Haiku totally sux

AppImage is a universal packaging format intended for use across different Linux distributions. Due to the lack of distributions with their own package managers, it is unnecessary on Haiku.

1 Like

I would love to be able to run applications directly from a zipped file, sort of seeing the contents of the zipped file as a folder (like on Windows).

I would also like installable compressed files (for non-packaged stories) like Zeta. The content of the zipped file was also simply unpacked there after boot via an installer.

Why unnecessary? Why everyone has to use HaikuPorter and the package management system? It is incredibly limiting for alternative approaches.

Isn’t that pretty much like what the blue folder icon in trackers deskbar is signifying? There is already a sort of similar approach now. Last I checked the installer for haiku is just copying mostly zips/packages over, unlike the full file structure it was before. Check out packagefs.

There is no need to use appimahe because the problem it tries to solve doesn’t exist on haiku: differing unstable ABI between OS flavors.

On Haiku you can simply ship an ELF and be sure that it will run, ressources can also be embedded directly so many additional files in shared locations like in linux are unecessary.

Additionally on Haiku the standard library oath included the lib/ directory next to the executable as the first entry, which means overwriting system libraries is trivail to do by simply putting your versions in the lib/ folder.

4 Likes

That’s nice to hear. It’s good to have alternatives.

yeah, comparing haiku to Linux, umm, you have to be a masochist who loves text commands in the terminal to actually use most Linux software, because the developers for most Linux os are to lazy to write GUIs or to underskilled to write quality software that os extensible. basically Linux is a python scripting os on desktop afaict with a few apps that kinda work , as long as you’re ok with library hell and infuriating breakage and incompatibly problems.

haiku is very much not that.

Haiku itself doesn’t suffer from this, but try to update third party libraries, we are already in some form of library hell there …

Unfortunately Haiku inherited some Linux problems by using ELF executable format that can be observed in recent mass ABI breakage caused by libshared.a symbol leak.

You are so arrogant that you don’t even know what you are talking about. :face_with_raised_eyebrow:

I can do things using command line much faster that using GUIs (I know that this is subjective, but that is). I use terminal in Haiku daily to run all the task I need (building my ports, upgrading the system, pushing to repositories etc).

Now back on topic: I don’t think that haiku sux at all. I was shocked when I managed to port some of my software with just a compilation. The truth in my opinion is that if there were more people helping (instead of complaining) we would have more ready to use software/ports.

And please, for god sake, stop this nonsensical OS war. We can use and love more than one OS without being idiots.

19 Likes

Yes, I realize this now. Sorry about the confusion. I’ll pay better attention and post where appropriate.

I use linux since 2004 and yet I don’t feel superior to nobody, why should I? :face_with_raised_eyebrow: If you don’t want to use the terminal that’s fine, go ahead and write your GUIs, nobody is stopping you. Or just don’t use it at all.

2 Likes

I hope your having fun, trolling for fish… :wink:

No, the developers are not DELIBERATELY making buggy releases… but they’ve been putting the cart before the horse for so long, they no longer have any practical way to set things right.

Sure my post isn’t SETTING any realistic goals, other than to STOP constantly adding and changing stuff… and protracting out Haiku R1 ad infinitum. Whatever happened to “Haiku R1 is going to be BeOS R5”? That’s what Haiku R1 was SUPPOSED to be! But the longer it took (because everyone wanted to “do their own thing”), the more technology changed and the more devs started chasing after “new shinies”. And what should have taken… let’s say 10 years at worst… is now going past 22 years and counting. Goodness… had ANYONE known it was going to take this long at the beginning, would ANYONE have signed on? Seriously.

Money, alone, is not the problem. FOCUS is. Always has been. Single people have developed their own “Hobby OS” in a few years. An OS with a GUI and web browser and apps and stuff. ONE PERSON doing EVERYTHING. Don’t tell me it’s not possible! And, somehow… for some unknown, unfathomable reason… with a dozen or more people onboard at various times… over the course of 22 freaking years… we still HAVEN’T reached R1 yet? Is Joe Biden running the show or something? :smiley:

If ALL we had was Haiku R1… and it was no better than BeOS R5… as was the original plan… and we had taken the same amount of time, to this day, getting to R2… as silly as it seems… this overarching complaint would be null and void. At least the original plan would have been adhered to. No one could say… “Why haven’t we reached R1 yet?” No such plan/promise was given for an R2, so complaining about how long it’s taken, would be a different argument.

People want to say “it doesn’t matter”, “use the previous beta”, or whatever. But that means the original plan/promise has been forsaken. People were essentially lied to. Made to believe something would happen that still hasn’t. And that just doesn’t matter?

This isn’t about idolizing “Haiku R1 = BeOS R5” as the end all, be all goal… never was. That’s idiocy. It was meant to be a specific goal developers could all rally around as an ATTAINABLE goal. A specific checkpoint of progress everyone could celebrate, when it was reached. Was it a deliberate mirage, created by Phipps, to create a big enough snowball that he could safely step out of the way, and watch it roll on and grow indefinitely as his brainchild? Something that people would believe was something that could be accomplished but he knew they’d never actually achieve? Who knows… but I have to wonder…

What ARE we trying to create now? What will R1 look like, if/when it’s ever finished? Will we be celebrating Haiku R1 achiving “BeOS R5” (I seriously doubt that, at this point) or will we have completely forgotten the reason we started this journey and celebrate the completion of something we barely recognize as having anything to do with BeOS or from whence it came?

What specific goals do any developers have concerning Haiku? Is there ever an end to the quest? Or is there just the fun of going after the next new shiny thing that catches their attention that happens to be associated with Haiku? And that “new shiny” delays R1 yet another beta, because of any number of reasons… and the cart rolls on, ever before the horse.

Oddly enough… as little as I care about Haiku these days (due to tons of breakage over the years)… I still find enough energy to try and be the voice of reason. Because I WANT to feel that enthusiasm again! Not that I’m anyone worth listening to. Afterall, I’m not a developer, so what do I know, right? Will Haiku R1 be completed by December of 2023 or 2025? Place your bets, ladies and gentlemen!

Aha the voice of reason? What’s the point of hitting on the developers here? What’s the point in calling the work of many years nonsense? I don’t know what your private life is like, but many here do what they work on haiku in their free time. In here free time with there family and children. Then also at work for more than half of the day.

Criticism is ok and I can even understand the frustration that there is still no r1, but the sound makes the music.

Have you ever really followed the development of haiku? When alpha 4.1 was out, haiku was very close to beOS Rr5, but a lot of things in the source code were closed and that had to be changed, mostly completely rewritten. Then came considerations about how to improve the installation of software and introduced the package management system.

Then came from the community, not from the Haiku core team, developers who ported QT, later also other non-native support for a more attractive haiku (we still wouldn’t have an office).

You can complain here and talk bad about everything, but honestly, can anyone imagine using the old unsorted folder structure again? Again everyone writes where he wants? For my part, no.

3 Likes

I really shouldn’t bite but couldn’t help but feed the troll.

First thing to remind you of Luposian is that all OSs are flawed. I’m sure we can all name at least 3 things we can’t stand about any given OS so its always a case of picking the OS that annoys you the least for any given purpose.

I personally don’t care in the slightest about R5 compat and I’d rather see the devs drop the gcc2 builds and R5 compat and just become a BeOS inspired OS. There are no BeOS apps I know of that I would want to run. That’s just my opinion and I don’t expect anyone to care about it.

I only ever run the nightlies because Haiku isn’t quite there yet for me to use as my main OS so I’m just here to test the latest features and report bugs.

Its not fair to complain about how long a free, largely hobby-based project is taking to complete. That’s some next level entitled behaviour. Do whatever you can to help or go use an OS that works for you. R1 might take another 5 or 10 years to release but so what? It doesn’t detract from the fact that Haiku is the best open source, non-Linux/BSD based desktop OS available.

9 Likes

I have been following Haiku since before it was called Haiku, back when the window drawing routines were first made and I noted the improvements revision by revision. I have been following BeOS since it was running on a BeBox. I was following PowerPC since it was first mentioned in a magazine article.

If THIS (alpha 4.1) is where the branch split off into nightmare… this THIS is where the decision should have been made to STOP! There’s nothing you can do about closed software. You do the best you can with what you have. Was it just applications or was it specific drivers? Or specific internal code intrinsic to BeOS’ functionality, at the time? The goal should always have been to only go as far as absolutely necessary to make Haiku R1 every bit the equal of BeOS R5 and no further (unless by sheer accident a change made it any bit better functionally). But not to suddenly get wow’d by better this or more exciting that. It’s known as “chasing bunny trails” and I do it quite frequently in conversation (did it a couple times just yesterday!). Because a point of conversation creates “a new shiny” and you get distracted by that and then you forget what the previous topic was you were talking about. But THAT is why I emphasize FOCUS!

I have absolutely no doubt, we could “retcon” (not the main branch… a separate copy of it, of course) Haiku to a previous revision, compare it with BeOS R5, check off every point of comparison, have people test it on their hardware, make sure it’s good and then PRINT! And then we’d have Haiku R1.

Binary compatibility was the main focus, was that ever achieved? If it was deemed impossible, then maybe it was abandoned somewhere along the way and I wasn’t paying enough attention. Maybe a 1:1 Haiku R1 = BeOS R5 match was finally seen as utterly impossible and finally abandoned and I missed the memo?

Maybe when an exact copy was deemed impossible, the devs had to find a new goal… and the path grew more and more fragmented like light rays through a prism. And, maybe, just maybe… THAT is why Haiku has never yet reached R1. Because there is no longer a single path to the goal. There are MANY paths to MANY goals… all headibng in the same general direction, but there is no singular specific goal as there once was. Haiku R1 is now a vaporous, intangible goal… a type of mirage… seemingly just an arms-length away, yet can never be touched… because the closer you get, the further away it draws you. Something to try for… endlessly… ad infinitum.

And if THAT be how we’ve gotten to where we are, I simply have to accept that and shut up. And I can do that. Maybe that’s why I lost so much interest… because, subconsciously, I knew what the outside world wasn’t telling me. Or maybe I just wasn’t paying enough attention and missed the memo. But it wasn’t for a lack of trying.

I am not a troll, I assure you. I am simply perceived as one. I do not rant just to stir up the pot, though others may accuse me of doing so. A troll derives pleasure only from causing strife, by whatever means necessary. That is not my intent whatsoever. I desire only to wake many from the slumber of complacency (if it is as I perceive it, lo these 2.2 decades later)… to remind them that the first goal should have been reached AGES ago and the question of “why hasn’t it?” is paramount.

At this point, yes… R5 compatibility is moot. Even ridiculous in the grands scheme of things. But it would have been a goal we actually REACHED and THEN springboarded into R2 and beyond. And, at this point, Haiku IS just a BeOS-inspired development project.

Whether or not an OS is “flawed” is not the point. We all know that. There is no “perfect OS”, because they are written by men, not God.

At this point, if I had a working Email client, .PDF viewer, printer driver for HP laser printers (even just the Laserjet 1022n), working sound, working audio file player, and a reliable web browser, working WiFi… I’d be a pretty happy camper… until the next update… when WiFi breaks AGAIN and I can’t get online to web browse. Or sound stops working or whatever.

The main cause of my discontent is all the breakage that occurs from revision to revision. I don’t even dare wait YEARS between revision updates, because what IF my perfectly working revision is rendered broken at that point… all those years of use and getting things just so and being so happy… ruined in an instant.

This is why I have lost almost all interest in Haiku… I have gotten beaten down so much for so long… I no longer can barely muster any interest at all… except when I see a tiny bit of discontent from someone else and I think “maybe they’ll listen to me NOW!”.

I want Haiku to be what it was always meant to be… a worthy sucessor to the BeOS mantle. But it cannot be that if it keeps getting delayed ad infinitum. If things keep breaking. It cannot reach a goal that it never reaches. At what point will the ONLY people who even care about Haiku are JUST the developers? How close are we to that point? Would any of us have ventured to stick around this long (now 22 years), if we KNEW that’s where we would end up? If a vote were cast, how many people would vote their fervent interest in Haiku? Honestly. How many people even still give BeOS, itself, a thought?

I do not say all this to troll… I say this to help spur realization of how far we’ve gotten off track. At least, unless I missed the memo that the goal that once was, was forsaken long ago, and now it’s every developer for themselves and we’ll see where we all end up and hope that it results in a “finished” Haiku R1.

No man is harder to wake from slumber, than that man who wishes to stay asleep.

As a very old follower, since the call for OpenBeOS to its transformation to Haiku, I can confirm that Luposian is not a troll :slightly_smiling_face:.
As the current state, Haiku is still in beta, therefore it can’t really “officially” support a “normal user” base. All the users are “power users” and therefore should be prepared to have an unstable product.
Many power users, like myself and Luposian I guess, are thinking that with some sacrifice (what Luposian calls FOCUS) an official release should be in short reach.
Developers want to provide a “good” product and it is understandable (should be the “perfect product” by Luposian), but if not in high integrity/safety places, you will notice that sellers often sell unfinished products and put all the pressure on the devs later LOL (a la Microsoft before). Consumers are in fact very tolerant.
At the end of the day it is still a lack of resource issue. Maybe taking a despotic approach, like in the OpenBSD project, might be a good idea to have that first release after more than 20 years. One person decides on what will be in the release or not, with consultation but without voting.

1 Like

Use the betas then? That is the entire point of point releases.

This is the current goal of R1: Custom Query – Haiku

In any case, the goals of what should be R1 won’t always align. But I thunn this will make R1 an even better release, it should be the release one can confidently recommend to friends and family.
I don’t think Haiku is there yet really, and I doubt it will be there soon. Every dev has their own personal list of things they consider important for the release. For me some of those things are a dark mode, working network printer support (autodiscover), wifi ssid roaming, compatibility with microsoft devices (autodiscover smb shares, good smb compat), compatibility with apple devices (airdrop/airplay), bluetooth support, wine support, opengl support for gpus, DANE validating email client (and better email support in general), DNSSEC resolver, webkit2.

Some of those to me are more important to include and others less so, the point is that every dev will have their own idea of what Haiku R1 should contain, if it is rushed by one dev the others will percieve it as such and flawed setting a bad precedent.

In the end: if the betas work for you that is great!
If you really care for a R1 release then work towards it if you like, by contributing to what you consider to be high priority usability issues for example, there is no express train to R1, but we are getting closer still.

1 Like