it has to be able to run on them to begin with and here is that old manpower problem again. I’m getting a Rasberry Pi and it’s one of these devices where it would be geek heaven to run an OS like Haiku.
I don’t know about that, I think the ‘free’ part is mostly a result of most Linux applications/software being started as a ‘scratch your own itch’ project rather than make money project and thus it is released as open source once the author has scratched his/her itch. Part of it is likely also out of gratefulness towards others who has done the same. However looking at for instance the Humble Bundle Indie games we see lots of Linux customers who always pay the most (per buyer) during the ‘pay what you want’ campaigns.
Of course if you want to sell an editor on a platform which has tons for free then yes, it’s going to be a tough sell and your application must really shine. That’s not just a Linux thing though, there’s a huge amount of FOSS on Windows aswell which kills off the market for lots of applications as there are often open source equivalents working just aswell if not better, which in turn is likely what killed off the shareware/adware market on Windows, that and of course piracy, no point in developing a photo retouching program aimed at the consumer level market when everyone and their parents have a pirated copy of Photoshop installed.
I think it would be great if Haiku would turn out to be a platform where software could be commercialized as I have no problem paying for software (as long as there’s no proprietary data format lock-in, I refuse to entrust my data to a format which deliberately obfuscates it to prevent me accessing it outside their application), but we are a long way from there. For Haiku to even become remotely economically viable as a commercial software platform it needs lots of users.
For me personally, the only thing ‘really’ preventing me from using Haiku as a full day-to-day OS is the lack of a ‘modern’ browser (Chromium would fit the bill nicely as it supports HTML5 and also allows wathing youtube using Webm and has tons of extensions), Inkscape, Blender (requires 3d acceleration to be really useable but there is a gallium port being worked on), Mypaint, and importantly an update mechanism (package manager). Development-wise Haiku caters to my needs as I mainly develop in C and Python, I’ve been dabbling with Go aswell and IIRC Bruno Albuquerque is working on a port so that could turn out nicely.
Given that none of these things are really ‘out of reach’ for Haiku I have strong hope of being able to use Haiku as my primary desktop.
[quote=leavengood]I’ve been thinking about that a lot, and I think if we can get R1 out sooner rather than later, and build a compelling “brand” for Haiku (whatever that may be), we will get some developers. It may not be like iOS, but I think there will be some people.
But we need to have a real market for Haiku, which could be as simple as a good, fast OS on cheaper and more environmentally friendly hardware (versus the monsters needed to run most other OSes well.) If we could provide some good “HaikuBox” hardware which fits that mold and works very well with Haiku, all the better. Sort of the anti-Mac, but in a good way.[/quote]
Unfortunately (as much as I love weird little platforms like this idea,) I think tying your fortunes to a hardware project is going to backfire big-time in this day and age; people who aren’t interested in buying a hobby system will see Haiku as just “the operating system for the HaikuBox,” so most of the interested parties you’ll get out of it will be hardware geeks, who are going to be more interested in the platform than the OS.
Though I’ve bagged on Linux earlier in this thread, I will say that one page Haiku should definitely take from it is the “port to absolutely anything” philosophy. I’ve read the rationale for focusing on x86 first, and it does make sense, but on the other hand, at this stage in the life of Linux it didn’t matter what you had, you could run Linux on basically anything 32-bit, from a 386 to an Amiga to a SPARC workstation. People joked about running it on your toaster, but it was hardly that much of an exaggeration; there was basically no barrier to entry.
Haiku already has a leg up on the easy-experimentation factor by virtue of being easily able to be run from a flash drive and not even touch the main hard drive, but it could be taken so much further. If you think that having a specific Haiku platform would help (and I think you might be right,) make that platform whatever people have on hand. Get the PowerPC port up and running, let people pull their old iMac out of storage and give Haiku a try, or put it on their 360, or whatever. Do an ARM port and let people find uses for that (I’ll save my anti-tablet rants for another thread, as long as it doesn’t affect the main Haiku UI.) Those chintzy $50-75 Windows CE ARM netbooks would be a cheap-and-easy way for people to get acquainted with Haiku, and the Raspberry Pi would certainly be more comfortable running Haiku than Debian, by my experience.
The more stuff people already have that can run Haiku, the less reason there is for them not to try it out.
Also this. I don’t even think Chromium is necessary (and I don’t trust Google as far as I could kick 'em,) if WebPositive were only filled out feature-wise - better keyboard shortcuts and some form of configurable ad-blocking are the key elements I can think of left to add. A package-manager would be nice, as long as it supports installing from downloaded package files just as easily as from a repository.
And one more thing: getting WINE for Haiku finished and polished should be considered absolutely critical. Native apps would always be better, but the simple fact is there’s always going to be stuff people would like to be able to run from their Windows systems. The less trouble it is for them to get that working, the less they’ll feel like “oh, this is interesting and all, but it’d just be too much trouble to switch.”
Oh, and as long as we’re making wish-lists: an easy way to maximize windows without going into a full-screen mode would be terrific.
I started using BeOS around 2000. I think originally I purchased my copy of version 5 from GoBe? Frankly I can’t remember. I can say that every so often, I sell my BeOS stuff on Ebay, only to regret it and buy another pro cd and BeOS bible. Currently I have Zeta 1.21 from Magnusoft.
