Dell’s project Sputnik was formed to create a Linux-based developer-focused distro that would run on specific hardware. Over the last 7 months Dell have been working to build a final product with developers and component suppliers using the Dell XPS 13 Ultrabook. Developer edition models went out in July.
There have always been thoughts that a real product would hit Dell’s product range but that’s been confirmed now by project manager Barton George.
We’ve only reported on three Linux-related Ultrabooks since we started Ultrabooknews. We highlighted the Novatech nfinity as a potential candidate as it was available without an OS. Then there was the Dell Sputnik project which is supposed to have been offering a product in ‘fall’ but hasn’t delivered yet. Finally there’s the complete off-the shelf solution being offered by ZaReason UltraLap 430 – an OEM design that we’re seeing under quite a few brands now. Arstechnica have gone hands-on with it and put together a review. They seem quite happy.
It’s not clear whether a Linux-based ultrathin with Ultrabook hardware specifications can be called an Ultrabook. Probably not because ZaReason don’t mention Ultrabook at all in their information on the Ultralap 430 – a Linux-based laptop built on Ultrabook hardware that’s now available. The Ultrabook name is not going to matter to this audience though.
We’re big supporters of what Dell and Canonical are doing with project Sputnik. It’s not just another Ubuntu-based Linux distro, it’s a Linux distro with funding and a defined hardware target and the fact that the hardware target is an Ultrabook makes it even better. If you’re a Linux fan, check out our first post on the subject and then, read-on here…
Dell have just kicked off a project with Ubuntu Canonical. Project Sputnik is a 6-month effort to try to pull together a solid Ubuntu build on a solid laptop, for developers. The Dell XPS13 has been chosen as the first target platform.
We’re constantly getting questions from developers about which Ultrabook to buy but the problems is the word ‘developers.’ I’m a developer myself but the only tool I use is VI as I hack my PHP and HTML for the Ultrabooknews product database. Other, more serious, developers need source code control, collaboration tools, compilers and integrated development environments. Project Sputnik is aiming to deliver a standard Ubuntu build (currently based on 12.04) and additional, downloadable profiles. So if you’re developing for the web, you might add in a web-focused profile. If you’re developing for Android, there could be a better profile for you.
The Open Pandora project has been an amazing project to follow. We first reported on the product, an OMAP-based Linux mini-laptop primarily focused on gaming emulation (but kitted out with some interesting features for anyone interesting in mobile computing) in Dec 2007 and in the last 4 years the product has gone through some amazing ups and downs with spec changes, production issues and community financing but it looks like they’ve made a break-through and that Open Pandoras will be shipping soon.
Update: The first devices off the production line are now being shipped. [9th March 2012]
The story would make a great book. We saw an update in Dec 2008 showing a prototype build and an Angstrom OS build and after a year of refinements it finally went into limited production in May 2010. 4000 units should have been produced before Feb 2011 but it didn’t happen.Â â€œ[The production company] communication has been terrible, the missed all the deadlines they set themselves and they have a failure rate of at least 25%.â€
On 12 July 2011 ‘EvilDragon’ the lead developer for the project wrote a post entitled ‘A fresh new start’ explaining how production was being stopped as a result of problems with the Texas-based production company. The search started for a new production company and by 27th of the same month they had found a candidate. Soon after, 70 investors had stepped forward and pledged nearly half a million Euros. Contracts were prepared and pre-orders started again.
The next months updates are worth reading in full over at the Openpandora news forum. There’s snow, hacking, sad news about a community member, delivery problems, contracts and more. It’s an amazing story that ends up with this fantastic post and video entitled â€œ100% success.â€
What a joy to watch.
The OpenPandora story isn’t over yet though. Mass production is due to start next month and after 4 years of waiting, the specifications don’t look as good as they used to. There’s software to write too. What you’ve got here though is an open-source, very efficientÂ handheld PC with a strong community behind it. It’s also a bit of history.
In support of the Open Pandora project I’ve put in an order andÂ I’ll do my best to give it airtime on UMPCPortal when it arrives. You can place a pre-order here. I’m sure there are many readers here who already have their orders in.
When I read Lukes request for mobile developer help I was impressed. This is Chippys Customer Challenge 2011 #3
I recently went to a local development meetup, and found myself content to use my phone during a ‘Hack Night’ to remotely access my desktop at home to type, and then browse using the phone’s browser. My goal would be to try to find a device that is ultra portable, but that would be a little more easy to do some of my regular development tasks on. It seems like there are a lot of options out there that might fit the bill, but I don’t know if people are really doing much development on more mobile devices rather than just laptops or netbooks.
It takes some patience, good eyes and even a little courage to mobile development work over a remote desktop solution on such a small screen. Luke is now looking for a solution that he can also use as a portable web server. I’m assuming that he also wants to do local development work too. It makes sense as relying on multiple network connections and ISPs to get to your work machine is a little wobbly! CPU and screen requirements can be quite high for software development but there might be a balance that can be done between local input and remote processing.
Luke mentioned Ubuntu which immediately cuts out a whole section of UMPCs from the toplist. Z-series Atom devices have never been well supported by Linux distributions although I know there are some workarounds with Ubuntu. Maybe it’s even baked-in by now. Can anyone comment on that?
Here’s what I’m thinking would suit Luke:
5-7â€ screen on X86 architecture at under 800gm. At least two USB ports and a VGA port. Local storage requirements relatively low. Touchscreen not really required. Tablet format OK with an external USB or BT keyboard.
The 5-7â€ category X86 category is non-existent right now and for value-for money, could you really beat something like the 900gm Asus Eee PC X101 with Meego for 170 Euro? It makes decisions really really hard.Â Mobility is the driver here though so where do we go with this? The good old, possibly beast ultra mobile PC ever, Samsung Q1 Ultra Premium. If you can find it, go for it Luke!
Other options you might consider are some of the newer, lightweight tablets. The AMD-based MSI Windpad 110 or Iconia Tab W500 for example. 10â€, yes, but well under 1KG.
Also keep an eye on new Oaktrail models running Meego. That might happen in the fourth quarter and would indicate a stable status for Linux on the Z-series CPUs. It’s a Fedora-based solution but I’m sure the GPU support will find it’s way over to other distro’s soon.
Luke is not the only one looking at ultra-mobile Linux. I know a few other people that are struggling with this too. If you wait until the end of September we should have some more information on the dual-core Cedar-Trail devices. Tablets based on these should be lighter and smaller than the Pine-Trail equivalents like the Gigabyte S1080, a N570-based device at 900gm for about 550 Euro here. Having said that, Oaktrail devices like the Viewpad 10 Pro, Lenovo Ideapad P1 and Fujitsu Q550 (which I see in Germany for just over 600 Euro now) are worth looking at if weight and battery life is more important. With both of these solutions though you’re looking at Power-VR graphics. I worry about Linux support for that.
There’s one other device you might consider. The Toshiba Libretto W100. Is the dual-screen device based on a dual-core U5400 CPU. It’s got standard Intel graphics and should work well as a web server and even for compiling due to its relatively powerful CPU. It can be noisy, warm and, lets be honest, very short on battery life (2hrs) but it’s actually the only ‘current’, small-form-factor ultra mobile PC on the market that isn’t running a Z-series CPU. Price: Around 700 Euros with UMTS (I’m looking at a European price here.) I wouldn’t mind one of these myself for a bit of mobile 480p video editing and hotel work.
As for older devices, the previously mentioned Samsung Q1 Ultra Premium is the one to keep an eye out for.Â It’s still holding it’s own thanks to some fantastic engineering from Samsung. There are a few VIA-C7-based devices out there too. They’re cheap and not too powerful (think low-end single-core Atom) and can be a little on the warm side but at least Linux support would be easy.
My choice? If I just wanted a lightweight Linux device for light duties, I’d probably be looking at the Asus Eee PC X101. 920gm, basic specs, small SSD storage and an unbeatable price of 169 Euro here in Europe.
Anyone out there already doing mobile development work on Linux with a UMPC? Let’s hear your opinion.
Today’s announcements by Nokia (there are many to sort through) have shocked a lot of people. The major focus here is that Nokia will now use Microsoft (Windows Phone 7) as their primary platform for smartphones. I want to emphasise that this is a revenue generating strategy. It doesn’t include disruptive computing devices which indicates the removal of risk elements within Nokias strategy.Â Symbian gets turned into a ‘franchise’ platform (cheap, stable and, probably, with less focus on corporate support.) Important for the financials is that R&D spend drops. Symbian â€“related spend drops away completely. MeeGo will get hit very hard here too. Whatever way you look at it, near-term investment in MeeGo from Nokia will drop.
This slide says it all.
The message is clear. MeeGo isn’t ready to be used for a smartphone platform in Nokias portfolio. Perhaps if Nokia had continued with Maemo, it would be ready now? Other potential partners in the MeeGo ecosystem will take note of the money spent on R&D by Nokia during this partnership and will look to see what Nokia develop over the next 12 months. Adding to the financial hit, this knocks confidence levels in MeeGo.
MeeGo remains in Nokias strategy but the message we see is that it will be used to experiment with the next generation of disruptive products. Open-source is gone from Nokia’s revenue-generating strategy. We’ve heard nothing about an expansion into tablets, smart-books or other non-phone devices so clearly, this indicates that either Nokia don’t want the financial markets to speculate about this or that they really don’t have a strategy at all here. Nokia have re-affirmed their commitment to delivering a Meego ‘Device’ this year and we suspect that this is an Intel-related commitment for a tablet in the 5-7â€ range to match focus on mobility, clear separation from WP7 devices and to match Intel’s Moorestown platform design limitations. Other MeeGo development work including chipset and industrial design (wait for it, this bit will hurt MeeGo fans) will be ‘repurposed’ in Windows phones.
Where does that leave MeeGo?
The Linux Foundation own the MeeGo brand, take care of the contributions and offer it out as an open-source solution. That hasn’t changed. Linaro, the ARM-focused organisation that can assist ARM product designers to match MeeGo to specific ARM-based platforms is still there. Nokia are still contributing. Intel are still contributing. Intel are still building platforms and services for MeeGo. MeeGo remains one of the best cross-product solutions based on Linux and is the only solution that includes dedicated hardware, development environment and (if AppUp for MeeGo launches at MWC as we expect) applications store. It is still the ‘complete stack’ solution I mentioned last week. What does happen is that Nokia now can’t be relied on as someone that will put a strong brand on a range of MeeGo products. Intel lost a launch partner.
Where does that leave Qt?
Qt will not be used on Windows Phone 7 devices. Without a doubt it waters down the proposition of developing for Qt and as a result, for MeeGo. Todays announcements reduces the potential of Qt to attract developers. On the plus-side, it probably removes OVI as a competing application store leaving Intel to focus on AppUp as the primary application store for MeeGo. A lack of direction for Qt is probably the most significant issue for MeeGo now.
Intel â€œremain committed â€œ
We asked Intel for a statement and we got this.
While we are disappointed with Nokia’s decision, Intel is not blinking on MeeGo. We remain committed and welcome Nokia’s continued contribution to MeeGo open source.
Our strategy has always been to provide choice when it comes to operating systems. MeeGo is one of those choices. We support a port of choice strategy that includes Windows, Android, and MeeGo. This is not changing.
Right now, Intel need to secure some significant product partners for MeeGo, Moorestown and Medfield and to shore-up the development ecosystem by pulling together partners that will also use Qt. Qt is now the burning platform which means AppUp on MeeGo is at risk too.
MWC starts in just a few days and we expect this to be a huge software event for Intel. MeeGo, Appup, IADP, AppLabs and other activities are being showcased. Intel, more than ever, need to use MWC to announce partners.
