Markus Thielmann's blog

CrunchPad announced to be dead

As TechCrunchs Michael Arrington reports, the CrunchPad is dead.

It was aimed to be a low-cost tablet PC, based on open source software. Since their Goal was to offer it for around 200$, it might have been a really successful gadget.

I'm still not convinced that this decision is final. It seems there are some serious issues about intellectual property rights and the distribution of the profits. So putting CrunchPad to death is a lose-lose situation. If you include possible customers, it's actually a lose-lose-lose situation.

Let's hope TechCrunch and Fusion Garage can sort things out.

RepRap

RepRap is an open source 3D "printer", which aims to become a full replicating machine.

Even though this idea might take another lifetime to succeed, I really like the idea of downloading (and printing) spare parts instead of ordering them.

It's also a great reminder on how much more important intellectual property will become in the future.

The future of social networking: Webfinger

Since a number of years, the so called "social networking" is dominated by a small number of service providers like Facebook or LinkedIn.

This centralized approach does have some serious problems, especially when it comes to information security and information protection. Beyond that, centralization constricts competition to very few providers, since it requires a certain amount of possible contacts using the same platform in order to be successful.

Webfinger: first step towards decentralization

In my opinion, Webfinger is currently the most underrated new technology.

Webfinger provides a simple way to attach information on additional data and services to an e-mail like identifier, like mail@markusthielmann.com. While this identifier does not necessarily need to be a working e-mail address, having the same identifier certainly has a few advantages.

Using Webfinger, our personal information manager application would only need a simple query, to fetch additional information on how to reach me via Wave, SIP (VoIP), XMPP (IM) or just get my profiles on various social networking sites.

The possibilities are endless, which makes Webfinger a great starting point for distributed social networking.

Who is behind webfinger?

Webfingers list of contributers is actually quite impressive:

Brad Fitzpatrick, Google Engineer, Founder of Livejournal and former Six Apart Chief Architekt.

Eran Hammer-Lahav, Yahoo Open Source Evangelist and editor for the core OAuth specification.

Chris Messina, well-known proponent of microformats and OAuth.

Blaine Cook, former lead developer of Twitter, principal co-author of the OAuth specification.

John Panzer, Tech Lead Manager at Google.

Joseph Smarr, Chief Technology Officer at Plaxo.

Google and Yahoo alreay implemented Webfinger for their e-mail services. If you own a @gmail.com, @googlemail.com or @yahoo.com e-mail address, a rudimentary Webfinger implementation already allows people to query your profile for more information.

Distributed networking to the rescue

Currently, the single most successful social networking and communication application is e-mail. One cornerstone of its success is its distributed character and its wide acceptance. I guess it's safe to assume, that nearly every Internet user has access to at least one e-mail account.

Technically, there's no reason why other social networking services (like activity streams) would require a centralized approach. With OpenID and OAuth, there are already established protocols for authentication and authorization. Google even combined both protocols as an OAuth extension to the OpenID standard in order to reduce complexity.

Building upon Webfingers glory and using well defined open standards like FOAF, OpenSocial, activity streams, Atom Activity, Portable Contacts or FeedSync, there's a great opportunity of actually replacing centralized social networking providers.

Project drivers

Currently, neither Google or Yahoo own a successful social networking provider. Googles Orkut flopped for most of the world, Yahoo! Mash didn't even leave beta status. Both company's would profit from an open, distributed social network based on e-mail like identifiers, since profile information could be easily integrated into their search products (like Google already did with their Social Graph API). Additional to that, it would make their e-mail products much more attractive.

There's a chance, that a lot more e-mail providers could jump on to this idea. Germany's biggest personal e-mail provider united internet already failed launching a social network named "UNDDU. They however already support Jabber/XMPP and might be interested in offering additional services to their customers.

Conclusion

Decentralized social networking has a lot of advantages:

  • Full control over the visability of your personal data
  • No need for joining multiple social networks in order to connect to your contacts
  • Ability to choose your identifier
  • Simple access to additional identities (e.g. private and business related identities

... and many more. I'd really love to see Webfinger based decentralized social networking within the next year. And chances aren't too bad this might actually happen.