I like Haiku, and I wish it would get some support - even if that’s commercial. I wouldn’t mind paying for it - say $100 a copy - as long as I say it going somewhere.
So here’s my list:
Pros:
Everything works on my Compaq Presario CQ56. The only thing I changed from stock was the wireless card. I put in an Intel Pro model - it works.
Fast and simple to use. Software doesn’t need everyone else’s library to run. It bothers me to have to install the KDE environment to run one app, when I’m using Gnome. It makes me feel dirty I like having one desktop environment. Less choices seem better in this regard.
Low requirements. I never understand why Microsoft’s requirements keep rising, except to encourage another computer purchase. It seems that Windows 98 doesn’t do much more than Windows 7 (at least not enough to justify that big of a difference).
Brings back that nostalgic BeOS feeling
Cons:
SOFTWARE! Unique things - I don’t need another office suite. I’m locked into MS office at work/school and don’t plan on learning every program’s shortcut, formatting rule, etc… Personally If I remember, WebPositive works with Google Docs but not with MS SkyDrive’s online office suite. It’d be nice if Haiku was supported in the Humble Bundle software packages. I picked up Crayon Physics from them - to see this on Haiku would be awesome!!
Community - Maybe I’m missing the boat on this one, but the community seems few and scattered between different forums, IRCs. Do we have a IRC channel for non-developer Haiku users? Of course, I’m in in the US in Michigan, so maybe most of you are in Europe?
Software development languages. So far I’ve taken a classes for SQL, Python, and now C#. C# is probably where I’ll stay. My time is limited to work, school and family so it would just confuse me to learn another.
question: How similar are C++ and C#??
Maybe YAB needs official support from Haiku. It seems to be a nice intermediate level language that might encourage more development.
The website - it’s getting stale. It’s a great format, but it needs a change. That might help revitalize Haiku.
A beta release - give us a new release.
But the biggest hurdle Haiku is facing (in my opinion) is purpose. We need a reason to use Haiku. Look, it’s great. It works good. It’s fast. But so what? What do I do with it? What can I do that I can’t already do with the other operating systems? Make it Haiku fun and exciting - and bring back Zsnake as a menu option
Let me say this first before I get flamed. I think the developers are doing everything that can and working as best they can with their limited numbers. I wish I could stop my hectic life to help out - but I can’t at least not now. This is probably some of the problem with Haiku. It’s fanbase was alot youger with BeOS - we’re only getting older. I myself now have a family, college, etc. I’m sure this applies to alot of us former BeOS users.
Now with that being said: The only way I think Haiku will survive is with money and lots of it. Personally, I don’t care if it’s closed source as long as I get a decent IDE like Mono’s IDE or YAB. Charge me $100 for a copy, I don’t care - just make it work.
Big ups to everyone who has written code for Haiku. It’s very exciting to actually be testing the OS on real computers now.
Ryan wrote:
But we need to have a real market for Haiku, which could be as simple as a good, fast OS on cheaper and more environmentally friendly hardware (versus the monsters needed to run most other OSes well.) If we could provide some good “HaikuBox” hardware which fits that mold and works very well with Haiku, all the better. Sort of the anti-Mac, but in a good way. <<
I totally agree with the guy who said get it to run on everything but the kitchen sink. I am a green geek into hardware reconditioning, and I’d love to be able to run Haiku on old Apple systems (for example), or handhelds (“smartphones”). In fact that brings up a project all open source OS crews could work on together - open firmware for as many kinds of hardware as possible, which allow any open source/ free code OS to be installed.
Having said that, one of the main reasons FUD stats understate the number of GNU/ Linux users is that most of them are installing on a machine sold with Windows pre-installed. The stats count them as Windows users, even if they aren’t. IsBut we need to have a real market for Haiku, which could be as simple as a good, fast OS on cheaper and more environmentally friendly hardware (versus the monsters needed to run most other OSes well.) If we could provide some good “HaikuBox” hardware which fits that mold and works very well with Haiku, all the better. Sort of the anti-Mac, but in a good way.
But yeah, until there is a good market for them, there is little incentive for most developers to develop for Haiku. Fortunately because we aren’t Linux there will be less of a “all software must be free” stigma (which isn’t necessarily a fault of Linux, that is just the sort of community that has built up around it.) Though maybe the Ubuntu software center has changed things some, I don’t know the data. I suspect companies like ZaReason which ship hardware with well-supported versions of GNU/Linux preinstalled are going to make a big difference there, and I see no reason not to work with hardware vendors (again, as many different kinds as possible) to get hardware pre-installed with Haiku on the market as soon as R1 is released.
But yeah, until there is a good market for them, there is little incentive for most developers to develop for Haiku. Fortunately because we aren’t Linux there will be less of a “all software must be free” stigma <<
All software should be free (libre) for the same reason that all speech should be free. That doesn’t stop people from choosing to charging money for forms of their speech (novels or songs for example), or choosing to give them away for free (gratis). So much of the misinformation around GNU/Linux and its communities stems from people’s inability or unwillingness to understand the difference between libre and gratis definitions of the word “free”. It’s really important that core devs on significant open source/ free code projects set an example, in fact, I would go so far as to say that whether or not they do is a measure of how significant their open source project is.
I like having one desktop environment. Less choices seem better in this regard.