What MeeGo can learn from Microsoft

I realize I’m courting controversy with the title, but for good reason.  I’m going to set aside any unsavory or otherwise questionable aspects of Microsoft business practices to focus on one that has worked very well for them and I believe can for MeeGo as well:

Developer outreach.

I’ve made it clear with this blog that the bulk of my information management and software development experience evolved in Microsoft business environments.  That naturally led to heavy involvement in Microsoft’s developer community, which included local and regional product launch and outreach events.

It’s no secret that Microsoft loves developersMSDN, Technet and the Microsoft Partner Network are successful examples of the company’s long romance with coders.

And one thing Microsoft does know how to do is throw a party.  I remember vividly the roaring 90s, when thousands of people packed the streets of downtown Dallas, Texas for the Windows NT Workstation 4.0 launch, each clutching free disks loaded not just with that cool new operating system but with fullblown Office as well.  And even when things slowed down after the 2000 tech bubble bust the parties continued, just a little smaller.  Great food, cool prize drawings, handouts of expensive software.  Oh, and key presentations too.

In recent years, however, Microsoft has further scaled down the face-to-face engagements.  There are many factors driving this but my guess is the general economic malaise is the biggest one.  Which is unfortunate for Microsoft, because this is the exact sort of time when companies need to work harder on presence.

And that brings us to MeeGo.  Even as part of an organization rather than corporation, MeeGo is going head-to-head against well-established commercial offerings.  One reason the prospects of companies like Microsoft are in decline is because those of open source players are on the rise.

I am concerned here though because of a response I received from MeeGo-partner Nokia when I raised the need for regional, physical community outreach (primarily for developers, testers and superusers): that before Nokia committed to such efforts, they needed to first be assured there was enough interest to justify the expense.

I can understand such hesitancy from any executive far removed from the subject, but I think it’s fairly safe to say that this is a circular scenario.  There certainly won’t be much interest if it isn’t seeded and cultivated.  Success will depend on investments made without fear.  Failure will be a self-fulfilling prophecy borne of inaction.

After talking with European and South American friends at KDE’s Akademy 2010 recently, I get the feeling this is largely a dilemma for the United States.  There’s plenty of the necessary activity in the rest of the world; it seems the US has been abandoned to RIM, Apple and Android.  If it’s in the MeeGo plan to leave that as the status quo, that’s one thing– but if Nokia, Intel and other partners still see the US as a viable market then MeeGo outreach here isn’t just a want it’s a desperate need.  And given the mostly-closed operating environment here, a US MeeGo push will have to be nothing short of spectacular as I’ve noted previously.

So the challenge goes out to MeeGo.  Of course it will be interesting to see who takes ownership of what activity.  Is it the responsibility of the Linux Foundation to manage this outreach?  Or would it be best managed according to need and participant interest, such as Nokia directly funding Qt and mobility-based events?

Some things that work for Microsoft don’t translate to MeeGo.  Microsoft makes quick friends of developers by bribing them with free copies of operating systems and development tools– lures that aren’t available to open source solution providers.  And there are as many differences between MeeGo and Microsoft as there are similarities.

That doesn’t let the MeeGo sponsors off the hook.  Nokia, Intel and others need to “kidnap” developers away from their current comfort zones.  They will have to come up with a convincing sales pitch for Qt development over the likes of Visual Studio (maybe not so difficult) as well as alternative incentives.  Free or discounted devices, CPU giveaways, free classes– that’s a good start.

Prying developers loose from commercial enterprise will take some work but it’s not an avoidable option, and it will take more than online appeals can muster.  MeeGo needs to physically get out in front of developers, and in a spectacular way.  Big, splashy campaigns.  Roadshows with lots of hands-on demonstrations.  Echoes of the upcoming MeeGo Conference 2010 in every major region, especially the US, Brazil and southern India.

An article I read today likens developer advocacy to Ohm’s Law.  Good analogy.  MeeGo will definitely need to electrify the developer community.


20 responses to “What MeeGo can learn from Microsoft

  1. Great post. I am a regular reader of your blog.

    I would like to add the following. If Nokia wants to attract developers from North America, then they need to do not only developer road shows but make sure they also advertise and actually sell their product in the region first. Many developers write applications thinking “Ok I bought this cool device. How can I make it better?” or “My friend has this new phone and wants to do X. Can I write software that will help him/her?” etc. I mean to say that the device and platform must stand on its merit FIRST before many developers consider developing for it.

    So build a solid product, make me want to own a MeeGo device, actually make it available in North America with proper 3G bands, advertise it here, make it available in shops and with carriers and make sure that its solid enough that I can actually recommend it to my friends. THEN I will buy it and develop for it.

    Take the N900. While its an attractive product for a hobbyist developer like me, Nokia never bothered to release a version compatible with HSPA here in Canada thus turning off lots of developers as they may not want to use a non-3G phone as their main device. And for students like me, buying a phone just for development purposes is not an option.

