growing pains

Nokia’s launch last year of the N900 mobile computer introduced a bit of disruption to their normal business model.  The ripple effect naturally propagated down to the lowest levels and had a huge impact on the semi-independent community,

Before subjecting you to an epic article, I want to point out that change was anticipated, but have to admit with some embarassment that when talk of adding cellular capability to Maemo devices gained traction, I was one of those naive souls downplaying the potential impact.  I assumed two things, both unfortunately proven incorrect:

  1. the consumer cost of a phone-capable tablet would be fairly low
  2. Nokia would platform the product and continue offering “slate” form factors with no cell phone embedded along with cell-enabled models

But others more prescient grasped very early that even if either or both of these played out (which they sadly didn’t), adding that GSM/UMTS radio introduced the devices to a largely different demographic… one that could significantly shift the priorities of

This became patently obvious even before the N900 reached the hands of eager early adopters.  Once the N900’s existence was certain to be more than rumor, questions about cellular frequencies, service providers and other phone-oriented subjects began to drown out the usual internet tablet talk.  And after ~300 devices were distributed to Maemo Summit 2009 attendees in Amsterdam, application ideas very quickly began swirling around what could be done with a much more connected and mobile Linux computer.  With repercussions both good and bad for the status quo, those ideas proliferated fast and far.

The instant success (in relative terms) of the Nokia N900 quickly indicated that the community was ill equipped to handle such a hot product.  Across the board, systems, policies and processes failed time after time to accomodate the exploding and changing population in every way.  This is not an indictment of anyone; rather, I’m setting the stage for solutions.

Recent lively discussion on both the mailing lists and discussion forum have identified the following as the most critical issues, in no particular order of importance:

  • Application testing
  • Bug reporting
  • Brainstorming
  • Forum management
  • Fragmentation of communications
  • Council purpose and effectiveness
  • Recognition of member contributions

Let’s break this down topic by topic:

Application testing

I want to call this the hottest of the bunch.  As can be seen in this Talk thread and others, many outspoken individuals in the community agree (more info can be found on the wiki).

The current process favors a small community crunching through a handful of utility-type applications with little or no danger to the devices and/or other applications.  Most of us following this topic know that these expectations are no longer valid with the advent of Maemo 5 and the N900.

Some community-oriented or independent coders may balk at a more rigorous, formal application testing process but we have reached the point where it is no longer an option.  Granted, one size does not fit all, so whatever new process is developed needs to avoid lumping applications into one indiscriminate bucket.  And that’s a part of the current problem anyway: deciding application readiness is carried out by testers voting “thumbs up” or “thumbs down”, a black-and-white system that blithely ignores application purpose, scope and behaviors.

Some progress has been made in reaching consensus on what’s needed, but so far there’s no comprehensive workflow diagram to drive an improvement project.  Hopefully someone with the necessary expertise (it’s not me, but I’m willing to help) will champion this and get it moving.  In the meantime, I’m suggesting we identify a temporary trusted gatekeeper or gatekeepers to manually expedite application testing so nothing stalls as we reinvent this mud-stuck wheel.

Bottom line, the Maemo repositories and, more importantly to Nokia’s future, the Ovi Store need filling… especially if Ovi’s relaunch is to be successful.

Bug Reporting

Utilizing’s Bugzilla bug reporting system has been mainly in the purview of a small set of diehard users and developers.  A separate login from (where new and casual users tend to congregate) along with an IT-grade interface throw up hurdles that the majority would rather not face.  Users accustomed to a world of wizards are looking for a friendlier approach to providing feedback… assuming they can be motivated in the first place.

I believe the solution is a web-based “user experience framework” that extends feedback and other participatory ventures beyond the traditional confines.  I want to bring bug reporting, application rating, developer donations and related activities to Maemo devices in a low-barrier format.  I have ideas along those lines I am developing and will be presenting here very soon.


