Google's announcement that it is working on a lightweight, Web-based operating system for netbooks, to be called Chrome OS, is a surprise only in its timing. As I wrote last September, when Google released the Chrome browser and Sergey Brin denied that its ambitions went beyond building a fast, simple browser:

Don't believe it for a second. Although the first version of Chrome has a half-finished feel and runs only on Windows, a close look at its features and underlying design reveals a far more dramatic goal. Chrome aims to take on not just Internet Explorer's 75% share of the browser market but Windows' dominance of the desktop itself.

Chrome was designed less as a competitor to the feature-rich Internet Explorer and Firefox than as a container for running Web-based applications. That made it, in effect, the user interface for a Web-based OS. Add a kernel (Google, unsurprisingly, is using the Linux kernel as the core), a window manager, and assorted other pieces of OS infrastructure and you can have a simple, fast, and robust operating system without a massive development effort.

After a long beta test, Mozilla today released Firefox 3.5 for windows, Mac, and Linux. Though the new version has some nice features, it's potential may only be revealed over time.

The latest edition of Firefox offers a considerable boost in speed, especially in running JavaScript programs. I don't do any formal testing, but it is definitely much faster that Microsoft's Internet Explorer 8 (which is not saying all that much) on Windows, and seems roughly on a par with Google's Chrome 2.0 and Apple's Safari 4.0.

We are running a Linux HTTPS Server, and wanted to add to the IP Monitor. The way we wanted to add was by using username and password and see if IP Monitor can actually login in to the server.

Well, that was quick.

Yesterday, I blogged about an interesting confrontation in the making between Microsoft and its fastest-growing Windows licensee: Acer. It's become clear that while Acer wants to light a fire under the new Netbook category of cheap computing devices to drive volume hardware sales, Microsoft--having largely coopted this low-end market from the early Linux-based variants that dominated the market a year ago--now would rather make netbooks go away and convince consumers they are nothing more than plain old PCs, albeit with somewhat smaller screens and somewhat higher price tags. Now, given comments made by Microsoft manager Steven Guggenheim at the Computex 2009 show in Taipei today, this is the official position of the company: that the phrase netbook is meaningless.

I doubt it will be so easy to get consumers to give up on the idea of buying a device costing $200 or $300, rather than $500 or more. Once established, new low price points rarely just disappear.

Dell may be cooking up a series of netbooks based on Android software. On May 6, software company Bsquare announced that "it is porting Adobe's Flash Lite 3.17 technology onto Dell Netbooks running Google's Android platform."

Dell hasn't officially announced any netbooks based on Android, which is an operating system developed by a slew of companies including Google. Dell hasn't yet returned my request for comment.

But I wouldn't be surprised if the Bsquare leak holds water. After all, Dell has long put out computers based on open-source operating system Linux. Some people prefer them. Plus, they can be cheaper than machines based on Microsoft's Windows. Android is just the latest variant of the open-source software, and lots of companies, including Hewlett-Packard, are testing it out. So, why not Dell?

Clearly, this is great news for Google, and bad news for Microsoft, the maker of Windows operating system that dominates the market today. If Dell jumps onto the Android bandwagon, the move could push other companies toward Android as well.

Surprisingly there doesn’t seem to be much discussion of this concept anywhere that I have searched, or I am just searching the wrong way.

With the new PCI (Payment Card Industry, https://www.pcisecuritystandards.org/) guidelines/rules regarding the handling of credit card information, it would seem important that there be a way to safely store call recordings that contain credit card numbers.

Has anyone incorporated any sort of encryption of their call recording files? Specifically the individual files.

Yes, one could encrypt the entire drive/partition, but I believe that would not protect the files while the drive/partition is mounted and active. Anyone with access to the system could still listen to the recordings.

Encrypting the individual files would make it safer to provide access to individual recordings for QA purposes as well. Only those authorized would be able to decrypt them and listen to them.

With some post-call handling, I can use PGP to encrypt the recordings, but listening to them afterwards becomes a bit of a pain for someone with little technical skills. If a client needs to QA a call, they have to then decrypt the file before they can play it. While not impossible to handle, it adds an extra level of complexity that not everyone appreciates.

I’m kind of hoping someone’s heard of or used some solution that would handle this.

We currently record in WAV with an eye towards converting to mp3, ogg, or flac for archival purposes. Since the encrypted files would not be playable with regular media players, I’m thinking of finding someone who could create a plugin for VLC or something like that, which would accept the encrypted file, decrypt it with the proper key, play it and then get rid of the temp data.

We currently use Trixbox CE 2.2, but are exploring an upgrade to a newer version. It would be preferable that any solution work regardless of the Trixbox/Asterisk version, though.

Thoughts and comments appreciated.

Thanks in advance.

DionV

We are pleased to announce a major new release of the Digium G.729 codec module for Asterisk. This release incorporates a number of changes, and in addition, includes a new performance testing tool which will make it easier for users to choose the best ‘flavor’ of the module for the particular system they are installing it on.

The changes are:

- Version Numbering

The module’s version numbering is now more in line with our other products; the numbers will be in the form ‘X.Y.Z_A.B.C’, with the ‘X.Y.Z’ component representing the version of Asterisk the module is intended to operate in, and the ‘A.B.C’ component representing the actual version of the codec module code itself. Please note that the 1.6.0 modules will work in any Asterisk 1.6 release from 1.6.0 onwards, until such time as a new version is needed; when that occurs, we will provide 1.6.0 modules *and* modules designed for the later releases.

- Optimization

The new modules were built using the latest (4.3) release of the GNU C compiler, and use a number of new optimization methods available in that release to significantly improve encoding/decoding performance. In addition, we now build the modules in both 32-bit and 64-bit varieties for each CPU flavor that support both modes, so that users with 64-bit CPUs running a 32-bit Linux installation can still have a version of the module optimized for their CPU type. GCC 4.3 also provides optimizations for some newer CPU families (Intel Core2, AMD Barcelona, and others), so we’ve made those flavors available with this release.

- Performance Testing

In the past, we’ve offered different CPU flavors of the module to hopefully provide the best encoding/decoding performance, but it was up to the user to install and test each flavor to determine which one provided the best performance. This process is difficult and time consuming, and did not always provide reliable results. To help with this situation, we’ve now released a tool called ‘benchg729′, which can be run on the target system and will execute encoding speed tests (using a real audio sample) for each CPU flavor that we offer for that platform, and then report the results and suggest the best performing module flavor for that system. Note that the use of this tool requires that the system have at least one valid channel license for the Digium
G.729 codec installed; it will not run without a license.

The new codec modules and benchg729 tool are now available at:

http://downloads.digium.com/pub/telephony/codec_g729

New codec modules are available for Asterisk 1.2, Asterisk 1.4 and Asterisk 1.6.x on both x86-32 and x86-64 Linux platforms, and the benchg729 tool is also available for x86-32 and x86-64 Linux platforms.

We hope these product updates improve your system performance, and as always, we thank you for supporting Asterisk and using Digium products!

Verizon Wireless launched a VOIP-based, Linux-powered home phone on Friday, the Verizon Hub. The Hub plugs into a home broadband line and acts as a family calendar, limited Web browser, messaging center, digital picture frame and, of course, a phone.

Steaming Open Cup: “To develop a visual representation in your mind, imagine your newly installed Ubuntu 8.10 with all its fascinating features. Now, try to imagine installing all that functionality into your phone. Is that possible? No. Not by a long shot.”

— Dec. 15, 2008 — VirtualLogix, which sells Linux-compatible “VLX” virtualization software designed for embedded devices, announced the selection of a new executive chairman after the departure of CEO Peter Richards.

— Dec. 15, 2008 — Ronetix announced the availability of new PowerPC firmware for its Linux-friendly JTAG/BDM emulator and flash programmer. The PEEDI JTAG/BDM Emulator and Flash Programmer now supports the Freescale PowerQUICC II Pro MPC83XX processor, says the vendor. …

The Free Software Foundation (FSF) has sued Cisco for copyright infringement. It says Cisco’s Linksys division uses GNU code and won’t honor the GPL 2 and LGPL 2 and 2.1 licenses that the GNU software’s distributed under and give customers the source code to the Cisco-modified embedded firmware that would of course let users in turn modify the software. FSF wants the profits that Cisco’s made …