  2. Koos Vriezen

    MS had a hard time pushing windows-3. The successful strategy IMO was to bring key apps, office, sqlserver, browser, mail etc and easy to use development tools like VB combined with the COM framework together with a stable backward API to this platorm by MS themselves. (Ironically the latter is hunting them to bring out a light OS, 20 years later.)
    Also that you could install it on cheep hardware, assembled yourself.
    Looks to me, android is taking this approach as well.

  3. Well, if you went to Nokia’s Maemo conference in Amsterdam you would have noticed that Nokia too isn’t shy of giving expensive presents to developers. Like a N900 for everybody while the device wasn’t in stores yet at that point.

    And if we developers are honest, if you went to a few GUADECs you know it wasn’t very hard to get a free N810, 770 and N800 either. Just passionately talk about your project that you are porting to the device to one the Nokia guys, and pretty soon they’ll most likely give you a device.

    But don’t tell that I told you that. Besides I don’t know how willing to give away devices they are today. Just try?

    • You’re right, and I was at Maemo Summit 2009. In this article I was speaking generally though and didn’t want to assume that the past experiences will be repeated. And anyway, it would be useful to explore other incentives as well.

  4. From the MeeGo side of things, Nokia might not be giving as much attention to the US, but Intel (and even Qualcomm) are. When I was at Uplinq, there was a nice buzz around MeeGo – not form the handheld side, but how it skillfully answers some of the service integration efforts many of these companies want to push.

    Could Nokia have done more in their session at Uplinq (wish I could have been there, was at another session and caught only the tail end where it was mostly the same stuff we’ve heard already)? Sure. But, I don’t know that the mindset of the US market is as good to crack for Nokia as a brand of MeeGo, as much as it will be for players with a bit more positive feel from both developers and consumers.

    That being said, I will also agree with others that given the right device with MeeGo – and a price and marketing push to boot – Nokia could ignite that passion to cause some of these parties to happen.

    Side question: there are only so many mobile areas where this could happen: Seattle, Silicon Valley, Austin, Chicago, Boston, NYC, and possibly Atlanta. Would there be enough energy generated around these cities (except Silicon Valley, NYC, and Boston) to support a new-fangled developer engagement party?

  5. Pingback: How Nokia can retake the US Market… and more « Tabula Crypticum

  6. allnameswereout

    Microsoft gives kids cheap candy, makes them addicted, then raises the price and anyone who also wants to make quality candy they attack with dirty tactics. They simply used the fact people required backwards compatibility and used this to (MS)DOS -> Windows 3.x -> Windows 9x -> Windows NT-based -> …

    In their server products, NT was simply far cheaper than UNIX in the 90s. They won the market, and then a competitor came (Linux-based products) and they started to bad mouth it, and use all kind of dirty tactics to minimize their damage (such as SCO, patents, flirting with competitors such as Novell, GPL-trolling, embrace & extend on Samba/Kerberos/… etc etc).

    However, getting backwards compatibility on embedded devices is really difficult and inefficient. So we tend to support the current protocols and make our own UI to them. This is the backwards compatibility Nokia has to provide, and they are partly doing this. My N900 now has a calendar supporting ICS.

    If you recognize the above, think of Xbox, and see how they managed to set up a product line in the console market. Right, ok, with losses they succeeded there. Now, do you believe people see Microsoft as a corporations which delivers stable products, with good usability? No, especially not in embedded/KISS aspects, and that is why their smartphone market isn’t strong, and why their search engine will never win from Google. They’re trying to improve it, but their usability is _horrible_. Much stuff is only accepted because it has only been that way, but still, something like c:\ makes far less sense compared to / the UNIX hierarchy.

    Because Maemo is Linux-based it follows the UNIX philosophies and is partly open source based hence follows open source philosophies. Stick to those instead. Not Microsoft. If anything, Nokia can learn from Apple. Apple also uses an open source UNIX stack and applies this on embedded products. They have successfully created a proprietary/commercial software market (“AppStore”) which allows developers to receive some income based on their work. For those who want more (cracked or open source), there is jailbreak. Nokia will provide this jailbreak for open source software by default, provide an easy app store framework based on APT and easy payment, and allow users to decide between a locked device or an open source device (much like in Symbian). I have faith in Nokia, but I don’t believe they will do everything right in 1 strike; their next strike. But to the reasonable customer that is OK: nothing is perfect. Not even a product. For if it were, there would never be any new successor.

    • The point was to take away the best practices from Microsoft that can be useful in an open-source context, and discard what isn’t. No matter what, Microsoft just flat excels at developer outreach. SOME of its practices can be used.

      • allnameswereout

        Well, which ones would be usable in an open source environment? Remember that open source community is less centralized. From an end user point of view the centralization, if anything, is starting at the distribution/vendor (for a developer it depends what they develop). They have much less resources than Microsoft. So, they (and tons of community sites) do have quality articles and databases, but they’re not coming from 1 centralized vendor.

        However, when you write: “Microsoft makes quick friends of developers by bribing them with free copies of operating systems and development tools– lures that aren’t available to open source solution providers.” you make a mistake. Open source already gives away their source code, and although not by definition usually also development tools, documentation, and a free copy of the OS. So that factor is BS.

        I already mentioned centralization (which UNIX vendors also each did).

        In my previous post I mentioned the fact they own the (desktop) market.