The Brainstorming tool and subforum were set up to provide a channel for ideas that didn’t quite meet the definition of bug solving but might be deemed worthy of consideration for near-term and future implementation.  Improving aspects of Maemo devices, software and even the forum itself are the typical core intentions.  Slow to take off at first, partially due to poor server performance, the process has recently caught fire and with that popularity comes controversy.  As a moderator I was instructing participants to follow a process that turned out to be unsustainable based on the current Brainstorm incarnation.

There’s been some intense discussion over just how this should work, but the easy analysis is that brainstorming isn’t really meant to be enacted over the internet.  The productive sessions I’ve led or participated in were conducted in person and followed the group development process of Forming/Storming/Norming/Performing.  A brief description from wikipedia:

…a model of group development, first proposed by Bruce Tuckman in 1965, who maintained that these phases are all necessary and inevitable in order for the team to grow, to face up to challenges, to tackle problems, to find solutions, to plan work, and to deliver results

The dilemma we face is that progressing into Norming and then Performing are difficult in a strictly virtual environment– especially one lacking project management capabilities.

This is another activity crying out for a process diagram at the very least, and that’s on my agenda and in  progress.

Forum management

An expanding and changing community not surprisingly stresses a forum structure.  Members accustomed to legacy needs find themselves at odds with newcomers drawn by new advents like the N900, and suddenly feel they don’t fit in their own “home”.  It isn’t easy to reconcile the needs of both camps, but ignoring the issue tends to result in splintered communities.  However, that’s not always a bad thing– if a community gets so large and diverse that finding common ground becomes impossible or impractical, spawning one or more highly-focused forums might not be a bad idea.

I’m told the Brainstorm subject I started on this is overly broad, so maybe it needs to be broken down into smaller chunks.  Regardless, periodic forum redesign is a natural consequence of a morphing membership so don’t be surprised if what this matter really requires is a comprehensive evaluation as opposed to picking at various parts and applying bandaids to the sore spots.

Fragmentation of communications

A long-running debate over where important community issues are better discussed too frequently misses the mark.  Different communications modes have different strengths… and weaknesses.  Choice of discussion venues also comes down to a participant’s preferences, location and availability.

Automating cross-pollination between venues is the ideal, but pending that sort of solution the community depends on facilitators to bridge the environments.  This isn’t really difficult and just requires diligence and desire.  More on this as it develops.

Council purpose and effectiveness

As I wrote not long ago, the current community council finds itself at more than one crossroads.  Both at the midpoint of our term and in the midst of a rapidly evolving environment, we have been struggling to find our way… or at least it feels that way to me.

Prior councils seemed more active and visible, but that could easily have been a byproduct of a smaller, more focused community.  Now on its fifth device (770, N800, N810, N810 WE, N900) the community is more diverse and dynamic than I’ve ever seen it.  The challenge of this council is to examine where we are at and then decide where we need to go.  Suggestions of longer terms (eg, 6 months to 12), frequent council meetings and fine-tuned focus on attainable goals are worthy of consideration.  I will update you all as we arrive at decisions… and as I’ve said before we are your council and open to suggestion!  (FYI, here’s a similar discussion by the previous council).

Recognition of member contributions

Karma has been another popular topic, an on-again-off-again carousel of calculations with the goal of designing a recognition system that encourages member contributions.  Philosophical doubts about the usefulness of such a system aside, the main struggle lies in juggling various contribution methods into positions of parity.  As with most debates supported largely by judgment calls, this one has been hard to settle.

The subject appears to be in a lull for now, which could mean participants are digesting the details and preparing their next interjection, or hopefully we are close to a consensus and just don’t realize it yet.  I think we’re close enough that most differences are relatively minor and easily resolved… and there’s nothing to keep us from monitoring the result and refining it as needed.


The main point to consider here is that we can’t be the first community to deal with these issues.  Therefore it makes sense to look to our peers for best practices and take advantage of someone else’s lessons learned.  No need to waste time on completely new inventions in an open source world.

One drawback we have faced is the failure of people to self-identify as process, role and item champions.  There are many factors contributing to this but I’ll just address the one that affected me personally: the fear that others might not see me as the right person for the job.  But at some point I finally realized I was creating a self-fulfilling prophecy and decided to assume whatever mantles fit and were unclaimed.  Community outreach was the main one for me.  Now the challenge is to get others to assert themselves and fight for their passions.

My intent is to stay on top of these items and use this blog to try and bridge the email and forum discussions.  Subscribe and stay informed!


On a lighter note, irreverent humor has returned to after a lengthy absence.  It’s good to see people cut loose once in a while and convert frustration into good old-fashioned pointless banter.  That has value, too.  😉


19 responses to “ growing pains

  1. Good post. Lots of great points.

    Can I say how much I hate being told “We discussed this before”. The community is a different place from what it was before: what worked then may not work now, and vice versa.

    And can we do more than talk, please?

    • Well… we’re trying, and that was part of my purpose with this article. Sometimes summing things up can help clarify what actions are needed, especially when you have as much diverse input as we’ve had on these topics.

  2. Thank you.
    Much food for thought.

    • Thanks for commenting!

      It’s obvious though that not everyone feels the same way. I’m always curious to know why anyone would find little or no value in a summary/analysis like this (or maybe just my execution). Unfortunately, so far there’s been no constructive feedback to guide me in improvement.

  3. @RevdKathy, however sometimes we HAVE discussed it before, in the last 2 months – i.e. AFTER the N900 launch, and e.g. when the problems with the QA system became obvious.

    Can I say how much I hate having the exact same discussions, with the same suggested answers, being held over and over again; whilst the people capable (or willing) to enact pending changes are trying to get around to it?

    • Agreed. Karma and testing have certainly been hashed over enough, and there should be enough material now to drive solutions. We can always monitor and improve as needed.

    • Well – I’m just getting into this community (owned a N900 for exactly 1 week), but in other open source communities I’ve been involved in (my previous phone was an openmoko freerunner) – comments like this usually mean that there isn’t an easy way to find the previous discussions…

      I must admit I have some trouble navigating around the site, and finding the interesting and important discussions… I had trouble finding any ‘getting started with the maemo community’ kinda messages…

      I didn’t even know there was a maemo council until today…

      • Thanks for the input Warren.

        We are definitely working on solving the issues you cite. As my write-up shows, there really have been growing pains and we’re just now getting solid actions wrapped around the gaps. Now that we’ve gotten the bulk of the discussion out of the way, going forward the fixes should come faster than they have in the past.

        Please check the following Talk topic to contact the council:

  4. Just in case the related links change– Quim had a relevant article in June 2008 that should be very useful for re-examination and benchmarking. I’ll go over it soon. Reposted:

  5. Pingback: Pinguins Móveis » Blog Archive » Pinguins soltos

  6. Can we move this discussion to the Mailing lists please?

  7. Hey! What happened? The little sarcasm closing tag didn’t show up in my post! I will have to add it here.


    That means I am just kidding about moving this to the mailing lists.

    I think the joke is kinda ruined now.

  8. Pingback: Maemo moments. | Maemo Nokia N900

  9. Thanks for a amazing piece of information. Keep the great work up.

  10. i like your topic, i will visit this site more often,
    i like your post specially about :
    Nokia’s launch last year of the N900 mobile computer introduced a bit of disruption to their normal business model. The ripple effect naturally propagated down to the lowest levels and had a huge impact on the semi-independent community,


  11. Hi, I would gladly contribute, especially in the project management part (leading a virtual world-wide project management office team) and probably some in the virtual community organization.

    The question is: how to contribute? I’m preparing a presentation, but at this moment one word typed takes 2 hours of checking around the webs of to see, if it is fitting 😉

    • Really, the single best place for anyone to help is wiki page creation/editing. In addition, there’s plenty of room for people who want to help with community outreach, which is about to start heating up soon (we’ve been in a fact-finding and process developing mode for a bit). But hang out in the Community section at, and join the maemo-community mailing list as well– you’ll be dodging opportunities before too long!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s