Stroller wrote: > One of the things that Risto was complaining about was the number of > distros for the Freerunner, and "they're all incomplete!!" > > His words "why don't the developers feel ok to contribute directly to > 2007.x and 2008.x but 'fork' their own distros?" echo my own complaints > a couple of months ago in my message "Community contributions to core > apps & features" 9 weeks ago. Let me provide a little history. back in October of 2007 We faced a decision on software. From the schedules we knew hardware would be ready before software. So the following choices where available to us.
1. Ship the hardware with a bootable linux. Absolute bare bones. a number of people thought this would make the community happy. 2. Ship with Qtopia if it was ready. A number of people supported this. 3. Take the existing Openhand project. Cut back the features to a bare minimum ( dialer, sms,contacts) and ship with that. #3 (2007.2) won out for a variety of reasons.It was not a perfect solution, by a long shot. It was clear any one of these paths would cause some sort of division, some sort of complaints. In parallel, we would define a new set of applications with a different look and feel and some new and interesting applications. this would be a long term project. The design would be as simple ( empty vessel to use shawns term) as possible, and we would encourage developers to add to it improve etc. It was clear that this too would cause division. I guess you could say we embraced fragmentation, well aware of the pitfalls. http://arstechnica.com/news.ars/post/20080918-openmoko-ceo-embrace-fragmentation-diversity-is-a-strength.html > > You & Sean have said that you want to follow your own vision in the > software side of things - developers can't supply patches for power-user > features (that make the UI more complex) and expect you to include them > in the "core" distro, if you're trying to produce a phone software suite > for grandmothers. That's not the Goal of 2008.x The design philosophy is to create something clean and simple and then let end users clutter it up to their hearts content And I really understand where you're coming from with > this - you have to sell loads more volume if you want your hardware > business to be successful. Well, actually not. I measure success in terms of profitability. At the current volumes and the current cost of operations we are where we need to be. ramping to grandma volume is a careful process that does not happen overnight. Crawl walk run. The pesky in and outs of cash flow and building out process of bigger sales, marketing and engineering organizations. > > So the only answer to this IS to have more distros. And really, anyone > complaining about the state of the current software stacks should have > been here 3 months ago. Back then it was "Openmoko shouldn't have > shipped broken hardware with this GPS bug!" Isn't that now all fixed in > the kernel drivers? No-one who sees how much the situation has improved > is complaining now. We actually like the fact that there are competing distros. The one unique thing we offer is the freedom to choose your distro and choose your carrier. In the US, for example you can choose Tmobile or ATT. > > I appreciate there's some room for compromise between "grandmothers" and > power-users on the state of the software. You can start with a basic > interface and have a framework so that extra features are only shown > once installed & configured by the advanced users. Yup, thats the plan. But there is no "one > true way" - if we look at the state of desktop window managers, we see > that. This is a relatively mature "market" - Gnome & KDE have both been > around and stable for several years. When Risto or some other newcomer > looks at Openmoko & the Freerunner, you cannot expect them to see a > "path" as clear, directed and well-signposted as that space. Agreed > > I applaud your effort - you're responding to criticism and asking how > you can fix the problem - but I can't see how a "community manager" can > change anything. You can't exactly deny there are several Freerunner > distros, or that they're all works in progress! facts are facts. The community manager cant change past facts. What can he do? That's an open question I pose to a creative community. Do I have ideas, sure; will some of them suck? most assuredly. Might some be good? I've had my lucky epiphanies. The one thing I'm sure of is that the community will think of things that I didnt. ( and yes, some ideas will suck and others will have merit) > > Build the product and the community will come to you! > It already has, it's just a little too early for everyone to see the > fruits of this. yes. my question is can we optimize the effort. Again, open question. negative feedback is as welcome ( in due course) as positive feedback.not to pat you on the back stroller, but when Sean and I talk we almost invariably discuss your perspective on things. > > We already have Michael Shiloh providing weekly community updates (ahem) > - IMO a "community manager" would just be a distraction from Openmoko's > real business. You should be concentrating on the hardware, and if > you're employing an additional member of staff then make it a kernel > programmer, so that your hardware runs more smoothly for the distros > that evolve around it. Or get FSO complete sooner, so that (again) all > the distros benefit. Well engineering hiring continues day in and day out. As VP of marketing it's part of my job to find ways to make use of this wonderful asset, the community> I'll give you an example. At linux world I faced a huge problem. Two booths. and a marketing staff of two. me and michael. And a sales staff of two. Whats missing? somebody with technical knowledge at the booth. Should I ask for an engineer to attend the trade show? Nope, I asked the community to step forward and man the booths with us. When the press came to talk to me, I pointed them at the community member who gave them the unvarnished truth. At first PR thought I was insane, later they changed their minds. Now, for example, I cant coordinate this kind of effort all the time for every show around the world. Is that a job for a full time trade show manager or a community manager? I don't know, I'm at the stage of kicking around ideas. Some of them should get kicked in the head, others in the butt. But its not a distraction. me dragging engineers off of projects to support trade shows is a distraction, to use a concrete example. > > Running a business is all about customer satisfaction, but you can't > keep EVERYONE happy. There will always be 1 or 2 who "don't get it", and > they just happen to be vocal about it. You've already satisfied 99% of > us with your open-source mobile phone platform - already so many people > are bringing their own ideas and (more importantly) work to that. Ignore > the whiners! I don't include Risto in that characterisation, but I don't > see how you can please him. The goal isnt to make everyone happy. there is not enough free beer for that. the goal is to optimize the value of the community. Looking for ideas. "There's no way" is an acceptable answer, in the end, after exploring ideas. But Steves rule for brainstorming is that during the brainstorming part, NO is not allowed. get the ideas out. get them on the table. then evaluate and toss out the bad ideas. It a difficult process to suspend judgement during the brainstorming session, but its vital to the process. Plenty of time for No's. > > In 6 months time you will have some amazing community distros for your > phones, and at least then the "incomplete" complaint will be satisfied. > Those that "don't get it", meanwhile, will have found something else to > complain about. This is the nature of open source. the complaint about incomplete distros was what got me thinking. I dont see a community manager as a solution to that. basically I read the article he pointed at, and it took me down an entirely different path. Tangent Man! Now the distro question obviously plays back into it. In what way? Dunno yet. If I knew or had made a decision i would post a job spec. But I dont know and havent made a decision, so I throw it out to the community for feedback. Yours is always welcome and taken to heart. > > Stroller. > > > > > > On 6 Oct 2008, at 03:37, Steve Mosher wrote: > >> Stroller let's assume it is Possible. I had a long chat with Sean >> today. We both read the community list daily and our number one topic >> of conversation was the "Lost community" thread. Sean asked me what >> I thought of having a community manager. ( he was reading my mind again) >> On one hand, I said, Stoller has some good points ( as always). It would >> be a bit like herding cats, and in someway we want interesting cats, >> wandering off to do things that A) we didnt think of and B) we disagree >> with. basically because we don't know everything. On the other hand, >> we do recognize the benefit to be had from a little bit of structure. >> I have my ideas about what a community manager would do to organize and >> mobilize, But before I put those ideas down, I'd like to throw it open >> to the community. Question: what functions do you see a community >> manager performing. Write his job spec. ( hint hint) > _______________________________________________ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community