[haiku-development] Re: Proposal from Begeistert: Getting Ready for the First Beta

  • From: Marcus Jacob <rossi@xxxxxxxxxxxxxxx>
  • To: "haiku-development@xxxxxxxxxxxxx" <haiku-development@xxxxxxxxxxxxx>
  • Date: Tue, 17 Sep 2013 11:01:34 +0200

On 17.09.2013, at 07:24, Stephan Aßmus <superstippi@xxxxxx> wrote:

> Am 17.09.2013 03:20, schrieb Sean Collins:
>> Stephan Aßmus wrote:
>>> And this all has a substantial technical side as well, where tools and
>>> workflows need to be built to make this all as painless as possible.
>>> This process and transition could take many months or even years.
>> If including package management into the OS introduced years of extra
>> development, then its inclusion has made a small problem into a
>> significantly larger one. The bigger one being that anyone willing to
>> donate to haiku, is going to give up and walk away. I am still highly
>> disturbed about the disposition of Karl at Haikuware and I understand
>> his frustration. You do have some responsibility to be up front with
>> your support groups. Frankly, they are tired of waiting.
> First of all, what I wrote are just my own impressions and ideas. Secondly, I 
> think the "waiting mode" of many in the community is probably the biggest 
> problem the project has. We don't need people who declare they want to help 
> and be told to search in Trac for the easy tickets. We need people who see 
> something that sucks in Haiku and actively work towards resolving that the 
> way they see how.
> What I tried to outline in my previous mails is a larger infrastructure that 
> harvests the possibilities brought to the table by PM. A large part of that 
> would even be outside of the Haiku code-base. We need web developers for 
> that. We need testers. People pouring energy and work into the representation 
> of software, writing reviews, translating descriptions, making screenshots, 
> keeping it all tidy and clean. There is an opportunity for people to get out 
> of "wait mode" and start contributing other resources than donations.
> When I donate money to Haiku Inc (which I only have done very seldomly, I 
> admit) my expectations are that they will put this money "to good use" to 
> further the project. And this has been fulfilled more than satisfactorily. 
> This is still an open source project. All what an organization like Haiku Inc 
> can actually "control" is how individual developers are supported to advance 
> some aspect of the code-base. If your expectation is to hold the 1.0 release 
> in hand within a certain time frame, then I am not sure how that expectation 
> could ever not be disappointed. There is neither enough money nor enough 
> developers to choose from.
> I have started coding the software center type application, initiated by a 
> mockup from Humdinger. It prompted me to envision how I would like this new 
> system of installing software to work. A lot of pieces are needed to work 
> together here. The actual PM framework in the OS is just one. There needs to 
> be some kind of online platform where users can contribute information about 
> software packages (screenshots, translations of descriptions, just plain 
> opinioned ratings). This will be a web application, similar to how Haiku 
> translations are handled. The software center app needs to pull information 
> from there. Building this infrastructure is all web development work. 
> Something I have mostly no clue about.
> If people step in, then maybe it all clicks into place in a few months from 
> now. Otherwise all we will have is pkgman on the command line and just the 
> official repository and no stable / current release channels and neither 
> rolling releases.
> Even with a family, it is possible to find an hour in the week here and 
> there. One needs to be motivated and there needs to be the notion that one 
> can rely on others to help. If someone had told me that nobody will come 
> along and code the web application side of the story, then I would have 
> probably never even started work on HaikuDepot. I just have this in my head 
> that someone or even someones will step in and start to work on the other 
> pieces of this vision.
> Best regards,
> -Stephan

Please allow me to add a few comments from an outsiders perspective. In my 
humble opinion Stephan is totally right with his observations about the 
community. As I see it, from just following the mailing lists from time to 
time, there are only very few really dedicated frequent contributors. However 
those really dedicated group of people managed to keep this project alife for a 
very long time, actually much longer than most people would have guessed 12 
years ago.

Except for a few quirks, they managed to produce an OS, which is actually 
running pretty stable most of the time, even though quite a few things are 
missing. Some of the missing parts are deemed essential, because they are 
needed to make Haiku usable as primary system, some things are needed to 
integrate nicely in todays infrastructure. However, how many of these things 
are really critical, knowing that probably nobody is using Haiku as their 
primary and only OS right now? My feeling is that those items are ranked based 
on the aforementioned perception that Haiku should be an OS, which stands on 
its own, rather than the reality that Haiku is an interesting project but 
mostly used as secondary OS only. Based on this, what should a Haiku release 
actually be? A stable version of what is available right now (or within a 
reasonable amount of time) or a version based on certian desired features, bug 
fixes, etc.?

Considering another comment made, who would be the audience of such a release? 
Is it the current community, which probably mostly is made up of the developers 
and a few highly interested and technical advanced people, who are fairly 
capable and probably mostly working of the latest repository or the nightly 
images anyway? Or is it the the rest of the world, which has no regular 
interest in Haiku anyhow and just peeks in occasionally, whenever there is a 

Basically this boils down to the question, whether a release is actually needed 
and if yes for what reason?

I strongly believe its not needed for the developers or the current 
community,nas anybody who falls into this group is using a fairly recent 
version anyway.

Is it needed for the general public? Probably not in technical terms, as there 
probably is no general public, which actually uses Haiku except for the curious 
short peek once in a while.

But all the above arguments are driven from a technical point of view. Now 
let's try to look at it from a marketing point of view. And yes I know, most 
engineers don't like the marketing people, as those tend to have no clue, 
present unrealistic schedules, demand weird and useless features and drop the 
interesting features. However the marketing people are also the ones who 
actually market the product, create the market or make the already existing 
market aware of the product and position it in a way that it gains traction and 
gets used. And let's face it, every engineers takes pride not only in the fact 
that he created something cool, but also in the fact that he created something 
cool, which is actually used by other people.

Therefore I think its important to not only judge release or not and when to 
release based on the given technical facts, but also from a marketing point of 
view. And since the last release is almost a year ago,  from a marketing point 
of view it makes sense to push a new release for this fall.

This will create a slight buzz again and might keep Haiku on the radar, even 
though all those mobile stuff seems to be much more interesting at the moment.


Other related posts: