WE RECOMMEND THE FOLLOWING PLANS

  • Advertise Here
  • Advertise Here
  • Advertise Here
  • Advertise Here

The Failed Promise of HTML5

Posted by eXactBot Hosting | News | Wednesday 29 October 2014 12:44 pm

From the ‘Standards That Aren’t’ files:

The W3C announced this week that the HTML5 specification is now an official recommendation. While I was an avid supporter of the HTML5 effort in the early days, seven years ago, you can count me among those that aren’t all that excited by the W3Cs announcement.

In the early days of the web, it was the W3C and the HTML standard that developers worked towards. The mess that was HTML 4 and HTML 4.1, was only abstracted in the early 2000’s by the fact there there was no browser vendor competition.

That changed thanks to Mozilla. Mozilla helped to innovate the web, not the W3C. It was Mozilla that first popularized the use of tabs (which are now commonplace), put an emphasis on JavaScript (that now powers most dynamic web HTML5 apps) and made the browser space exciting again.

The W3C and the HTML5 specification is not an innovation engine. Rather it is in my opinion that the W3C’s recommendation is a statement after the fact. Four years ago, when I still actively believed that the W3C was on the right track, I wrote a story about how HTML5 was already broken.

The problem then as it is now, is that browser vendors call the shots, not the W3C.

Even now with HTML5 an official recommended standard, it’s not implemented uniformly by the browser vendors. Google for example, has taken multiple steps to ensure that certain applications only work in Chrome and not other browsers (unless those browser use extra add-ons). That’s not necessarily the fault of the standard, but it does speak to the failed promise.

The promise of HTML is a standard that browser vendors and web developers can look too for building technology and sites. The challenge is that the implementation remains somewhat fragmented.

I don’t blame browser vendors entirely for the fragmentation either, I blame the standards process. In December of 2012, I wrote about the last call for HTML5, that was nearly two years ago. Work on HTML5 has been ongoing since at least 2007.

The modern web hasn’t stood still for the last 7 years. In the modern web, browser vendors iterate every six to ten weeks.

Certainly the use of the Canvas tag has changed the web, for the better. HTML5 enables the use of video and audio (though codecs are still a challenge). But when I think of the goodness of the modern web, much of it is enabled through faster CPUs, more powerful JavaScript engines and Cascading Style Sheets that enable forms of layout that were un-imagineable in the HTML 4 era.

My challenge with the way the W3C works and the real world is that instead of creating standards; W3C strives to enable consensus, with limited success.

As a standard HTML5 is also somewhat questionable. Try building an HTML5 website using the published specs and see if it will work and render the same in IE, Safari, Chrome, Firefox on both desktop, mobile, Linux, Windows and Mac OS X — I dare you.

As I see it, web standards are now evolving every six to eight weeks and the W3C is merely a bystander in the process.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

The Failed Promise of HTML5

Posted by eXactBot Hosting | News | Wednesday 29 October 2014 12:44 pm

From the ‘Standards That Aren’t’ files:

The W3C announced this week that the HTML5 specification is now an official recommendation. While I was an avid supporter of the HTML5 effort in the early days, seven years ago, you can count me among those that aren’t all that excited by the W3Cs announcement.

In the early days of the web, it was the W3C and the HTML standard that developers worked towards. The mess that was HTML 4 and HTML 4.1, was only abstracted in the early 2000’s by the fact there there was no browser vendor competition.

That changed thanks to Mozilla. Mozilla helped to innovate the web, not the W3C. It was Mozilla that first popularized the use of tabs (which are now commonplace), put an emphasis on JavaScript (that now powers most dynamic web HTML5 apps) and made the browser space exciting again.

The W3C and the HTML5 specification is not an innovation engine. Rather it is in my opinion that the W3C’s recommendation is a statement after the fact. Four years ago, when I still actively believed that the W3C was on the right track, I wrote a story about how HTML5 was already broken.

The problem then as it is now, is that browser vendors call the shots, not the W3C.

Even now with HTML5 an official recommended standard, it’s not implemented uniformly by the browser vendors. Google for example, has taken multiple steps to ensure that certain applications only work in Chrome and not other browsers (unless those browser use extra add-ons). That’s not necessarily the fault of the standard, but it does speak to the failed promise.

The promise of HTML is a standard that browser vendors and web developers can look too for building technology and sites. The challenge is that the implementation remains somewhat fragmented.

I don’t blame browser vendors entirely for the fragmentation either, I blame the standards process. In December of 2012, I wrote about the last call for HTML5, that was nearly two years ago. Work on HTML5 has been ongoing since at least 2007.

The modern web hasn’t stood still for the last 7 years. In the modern web, browser vendors iterate every six to ten weeks.

Certainly the use of the Canvas tag has changed the web, for the better. HTML5 enables the use of video and audio (though codecs are still a challenge). But when I think of the goodness of the modern web, much of it is enabled through faster CPUs, more powerful JavaScript engines and Cascading Style Sheets that enable forms of layout that were un-imagineable in the HTML 4 era.