Stay tuned to Carrypad and we continue to follow this important story over the next week.
Linpus, a company that has been working on Linux distributions for netbooks for a number of years under their ‘Linpus Lite’ brand have been invited to MWC to demonstrate their current offering in the MeeGo and Qt booths. The solution is targeted at manufacturers of netbooks and tablets based on MeeGo. Like MeeGo, the Linpus solution will be a ‘base’ on which to build on through contractual work by Linpus. We’ve seen a tablet UI before but this is more than that.
Unfortunately, the target device, a Lenovo S10-3T is out of the studio right now so I couldn’t get everything up and working well enough for a full video demo but I did boot it on the Tegatech Tega V2 tablet where I was able to test the UI and apps, a key part of the Linpus solution.
Note that this is a presentation suite rather than a full, polished OS but it gives you an idea of the sort of work the OS distributors (aka Operating Systems Vendors, OSVs) are prepared to put in with MeeGo. You’ll see an on-screen keyboard, an active desktop and a glimpse of the applications that Linpus have developed for the solution. It’s far from customer-ready but it’s worlds apart from what you’ve seen on MeeGo builds so far.
The video below is also available in 480p resolution.
It there is one thing we know about JKK of JKKmobile, it’s his uncanny ability to cover huge amounts of distance at technology shows and exhibitions to produce first-look videos, by the bucket load so when he posts the following about Amtek prepping not 1 or 2 tablets but an incredible 5, we know we are in for a interesting Computex;
Firstly there is the iTablet Speed-Lite (AE03), packing a Nvidia Tegra 2 1Ghz, 1GB RAM, 2MP camera, 10.1â€ multi touch display and WiFi. Then there is the iTablet Ex-Lite II (AE04), wielding a Freescale 800MHz CPU, 512MB RAM, 2MP camera, 10.1â€ multi touch display and WiFi also. Both of these come with a choice of either WindowsCE 6.0R3/7.0, Android 2.0 or Linux and have options for bluetooth, 3G connectivity and GPS.
To add to these is the iTablet Lite (TZ10), sporting an IntelÂ® Atom Menlow-XL Z530 1.6GHz CPU, 2GB of RAM, 10.1â€ multi touch display, 2MP camera, WiFi, 4200mAh battery, Windows 7 and a choice of either a 1.8â€ IDE HDD or a 64GB SSD.
Then the icing on the tablet cake in the form of T23A CULV tablet PC and the T23x CULV Series. All of which have Core 2 Duo 743 1.30GHz Intel Montevina Platform CPU’s, 2GB of RAM, 1.3MP cameras, 2.5â€ SATA HDD’s from 160GB to 320GB and coming with a choice of either a 12.1â€ XGA digitizer, touch or digitizer and touch screens.
Brad at Liliputing has put up a first-look video of Meego 1.0 for netbooks. Clearly the UI is based on Meego as is still using clutter instead of Qt and there don’t seem to be many major changes apart from Chrome being used as the browser. Brad does report that it’s fast though
As I write, i’m installing to an MSI Wind and will do some testing. I’ll also test the image out on a Menlow device. I doubt it will work but let’s see!
Bringing a computing device to the market is, quite frankly, a nightmare. 90% of the work is in the last 10% of the detail and timescales will always slip. For the first-timer, the success rate must be pretty low. For a community project, the success rate must be even lower so we’re truly happy to see that the Open Pandora project is just a few steps away from the finishing line. (Actually, a new starting line!)
Production of the cases was due to start in time for Christmas shipping but a hiccup on the case has held things up. Fortunately, some production sample units are out there and the videos keep rolling in.
The Open Pandora product is aimed at hobbyists and hackers who want a relatively high-powered product for gaming-focused software projects. The device is just 140mm wide but contains a relatively powerful TI OMAP 3 board inside. That’s as good as the best ARM-based devices on the market right now!