30.6%

I haven't seen the December issue of Linux Journal myself (yet!), but according to this blog entry KDE's desktop offering took 30.6% of the votes. That is twice the votes the runner up received which is a significant show of support from the Linux community. Thanks to everyone who voted, and congratulations to everyone who has helped make KDE's Plasma Desktop a reliable, performant and useful solution.

We are a diverse bunch in Free software and there are usually more than just a few options available to us. The multiplicity of options brings several benefits: competition, lessened risk due to single points of failure and an ability to focus on different use cases without devolving into lowest common denominator software to name three. However, sometimes there is more diversity than the user or developer base can support successfully. In such cases, the diversity prevents any one solution from gaining critical mass limiting the quality and capabilities of the software.

On the desktop side, having seen a variety of reliable industry numbers (not all of which were public, sadly) it was my understanding that Free software desktops were growing at a significant pace in the first half-decade of this century.  Since then, we've hit a stall point. There still is growth happening, but it has not been even. There are many reasons for this, the biggest of which are unrelated to the diversity we see in Free software desktop environments. 

However, the lack of a clear mandate from the users of Free software on the desktop has prevented any one desktop environment achieving the critical mass that would help to push through the ceiling we currently are pressed up against. Just how fractured support is can be seen clearly from the Linux Journal survey, which lines up quite neatly against every other survey I've seen in the last year.

During the time when there was more significant market growth for the Free software desktop, KDE software usually polled at 60%+ of the user base. The biggest shift in these numbers happened during the KDE 3.x desktop releases, well prior to the 4.x releases. Since that time, the numbers for KDE's desktop environment has remained generally constant. There was a dip right after the 4.0 release for reasons that have be well covered, but that was a set of issues which the community addressed on the way to realizing the vision for KDE Plasma. For for the last several years we've been idling in that 30-40% range of support.

This raises in my mind what is probably the obvious question: if KDE were to want to build on that 30%+ support shown by Linux Journal readers, what could we do to achieve that?

That is not a glib question that can be answered with "well, if someone just fixed bug X" type simplifications or  tired (and largely debunked) criticisms (e.g. "developers don't listen to users", "KDE software is 'bloated' relative to other options"). The answers (plural) will most likely lie within the social dynamics of the Free software user community rather than individualized wishes and guessing games. They are likely to not even be technological; I suspect most of the answers are social and industry related.

In closing, 30.6% is a terrific showing and shows KDE's central role in Free software desktop environments. But let's not stop there; let's us that as inspiration to go further.

Read More

Improv and KDE

When I announced the Improv ARM computer on Monday, I did it on my blog which is also syndicated to Planet KDE. That's because there is a very significant link between Improv and KDE.

People are shifting their computing time from traditional large laptops and workstation desktop systems across a breadth of devices. Not only are people using tablets and phones more, they are also using online services more and more.

For that least several years this trend has weighed on my mind in relation to desktop oriented Free software and, in particular, KDE software. How can we ensure KDE software can be where people are using computers today and tomorrow? I'll call this the Big Question.

The answer is not an easy one, in that it's actually a set of answers and it implies work for all of us. Some project teams within KDE, such as Plasma, Calligra, Marble and Kontact (among others) have been starting on addressing the Big Question. We've been creating software that can scale down to smaller hardware and creating user interfaces that work with various interaction patterns, such as touch, in addition to our traditional focus on the WIMP paradigm. Some have also targetted platforms and operating systems that are fairly foreign territory to be able to access the audience that use them; the port of Qt to Android can be seen as an effort in this line.

This is a lot harder, however, without hardware in your hands that you can easily work with to test your software on. So the Big Question leads to the Small Question: What hackable ARM-based hardware exists that supports KDE software out of the box?

Until Improv, the answer was simple: none.

Improv will ship to people with KDE libraries as well as Plasma Workspaces included. We will support that software, take bug reports, issue updates, etc. Community support is something we are fostering even now as well. We have an open build service farm that we've been using for quite a while now to create KDE packages for Mer OS for both ARM and Intel targets. There's even a Mer SDK you can install locally to get started on image creation if that's what you're interested in doing.

So if you'd like to be able to see how your software runs on a small ARM device; if you'd like to see how thin clients accessing your applications might perform; if you'd like to experiment with alternative form factors .. Improv is there for us.

Being able to offer a full computer for $75 that can run KDE software to an acceptable level, and in the process give a good impression of it to those using it, could be invaluable in promoting KDE technology to students and makers alike.

Improv is also great for non-GUI development. It makes an awesome whisper-quiet, low power server for the home or small office. These kinds of services, properly integrated with rich client software, is an area which will become increasingly important to the future of KDE and other Free software desktop software.