My challenge with the way the W3C works and the real world is that instead of creating standards; W3C strives to enable consensus, with limited success.

As a standard HTML5 is also somewhat questionable. Try building an HTML5 website using the published specs and see if it will work and render the same in IE, Safari, Chrome, Firefox on both desktop, mobile, Linux, Windows and Mac OS X — I dare you.

As I see it, web standards are now evolving every six to eight weeks and the W3C is merely a bystander in the process.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

PHP 5.6.2 and 5.4.34 Update for Critical Security Flaws

Posted by eXactBot Hosting | News | Monday 20 October 2014 10:55 am

While much of the security world is consumed with the latest branded vulnerability (last week it was POODLE), the open-source PHP programming language fixed some very serious bugs.php

PHP is widely deployed across the Internet and is the language used to power much of the world’s leading Content Management Systems (CMS) and blogs (including this one).

In the PHP 5.6.2 update, four security vulnerabilities are being fixed including: CVE-2014-3668, CVE-2014-3669 and CVE-2014-3670. Bug #68089 does not yet have a CVE number but it’s a non-trivial Null byte injection flaw.

PHP 5.4.34 is being patched for six vulnerabilities including CVE-2014-3668, CVE-2014-3669 and CVE-2014-3670. The non-CVE number issues include bug #66242, 67985, 68089 and 41631.

Across both PHP 5.4.x and PHP 5.6 updates, the CVE-2014-3669 is one of the most serious.

“An integer overflow flaw in PHP’s unserialize() function was reported, a Red Hat security advisory warns. “If unserialize() were used on untrusted data, this issue could lead to a crash or potentially information disclosure.”

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

Firefox Hello Not Working and Mozilla Claims the Bug is Invalid

Posted by eXactBot Hosting | News | Sunday 19 October 2014 10:50 am

Mozilla announced the Firefox 34 Beta release on October 17 and a key highlight is the new Firefox Hello feature. Firefox Hello is supposed to enable users to simply use the browser to be able to call each other.

It’s a feature that leverages the emerging WebRTC standard and a feature that I was veryfirefox eager to try out.

Problem is it, that it didn’t work on the first try.

So I tried again on Mac and then again on a Windows machine too (my first try is always Linux) and still no dice. The feature just didn’t show up.

So I complained to @firefox on Twitter and still no response. Apparently responding to features not working is not within Mozilla’s social mandate of creating an open web.

A little bit of research let me to Buzilla entry 1083525, titled “Hello button does not appear in palette if customized prior to FFx34”

That bug provides details on a workaround that worked for me.

As per Adam Roach:

1. Start an older released version of Firefox (e.g. 33)

2. Customize the toolbar buttons

3. Start Beta 34 using the same profile

4. Go to about:config and change “loop:throttled” to “false”

5. Restart Beta 34

6. Open the customization palette.

Unfortunately another Mozilla developer has deemed this bug to be invalid. I reached out to Mozilla PR on this issue too and got back a somewhat unsatisfactory answer.

“Firefox Beta users will be seeing our experimental Firefox Hello feature appear over the next few weeks, as we slowly roll it out,” Chad Weiner, Director of Product Management at Mozilla stated. “This is done in stages so as to allow us to test the experience carefully, as a result not everyone will see the feature at the same time. We will work on any bugs raised as part of the ongoing Firefox Hello experiment.”

So, we have a new feature that doesn’t show up for some users, with a bug that apparently isn’t valid and Mozilla product management cautioning about a staged deployment.

On Bugzilla, developer Matthew N responded to me with a somewhat more technically articulate answer :

“This bug is about what happens when Loop is unthrottled or if your build gets though the throttling. It’s intentional that you won’t see it at all until that point.”

Truth is of course, that this is a beta release, so bugs are to be expected. I for one hope that Mozilla developers recognize that announcing a feature that isn’t available to the users you want to be testing your builds, is not necessarily a good best practice.

I’m cautiously optimistic that any bugs and the loop throttling issue are all resolved in the next five or so weeks, so the GA release of Firefox 34 will truly enable a new era for the modern web.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

Firefox Hello Not Working and Mozilla Claims the Bug is Invalid

Posted by eXactBot Hosting | News | Sunday 19 October 2014 10:50 am

Mozilla announced the Firefox 34 Beta release on October 17 and a key highlight is the new Firefox Hello feature. Firefox Hello is supposed to enable users to simply use the browser to be able to call each other.

It’s a feature that leverages the emerging WebRTC standard and a feature that I was veryfirefox eager to try out.

Problem is it, that it didn’t work on the first try.

So I tried again on Mac and then again on a Windows machine too (my first try is always Linux) and still no dice. The feature just didn’t show up.

So I complained to @firefox on Twitter and still no response. Apparently responding to features not working is not within Mozilla’s social mandate of creating an open web.

A little bit of research let me to Buzilla entry 1083525, titled “Hello button does not appear in palette if customized prior to FFx34”

That bug provides details on a workaround that worked for me.

As per Adam Roach:

1. Start an older released version of Firefox (e.g. 33)

2. Customize the toolbar buttons

3. Start Beta 34 using the same profile

4. Go to about:config and change “loop:throttled” to “false”

5. Restart Beta 34

6. Open the customization palette.

Unfortunately another Mozilla developer has deemed this bug to be invalid. I reached out to Mozilla PR on this issue too and got back a somewhat unsatisfactory answer.

“Firefox Beta users will be seeing our experimental Firefox Hello feature appear over the next few weeks, as we slowly roll it out,” Chad Weiner, Director of Product Management at Mozilla stated. “This is done in stages so as to allow us to test the experience carefully, as a result not everyone will see the feature at the same time. We will work on any bugs raised as part of the ongoing Firefox Hello experiment.”

So, we have a new feature that doesn’t show up for some users, with a bug that apparently isn’t valid and Mozilla product management cautioning about a staged deployment.

On Bugzilla, developer Matthew N responded to me with a somewhat more technically articulate answer :

“This bug is about what happens when Loop is unthrottled or if your build gets though the throttling. It’s intentional that you won’t see it at all until that point.”

Truth is of course, that this is a beta release, so bugs are to be expected. I for one hope that Mozilla developers recognize that announcing a feature that isn’t available to the users you want to be testing your builds, is not necessarily a good best practice.

I’m cautiously optimistic that any bugs and the loop throttling issue are all resolved in the next five or so weeks, so the GA release of Firefox 34 will truly enable a new era for the modern web.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

Docker 1.3 Improves Container Security

Posted by eXactBot Hosting | News | Friday 17 October 2014 10:24 am

The open-source Docker container virtualization project is out this week with the Docker Engine 1.3 release, providing some really important security features.

Docker Founder Solomon Hykes detailed many of the big updates in Docker Enginer 1.3 during a Linuxcon Dockerkeynote in August.

Among the big additions is the ability to check images with the using of a digital signature. By having the digital signature, it provides users with an additional layer of confidence to know that an image has not been tampered with.

There are also new security labeling options that can enable control with SELinux and AppArmor. Both SELinux and AppArmor provide fine-grained system controls for what running processes can and cannot do on a system.

Looking beyond just security, Docker Engine 1.3 also has one really interesting new command that should be a major boost for container orchestration and control. That command is ‘docker-exec’ and it enables an administrator to run a process in an existing, active container.

“With docker exec, you’ll be able to do things like add or remove devices from running containers, debug running containers, and run commands that are not part of the container’s static specification,” the Docker 1.3 release notes states.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

3.8 Million Raspberry Pi Linux Computers Sold

Posted by eXactBot Hosting | News | Monday 13 October 2014 9:44 am

From the ‘Little Computer that Could’ files:

I own three Raspberry Pi’s (two B’s and one B+) and many people I know also own one or more Pis. All those Pi add up and now the Raspberry Pi Foundation says that it has sold 3.8 million units.

That’s a whole lot of Pi.

The Raspberry Pi was never supposed to be a massive volume seller. It was supposed to be a teaching and educational tool to help get kids (and adults) interested in development and maker culture.

A year ago at LinuxCon NewOrleans, Raspberry Pi Founder Eben Upton said that 1.7 million units had been sold. That means that in only a year, another 2 million units were sold.

By any estimation, the Raspberry Pi has been a stunning success.

There are no shortage of projects that a Raspberry Pi can enable. From the ever-popular media server (RaspBMC and OpenElec), to security (Onion Pi) and even as an air quality monitoring system (Air Pi).

What the Raspberry Pi has proven is that if you make low-cost, yet powerful, computing accessible, people will use it for all manner of wonderful things. The low-cost hardware of course is the key enabler, but it’s important to remember that so too is the ARM silicon and the open-source operating systems (Linux!) that run on that silicon.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist

OpenStack Juno Races to Completion as RC2s are Released

Posted by eXactBot Hosting | News | Wednesday 8 October 2014 8:42 am

From the ‘Not Far Now’ files:

The open-source OpenStack cloud platform is *almost* at its next major release, known as Juno — but first there all the release candidates.

Earlier today. the Ceilometer monitoring project released its RC2 milestone for Juno. The openstackKeystone project alsoreleased its RC2.

Yesterday, Sahara, the project formerly known as Savanah released its RC2. Sahara is particularly noteworthy for OpenStack Juno as this release will make Sahara’s official debut as an integrated project.

“Unless new release-critical issues are found that warrant a release candidate respin, this RC2 will be formally released as Sahara 2014.2 on October 16,” OpenStack Director of Engineering Thierry Carrez wrote.

Other project including Neutron, Horizon, Trove, Nova, Heat, Cinder, Glance and Swift all issued their RC1 releases last week.

Unless any last minute bugs show up, those RC1s will all be consider the final releases for Juno as well on October 16.

Sean Michael Kerner is a senior editor at InternetNews.com. Follow him on Twitter @TechJournalist