[openbeos] Re: Openness

  • From: "Jorge G. Mare (a.k.a. Koki)" <koki@xxxxxxxxxxxxxx>
  • To: openbeos@xxxxxxxxxxxxx
  • Date: Sat, 12 May 2007 12:16:44 -0700

Hi Michael,

Michael Phipps wrote:
The admin team is the group of people who have *earned* the right to direct Haiku. They have put in blood, sweat and tears to get us to where we are today. You don't join the admin team, you get invited. :-)

I don't doubt that many of the admins have put in blood, sweat and tears into Haiku, and everyone is grateful for that. The issue being raised here is different. If you read the last paragraph of the original post at...

http://www.freelists.org/archives/openbeos/05-2007/msg00100.html

...you will see that the point being is one of lack of transparency, and the notion that the project leadership *also* needs to earn the trust (and respect, I may add) of the community, and that more transparency as to what goes on within the admin group is needed for that to happen.

And while I don't doubt that many of the developers have earned their place, the "get invited" part is reminiscent of an untouchable elite mentality that, IMHO, does not convey a sense of openness at all.

What goes on at the admin team meetings is, generally, closed to the public. Why? Because it allows the admins to discuss without public comment. We can have crazy ideas and shoot each other down. We use it as a forum for working out details on things like distro guidelines without having to do it in public over 10,000 posts.

Is closed bad? I don't think so. It keeps the signal:noise ratio very high (or at least it intends to). It means that the people who have the right to make those decisions can do so privately. Note, too, that EVERY OSS GROUP OUT THERE HAS THIS. BSD doesn't publish their minutes. Linus has a mailing list similar to our haiku-admin list. Look at the Red Cross, even - you can't get minutes of their board meetings. They are closed.

One thing is to have discussions in seclusion in order to achieve focus (which is good), but another is to keep the community totally out of the loop (which is bad). This is what happens with Haiku: a lot of discussions take place on the admin list or on IRC meetings, but very little trickles down to the rest of us.

You mention other OSS projects, and I have to say that, in comparison, Haiku pales in terms of openness. Remember when I once mentioned to you how KDE discussed their release roadmap on a public mailing list as a good example of openness?

http://mail.kde.org/pipermail/release-team/2007-March/000071.html

That's what being really open is about.

Don't know what to say about the Red Cross example. How is it relevant to an open source software project?

Having said that, Haiku tends to be more open then anyone else, IMHO. Generally, when a decision is made, it gets published.

Michael, come on, this is an open source project. That you publish decisions is a given. The lack of transparency that the original poster pointed out is what makes the project look closed. It is a perception from the outside, and you have to look it from that perspective (not from the inside) in order to recognize it.

As far as the updates to the donations pages... Yes, I made them a couple of days ago. I didn't announce them because I wanted to have a few people make their way through them by accident first to see if there were any problems before I announced them. It was *intentional* and *planned* that I would announce them when I was sure how they were working. There is one small issue (people who make 1 time donations are getting redirected to the t-shirt page, if you want to know) that I need to resolve.

As far as the t-shirts go... There is a very good reason for us to do this ourselves. Several, actually. 1) Printing t-shirts is generally cheaper by the case. 72 of them, in particular. When I did the shirts for WalterCon, I needed 15 or 20 of them. By printing 72, I saved a TON of money per shirt. So WalterCon was cheaper, and we got shirts to sell. 2) The revenue difference is a factor of 2. CafePress charges somewhere around $12 for the quality and printing of shirt that we do. The shirts that I have here are less than $6.

Yes, but in a CafePress-like shop you can offer a lot of stuff more than just T-shirts, so your overall potential for revenue is much greater, which is what matters.

All of which sounds like I am a greedy slimeball. :-D Except:

3) There are rules about charities selling items. The goal of the rules is that you can't be a charity so that you can run a store - you have to be a real charity that, oh, by the way, happens to run a store. This is why Public TV stations (for those in the US) ask for a $60 donation to get a CD. Because the value of the CD is so low that almost the entire $60 can be counted as a donation. This all factors into some of the end of the year paperwork that I have to do to prove that we are still a public charity.

So, to sum it up, if we chose to go with something like Cafe Press, each sale would EITHER have to be a) a lot more expensive or b) we would need a lot of donations to keep the cost of goods:total revenue in the neighborhood that the IRS wants to see.

Speaking of donations... We have talked, on and off, about posting a list of donors. Every time I have asked actual donors what they wanted, it has been a split on the order of: 1/2 - I want to remain anonymous, 1/3 - I don't care, 17% - I want to be recognized. That's why we don't maintain a public list. It doesn't make sense for the low percentage of people who want it.

As far as bounties... This horse just won't die. :-/ All I will say is "look at the fruit". Not a single bounty came even close to completion except USB which was already underway. Some of them couldn't have even been done, legally (Java). I am not opposed to the CONCEPT, but it takes work to ensure that the bounty is possible, legal, and that the money matches the effort. I have asked many people, privately, to put together a list of the drivers that we need, the sources of specs/source code, etc. I haven't had a response yet that gets me what I would think would be wise/necessary to proceed. So we haven't. It is WORSE to do something halfway than to not do it at all, especially when you are asking for money.

This horse does not die because it is the best example of what is wrong with Haiku.

I will tell you what the fruit actually was: an individual was very eager help, but that was forced to do it all alone due to a lack of response from Haiku, and that ended up being demoralized and eventually driven away by the lack of recognition and even some unwarranted chastising on a public mailing list from the very same people he had helped with his effort.

The fruit was also the $2,000 raised by a dedicated individual alone, donated by people eager to make contributions to Haiku, and that went into the Haiku Inc. bank account.

You can argue that about the details of how bounties should be done, but that's not really the point. The point to be made here is that somebody offered himself to work with Haiku, and that Haiku failed to capitalize on a very unique opportunity to turn somebody's effort into a sustainable bounty program.

Do you really think there is no lesson to learn from here?

Cheers,

Koki


Other related posts: