Microsoft has been going out of its way to tick off its partners.
First, Windows 8 has an interface, Metro, that only a mother could love. Metro will require Windows users to re-learn everything they know about how to use Windows. Then, Microsoft announced Surface, a vaporware tablet that leaves all its partners’ Windows 8 tablet plans in disarray. , Finally, adding insult to injury, Microsoft stabbed its smartphone partners in the back by announcing Windows Phone 8, which made all currently shipping Windows phones obsolete. So, if you’re in the PC business do you really want to work with Microsoft or is it finally time to look for a partner that really wants to work with you rather than use you?
I think it’s time for Dell, HP, Lenovo, and all the other big-time PC vendors to finally start taking the Linux desktop seriously. It’s clear that Microsoft’s agenda no longer is running in parallel with their plans.
Shifting to Linux won’t be easy. I’m sorry to say that in 2012 there are only two significant Linux desktop/tablet operating systems for original equipment manufacturer (OEM) to consider for partnering: Canonical of Ubuntu fame, and Google with Android and Chrome OS.
Yes there are many other Linux desktop distributions. Yes, some of them may be better. I, for example, favor Mint 13. But, Mint, while it finally has a partner shipping Linux Mint-powered PCs, and the other small Linux distributors aren’t big enough for the major OEMs to take seriously. The other big name Linux companies, Red Hat and SUSE, are now focused on servers.
Ubuntu, on the other hand, not only has been courting OEMs for years, it’s actually been shipping Ubuntu-powered laptops and desktops from companies like Dell for years. When Mark Shuttleworth, Canonical’s CEO, said recently that he expected 20-million PCs to ship this year with Ubuntu, he wasn’t blowing hot air.
Shuttleworth was, however, not talking just about the North American and European Union market, but the world market. It’s in China and India where Canonical, with its partner Dell have found that people really will buy PCs without Windows. I know for a fact that Canonical would be more than happy to work with other OEMs and bring the Ubuntu Linux desktop to Western markets.
Are you still under the delusion that Linux is too hard? That once people go Windows they won’t look at anything else? Please, meet my now 80-year old mother-in-law who’s a happy Ubuntu 12.04 user.
The major OEMs already have experience in working with Google and Android. Since Android is the hottest selling smartphone operating system on the planet, Google must be doing something right. There’s no reason Chrome OS-powered Chromebooks can’t be the next step in desktop evolution.
Think about it. Chrome OS is just the popular Chrome Web browser running on Linux. If you know how to use a Web browser you can use a Chromebook. Unlike Windows 8’s Metro there is no learning curve what-so-ever.
Chrome OS’ big problem is that it requires an Internet connection to show its stuff. It is, after all, the first significant cloud-based desktop operating system. But, how much work can you get done now with your Windows PC without an Internet connection? If you’re honest you know that the answer is: “Not much.”
Besides, Chrome OS’ offline capabilities are improving. You can already use GMail off-line. It also looks like Google will be rolling out offline Google Docs for Chrome OS this week at their annual show of shows Google I/O.
Now, let’s take a long, hard look at the situation. Microsoft is showing itself to be no friend to its partners and Windows 8, like Vista before it, looks to be a flop in the making. But, if the hardware vendors start offering a Linux-based product lines they’ll increase their razor-thin margins, work with partners who want to work with them, and be able to offer customers attractive and secure operating systems that actually require less training than Windows 8 will.
Heck, thanks to Ballmer’s desktop and partner mis-steps maybe we finally will see a year of the Linux desktop after all!
While not a major release, RHEL 6.3 does include enhancements to take advantage of the most recent advancement from hardware originial equipment manufacturers (OEMs). This includes updated device drivers for a multitude of peripherals, and also various features like compiler optimization for the Intel Xeon E5 processor family.
It also includes the following new features:
Developer Tools: The new RHEL supports OpenJDK 7. This is the Oracle-approved open-source Java Standard Edition (SE).
Security: Users can now use two-factor authentication for securely accessing RHEL. It also includes advanced encryption capabilities so data blocks can be encrypted in parallel by taking advantage of underlying multi-processor capabilities. This is supported by the introduction of AES-CTR (Advanced Encryption Standard Counter Mode) cipher for OpenSSH, the secure connection network tool
Scalability: RHEL can now support up to 150 virtual CPUs (vCPUs) per guest. By comparison, Red Hat states that is “significantly higher than the 32 vCPU per guest limit for VMware ESX 5.0.” In addition KVM virtual machine operating systems guests can access up to 2 terabytes of RAM.
Storage: RHEL;s Logical Volume Manager (LVM) now provides support for RAID levels 4, 5, and 6 to simplify overall storage administration by consolidating all management functions, such as creating and re-sizing volumes, deploying RAID, and taking snapshots into a single interface.
You can also now deploy RHEL as a Fibre Channel over Ethernet (FCoE) istorage target server. Red Hat claims that this will provide higher levels of reliability and performance that you get with native Fibre Channel but at a significantly lower cost.
Taken as a whole, there’s nothing that jumps out at you, but the over-all message is that RHEL is better than ever not just as a high-end server operating system, but as a platform for virtual machines as well. If you need a top of the line server, RHEL is continuing to give you reasons to give it serious consideration.
The Voodoo Economics of Hypervisors
Red Hat carefully repositions CloudForms as open hybrid cloud management platform
Red Hat celebrates 10 years of Red Hat Enterprise Linux
Fedora 17 & GNOME 3.4: Return to a useful Linux desktop (Review)
If you buy a Windows 8 or Windows RT computer or tablet, yes even Surface, it will come with secure boot enabled by default in their replacement for the BIOS, Unified Extensible Firmware Interface (UEFI). I doubt that will actually make them more secure, but it’s always crystal clear that it will make it much harder to boot Linux or any other operating system, such as Windows XP or 7, on them. Fedora came up with a way to get around this problem and Ubuntu Linux has come up with its own solution to the Windows 8 lock box as well (PDF Link). Fedora’s developers, however, don’t like Ubuntu’s answer.
In a blog posting Matthew Garrett, a developer for Red Hat, Fedora’s parent company, wrote Ubuntu’s UEFI requirements are “basically the same set of requirements as Microsoft have, except with an Ubuntu key instead of a Microsoft one.”
Garrett continued, “The significant difference between the Ubuntu approach and the Microsoft approach is that there’s no indication that Canonical will be offering any kind of signing service. A system carrying only the Ubuntu signing key will conform to these requirements and may be certified by Canonical, but will not boot any OS other than Ubuntu unless the user disables secure boot or imports their own key database. That is, a certified Ubuntu system may be more locked down than a certified Windows 8 system.”
Garrett admits, “Practically speaking this probably isn’t an issue for desktops, because you’ll need to carry the Microsoft key in order to validate drivers on any PCI cards. But laptops are unlikely to run external option ROMs, so mobile hardware would be viable with only the Ubuntu key.”
He sees two possible solutions to this, but neither are ideal:
1. Canonical could offer a signing service. Expensive and awkward, but obviously achievable. However, this isn’t a great solution. The Authenticode format used for secure boot signing only permits a single signature. Anything signed with the Ubuntu key cannot also be signed with any other key. So if, say, Fedora wanted to install on these systems without disabling secure boot first, you’d need to have two sets of install media - one signed with the Ubuntu key for Ubuntu hardware, one signed with the Microsoft key for Windows hardware.
2. Require that ODMs (original design manufacturer) include the Microsoft key as well as the Ubuntu key. This maintains compatibility with other operating systems.
“This kind of problem is why we didn’t argue for a Fedora-specific signing key,” concluded Garrett. “While it would have avoided a dependence on Microsoft, it would have created an entirely different kind of vendor lock-in.”
First, Shuttleworth isn’t happy with Ubuntu or Fedora’s current answers to Microsoft’s attempt to lock-in users to Windows 8. Shuttleworth said, “We’ve been working to provide an alternative to the Microsoft key, so
that the entire free software ecosystem is not dependent on Microsoft’s goodwill for access to modern PC hardware. We originally flagged the UEFI/Secure Boot transition as a major problem for free software, we lead the efforts to shape the specification in a more industry-friendly way, and we’re pressing OEM partners for options that will be more broadly acceptable than Red Hat’s approach.”
Indeed, the Red Hat/Fedora answer, which uses Microsoft’s own secure boot key signing service, annoys many Linux users. But as Linus Torvalds, who has no low for how Microsoft is using UEFI to block Linux, recently told me, “Signing is a tool in the tool-box, but it’s not solving all the security problems, and while I think some people are a bit too concerned about it, it’s true that it can be mis-used.”
Shuttleworth wishes he has a better answer, but at this point he doesn’t. He continued, “Secure Boot retains flaws in its design that will ultimately mandate that Microsoft’s key is on every PC (because of core UEFI driver signing). That, and the inability of Secure Boot to support multiple signatures on critical elements means that options are limited but we continue to seek a better result.”
That better solution, Canonical commercial engineering director Victor Tuson Palau suggested last year, would include: “systems manufacturers including a mechanism for configuring your own list of approved software. This will allow you to run Windows 8 and Linux at the same time in your PC with Secure Boot “ON”. This should also include you being able to try new software from a USB stick or DVD.”
Palau added, “With the ability for users to configure Secure Boot, it will become harder for non-techie users to install, or even try, any other operating system besides the one that was loaded on the PC when you bought it. For this reason, we recommend that PCs include a User Interface to easily enable or disable Secure Boot.”
I think anyone who’s serious about Linux desktop agreement would agree on these points. Linux developers would be better off co-ordinating their efforts to get ODMs and OEMs to work together on an open UEFI Secure Boot solution, such as the Linux Foundation proposed last year, than in bickering with each other. In the end, if we squabble among ourselves over the best ways to address Microsoft’s attempt to lock Linux out of the desktop instead of working on a unified response to UEFI Secure Boot the only real winner will be Microsoft.
A few decades back I was working at Goddard Space Flight Center. I’m sorry to say that I left just before some people I’d met, Don Becker and Thomas Sterling, built the first Linux cluster, Beowulf. They didn’t know it, but by making a cheap cluster from 16 486DXs processors and 10Mbps Ethernet, they were creating the ancestor to today’s Linux supercomputers. Now, not 20 years later, well over 90% of the world top 500 supercomputers are running Linux.
The new supercomputer champion of champions, according to the TOP500 list of the world’s top supercomputers is Sequoia. This IBM BlueGene/Q system installed at the Department of Energy’s Lawrence Livermore National Laboratory achieved an impressive 16.32 petaflop per second (Pflop/s) on the Linpack benchmark using 1,572,864 cores.” That’s 16.32 quadrillion floating-point operations per second). The operating system? Linux of course.
Second place goes to Japan’s Fujitsu’s “K Computer.” It’s installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan, with 10.51 Pflop. It uses 705,024 SPARC64 processing cores. The hardware may be Sun/Oracle in design, but the operating system is Linux, not Solaris or OpenSolaris.
Indeed, Solaris isn’t on the top supercomputer list at all and OpenSolaris only has one system. Microsoft does a little better than that. There are two top 500 supercomputers that run Windows HPC 2008. The only operating system that even pretends to give Linux any competition if IBM’s house brand of Unix: AIX with 22 systems, aka 4.4% of the world’s fastest supercomputers.
If you just glance at the Top 500 chart you might be fooled into thinking Linux isn’t quite as dominant as it really is. If you look closer though you’ll see that after AIX instead of the customized Linux that makes up the bulk of the list, there are numerous specific Linux distributions. At the top of this list you’ll find SUSE Linux Enterprise Server 9 with 11 systems; SLES 10 with 8; and Cray Linux with 7. The supercomputer world really does belong to Linux.
What’s even more amazing than Linux’s total domination of supercomputing is how fast these Linux-powered thinking machines are getting faster. Only six-months ago, in the last round-up of supercomputers, the combined performance of all the supercomputers was 74.2 Pflop/s. Now, the total performance of all the systems on the list is 123.4 Pflop/s. Can you say fast? I knew you could!
To look at another way, 20 of the supercomputers on the latest list reached performance levels of 1 Pflop/s or more. The first supercomputer to break the petaflop barrier, IBM’s RoadRunner running Linux of course, only did it in 2008 The No. 500 machine on the list? Its 60.8 teraflop/s would have made it number 332 last year.
Looking ahead, the next goal to crack will be exaflop supercomputers (PDF Link). An Exaflop is a thousand petaflops.
Since Sequoia just cracked 16 Pflop/s that may sound like a goal for the 2020s or even the 2030s. Intel thinks it can get supercomputers there with its new Xeon Phi processor family by 2018. I’m not sure they’ll be able to do it that fast. I do know though what operating system the first exaflop capable supercomputer will be running though. It will be running Linux.
The “Folsom” version of OpenStack, one of the most anticipated open source releases this year, is expected for release on September 27.
During a call today to discuss the OpenStack Summit this fall, organizers say the late September release timeframe dovetails nicely with the October conference in San Diego. It will be the last time that Rackspace – one of the founders of the open source cloud operating system – will be listed as the organizer of the event since the newly-formed foundation will be up and running by then.
Openstack was initially launched in July of 2010 by Rackspace Hosting and NASA and to date has attracted more than 150 backers including Red Hat and SUSE, IBM, Dell, HP and even VMware.
The “Essex” version of OpenStack was released in April of 2012 with “Nova” Compute, “Horizon” Dashboard, “Glance” imaging, “Swift” Object Store and “Keystone” Identity services.
By far, the most anticipated new feature of the next OpenStack release is the new network service known as “Quantum,” which controls network virtualization, developers say.
Quantum provides network connectivity as a service between devices such as NICs that are managed by OpenStack services. Ultimately, it is described by one Nicira executive as a building block for sophsiticated coud network topologies.
It provides a standardized interface for building and managing virtual networks and can plug into SDN components such as OpenFlow. It is not SDN but can transform anything into SDN, others say.
“Project Quantum was incubated during the Essex release and aims to provide an automated framework for managing data center network activities. Quantum is a plug-in based service that manages common network administrative tasks, from creating ports and routes to configuring VLANs,” according to a press release issued by OpenStack when Essex shipped in April.
“Many users have been deploying OpenStack clouds with the Quantum networking service during the incubation phase, and Quantum is expected to become a core part of OpenStack in the “Folsom” release expected Fall 2012.”
OpenStack backers say “Folsom” offers an abundance of other capabilities including virtual networking services and high availability.
“There are quite a few real-world benefits of the development that’s happening on the Folsom release of OpenStack,” said Christopher MacGown, CTO of Piston Cloud Computing. “With the on-going modularization of OpenStack Compute and the newly created volume service, we’re seeing expanded support for traditional SAN and NAS from the usual suspects — EMC, NetApp, and Nexenta.
“There’s work being done to make the existing Nova config-drive more viable for tools such as Canonical’s CloudInit, Chef, or Puppet,” MacGown added. “OpenStack’s image service has a proposed tool to add image auto-detection to enable users with an existing repository of supported virtual machine images to automatically enable them for use in OpenStack Compute. The Authentication service is seeing improvements in multi-factor authentication, public-key infrastructure support, and IPv6 support.”
The so-called war of open source cloud computing platforms has been heating up since Citrix donated CloudStack to the Apache Software Foundation this spring. Still, observers say there’s plenty of room for competition in a cloud compuitng market that is growing exponentially.
“OpenStack is younger and may benefit from a broader, more diverse community that includes heavyweights in hardware, networking, storage, systems management and other key areas,” said Jay Lyman, a senior analyst at 451 Research.
“When OpenStack was established in 2010, people were asking how many open source cloud options we needed since Eucalyptus already existed. Today, we see more and more open source options, vendors and communities emerging in IaaS, PaaS and other subcategories of cloud computing. This matches customer demand for a marketplace of cloud services where they can pick and choose the pieces they want for their cloud implementation, whether private, public or hybrid. Given that open source software underlies a majority of cloud computing technology, it’s no surprise there is high demand for open source among these pieces they are picking and choosing. “
Here’s the best part of Microsoft’s lame attempt to surprise us with a significant announcement:
Windows Embedded, Drone Edition
IPad for Xbox
Clippy for Metro
Windows Vista 8 Metro Edition
Microsoft Works for Blackberry 10
Vista, Second Edition
Those are among the names that the bored tech press came up with for Microsoft’s new product on Twitter while we were waiting for Microsoft to get its act together and make its announcement.
In the Los Angles-based event, Steve Ballmer announced Microsoft is selling a Windows RT and Windows 8 tablets, the re-branded Microsoft Surface. It’s a 10.6-inch—about an inch bigger than an iPad–tablet with a keyboard and touchpad on its built-in cover.
That sounds cool, but I really wonder just how sturdy it can be in real life. In addition, I’ve seen hybrid laptop and tablet before and I’m still waiting for one that actually delivers. One thing I will note to its credit is that Microsoft promises that it will work with both a stylus, for fine detail work, and with your finger when you’re trying to use the Windows 8 klutzy Metro’s interface.
The Windows 8 model, which runs a 3rd generation Core i5 (Ivy Bridge) is a cross between an ultrabook and a tablet. The Windows RT model uses, of course, an ARM processor. For a display, Microsoft is offering something it calls ClearType HD. This is meant, of course, to compete with Apple’s Retina Display. The RT tablet will come with 32 to 64GBs of storage, while the Windows 8 version will come in 64 and 128GB models. Battery life? We don’t have a clue yet.
Microsoft fans say the the new Surface units are like a combination MacBook Air and iPad. I say it’s Microsoft desperately trying to pull attention away from Apple’s products. Lots of luck with that guys.
I also must note that in two ways this was a very typical Microsoft vaporware announcement: promise the world but don’t mention when it will be available or what it will cost.
How vapor is it? During the demo to the hand-picked press and employee audience, the demo tablet crashed when Steven Sinofsky, president of the Windows division tried to run a Netflix application.
Others agree, Surface is a blunder
While Microsoft refused to announce any pricing, the company did say it would be competitive with comparable ARM tab/Intel Ultrabook-class PC. That would put the RT device at about $500 and the Ultrabook at around $900.
This make me wonder what Microsoft’s original equipment manufacturer (OEM) partners are going to think about this? If I were Dell, HP, et. al., I wouldn’t be one bit happy. As for Apple and the Android vendors, I wouldn’t be in the least bit worried.
No, that’s not my prediction for Microsoft’s mysterious Monday announcement. No, it’s what Microsoft is already doing with last week’s unexpected release of Skype 4 for Linux. Microsoft–Microsoft!–of all companies has just shipped its first mass-market, end-user Linux desktop program.
It really wasn’t surprising that Microsoft saw the light of Linux on servers when they started supporting major Linux distributions — CentOS, openSUSE, SUSE Linux, and Ubuntu — on Windows Azure. Ballmer and the rest of Microsoft’s brass may not like it one darn bit, but they know that people want Linux servers on the cloud so they had to give it to them.
In fact, Microsoft itself is using Linux for its services. Ironically enough, Microsoft has moved Skype from its peer-to-peer (P2P) architecture to one built around… wait for it: Linux servers.
Earlier this year, Immunity Security’s senior security researcher Kostya Kortchinsky discovered that Microsoft had replaced Skype’s network of “supernodes” Skype user PCs with sufficient bandwidth, processing horsepower, and system resources to handle Voice over Internet Protocol (VoIP) calls and traffic control with 10,000 Microsoft/Skype hosted supernodes. According to Kortchinsky, and later tacitly confirmed by Microsoft, these new Skype servers are running Linux with grsecurity, server patches.
The desktop is another story though. While I use the Linux desktop, Linux Mint 13 for the most part, every day, I know there aren’t that many of us. Even now, hardware vendors like Nvidia don’t give Linux anything like enough support and if you want to buy a laptop or desktop with Linux pre-installed you need to look to small vendors such as System76 and ZaReason.
Despite that though Microsoft is finally offering a desktop program for Linux. If you consider Microsoft’s long bitter history with Linux, that’s amazing. There’s only one reason why they’d do it: They think that they can make money from it.
So, perhaps the Linux desktop is indeed bigger than the 1% of the market it’s usually given. Indeed, it appears that the Linux desktop has been growing over the last year.