Improv is a product that can open the doors to the world of ubiquitous, device-centric computing for KDE. No more waiting for a big vendor to be kind to us and take our needs into consideration, no more buying devices never intended to run KDE software and try and shoehorn it in.

This is why Improv should be of prime interest to, and receive the support, of the KDE community and those who care about the success of it.

Read More

Introducing Improv

Improv: the powerful, open hardware development board

Make·Play·Live is happy to share some great news with you today: our first hardware product, Improv, is available for order today. As the name suggests, Improv is all about making new things. It is the perfect board for prototyping and creating small, powerful devices.

A combination of three attributes sets Improv apart from the crowd:

1. Power: Dual core CPU, lots of storage, powerful GPU and modern software
2. Modularity: Improv is actually two plug-and-play parts: a CPU card and a feature board
3. Community: The feature board is freely licensed as Open Hardware, is supported with community infrastructure and contributes to the technologies on which Improv is based.

In broad strokes, this is what makes Improv unique and exciting. It is the perfect starting point for your hardware projects.

A view of Improv  from the business end of the card

Power

The hardware of Improv is extremely capable: a dual-core ARM® Cortex™-A7 System on Chip (SoC) running at 1Ghz, 1 GB of RAM, 4 GB of on-board NAND flash and a powerful OpenGL ES GPU. To access all of this hardware goodness there are a variety of ports: 2 USB2 ports (one fullsize host, one micro OTG), SD card reader, HDMI, ethernet (10/100, though the feature card has a Gigabit connector; more on that below), SATA, i2c, VGA/TTL and 8 GPIO pins. The entire device weighs less than 100 grams, is passively cooled and fits in your hand.

Improv comes pre-installed with Mer OS, sporting a recent Linux kernel, systemd and a wide variety of software tools. By default it boots into console, so if you are making a headless device you needn't worry about extra overhead running that you don't need. If you are going to hook it up to a screen (or two), then you have an amazing starting point with choices such as X.org, Wayland, Qt4, Qt5 and a full complement of KDE libraries and Plasma Workspaces.

Improv unplugged undocked

Modularity

Improv takes advantage of the open EOMA68 standard to deliver a unique design: the SoC, RAM and storage live on one card (the "CPU card"), the feature ports are on a PCB it docks with (the "feature board"). The two dock securely together with the CPU card sitting under the feature board nestled in a pair of rails; they are undocked from each other by pushing a mechanical ejector button.

This means that you can use the CPU card separately from the feature board, have multiple CPU cards for different projects or even upgrade your Improv as new CPU and feature boards become available. Not only is this better for the environment, but it gives you ultimate flexibility.

The software is similarly easy to bend to your will. You can boot into the included Linux-based operating system and decide whether to stay at console or fire up a full OpenGL accelerated graphical environment. Or you can choose to boot something else entirely from the SD card slot, via the USB OTG port or by flashing a new bootloader and/or OS image to the built-in NAND.

Additional hardware add-ons such as VGA connectors, keyboard kits and cases are also in development and rely on the openness and modularity of Improv. Through its modular design, Improv is designed to last and grow.

More than just a product, Improv promotes community

Community

Improv is the result of a community effort. Collaboration within the EOMA68, Linux netbook and ARM communities helped to deliver and prove the hardware design, while working with software communities such as Mer, KDE and Linux have allowed us to deliver amazing software on top of that hardware.

Production, branding, marketing and retail are all enabled by collaborative interaction within Make·Play·Live's Partner Network. Via this network, we bring together individual community participants as well as entrepreneurial and corporate support, without which Improv would not be possible.

The schematics for the feature board are licensed under the GPL and will be made available in tandem with shipping so that you can learn about the design and even extend it in new ways. We are also ready to help motivated makers who come up with great ideas for new feature boards, add-ons and entire devices through the process of prototyping, manufacturing and delivery. This is not a device you buy and are then left on your own with. Improv is a starting point from which to create amazing things that can take flight.

To support you in this, we provide sleek online discussion forums, the Open Hardware Registry (OHR) and the Mer Community Open Build Service (COBS). The forums are administered by Make·Play·Live community members and provide a place for people to discuss projects, ask questions and share answers. OHR allows makers and vendors to register globally unique IDs for their hardware projects, much as the USB-IF does for USB, only without the fees, bureaucracy and limited ID space. The Mer COBS, which is jointly sponsored by Make·Play·Live and Jolla, provides an effective way to build and distribute software for Improv.

This is only the start of the community infrastructure to support your needs as makers. Videos on Youtube showing tricks, techniques and project ideas for Improv will appear weekly. These will feature people who were key in the development of Improv; in time, we will also feature videos highlighting community contributions. Future plans include infrastructure to help you share your hardware projects with others, a community driven knowledge-base and local user groups supported with project concepts and planning.

Creating sustainable models

One of the biggest challenges we have identified in the free software and open hardware movements when it comes to consumer and mobile devices is sustainability. Hardware costs serious money to produce and involves engaging with a supply chain that can be challenging to work with. We know, because we have spent the last two years working on some ambitious open hardware projects.

From that experience we have built a network of hardware, software, procurement, manufacturing and branding expertise. Starting with Improv, we are opening up that expertise to others who would like to turn ideas into reality.

Improv is also an opportunity for us to invest resources into relevant free software and open hardware development. The Mer Open Build Service, for instance, will be partially funded by proceeds from Improv. Software developers and open hardware designers working on similar projects will receive continued sponsorship and funding thanks to sales of Improv. Ultimately, we expect new products to spring up as others use Improv, continuing the cycle of building sustainable models in the open device world.

When you purchase an Improv you are not only getting a truly great device to work on, you are also supporting the process of creating quality free software, open hardware and the services which will enable others to do the same.

Elegance, freedom and power in one package.

"Shut up and take my money!"

Improv retails for US$75 plus applicable shipping and taxes. That includes the feature board, the CPU card and a commitment to engaging and fun makery from us to you.

We will start shipping to North America and Europe near the end of January on a first-ordered, first-shipped basis. Needless to say, each manufacturing run produces a limited number of Improvs and we expect the first lot to sell out quickly. To order your Improv today and get in on the first public shipment of devices, visit the Improv information page and click on the purchase button.

Read More

Kate has won

A couple of weeks ago Martin Gräßlin noted that I use vim and not Kate in a G+ discussion. Suitable shown up, II decided to give Kate another serious run. I do this every couple years, and it has indeed gotten better every time I've used it, but ... you have to understand that I'm fighting a long term addiction to vim here ;)

So I checked myself into editor rehab and forced myself to use Kate on two projects recently. I have to admit that I'm a convert. A lot of the keyboard controls I'm used to from vim translate over to Kate; if I turn on the vi mode in Kate it gets even better.

What really sold me though were two things: sessions, and the built in command lines (plural).

Kate lets you give am editing session a name. You can open and close sessions, and Kate tracks the files you have open, etc. There is also a Plasma widget and krunner integration, so I can now hit alt+f2 and type "kate" to see all my saved sessions. With autocompletion, of course! So: alt+f2, "kate mpl", , bam! right directory, 30+ files open. Beauty.

The other must-have for me are the command lines. I live and die by the terminal, so having one around is crucial. Kate allows having a Konsole session right in the editor window. This is collapsable, so it doesn't waste space when i don't need it, but is always right there when I do. The killer feature is that it follows me around as I move between files in the GUI. I can do all my git foo, scp'ing, grep/find'ing, etc. in that window, of course, but perhaps the best thing is that `kate ` opens the file in the kate window, adding it to the session. My workflow is preserved!

There is also another command line in Kate: the Kate command line. This lets you tell Kate to do things with nifty little commands. As a vimmer, this feels so very, very natural. :)

A lot of things I'm used to doing quickly in vim can be done quickly in Kate. It's not 100%, but then Kate is better at some things compared to the default vim, too. For instance, Kate has the filebrowser sidebar with all the usual KDE goodness such as breadcrumbs, autocomplete, etc etc. Kate is fast, scriptable, detects files changed behind its back, autorecovers if your X session dies (not that that ever happens to me when working on things like the lock screen with X screensavers *cough* *cough*). It also has those nifty over scrollbars and various gewgadgetry that is just little sprinkles on the cake.

Not that I won't use vim from time to time still ... (he writes, looking at his various ssh sessions in various konsole tabs...)

Read More

what trains are for

Today I had to go to Milan .. and back .. by train. That's a total of eight hours planted in a moving seat. I won't explain why I had to do this crazy little trip (it's related to our announcement on Monday) but there I was .. at least I had a power plug and a table to myself. Which meant I could catch up on various bits of work, and when I needed a break from that I happily hacked on random Plasma bits.

One thing I did was clean up some issues with the lock screen. The new QML based lock screen would show the password dialog even if you had set it to not require a password. There was also some fun you could have by connecting a second monitor, something I only noticed while reading the code fixing the first issue.

One tester noted a bug in my first run at this, so this evening once back home I spent some time pounding on it in every direction I could. I can no longer get it to misbehave, so that's a good thing.

There's also a patch on reviewboard, thanks to Thomas Lübking, that should get pushed shortly that improves the screenlocker when using multiple screens and the (legacy) X screensavers.

Given the nature of the screenlocker, it would be great if those of you testing pre-releases could also keep an eye on the screenlocker for new unwanted behaviors that may be introduced by these changes. As I said, it works well here and it's been tested by at least one other person already ... but more eyeballs equals fewer regressions.

In any case, that's one more set of issues that are cleared up for the Plasma Workspace 4.11 long term release. The fixes are not streaming in at an alarming rate, but they are trickling in at a nice steady pace, which is the entire point of these things.

Read More

bodega: partners, aggregating audiences and YOU

I did a quick screencast today showing what "partners" are in Bodega and how they work. It's one of the many ways that Bodega is a little bit different from other similar systems and as such a screencast can't hurt to help people understand how it works. I will cover other aspects of Bodega in future, so feedback on what would make these videos more useful to you is welcome.

A primary concept with Bodega is that of audience aggregation. Many of us have audiences, but few of us have big audiences, and even fewer have huge audiences. Unfortunately, the economics of selling content only works with larger audiences and even for those who aren't selling stuff it's often more of an incentive when your audience is larger.

Since Bodega can host all kinds of content and then show only selections of that content, it naturally allows for content aggregation. To understand why imagine that there are three solitaire card games which have configurable backgrounds, card deck graphics and game rules. Each of them uses similar image files to decorate the cards shown on screen, but each of them uses a different file format to describe how the games work. Now imagine that all three of these projects integrate Bodega into their game to allow people to get new card decks, backgrounds and game rules.

In Bodega, people could happily publish backgrounds and card decks. The backgrounds might just be regular ol' wallpapers, or they might be a specific type of image for board game "mats". Other people could publish game rules that are specific to one of the three card games. While the backgrounds and card decks would have a common tag noting what they were, these game rules would be tagged with the applicable game.

A card game developer would then create a store in Bodega that shows all backgrounds, all card decks .. but only the game rules for their game. Each of these could be put into a separate channel in their store so when the person playing Mystery Science Solitaire 3k presses the "Download card decks.." button they see just card decks and not backgrounds or game rules.

Everyone is happy .. but something interesting has happened in the process. Instead of trying to motivate an artist to make card decks for your special card game, they just have to be motivated to make a card deck for any of the card games. Perhaps they'll be more motivated if they know that their work will show up in multiple applications and therefore get in front of more people. Probably, right?

In such a case, the artist doesn't even need to know about Mystery Science Solitaire 3k to make it better, so the app developer also wins.

Now .. forget about card games for a moment and think about entire Linux distributions. Or independent musicians and authors. Or small device makers. Yeah.

Together we can aggregate audiences and increase the sustainability of each of our interests in the process. The key word is together.

We're going to be writing a "integrating a Bodega store into your system" HOWTO and then I'll be making it my mission for Q1 2014 to get as many projects, people and companies doing exactly that. Between now and then, I'm looking for early adopters that we can work more closely with to refine that HOWTO.

If you are involved with a project that you think Bodega integration would be perfect for, please contact me by email at aseigo at kde.org. Your project can be web based, focused on the traditional desktop or mobile; it can be a single application or a bigger system. All I ask is that it has had a public release and has some user base already, even if small.

Read More

#merweek

Make·Play·Live's website is counting down to ... ?

As Dario Freddi noted in his G+ stream today, the week of the 25th is shaping up to be #merweek. There will be not one, not two but three product releases that week featuring Mer OS over the course of the first three days of the week from three different companies, all of whom contribute to Mer OS. It is going to be an exciting, fast, hot week for anyone interested in embedded and mobile devices, whether as a user or a maker. 

We'll be kicking off the festivities on Monday morning at 17:00 UTC (09:00 PST) as we unveil the first Make·Play·Live product, at long last. The only thing I can reveal right now, other than it uses Mer OS by default, is that it is not a tablet. 

For everyone waiting for Vivaldi: We have not discarded the tablet concept. In fact, the final product is frustratingly close to completion but we missed the the holiday season time window. The Vivaldi journey, however, was crucial for informing the product we will be making available on the 25th.

Supporting this announcement will be a brand new Make·Play·Live web site, which currently is counting down the seconds until 17:00 UTC on Monday the 25th. This new site will serve as a visible hub for the business network we've been putting together throughout 2013, a place for those looking to harness the capabilities of the Bodega content system to get hooked in and productive as well as a community gathering and support node for .. well .. you'll find out on the 25th. ;)

The day after that, Ispirata will be releasing a device development system designed for embedded systems needing quick development and long term support which they've christened Hemera. The day after that Jolla has their product release.

It's going to be one hell of a week ... #merweek

Read More