        What is left is lack of QA on documentation, hegemony of / hierarchy, and configuration files, and of course the many languages available.

        Am I forgetting anything?

        If you take into account the size of Maemo project are they really doing that bad on above matters?

      • Microsoft normally charges for its operating systems and tools. There is usually no charge for Linux equivalents. Therefore, no value in providing Linux-based items to event attendees as any sort of incentive. So, no, the factor isn’t BS. The Linux world can’t offer free software as a hook.

        And don’t mix Maemo with MeeGo. My original concerns were indeed once directed at Maemo, but that’s now moot. The question is how to create and manage effective community outreach under the MeeGo umbrella. There are certain best practices that are MUSTs if an open source solution is going to go head-to-head with commercial enterprise. Big question is, who takes responsibility for what outreach activity in the unusual MeeGo model? More and more I’m thinking the Linux Foundation would be the best candidate overall.

        I must say I’m surprised at this sudden outburst of aggressive posting from you though… ???

      • allnameswereout

        Yes you’re right if you give something for free which was previously available for fee it appears to some as an offer not to be refused whereas something which is given away from free is not valued since it doesn’t cost anything. Ie. there must be some kind of drawback, hidden cost. This is a fallacy though. A common one, but a fallacy. To put in a different context: the Maemo SDK is given away for free, while normally SDKs aren’t for free. If you also consider piracy the impact of Microsoft giving away for free their development tools isn’t very high. Also, remember that Microsoft catches up with their competitor (who already delivers it for free) by finally providing it for free.

        From what I understand Maemo 6 = MeeGo Mobile ARM edition and MeeGo is just the upstream project which is an embedded UI-based Linux distribution using tons of frameworks compatible with Linux desktop/server. What I don’t understand is whether Nokia calls the OS Maemo 6 or MeeGo 1.x and which platform site will be used. Is all the wiki content useless and out of sync once again when next product delivers?

      • allnameswereout

        Here is a sysvinit replacement inspired by Apple’s launchd http://0pointer.de/blog/projects/systemd.html by same author as PulseAudio. It will allow a more parallelized startup of the OS, with services being started on demand. This also means that, for example, SSHd wouldn’t be fired up until it is necessary. Projects like this is what, I believe, push the Linux ecosystem (and in this case embedded market) forward. I remember vividly, being at FOSDEM, hearing DBUS, Udev, and HAL being explained. Nowadays they’re part of the Linux ecosystem, and you probably don’t know better, but back then this was ‘evolutionary’ and ‘eureka!’. The power of Linux is that this software can be integrated in every Linux ecosystem, to slowly but surely make a better OS. This is an example of where MeeGo could become technologically better (which is a different aspect than e.g. development or documentation).

      • Microsoft has given me thousands of dollars worth of product over the years, most of it tied to event attendance. That’s not trivial. It’s been an effective form of bribery.

        I’m just saying the open source endeavors, MeeGo in particular, could stand to find some grounds for competition. What are some ideas?

  7. allnameswereout

    Flattr micropayment system http://raphaelhertzog.com/2010/08/12/how-to-make-110-eur-in-one-month-with-flattr-foss/

    I suppose many ideas are published on Maemo Brainstorm?

    If I merely read Planet (Planet Ubuntu, Planet FreeBSD, …) I already see tons of interesting tidbits or projects which don’t make the regular news. Great way to waste time on quality blog posts.

    Anyway, heres 2 simple rules 1) learn from your competitors where possible 2) do your own thing. Rule 2 is the main one, but before you do your own thing you must follow Rule 1. Rule 1 doesn’t mean you copy, it means you learn (and may copy in some % from 0 to 100%).

    Now, given Rule 2, open source mostly is community-driven and we have to work from that. The cathedral-based is corporation specific, and in cases such as Microsoft a totally different, proprietary OS.

    So, I am not saying there is nothing we can learn from Microsoft at all. I just don’t see what we can learn from them, and like in sports (competition) you have to start from within your own power and limit.

    However, MeeGo is hybrid proprietary/open source. The platform tries to cater to both (like Android, I suppose). If Flattr platform is for open source software and something akin to AppStore/Ovi is for proprietary software I’d say that, for the particular payment aspect, you serve _both_ worlds best. Same like APT + Ovi for the package management/platform.

    • My main focus with this article was on developer outreach. MeeGo needs to steal developers. Commercial enterprises like Microsoft have found effective means of bribery to rope in developers… what will MeeGo (partners) do? What incentives can be provided to hook developers into attending events and then following up by producing code for MeeGo? Certainly covered travel expense is one carrot to dangle, but that isn’t so effective for regional events– which I believe MeeGo (partners) will need to hold. But yeah, a solid micropayment infrastructure can help, as we have been pushing with the MeeGo User Engagement Framework project.

      Anyway just the structure and governance model of MeeGo makes developer outreach… intreresting to say the least…

      EDIT: free Qt certified developer exams would be one easy incentive.

  8. Well, with the MeeGo Conference 2010 now behind us I can safely say that the MeeGo partners just might be getting it. 😉

  9. Having been through a whole day at MIX11, looks like Microsoft still has it!

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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