Will Linux on the desktop ever threaten to catch up with Windows on the desktop? Nah. But, it does seem that Microsoft, for the first time ever, thinks there’s enough Linux desktop users out there that they’re worth supporting. Who would have thought it? Can Microsoft Office for Linux be far behind!!?
You could have knocked me over with a feather. The last thing I expected was to see a new version of Skype, the popular Voice-over-Internet Protocol (VoIP) program, arrive for Linux. But, just in time for Skype’s new in-your-face advertising program, Skype 4.0 for Linux has arrived.
According to Microsoft, there are four major changes in this Skype release. These are:
In addition, there are numerous minor improvements. These include:
Microsoft also warns “the very first time you start Skype for Linux 4.0 might take a few minutes (depending on how lengthy your chat history is). In the event, as I’ve started to tinker with the new Skype, Skype on Linux Mint 13 on my Lenovo ThinkPad T520 took less than a 30-seconds.
Formally, the new Skype for Linux is available for the 32 and 64-bit versions of Ubuntu 10.04 and Debian 6.0 and the 32-bit versions of Fedora 16 and openSUSE 12.1 . From my own experience I can also say that it will work on later versions of Ubuntu and related Linux distributions. The overall requirements are minimal: Qt 4.6.0, D-Bus 1.0.0, libasound2 1.0.18 with both PulseAudio 1.0 and BlueZ 4.0.0 being optional. Without a source code option, though, you’re much stuck with the Debian/Ubuntu, Red Hat, and SUSE Linux families.
OpenSUSE’s version 12.2 will once again be delayed and top developers are planning a reorg of the open source development project — but it may not be a bad thing.
In a critique of the current state of affairs, openSUSE release manager Stephen “Coolo” Kulow issued an email today noting that RC1 will be delayed and he called for a community-wide discussion on how to move forward with what appears to be a growing community.
“Many people have noticed that the milestones and the beta for this openSUSE release have been delayed or even canceled like Milestone 4. Now the RC is planned to go out Thursday – but that unlikely “to happen as Factory, our development project, is still far too unstable,” Kulow wrote. “One thing is certain: the openSUSE 12.2 release won’t see the light of day on July 11th.”
“Pretty much every milestone of openSUSE 12.2 has been delayed or even canceled,” he noted. “ Compared to the preliminary schedule, milestones 1 to 3 were only about one to two weeks late – but milestone 4 had to be canceled and even Beta 1 was 2 weeks late. Release candidate 1 won’t make it either – to get Factory close to a releasable state we’d need to think about a serious delay.”
openSuse is to SUSE what Fedora is to Red Hat. Version 12.1 – the first update since SUSE was spun off from Novell — debuted last November.
openSUSE has had other problems in 2012.
One spokesman for the group said openSUSE has important assets including the Open Build Service and Tumbleweed updating system but that recent delays are a “wake-up” call and “an opportunity to find new directions.
Some had hoped that the development project affiliated with the No. 2 distribution would attract more developers after Attachmate purchased SUSE from Novell and established it as an independent unit focused exclusively on Linux. “We’ve grown and our current way of working doesn’t scale anymore,” one spokesman wrote about Kulow’s announcement, which went out today.
VMware is on the open source move again, this time launching a new project called “Serengeti” aimed at enabling enterprises to quickly “deploy, manage and scale” Hadoop in virtual and cloud environments.
The project is available on GitHub as a “one-click” toolkit under an Apache 2.0 license. VMWare is also offering a version compiled as a binary in a virtual appliance that can run on vSphere.
VMware worked with Cloudera, Greenplum, HortonWorks and MapR to develop the code, which will help automate the deployment of Hadoop workloads on virtual infrastructure, VMware said.
In addition to its latest open source project, VMware said it will contribute extensions to Apache Hadoop that make components more virtualization aware, which will better support scalability and performance in virtual environments, the company said.
VMware also announced an update to another existing open source project known as Spring for Apache Hadoop.
The project, which debuted in February, makes it easier for developers to build distributed applications with Hadoop. The updates allow for integration with HBase databases, Cascading and Hadoop security. It is also available under an Apache 2.0 license.
There’s no doubt about it. Android, especially Ice Cream Sandwich (ICS), version 4.0, already offers more than what is coming in Apple’s forthcoming iOS 6. But, Android has its own flaws.
True, as Tom Henderson, principal researcher for ExtremeLabs and a colleague, told me, there’s a “Schwarzschild radius surrounding Apple. It’s not just a reality distortion field; it’s a whole new dimension. Inside, time slows and light never escapes– as time compresses to an amorphous mass.
“Coddled, stroked, and massaged,” Henderson continued, “Apple users start to sincerely believe the distortions regarding the economic life, the convenience, and the subtle beauties of their myriad products. Unknowingly, they sacrifice their time, their money, their privacy, and soon, their very souls. Comparing Apple with Android, the parallels to Syria and North Korea come to mind, despot-led personality cults.”
I wouldn’t go that far. While I prefer Android, I can enjoy using iOS devices as well. Besides, Android fans can be blind to its faults just as much as the most besotted Apple fan.
For example, it’s true that ICS has all the features that iOS 6 will eventually have, but you can only find ICS on 7.1 percent of all currently running Android devices. Talk to any serious Android user, and you’ll soon hear complaints about how they can’t update their systems.
You name an Android vendor-HTC, Motorola, Samsung, etc. -and I can find you a customer who can’t update their smartphone or tablet to the latest and greatest version of the operating system. The techie Android fanboy response to this problem is just “ROOT IT.” It’s not that easy.
First, the vast majority of Android users are as about as able to root their smartphone as I am to run a marathon. Second, alternative Android device firmwares don’t always work with every device. Even the best of them, Cyanogen ICS, can have trouble with some devices.
Besides, while Cyanogen supports many smartphones and tablets, it doesn’t support all of them.
For example, there’s still no stable CyanogenMod 7 (Android 3.x) firmware for Barnes & Noble’s Nook Tablet. Sometimes even when there is support, such as there is for the popular Samsung Galaxy Tab 10.1, there are driver troubles that keep the camera from working for many users.
Another issue is consistency. When you buy an iPhone or an iPad you know exactly what the interface is going to work and look like. With Android devices, you never know quite what you’re going to get. We talk about ICS as if it’s one thing-and it is from a developer’s viewpoint-but ICS on different phones such as the HTC One X doesn’t look or feel much like say the Samsung Galaxy S III.
A related issue is that the iOS interface is simply cleaner and more user-friendly than any Android interface I’d yet to see. One of Apple’s slogans is “It just works.” Well, actually sometimes it doesn’t work. ITunes, for example, has been annoying me for years now. But, when it comes to device interfaces, iOS does just work. Android implementations, far too often, doesn’t.
So, yes, Android does more today than Apple’s iOS promises to do tomorrow, but that’s only part of the story. The full story includes that iOS is very polished and very closed, while Android is somewhat messy and very open. To me, it’s that last bit-that Apple is purely proprietary while Android is largely open source-based-that insures that I’m going to continue to use Android devices.
Now, if only Google can get everyone on the same page with updates and the interface, I’ll be perfectly happy!
One developer is forwarding his alternative app framework to MapReduce to make big data management in the Hadoop era easier.
Concurrent CEO and Founder Chris Wensel is the creator of an open source data workflow API, Cascading, which is used by Twitter, Amazon and Razorfish.
Last week, the company was officially launched as was Cascading 2.0, which is now available under an Apache 2.0 license.
Wensel sees growing adoption of the API as big data management explodes. He created Cascading to help him develop complex Hadoop applications easier and first released the code in 2007.
“I was writing Hadoop applications and it was an extremely painful process. I started writing a framework to give me a different model . MapReduce is a simple way to parallelize data computations but it’s hard to solve real problems,” Wenzel said in a recent interview.
True Ventures and Rembrandt announced $900,000 in seed funding last year.
“We have seen a lot of innovation and investment in the data center at the storage, database, and around new technologies like Hadoop to accommodate the exploding data growth. We think a new equally important category will also emerge around data processing and management in order to give context to these growing volumes of data,” wrote True Ventures’ general partner Puneet Agarwal in a bog last year.
In a press release, Concurrent announced that using the upgraded 2.0 API, now available under an open source license: