RealTime IT News

Blog Archives

Mozilla Jetpack goes private with 0.4 SDK

By Sean Kerner   |    May 28, 2010

jetpack_smaller.jpg
From the 'Private Add-ons' files:

Mozilla developers are continuing to work on the next generation Jetpack add-on technology and this week released the 0.4 milestone.

While the Jetpack SDK is still a work in progress, it is set to be an integrated part of the upcoming Firefox 4 release -- though that release is still months away.

With the 0.3 Jetpack SDK context awareness was added. With this new 0.4 SDK, Jetpack is adding a few really interesting capabilities among them is the ability to detect if a browser is in Private Browsing Mode (aka porn mode).

The ability for Jetpack to detect and work with Private Browsing is an interesting one. Potentially I could see that it will mean that Jetpack add-ons will be able to work in tandem with Private Browsing such that cookies and other history information isn't stored when they shouldn't be.

Sugar on a Stick v.4 gets sweeter with Fedora Linux

By Sean Kerner   |    May 28, 2010

fedora-sugar-small.gif
From the 'Linux Spins' files:

One of the important things to remember with Fedora Linux is that it is a platform on which solutions can be built. One such solution is the Sugar On A Stick Fedora Linux spin which adds the Sugar learning environment originally built for the One Laptop Per Child effort to Fedora and packages it up so it'll run on a USB stick.

This week, following the release the Fedora 13, the Sugar On A Stick v.3 release came out, codenamed 'Mirabelle'. The Mirabelle release follows the Blackberry v.2 release that came out last December.

So what's new?

Well obviously Mirabelle benefits from all the changes in Fedora 13 for stability and ease of use. On the Sugar side the new release is now making it easier for users to customize the activities/applications that run on the stick.

Novell openSUSE 11.3 and Build Service near release

By Sean Kerner   |    May 26, 2010

opensuse.small.jpg
From the 'Gecko On The Move' files:

Novell's openSUSE developers have been a busy bunch this week.

Today openSUSE 11.3 milestone 7 is out which is the last stop ahead of a Release Candidate in June and a final release in July. The openSUSE Build Service 2.0 Beta 1 came out yesterday with a final release now set for June 10.

The two releases are somewhat linked in a way.

The openSUSE Build Service is a really neat offering was first announced back in 2007 and hit its 1.0 release in 2008. What the build service enables is any open source develop to build Linux packages for openSUSE as well as rival distros Debian, Ubuntu, Fedora and Mandriva. The Build Service is also used in the actual building of the openSUSE distribution itself.

With the 2.0 version, new enhancements are set to make the whole developer experience easier and faster.

"The new Web UI makes your work more efficient as stuff is where it is expected, it loads faster and gives you a better access to projects, packages and meta data," Andreas Jaeger, program manager at openSUSE wrote in a mailing list posting.

Is VP8 open source?

By Sean Kerner   |    May 25, 2010

webm.png
From the 'Open Source Rules' files:

One of the big pieces of news out of the Google I/O conference last week was the open sourcing of the VP8 video codec.

With VP8, the promise from Google is a video codec on par with H.264 that will be available royalty-free. It's a good idea, but there might be a problem.

The license that Google has chosen for VP8 isn't technically an approved license from the OSI (Open Source Initiative), the group that decides what is and what isn't a bona fide open source license.  At least that's the view of former Sun Chief Open Source Officer Simon Phipps.

The actual license that Google is using hasn't been submitted and approved by the OSI, which is the big issue. At it's core, as far as I can tell in my own personal opinion, the VP8 license should be considered open source in spirit, if not Open Source by definition. Google is making the code available and making a royalty patent license to any user. On the surface that looks to me to be as good as what the Theora codec is currently providing.

Google Chrome hits version 6 ?!

By Sean Kerner   |    May 21, 2010

googlechromologo.jpg
From the 'Fastest Browser Release Cycle Ever' files:

As a Chrome user, I don't actually notice the version number I use all that much, since it's always auto-updated to the latest version (as opposed to say Firefox where users actually have click something to update). That's part of the reason why I was somewhat surprised to see Chrome version 6 as the latest dev-channel release.

Didn't Google just hit version 5 in February?

Well as of May 20th, Chrome dev-channel is at version 6.0.408.1. That's right Chrome has advanced another full version number in about four months - for devs at least.

Without a doubt, this is one of the fastest browsing naming efforts I have ever seen. At this rate Chrome 8 or 9 will be out by the end of the year. Is Google trying to have Chrome 9 out at the same time as IE 9?

As for why Google has made the version number jump - it's a question that I'm not the only one asking. Multiple commentators on the Google Chrome release blog have the same question.

Basically what the answer comes down to is Google's project release methodology for Chrome. The dev-channel puts out weekly releases (sometimes more if there is a particularly nasty bug fix required).

"The Dev tree becomes 6.0 after 5.0 is essentially done and the new 6.0 features then begin to show up and be tested," commenter Chris Jones wrote. "People keep forgetting that we're pretty nearly running real time for feature additions. They don't add a bunch of features and sit on it as they're doing that for a big "all new features" dev release.

The 6.0 is indicative of what's about to be added in upcoming weeks. Do notice that the 5.0 new features had pretty much tapered off weeks ago."

The Chrome 5 Beta came out at the beginning of May, and with it came  the inclusion of an integrated Flash player as well as some serious performance gains.

Is it enough to say that a new version number is now warranted? Personally I don't think so. That said, Chrome does have a rapid pace of innovation, and Google's ridiculously rapid numbering of new browsers helps to fuel that too.

It's time for an Open Web App store

By Sean Kerner   |    May 20, 2010

cwslogo-chrome.png
From the 'Walled Garden' files:

Yesterday, Google announced its chrome web store as a way to highlight web apps for Chrome.

The desire for the app store is model is one that Apple has already proven to be successful (albeit tightly controlled). With Google's entry into the space for web apps, there are some larger questions that need to be asked  about the need for app stores and the need for openness in those stores.

Mozilla is also looking at the app store model with critical eyes. In a blog post today, Mozilla developer Jay Sulivan outlined a number of high-level principles that should guide the Open Web App store model.

Among Sullivan's principles is that the open app store hosts standards based apps that don't lock-in users. He also has a few great ideas on the whole approval process for app entry to an open store.

Firefox FUD not lagging

By Sean Kerner   |    May 19, 2010

sr-firefox3.jpg
From the 'Ex-Employee FUD' files:

Five years ago, most Windows users were only aware of IE  as a web browser that was available to them. Since then Firefox versions 1, 2, 3, 3.5 and 3.6  have altered the landscape forcing Microsoft to innovate IE in versions 7 and 8 and creating new competition from Apple Safari and Google Chrome.

While Firefox still trails IE and isn't likely to ever regain the dominant status that its predecessor Netscape once had, it is still by most accounts and measures the number two browser (by market share) in the world.

Can Firefox's innovation and growth curve continue?

In a comment attributed to former Firefox developer Blake Ross, apparently not. Ross allegedly thinks that Mozilla has become too passive in order to release products quickly. I personally don't agree with that assessment of Firefox.

Fedora 13 delay fixes Linux GRUB bug

By Sean Kerner   |    May 18, 2010

fedora-logo.png
From the 'Bad GRUB' files:

No, Fedora 13 Linux is not coming out today (and if you read my blog last week you'd know that!). The release was pushed back by a week to fix a number of blocker bugs, among them is an interesting one that many users might simply consider to be just a papercut nuisance.

GRUB (GRand Unified Bootloader), the loader that most Linux users see when they first boot up, had a very small issue in it that might have affected users. For some users (myself included) GRUB is a critical screen where you can select which operating system you want to load. That could be a different Linux kernel on the same OS, a different Linux distribution altogether or even Windows.

Usually the way GRUB works by default is the user gets to select which kernel/OS they want to load with a time delay before a default kernel OS loads. That's where the GRUB bug comes into play.

"GRUB bootloader should have a
few seconds delay on a multi-boot setup," Rahul Sundaram
wrote in a Red Hat bugzilla entry. "GRUB by default is configured in Fedora to not show the menu and have zero delay but in a multi-boot environment where Anaconda detects Windows and configures GRUB to boot Windows as a alternative, GRUB was configured by Anaconda to have a few seconds delay and show the menu so that the user can choose between Fedora and Windows. Unfortunately this is not the case with Fedora 13 RC2 anymore. This results in a non-technical user unable to boot into Windows after installing Fedora 13."

So to recap, one of the reasons for the Fedora 13 delay was to help new users that want to run both Windows and Linux on the same computer. Yes it's a small bug, but it could have been a big deal for new users. Making the new user Linux experience as easy as possible, especially for those coming from Windows is a good idea and one that will likely help Linux adoption.

Zend Raises Another $9 million - For What?

By Sean Kerner   |    May 17, 2010

zendlogo.gif
From the 'Funding Mature Companies' files:

Zend, one of the lead commercial PHP vendors, today announced that they have received another $9 million in venture capital funding, led by Greylock Partners.

While that certainly sounds like good news on the surface - an open source vendor getting money etc - Zend isn't exactly a startup still is it? Zend has been a successful (as far as I can tell) vendor for over 6 years.

According to Zend, they had record sales in 2009, adding 4,000 news customers. They now claim to have more than 1 million registered users for its PHP solutions which include the Zend Studio IDE and Zend Server PHP efforts.

So my question is - if Zend is already successful, both in terms of technology and customers why does it still need additional venture funding? At this point, shouldn't revenue and cash flow from operations be enough to keep the company going?

Pidgin's :) MSN Issue Triggers Open Source IM Update

By Sean Kerner   |    May 13, 2010

pidgingif.gif
From the 'Smiles That Crash' files:

The open source pidgin instant messaging client has a new update today with version 2.7.0.

Pidgin is an important open source project in that it is included in multiple Linux distributions including Red Hat Enterprise Linux and is also available for Windows users as well.

While Pidgin is often updated for protocol and bug fixes (and there are lots in Pidgin 2.7.0), one key item on the changelog stood out to me.  It's a security fix for CVE-2010-1624 which is a custom emoticon remote crash for the MSN messenger protocol.

"A vulnerability was discovered in libpurple's MSN protocol plugin that
can cause a denial of service (crash) due to insufficient validation of
certain SLP packets related to custom emoticons," Pidgin's security advisory on the issue states. "An attacker could use
this vulnerability to remotely crash a client using libpurple for MSN."

So no, it's not the simple :) smiley face emoticon that could have triggered the vulnerability (that's not a custom emoticon). That said, it never ceases to amaze me how seemingly harmless items (it's an emoticon!!) can lead to security vulnerabilities.

Fedora 13 Linux Release Slips By A Week

By Sean Kerner   |    May 12, 2010

fedora-logo.png
From the 'Go/No Go' files:

T-7 and counting....no wait make that T-14 for the launch of Fedora 13, codenamed Goddard.

Fedora 13 had been scheduled for release next week, and is now being pushed back by a week to Tuesday May 25th. The release was at a go/no go decision point based on the quality release criteria in place for Fedora.

I spoke with Fedora Project Leader Paul Frields the other week and he mentioned to me that Fedora 13's release date would slip if the blocker bugs had not been fixed in time. For Fedora releasing on time is less important than releasing a Linux distribution that works.

Blocker bugs are by definition bugs that block a release because they're in some way blocking user/system functionality in a negative way. Sure there are always bugs beyond the blocker that often get fixed in the days/weeks following a release -- but the point is that the 'out-of-the-box' experience for an upgrade or new install should just work.

The transparency in Fedora release process is one that I personally
admire a great deal. It provides insight into how the distribution
release process is determined, not by arbitrary decisions but rather by
adherence to pre-determined criteria.

So congratulations to Fedora and its leadership for taking the time and care to ensure that the Fedora 13 release works on its release day. I have little doubt that the Fedora community will mind waiting the extra 7 days while the bugs get fixed. Releasing when ready is always the best policy.

Fedora 14 Linux Headed To Laughlin. Next Stop Vegas?

By Sean Kerner   |    May 11, 2010

fedora-logo.png
From the 'Linux Geography' files:

We now know the name for Fedora 14, the next major version of Fedora which should be out by the end of 2010. Fedora contributors have voted and they voted for the name -- Laughlin.

Laughlin got 610 votes with Laramie coming in second at 507. My personal favorite -- Fytnargin -- came in dead last at 403 votes.

Fedora is currently in the last stages of the release process for Fedora 13, codenamed Goddard, due out by the end of this month. The way that Fedora naming works is that there is supposed to be some kind of connection (however loose) between the name of one release and the next.

So what's the connection between Goddard and Laughlin?

"Robert Goddard was a professor of physics, and so is Robert Laughlin," Fedora's wiki states."He was awarded the 1998 Nobel Prize in physics for his explanation of the fractional quantum Hall effect. Moreover he argues for emergence which is a concept that says 'The whole is more than the sum of its parts'. Fedora is more than the sum of its software."

Laughlin is also a town in Nevada, which easily sets up at one name voting option for Fedora 15 in 2011. How about the Fedora 15 Vegas Release? Viva Fedora?

Peppermint Linux Mixes Ubuntu, LXDE and Prism

By Sean Kerner   |    May 10, 2010

peppermint_small.jpg
From the 'Yet Another Distro' files:

A new Linux distro is officially hitting its 1.0 release today, but you'll have to forgive me for not being too excited. Peppermint OS is now out, with the aim of being a fast, cloud focused Linux operating system.

A quick look under the hood reveals that it's mostly a mashup of technologies that are available (and I'm already running), with an Ubuntu base (by way of the Linux Mint Ubuntu derivative) and leveraging the lightweight LXDE desktop. Ubuntu by default uses GNOME which is bulkier and consumes a larger resource footprint than LXDE.

If you happen to be running Ubuntu today, get LXDE and you'll notice a difference (especially on older hardware). LXDE also is available in other distro repositories, including Fedora -- so if you're looking for a desktop speed boost and are already running a distro, don't be too quick to repartition your drive for a peppermint install.

The cloud application part of peppermint also caught my attention, but a look under the hook reveals that it's also something that existing Linux -- and yes even Windows users -- are already using with Mozilla Prism.

Mozilla Prism is a Mozilla add-on that creates a standalone browser window for an application. It works very well and I personally have a kiosk type setup in my own environment that is all Prism based.

Google Chrome also offers a similar capability for a standalone browser/app setup and is at the core of Google's Chrome OS plans which is -- a cloud focused OS.

Don't get me wrong here though, Peppermint OS is an interesting idea and one takes advantage of the current hot trends in computing. Other have kinda/sorta tried this route before (remember gOS?) and it's certainly a good idea. Just be aware that you don't necessarily have to switch distro's to get the benefits of a 'cloud OS' though, with a few tweaks you may be able to leverage what you're already using.

Mozilla Jetpack 0.3 SDK Provides Context

By Sean Kerner   |    May 07, 2010

jetpack_smaller.jpg
From the 'Next Gen Add-Ons' files:

I've had an interest in the Mozilla Jetpack effort for about a year now. It's a project that will eventually represent the next generation of add-on extension technology for Firefox.

It's also a project that has undergone some change recently as the project shifts to an SDK that that has recently hit version 0.3.

The big new item in the 0.3 Jetpack SDK is a context menu API which is one of the cooler things I've seen yet out of Jetpack.

"The context-menu API lets add-ons add items to the context menu for
web pages, so add-ons can give users additional options when they
context-click on elements in pages (links, images, etc.) or the pages
themselves," Mozilla developer Myk Melez blogged.

Mozilla staffer Daniel Buchner has put up a post on how the context menu API can be leveraged to build a menu that will let users launch a web image editing tool. I personally can see a whole lot of opportunities where this API can be easily leveraged.

While the ability to add context menus to a side (or an operating system
even) is hardly a new concept, with Jetpack it now becomes (in my
opinion) trivially easy to build.

Work is still ongoing for Jetpack with a 0.4 SDK release in the works. Frankly I can hardly wait until this tech finally gets baked into a mainstream Firefox release so that it gets the full attention it deserves.

What Will Fedora 14 Linux be Named?

