Linux Controversies of 2011: Does Richard Stallman Still Matter?By Sean Michael Kerner | December 29, 2011
From the 'Free as in Freedom' files:
One of controversies that re-emerged (if it in fact had ever actually sub-merged) in 2011, is the relevance of Richard Stallman (RMS), the Father of Free Software.
RMS was the key spark that ignited the fantastic world of FOSS that we enjoy today, he is the man behind the GPL and GNU. His contributions to the early days of Free and Open Source Software are well known and (hopefully) appreciated as well.
However there also has long been a contingent that doesn't agree with RMS or his views. In 2011, they all rose to the surface lambasting him over his two sentence comment about the untimely passing of the late founder of Apple, Steve Jobs.
The relevancy of RMS and his 'fanatical' views was questioned in blogs and editorials big and small, anchored on his comments about Job's demise.
From my own personal perspective, over the course of 2011 I tried at multiple points to see if some of the leaders of the Linux community would similarly say that RMS is no longer relevant.
I asked Jim Zemlin, Executive Director of the Linux Foundation and he wouldn't say anything negative about RMS. I also asked Jim Whitehurst, CEO of Red Hat and got a similar response. Both men still see RMS as a relevant figure and respect the man.
For the record, so do I.
While RMS can seem abrasive at times, regardless of the year (or decade) he has held true to his beliefs. While some were offended at his Jobs comments, others simply noted that's just the way RMS is.
RMS remains the pillar of Free Software and its philosophy. Certainly the popularity of BSD and Apache style licenses have somewhat eroded GPL as the be-all-and-end all for FOSS licenses. Yet, GPL still remains a powerful force and regardless of what some stats vendors might tell you, I suspect it will remain a powerful force for years to come – and with it RMS and his views.
Linux Controversies of 2011: Is Microsoft Out to Get Linux ?By Sean Michael Kerner | December 28, 2011
From the 'Patent FUD' files:
In the early days of Linux the anti-Microsoft controversies and conspiracies were everywhere.
In recent years, Microsoft has come to an accommodation of sorts with open source and Linux. Though the deals that they have made - that accommodation - continued to be a source of significant controversy in 2011.
Back in 2007, Microsoft first alleged that open source software infringes on at least 235 Microsoft patents. That allegation led to multiple companies including Novell (and now SUSE) signing patent deals with Microsoft.
In 2011, Microsoft's patent focus seemed to sharpen on Android. Except for Motorola, Microsoft now has every major Android phone vendor in some kind of deal over intellectual property.
No, Microsoft did not pursue a 'SCOsource' type license going after individual Linux users or even distros. Microsoft has chased the money and gone after big consumer electronics vendors.Their 'SCOsource' is all about the big Android vendors.
Is this still the same battle against Linux from years past?
That was a big controversy in 2011 and will remain one in 2012 (and likely for years to come). Microsoft will continue to protect what it sees as its own intellectual property. The Linux community in turn will continue to ask for proof in instead of just FUD.
Linux Controversies of 2011: Linux Mint is the Most Popular DistroBy Sean Michael Kerner | December 27, 2011
From the 'Top Controversies of 2011' files:
The decline of Ubuntu and the corresponding rise of Linux Mint is one of the bigger Linux controversies of 2011.
Ubuntu's introduction of Unity alienated a non-trivial portion of the Ubuntu user-base which then went looking for a new home – and found it in Linux Mint.
How big was the exodus from Ubuntu to Mint?
That's difficult if not impossible to accurately measure. Plenty of 'experts' liked to highlight the 'fact' that Mint is the top Distro as ranked by Distrowatch. Distrowatch however is not a measure of users or even downloads, it's just a measure of the relative popularity of a given distro's page on the Distrowatch site.
It is likely that Mint has picked up a good number of former Ubuntu users. It is also likely that Mint's popularity surged in 2011. It is not clear and also not entirely likely in my opinion that Mint is the most popular distro in terms of total number of users.