Low cost and portable plug computing

A few months ago, Marvell introduced plug computing. In essence, plug computers provide services to your home network, with a very low power profile. While this is a great idea from a developer point of few, it's not that easy to sell the idea to consumers. Most people still think of IT in terms of "on" and "off".

So I'd love to see the idea of plug computing expanded to the desktop world.

ARM recently stated their interest in capturing the market for $200 netbooks. One keystone of this goal is Canonicals commitment to bring Ubuntu to ARM based computers. Here's an Computex 2009 interview with Bob Morris, Director of Mobile Computing at ARM Holdings, who talks about Canonical's success with porting Ubuntu 9.04 to the ARM infrastructure.

Here's Chris Kenyon, Director of OEM Services at Canonical talking about the Canonical ARM partnership:

While there is a growing demand for netbooks, it's not just the portability which attracts customers. It's also the low price for computers, capable of doing most of every-days work: internet access, office applications and entertainment. With rising energy costs and a growing ecological consciousness, energy consumption will be an important factor for future investments. With Zii's announcement of the ZMS-05 based Egg handheld, I recently noticed how far mobile computing has come. It's not only a few hundred times faster than my first PC (a Schneider PC 1512 DD), they even close in on the performance of my current computer. To be precise: In difference to the Egg, my computer isn't even capable of offering 1080p HD without chopping.

So why shouldn't a customer be able to use a handheld computer for home computing?


Image based upon Flat_monitor.svg (CC-BY-SA 3.0) and Multimedia-player-iphone.svg (CC-BY-SA 3.0, 2.5 and 1). Feel free to use it accordingly.

All it takes from my point of view, is a (preferably industry standard) docking connector and a monitor with some i/o modules integrated (basically just USB, Ethernet and HDMI/DVI). After docking your PDA/smartphone to your monitor, your mobile device would just start a desktop environment. There wouldn't even be a reason to reboot the device, since I don't see a reason why Android and Ubuntu shouldn't be able to use the same kernel. If you need more hard disc space, just add an USB-drive to the monitor. You would be able to use the handheld display for additional information, like your RSS feeds. This would also work for your TV, just add a USB DVB-x device and a USB-harddisk to the TV and use your handheld as a media center. If you visit friends for a movie night, no need for a handful of DVDs. Just bring your mobile.

There's one catch: You would most certainly need more RAM for desktop usage than a typical "cutting edge" handheld has to offer. This isn't primarily a matter of costs, but of energy consumption. Luckily, there are already ideas on energy efficient RAM handling for mobile devices.

With falling prices for mobile equipment and subsidies from the phone network operators, this would enable more people than ever to use modern technology on their desktops. There's even a sane upgrade path for consumers with existing monitors and TVs: Just offer a stand-alone docking station. Since it would be basically a I/O-Board with a case, prices could be as low as $10-$20.

Since Ubuntu is capable of running Android applications, this might make Android even more interesting for game developers. Which would solve the problem of having only a few commercial grade games available on Linux.

Zii unveils the Zii Egg

In the last months, Zii gained a lot of interest for their StemCell computing announcement. Basically, the concept of the ZMS-05 chip sounds like a FPGA, combined with some sort of dynamic frequency scaling. Which is still not a good enough explanation, since they don't need to change frequency (just switching parts of the processor "off") and the reprogramming of the chip takes milliseconds, not seconds.

Since this all sounds to good to be true, most comments on these claims are quite reserved.

Today, Zii introduced the Zii Egg.

I'm quite sold: A handheld offering full HD resolution to a TV? How great is that? There are still a lot of questions open, one of them is a real world check on their energy efficiency claims.

If half of their claims are true, I'm quite convinced that this will change the whole chip industry. Not only for handhelds, but all kinds of mobile equipment, desktop computers and servers. Having a ultra scalable architecture, able to hard-wire tasks on the fly and just switch off if there's nothing to do still sounds to good to be true.

Let's see how this works out. If you see Intel offering a takeover or Microsoft starting to port Windows to the ZMS-O5, the claims might be true.

Syndicate content