By Sean Kerner   |    May 06, 2010

fedora-logo.png
From the "Open Source Democracy' files:

Some Linux distributions are named by benevolent-dictators-for-life (Ubuntu). In the case of Fedora, the choice of distro name is one that is voted on (and suggested) by the community.

With Fedora 13 (Goddard) nearing release this month, the Fedora community is now tasked with voting a name for Fedora 14. The list has now been narrowed down to six names which community members get to vote on until 10 May 2010.

The six names up for vote are:

  • Fytnargin
  • Hoppin
  • Laramie
  • Laughlin
  • Mitikas
  • Ventno

Tough list, but I like Fytnargin. The rationale for the name choice is an interesting one.

"Goddard is a mountain with elevation 4134m and so is Fytnargin in Russia," Fedora's name suggestion wiki states.

You see, there has to be some kind of connection between one release name and another. No, the name of the release doesn't have any direct correlation with the technology, but it's an interesting tangent to follow.

Why Did Tom Noonan Leave IBM ISS?

By Sean Kerner   |    May 05, 2010

ibm.big.jpg
From the 'Walking from $1.3 Billion' files:

Tom Noonan successfully sold his former company Internet Security Systems (ISS) to IBM for $1.3 billion in 2006.

Some 18 months after the acquisition closed, he left the company and this year officially took the helm of energy management startup JouleX.

I recently met up with Noonan to talk with him about JouleX, but I also managed to squeeze in a short conversation about his time at ISS and why he left IBM. While it's not uncommon for founders to leave their companies after being acquired, I'm always curious to dig a little deeper and find out how the acquisition and integration actually happened.

In Noonan's case, he told me he was offered some monies by IBM to stick around, but in the end it wasn't enough for him to stay. He also explains how ISS operations have been integrated into IBM and overall in his view IBM appears to be doing well with ISS so far.

Check out the clip from my interview with Noonan, for all the details:

Riverbed and the Open Source Flamebox

By Sean Kerner   |    May 04, 2010

riverbed-small.gif

From the 'Open Source in Networking' files:

While at the Interop Las Vegas show, I had the opportunity to sit down with Riverbed's Chief Scientist Mark Day.

As you might expect, we chatted about the cloud and Riverbed's core WAN acceleration market which continues to grow.

We also talked about the open source model and how it applies to a proprietary networking vendor like Riverbed.

Yes, Riverbed in some ways leverages open source and also contributes some testing components back. One of their bigger open source projects is something that internally carries the codename of 'Flamebox'.

Here's a clip of the conversation I had with Day:

Are Apple's Open Source Theora Claims Patent FUD?

By Sean Kerner   |    May 03, 2010

theora_org.png
From the 'Allegedly Written by Apple's CEO' files:

There has been a bit of hype in recent days about video codecs and potential patent infringements. My colleague David Needle has published a great story, about an alleged claim made by Apple CEO Steve Jobs that the Theora open source video codec is at risk from vendors that are assembling a patent pool against it.

Is this all FUD? Or is there something that Theora users and developers need to worry about?

The answer is NOT quite as clear cut as you might think. At least one Theora developer isn't worried (yet), based largely on the lack of credible evidence so far. Jobs' claims was reprinted on other websites and was not directly posted on Apple.com (unlike Jobs tirade against Flash last week).  Theora developer Gregory Maxwell noted in a Theora developer list posting that it would also have made sense for someone from Apple (or the assembled patent pool) to have contacted Theora about the claims.

"Since the developers of Theora have received no such contact, I can
only conclude that no such effort is being undertaken and that the
quoted statement is either a forgery, the result of a
misunderstanding, or that the statement may be indicative of a
dishonest and anti-competitive collusion by Apple and other H.264
patent holders to interfere which the development, promotion, and
utilization of unencumbered media standards," Maxwell wrote.

Maxwell's comments aside, I know that I personally contacted Theora's handlers (Xiph.org) at multiple points on Friday about the alleged patent issue and did not receive the courtesy of any response from them. If Apple or other patent holder had a similiar experience, perhaps they have been contacted and Maxwell just doesn't know about it (yet).