It’s interesting to see IBM add a Web app component to Symphony, much as Microsoft has done with its own Office 2010 and Office Web Apps offerings. I tend to use a combination of Web and desktop-native apps in my daily work–both platforms offer certain advantages, so why not take advantage of both?
After spending about 20 minutes of wandering within LotusLive Symphony–a few minutes of which I’ve embedded for your viewing below–it seems that IBM’s new Web apps, which include word processing and spreadsheet components, are off to a decent enough start.
I tried the new Web-based Symphony out with one of Microsoft’s Word-formatted reviewer guides–these tend to be heavily formatted, so they offer a decent test of applications’ Office-format fidelity chops. I found various small formatting issues, which is the standard I’ve come to expect. One feature that caught my eye was a task assignment capability I’ve not seen in competing Web-based products.
What’s your take on Web-based versions of existing desktop apps? Me-too feature of the new decade, wise embrace of the Web, or something else?
The latest Linux-based OS from Red Hat offers a strong foundation for hosting virtual workloads, complete with distinctive capabilities such as security features rooted in SE Linux.
Convirture’s vCenter workalike builds on the virtualization foundation offered by Linux server operating systems such as those from Red Hat and Ubuntu, with an easy-to-use Web-based management interface and separate open source and enterprise product options.
About four years ago, I wrote a blog post (since lost, apparently, to the sands of blog platform migration) entitled “What Is Fedora’s Prime Directive?” At issue, more or less, was whether it was appropriate for the Fedora project to push an Xorg modification that stood to deliver benefits to users of open source graphics drivers at the cost of disrupting the systems of closed-source graphics driver users.
Less important to me than the particulars of that issue was the way the project reacted to the problem, and what the dustup meant for the ongoing questions around the mission of Fedora. While that particular dustup is long gone, some of the basic questions remain, as highlighted last week in a flareup around a particular Fedora 14 bug involving Adobe’s Flash plugin.
In short, the developers behind the glibc C library project on which most, if not all Linux distributions depend changed its implementation of a particular function in pursuit of potential performance benefits on certain processors. The change, while in keeping with the intended, and well-documented, use of the function, caused problems for various sloppily-coded software components, most prominent of which was Flash.
A bug was filed by a Fedora user who experienced problems with Flash on Fedora 14, and several other users who encountered it, among them Linus Torvalds himself, chimed in on the discussion. Also, many tweets were fired to and fro on the issue, though I don’t think anyone offered up a pithy hashtag on the subject for our convenience.
You can check out this short youtube video I recorded to see the issue in action. In the video, I’m running a brand-new instance of Fedora 14 64-bit from the project’s LiveCD, I cruise to Red Hat’s Web site, I encounter content that requires Flash to view, install Flash, and hit the issue called out in the bug report. You’ll need to have your sound turned on–this is an audio issue. The borked sound begins about halfway through.
The bug itself was marked: CLOSED NOTABUG, with the rationale being that the developers of the Flash applet and other affected applications were in the wrong for using this function inappropriately, so it was their problem, not Fedora’s problem. Simple. Done and done.
Of course, part of the point of a Linux distribution is that it acts as a buffer between upstream projects and individual users. Yes, we could all roll our own Linux-based OSes from thousands of different open source projects, and deal with the integration ourselves, but performing these chores is how Red Hat earns its money, and it’s how Fedora–while a free, community-supported project rather than a product–earns its mindshare.
That is, of course, unless what many Fedora critics have long claimed is true–that Fedora is simply a bleeding-edge test bed for the release that Red Hat gets paid for, and that Red Hat would not likely push out into the world with a broken Flash implementation–regardless of who was to blame for the brokenness.
It’s perfectly fine for Fedora to be just such a distribution–YMMV, caveat “emptor,” etc. However, the Fedora Project projects itself as much more than this. Here’s the tag line from the fedoraproject.org Web site:
Fedora is a fast, stable, and powerful operating system for everyday use built by a worldwide community of friends. It’s completely free to use, study, and share.
I don’t mean to suggest that the Fedora project or Red Hat is acting in bad faith, or hiding at all its true intentions. Rather, the explanation now, as in past years, seems to me to be that as an open source project, Fedora houses a mixture of the attitudes and motivations of its contributors, with a tilt toward those of Red Hat, as Red Hat is the project’s biggest presence.
For Red Hat, it makes sense to push the envelope with Fedora, and to allow its slower-moving enterprise releases to benefit from what often amounts to creative destruction in the Fedora cycles.
Now, less clear to me is whether this arrangement pays adequate dividends for individual backers and users of Fedora (who may want to use Flash without implementing wacky workarounds), particularly given the presence of plenty of Fedora alternatives. Fortunately for those individuals, the switching costs between different Linux distribution options are fairly low, so they can vote with their feet.
Lately, I’ve had Mac on the brain—a state that’s stemmed in parts from P. J. Connolly’s coverage of Microsoft’s Office 2011 for the Mac, from Apple’s recent “Back to the Mac” event at its Cupertino headquarters, and from Apple’s disclosure that the increasingly consumer-oriented company plans to drop its most enterprise-oriented product, the XServe.
In particular, I’ve been considering where Apple’s insistence on tight(ening) control of its hardware, software, and third party application stack makes sense in an enterprise context.