That doesn't mean that Mint isn't a great distro, because it is. Mint developers respect the tens of thousands of Linux desktop users that want and/or need a more traditional GNOME type experience. This is a controversy brought on by the vehement dislike of Unity and the lack of sensitivity from Ubuntu's leadership about its shortcomings.
In 2012, with Mint now moving a new direction with the Cinammon desktop, expect to see more 'Mint is the best' type hoopla in the New Year.
Is Mozilla Really Getting $300 Million from Google?By Sean Michael Kerner | December 22, 2011
From the 'How Much is an Open Source Browser Worth?' files:
There is a report out that claims that the renewed Google Mozilla deal is worth $300 Million per year.
WHAT? Yup I did a double take when I saw that figure too. Sure I'm a believer in Mozilla, but $300 million sure does seem like an obscene amount amount...or does it?
The Google search deal - as I understood it in the past - paid Mozilla certain amount of money on a per search query basis. In 2010, Google paid Mozilla $123 Million so the new deal could be a 3-fold increase.
Neither Google Nor Mozilla have ever really made public exactly how the previous deal works, but back in 2007 I speculated on a formula that neither Google nor Mozilla have ever challenged me on. The way I see it, Google is paying Mozilla as much as 5 cents per query and as little as a penny. With this new deal, I suspect that there is not also a 'lump sum' payment as well that keeps Google as the default search and on the default Firefox start page as well.
It's likely that 'lump sum' piece where the big jump has come in for the revenue forecast - if the original report is accurate.
The only other justification for the higher value deal in pure math terms would be that either Mozilla's users have grown by 3x (they haven't) or they're doing more searches (they're likely not) or the value per query is higher (also not likely).
There is speculation that Microsoft and Yahoo bid for the space too. But hey, we all know that it's simple operation in Firefox to change the default search provider or home page.
For Google, there is no question in my mind that they aren't just paying for queries anymore. They're paying a premium to help protect their own interests and they're paying a premium as a way to help fund Mozilla innovation.
Mozilla's Innovation ultimately makes the web a better place for us all, which is a good thing for Google too.
Another GNOME Fork? Can Cinnamon Survive?By Sean Michael Kerner | December 22, 2011
From the 'RIP MATE' files:
Lots of people didn't (and still don't) like GNOME 3, which has helped the popularity of Linux Mint.
Linux Mint has not adopted the new interface instead going their own route with the MATE fork and the MGSE (Mint GNOME Shell Extensions). Apparently that's not enough and now Mint Founder Clement Lefebvre has launched a new effort to create a new desktop called Cinnamon.
Yes it's aggressive. But it makes more sense (to me) than Mark Shuttleworth escapade to oblivion, the abomination known as Unity. Lefebvre is trying to give users want they want -- a better GNOME 2.x type of experience and not some whacked out interface that is more 'usable' to some unknown group of mystical usability experts.
Now to be accurate, Cinnamon is not an entirely new technology - Lefebvre is smart in that he's still leveraging the power of GNOME.
The README for the Cinnamon project states that:
Cinnamon provides core user interface functions for the GNOME 3 desktop, -like switching to windows and launching applications. Cinnamon takes -advantage of the capabilities of modern graphics hardware and introduces -innovative user interface concepts to provide a visually attractive and -easy to use experience.
Makes sense to me, and hey it will still be compatible with other GNOME apps too. The fact that this is being developed in the open on GitHub (here's the project) is great too. As this matured, I'd sure love to see RPM packages of this for Fedora/openSUSE too.
Cinnamon is what open source is all about - here's a developer and community with an 'itch to scratch' - code is open source and we the users get to choose what we want to run.
Why Firefox 9 Is Not in Red Hat Enterprise LinuxBy Sean Michael Kerner | December 22, 2011
From the 'Rapid Releases Aren't Good For Everyone' files:
Mozilla faced its fair share of criticism this year when they moved to the rapid release cycle. Enterprises argued that they couldn't update as fast as Firefox is being released, with new browsers out every six weeks.
As it turns out, Linux vendor Red Hat isn't updating with every Firefox release either.
Red Hat Enterprise Linux (RHEL) is the industry leader for Linux at this point and they're taking a different approach to updating Firefox. I recently spoke with Tim Burke, Vice-President of Linux Engineering at Red Hat about the Firefox issue.
"Six weeks is just too fast for RHEL, we don't just take projects from upstream," Burke told me. "When it comes to RHEL we have been updating Firefox at most once a year. In the interim, we will cherry pick security fixes. We strike a middle ground between stability, security and access to rapid innovation."
Fedora is another story, taking a more rapid approach, though still not as rapid as Mozilla's own cycle. Burke explained that Red Hat will work Firefox updates via the Fedora community to gain confidence and ring out bugs.
Now while Red Hat doesn't follow Mozilla for the version changes, they are tracking security very closely. The 'cherry picking' of security fixes is something that Burke stressed is a strength of Red Hat.
"This is what we do, that's fundamentally the power of Red Hat in any component from kernel to desktop," Burke said. "As part of enterprise hardening we leverage the fact that we're in the code and we know and it's only possible to do this level of cherry picking due to the depth of experience and contributions."
At this point, Mozilla is still maintaining the 3.6.x branch, but the plan is for an enterprise support version with 42 weeks of support. Time will tell when that actually comes into play and whether or not that's the model that Red Hat will embrace, though I suspect they will. In the meantime, they've got a system that works for them. The challenge from my perspective is that the enteprise users aren't likely running the fastest/best Firefox they can. That said, any user can just download and use the new stuff if they want (assuming of course the enterprise/system policy restricts that).
CentOS 6.2 Already?By Sean Michael Kerner | December 21, 2011
From the "Open Source Project That Could' files:
CentOS is a project that many people (myself included) rely on as a way to get all the goodness of Red Hat Enterprise Linux, without the cost (or support - but that's another story). That's probably why there was a lot of concern when it took CentOS so very long to put out the 6.0 release.
CentOS 6.0 trailed the upstream Red Hat Enterprise Linux release by at least 8 months. It finally debuted in July and when it did, because it was so late, it was already out-of-date in terms of security. That's why in September, CentOS went with a continuos repository to help get things on track.
CentOS is now very much on track with the RHEL upstream.
RHEL 6.2 was released by Red Hat the first week of December. Barely two weeks later and CentOS 6.2 is now out too. That's a fantastic achievement. Now to be fair, the 6.0 release was likely more challenging due to the way that Red Hat packaged things, but the speed of the CentOS 6.2 release clearly indicates to me that there is a good process in place now.
CentOS isn't the first RHEL clone to hit 6.2, Oracle Linux 6.2 was out a week earlier, but hey Oracle in a multi-billion dollar company. CentOS is a community project. For the community to rally and get it together, fixing the process and the release cycle is a tremendously positive message about the power of open source -- and of course, about the power of the 'itch'.
CentOS scratches an itch that many people have and when there is an itch, developers/community are going to want to scratch.
Congrats on the release CentOS and I personally hope that this is the patch that we'll see for some time to come.
Why Do Developers Contribute Code as Open Source?By Sean Michael Kerner | December 19, 2011
From the 'Why We Fight' files:
There are a lot of obvious reasons, why open source makes sense. This isn't a new topic, but every so often another 'new' survey emerges that reminds us of the core fundamentals.
This time it's a new survey from the Outercurve Foundation -- which used to be known as the CodePlex Foundation - aka. the Microsoft sponsored open source foundation. But no, don't worry, no FUD mongering here, Outercurve is solidly in the open source camp.
The study found that 90 percent of their survey respondents were using open source. The top reason why? 80 percent said it saves time and money.
So why contribute?
Apparently it's not just about scratching the 'itch' as I firmly believe. Rather 44 percent of respondents to the Outercurve study said they contributed their stuff as open source, "... to improve their careers and credibility."
That's right open source is good for your career. But hey you knew that already right? (heck without open source, I wouldn't have a career let alone one that I could improve.)
Beyond that developers are also of course contributing for the 'itch' as well.
"We think it’s significant to note half of respondents contribute back to open source because they want to improve the projects they rely on,” said Paula Hunter, executive director of the Outercurve Foundation in a statement. "We’ve seen this positive energy and commitment to open source in a number of the foundation’s projects."
Sean Michael Kerner is a senior editor at InternetNews.com, the news service of the IT Business Edge Network, the network for technology professionals. Follow him on Twitter
Red Hat: ARM in Linux Is Not Ready for Prime TimeBy Sean Michael Kerner | December 16, 2011
From the 'Twist my ARM' files:
According to some vendors, one of the most exciting new avenues for Linux in 2011 is the ARM architecture.
Ubuntu is all over it and so is HP with their Project Moonshot effort using Calxeda ARM technology. While Ubuntu and HP are optimistic, Red Hat is more...realistic.
I recently had a conversation with Tim Burke, vice president of Linux Engineering at Red Hat - and to make a long story short - his views on ARM are very different than Ubuntu.
"ARM in Linux is not ready for prime time," Burke said.
However, he did note that Red Hat is helping to lead the effort to get standards for hardware enumeration. They are also working on some ARM scalability issues. At presents he just doesn't see a fit for ARM in the enterprise. He does however see it as being suited for Fedora - at least for now.
Considering the mission critical systems that Red Hat supports and the need for certifications, I can see where Burke is coming from. ARM might be a possible future for Linux, but the reality is that it's likely a bit too early for the enterprise. But hey, I'm happy to be proved wrong and it will be very interesting to see what the Calxeda/HP/Ubuntu combination is actually able to deliver in 2012.
WordPress 3.3 Improves Usability in Open Source CMSBy Sean Michael Kerner | December 13, 2011
From the 'More Than Just a Blog' files:
Yeaah, I know, WordPress is blogging software right? Well yes and no. WordPress has evolved to become the most popular open source Content Management System (CMS) in recent years besting Drupal and Joomla (as far as I can tell) in terms of sheer user numbers (admittedly not the best metric). WordPress itself claims a staggering 65 million downloads since the 3.0 release.
One reason why WordPress continues to gain fans (and you can count me among them) is their predictable and relatively non-disruptive update cycle. This week, WordPress 3.3 came out roughly six months after the 3.2 update. With 3.3, there are a lot of little improvements that add up to whole lot for the user experience.
In limited testing, one of the first things that jumped out at me was the improved admin toolbar that shows up whenever you're logged into your site. The admin toolbar now makes it easier to see what updates you need to make and if you're running a cache (who isn't) you get the option to delete it. For new (and old) users, there is improved contextual help which is a nice touch.
The 3.3 release also fixes a number of usability flaw that have plagued recent WordPress releases.
The way you used to get media (video/audio/pics) into WordPress was with separate icons for different media types. Now that has change with a new drag and drop unified media updloader. The other old usability flaw fixed in 3.3 is instead of having to click a menu item to see sub-menu (i.e Themes under the Appearance tab), now it's just a hover menu. -- So you can save your fingers the extra click.
Sure the changes in WordPress 3.3 are relatively small changes but don't doubt for a second that they provide big usability improvements.
That's why WordPress does so well. It has always been a easy to use system, but the community of developers keep on finding way to make easy even easier.
Facebook Accelerates PHP with HipHop Virtual MachineBy Sean Michael Kerner | December 12, 2011
From the 'Evolution of PHP' files:
Facebook is a big user of PHP throughout its infrastructure, but they don't use quite the same same PHP that the rest of us use. A year and a half ago they launched HipHop,their own open source runtime for PHP.
Last week, they officially launched the HipHop virtual machine (hhvm) which will make HipHop PHP execution faster than ever.
All this work on what is kinda/sorta a fork of PHP makes me (and others) wonder, why does Facebook bother? If PHP isn't up to the task, then why not use a different language?
In a Facebook comment, Facebook engineer Jaap Weel said:
"..when confronted with an enormous existing PHP codebase, it really is a better idea to improve the PHP implementation than to attempt to rewrite the whole thing."
HipHop isn't precisely a fork of PHP either - HipHop's goal isn't about changing the PHP language itself - it's all about helping PHP scale.
"There are two different pieces to keep in mind, one is the PHP language itself, then there is the runtime that actually goes and interprets the language and runs it," David Recordon, Facebook's senior open programs manager at the time of the original HipHop release told me. "What we've done is we've implemented the PHP 5.2 language with a few features removed. Our plan is to keep the language the same, but what we've changed is the underlying runtime and the process of going and transforming the source code into C++ and then compiling it and pushing out the compiled binary."
To date, I personally have seen nothing from the HipHop effort land in the broader/general open source PHP community mainline. That said, Facebook is doing most of their code drops in the open on GitHub, so hey php.net devs could take advantage of things if it was possible. That leads me to my next issue, is Facebook so unique an environment that HipHop has no external value?
Facebook's OpenCompute effort is helping to bring open source ideas to the data center, using Facebook's own infrastructure as a use case, so why can't the same hold for PHP?
HP Open Sources webOS - But that doesn't mean webOS will live.By Sean Michael Kerner | December 09, 2011
From the 'Open Source Lifeline' files:
The day that webOS was officially killed was a shocking one for me. I remember being at LinuxCon, and it was just a few short hours after an awesome keynote from HP on webOS that the axe fell.
Now HP is giving webOS a second life - by open sourcing it.
"webOS is the only platform designed from the ground up to be mobile, cloud-connected and scalable," said Meg Whitman, HP president and chief executive officer said in a statement. "By contributing this innovation, HP unleashes the creativity of the open source community to advance a new generation of applications and devices."
No, we don't know if HP will also seed a development community.
No, we don't know where this project is going - ie. to Apache or just as an HP project.
Really we don't know much, other than the fact that HP has announced their intention to open source the technology. This is kinda like the tech equivalent of a Hail Mary pass, HP doesn't really know so they're throwing the tech out there.
That's a good thing. Without even the Hail Mary, webOS was dead. With the 'promise' of open source webOS, someone could theoretically keep it alive.
That said, without a strong community, I suspect that people will just take bits and pieces of webOS to use for their own efforts. Assuming licenses and IP issues permit that...
As proprietary tech, webOS and the TouchPad were a dead end, no there is a glimmer of hope. But it is just a glimmer. HP has not fully committed to webOS or tablet development and without them it's questionable if this effort will advance. Sure it will survive, but in the mobile world, survival isn't the name of the game -- evolution is.
Co-Founder of GNOME Project Still Uses Linux EverydayBy Sean Michael Kerner | December 07, 2011
From the 'Desktop Linux Titans' files:
Miguel de Icaza started the GNOME project back in 1997 along with Federico Mena. In recent years, my interactions with De Icaza have been all about Mono, the open source implementation of .NET. Even more recently, De Icaza's focus has narrowed further with his new startup Xamarin that is built around mobile development.
In a recent conversation I had with De Icaza, he told me that he was 'heartbroken' that his company is not doing Linux desktop stuff. That said, he also noted that mobile developers are very pragmatic.
"In the Linux world, Mono was always the subject of a huge anti-Microsoft debate," De Icaza told me. "One thing that has magically vanished from my life is participation in Internet flame fests about how evil Microsoft is, so now I can actually focus on products and not flame wars."
Though Xamarin isn't a Linux desktop company, De Icaza told me he's an active GNOME 3 user at home and he likes it very much.
Mozilla's Google Deal is NOT Dead.By Sean Michael Kerner | December 05, 2011
From the 'Don't always believe what you read' files:
There has been some 'confusion' in recent days about the status of Mozilla's relationship with Google. Some have speculated (incorrectly) that the deal is over, leaving Mozilla without its chief source of revenue.
That's simply not the case. Just ask Mozilla. They just released a new statement on the issue which will clear up some of the misconceptions.
"Our search relationship with Google remains positive for both of us. We are in active negotiations and have nothing further to announce at this time. We have every confidence that search partnerships will continue to be a strong and growing generator of revenue for the foreseeable future. "
No, that's not a formal announcement of a straight deal renewal, but it is a very strong indication that one will be signed.
That's good enough for me.
The simple fact of the matter is that neither Google nor Mozilla can afford to have this deal lapse. For Mozilla, without Google, they wouldn't have the cash flow to fund the staff and operations. Sure they have assets and reserves, but not a whole lot of unique inbound cash without Google.
For Google, without the Mozilla partnership (which gives Mozilla a few pennies on every search conducted from Firefox), they'd raise the ire of the U.S. FTC and Congress. Google is already walking a thin line on anti-trust type issues and it is critical that they continue to support Firefox (a competitor in some respects) to provide an answer.
As well, Google benefits from Mozilla's development efforts which are all open source. Innovations that come from Mozilla can and do help Google.
It's a win win deal and no matter how the deal is restructured to account for new realities, none of us should have any doubts about it continuing for the immediate future.
Will GNOME 3.4 Save the Linux Desktop?By Sean Michael Kerner | December 02, 2011
From the 'I Miss GNOME 2.x' Files:
There are a lot of people that don't like the changes brought to the Linux desktop with GNOME 3. Among those people is none other than the father of Linux himself, Linus Torvalds.
But that could be changing as GNOME 3.4 development work continues and could build on some improvements that first debuted in GNOME 3.2
"Hey, with gnome-tweak-tool and the dock extension, gnome-3.2 is starting to look almost usable," Linus Torvalds wrote in a Google+ posting." Now I just hope those things become part of the standard gnome shell setup and made available in the regular "system config" thing rather than hidden off…Or would that be too close to "Ok, we admit we were wRong" and thus not politically acceptable?"
Intel developer Dirk Hohndel added in a comment the he is close to finding Gnome3 usable.
"I filed a few bugs for the most annoying things that still bother me - but my hope is that by Gnome 3.4 we'll forget about the disaster that was Gnome 3.0. 3.2 + the extensions is on the right path," Hohndel commented.
Going a step further GNOME developer Allan Day has started the 'Every Detail Matters' effort to help clean up GNOME 3.4. It kinda/sorta reminds me of the Ubuntu Papercuts effort and has a similar end goal – to make the system more usable.
"For the first round of Every Detail Matters, we will be focusing on the Activities Overview," Day blogged. "This is obviously an important part of GNOME 3, and there are lots of small things that can be done that will make a big difference."
We're still months away from GNOME 3.4 and until then, I suspect that Linux Mint and it's 'classic' view of the desktop will continue to win over users. But hey, GNOME devs are working on their 'issues' and I strongly suspect that the next version of GNOME 3.x will make a lot more users think twice about the Shell.
Linux Server Revenues Grow, but Unix Still has More ShareBy Sean Michael Kerner | December 01, 2011
From the 'More Unix Share Left to Take' files:
Ever since, I first started with Linux in the late 90's, I've been hearing Linux vendors tell me the opportunity is in taking share from Unix.
Over the years, Linux certainly has taken a whole lot of share from Unix, but as it turns out, there is still room to go.
According to IDC's Q3 2011 Server stats, Linux server revenues for the quarter grew to $2.3 billion. That's a 12.3 percent year-over-year gain. Overall, Linux holds an 18.6 share of total server revenues.
Not too shabby.
In contrast, Unix server sales hit $2.6 for the third quarter, growing at a significantly slower rate than Linux of only 1.6 percnetyear-over-year growth. Unix server revenues hold 20.1 percent of global server revenues.
So to recap: Linux is still growing fast than Unix, but Unix revenues are still higher and hold a larger percentage of total server revenues. Considering that Unix server (in nearly all cases) cost more than x86 based Linux ones, this isn't a terrible surprise.
What is interesting to note though is that after a decade of chipping away at Unix's market share, the way I see the numbers trending, Linux will surpass Unix, sooner rather than later.