The Apple event offered the public an early peek at the upcoming version of the company’s OS X 10.7 operating system. The new product, which is set to ship next summer under the code name, “Lion,” caught my attention in a way that no OS X release has done since Apple embraced Intel’s x86 processors.
I’ve been consistently lukewarm in my reception of most OS X releases because for me, the tight control that Apple demands over OS X and OS X Server haven’t come along with enough benefits to outweigh the limitations.
For instance, how can an enterprise IT administrator take seriously a server OS that’s banned, arbitrarily, from partaking in virtualization, the biggest server technology of the past several years? In a world in where server workloads are going virtual and headed for the clouds, it isn’t worth investing one’s time in an OS banned from running on alternate platforms.
What’s catching my eye about Lion, however, is Steve Jobs’ promise to make OS X more iOS-like in its function and management. While OS X fails sufficiently to outshine its more flexible OS rivals, iOS is an all together different matter.
When Apple’s mobile platform debuted, it blew away the competition and far outshone every other smartphone or handheld computer out there. And while Apple’s rivals have certainly upped their game, iOS continues to impress me.
For instance, Apple’s strategies for managing the limited resources of mobile devices by suspending and resuming applications as needed, and by offering up a set of background APIs intrigue me. After all, mobile devices aren’t the only sort of computing clients that suffer from hardware limits, and throwing more RAM at our problems isn’t always feasible.
What’s more, Apple’s App Store model for vetting and deploying applications, while certainly limiting, could actually work well in an enterprise setting–provided, of course, that an enterprise’s own administrators could control the system and make their own vetting choices.
As a Linux aficionado, I cringe over the hood welded shut nature of Apple’s products, but in a managed PC setting, I’m a huge fan of client control technologies such as SE Linux and Application Whitelisting. The client as appliance model stands to save IT time fiddling with desktops and free up resources better directed at driving business value.
Obviously, I’m making some leaps here. As currently situated, Apple and its platforms are not really tailored for, or aimed at, enterprises. With that said, news of enterprise uptake of the iPad keeps coming, and, through reported support deals with Unisys and local resellers, Apple appears to be taking business sales and support more seriously.
Even if Apple opts not to embrace the enterprise (a continued holdout against virtualization would be a dealbreaker) a more iOS-like OS X could still deliver more locked-down and tightly-manageable clients for companies through the same forces of competitive osmosis that’s led smartphone vendors to remake their wares in the iPhone’s image.
IBM’s Eclipse-flavored spin of the OpenOffice.org productivity suite undergoes a major update under the covers, moving up from the 1.x version on which it was originally based to sync up with version 3.x of the open-source project.
Fedora 14, the latest release of Red Hat’s fast-moving, community-supported Linux distribution, hit the Internet earlier this month bearing its typical crop of updated open-source software applications, with a particular focus on updated developer tools, such as the latest versions of the Eclipse and Netbeans Integrated Development Environments.
From an IT columnist perspective, Oracle’s acquisition of Sun Microsystems is a gift that keeps on giving. As the enterprise software giant works its way through digesting Sun’s many hardware platforms, software products, intellectual property holdings, and open source communities, there’s no shortage of fresh topics to cover.
Last week, another such topic presented itself, when a group of vendors and individuals launched LibreOffice, a fork of the OpenOffice.org productivity suite that Sun first shipped in 2002. The group also announced the creation of the nonprofit Document Foundation to maintain the breakaway office suite project.
Where Sun Microsystems had been bent on building itself out as a grand steward of open source projects, Oracle has pursued a spartan approach toward the projects it inherited from Sun. For instance, Oracle has continued development on Solaris while allowing the open source project OpenSolaris to sink beneath the waves.
Similarly Oracle hasn’t ignored OpenOffice.org—the company has selected “Oracle Open Office” as the name for the commercial version of the suite, in lieu of the StarOffice moniker that Sun used for its paid edition of the suite.
Given Oracle’s ambivalent stance toward running open source projects for their own sake, and taking into account the central role of OpenOffice.org on the Linux desktop, it’s not surprising that major Linux and OpenOffice.org distributors Canonical, Novell and Red Hat—all of which are supporting the Document Foundation—would prefer a more certain home for the suite.
It’s not that OpenOffice.org under Oracle appears at risk of moving backward—it seems to me that development on the suite under Oracle has continued at a pace that seems more or less the same as before the Sun acquisition.
The trouble is that as a project, OpenOffice.org needs to do more than maintain its pace. Even with Sun’s open source enthusiasm, development on the project merely crept along—particularly when compared to Mozilla’s Firefox. For instance, while the suite’s chief target, MS Office, has begun making a move to the Web, OpenOffice.org showed no such signs.
Instead, the Web-ward growth that was absent from Sun’s roadmap has indeed begun, in the form of an upcoming Oracle Cloud Office product that’s no more open (in terms of source or of development) than are the Web office offerings from Microsoft, Google, and Zoho.
“Fork” is often treated as a dirty word in open source circles, but project splits can be very effective, as was the case when, in 2004, the X.org project split off from the XFree86 project that developed the graphics layer for Linux and it cousins, delivering a much-needed shot of life into those OSes.
I’m hoping that the creation of the Document Foundation, and of the LibreOffice project, will enable the developers and backers of the code base to blow up the project and remake it into a faster-moving, more broadly accessible, and more relevant project moving forward.
jebpages 