I’ve been following the Android Linux phone phenomenon with interest since it began, even going so far as to download the SDK and start learning to write Android apps. All I had left to do was choose the Android phone I wanted for myself. First I wanted the HTC Magic, but then I saw the Hero video on-line and fell in love. Since Australia is the A$$ end of the world as far as tech companies are concerned, it took so long for the Hero to arrive here that before they did, I fell in love with the Sony Ericsson Xperia X10, it had twice the screen resolution twice the CPU power, a fantastic camera with flash and a flashy interface. I wanted one badly. When it was announced, it was the most technically advanced android phone on the market.

Fast forward months and months and I don’t have one yet. Why? because nobody else does either. When Sony-Ericsson announced the phone, it’s features were unrivalled. Now however, it still isn’t out, but it’s features are no longer outstanding. Motorola’s Droid, the Google Nexus one and others have most or all of the features of the Xperia X10, the main difference being that you can actually buy a Droid or Nexus one right now if you want to. (not in Australia yet, but you get the idea.)

An added sting is that the Motorola Droid (2.0.1) and Google Nexus one (2.1) are shipping with much newer versions of Android than the Xperia X10 (1.6) so in some ways it’s behind the 8 ball before it even joins the race. Why have all the other phone companies that wanted to, been able to design, manufacture and release outstanding Android phones, except Ericsson? There is already talk about Snapdragon 1.5ghz CPU Android phones, so unless the X10 gets out the door really soon, it will totally underwhelm the market when it finally does arrive.

Don’t get me wrong, the specs on the Xperia X10 still make my mouth water, particularly the screen and the camera, but if I’m going to spend big bucks on a phone, I’ll want it to be at the forefront of technology for at least a few months and the way it looks to me now, by the time it does get here, it will already be outdated. This phone had the potential to change the game for Sony-Ericsson, and now it is likely to under perform. Not because it’s a bad phone, just because the hype is gone and it isnt’ really special anymore.

Sad really, I had my heart set on one of these things and now I’m looking into buying a Google Nexus one.
Sony-Ericsson have had a hard time of it lately and if this is any indication, it’s not difficult to see why in my opinion. Building hype is great, but if you don’t actually sell the phone at the hypes peak, it declines. That is what happened here, the Xperia X10 took so long to come to market that it was overtaken by Motorola and Google who are now beneifiting from much of the hype generated by the X10 video’s and reviews all over the net.

I think these phones and the iphone are the future of the web for us, or at least a big part of it. There are a lot more mobile phones in the world than there are computers so the market is huge for the mobile Internet. My prediction is that 5 years from now, 60+ percent of the smart-phone market will be running on Linux/Unix based phones. Most of these will be used for browsing the net as well as social networking. It’s worth making sure your sites look OK on them.

It seems that not all of the users of Firefox 3 were previous users of Firefox 2 upgrading to the newer version. Prior to the release of Firefox 3, the Mozilla browser had roughly 48% of our browser marketshare, making it the clear leader. Post Firefox 3 release, our Mozilla Firefox usage is up to nearly 65%. It remains to be seen how this trend shows up on other sites but it’s looking good for Firefox to make up some significant ground overall. We’ll know more not long after the end of July.

In less than 24 hours, the latest release of Firefox has been downloaded more than 8 million (8000,000) times.

So far the reviews have all been positive, and while many themes and extensions are not yet available for 3, many have already been ported over and more are available every day. Additionally the default theme for Firefox 3 is much more eye catching than the corporate looking (read: boring) theme of Firefox 2.

Testing here at HTMLfixit.com over the release candidates and the final release show it to be an order of magnitude faster than Firefox 2 to load and to render, even on sub standard computers. It also doesn’t significantly increase it’s memory use over time as Firefox 2 did. The list of new and improved features is as long as your arm and has been detailed on dozens of sites so I won’t go into it here.

Interestingly Firefox 3 already accounts for nearly 10 percent of HTMLfixIT’s traffic, which goes to show that a good many of our techie visitors already know about it.

Secunia have reported that more flaws were found in Redhat Linux (633) than in Windows (123), but even a blind man can see it is nowhere near a fair comparison.

Redhat is made up of the core operating system, and thousands of third party applications that people can choose to install. (or not). 99% of the 633 security flaws found in Redhat Linux were in the third party applications, only 1% were in the core OS.

Windows however, only had 123 bugs, but 96% of them were in the core operating system. Since 3rd party apps are not supplied or supported by Microsoft however, all of their bugs did not get added to the total as they did in Redhat’s case.

Does anyone else think that this is perhaps not a fair comparison? I can tell you one thing, I’d rather have a core OS with 1% of 633 flaws (6.33), than one with 96% of 123 flaws 118.08. The OS results could just have easily been put “Windows had 118.08 more OS security flaws than Redhat Linux.”

With regards to Firefox, they also seem to be counting flaws that Mozilla have found themselves. We know they are not doing the same for IE, because Microsoft don’t announce flaws they find themselves. Again, not really a fair comparison.

Interesting however, is the patching statistics for IE and Firefox.

Out of eight zero-day bugs reported for Firefox in 2007, five have been patched, three of those in just over a week. Out of 10 zero-day IE bugs, only three were patched and the shortest patch time was 85 days.

(taken from here)

Microsoft’s best patch result was 85 days to release and only 3 out of 10 flaws patched, verses 5 out of 8 and just over a week for Firefox.

Statistics are all good and interesting, but taken in the wrong light, can paint a picture that is dangerously incorrect.

Microsoft risks alienating web developers with their unwillingness or inability to conform to standards and their secrecy surrounding Internet Explorer 8. I must admit to some amazement that they got to version 8 without web developers realizing that this is SOP for Microsoft and something they have done from the beginning. It seems that the only time you can really expect any openness, is when they are coming from behind and need some traction. When defending monopoly market share, they don’t seem to be able to do more than talk about interoperability, standards and the like. (ODF anyone?). Luckily for us, due to the ongoing success of Apple Safari & Mozilla Firefox, not to mention the growing range of Linux PC’s and laptops sold by giants like Asus and Walmart, Microsoft’s monopoly of the web browser is rather quickly being whittled away. That is great news as the Internet was designed to be useful to everyone, not to be held hostage by a commercial entity seemingly concerned with nothing but their own profits. (There is nothing wrong with being concerned about profits, as long as the actions stemming from such concern doesn’t prevent everyone else from striving to the same position.)

In other Microsoft news, apparently their hardware is just as vulnerable as their software has proven to be over the years. Using a Microsoft wireless keyboard can get your machine and all passwords used on it handed over to people some distance from you. Worse your machine need not even be connected to any network to be vulnerable. Apparently all you need do is use a Microsoft wireless keyboard and you are vulnerable.

In hardware news, Western Digital has released a fashionable range of external (and NAS) hard drives called “My Book”. They look cool, but apparently Western Digital has assumed that you want to pay them to police your files on your network. The client software (WD Anywhere Access) for the drives will not let you share a wide range of multimedia files on your network. Apparently you are to pay them handsomely for the privilege of the software you purchased presuming you are a criminal.

Lastly, could this finally be the time for of Desktop Linux getting widespread adoption? First we had Tivo running Linux, now we have a range of Linux PC’s showing up in huge shopping chains, then Google makes a bid to have Linux become the standard in mobile communication Operating Systems. Linux seems to be popping up everywhere. The majority of the top 500 super computers are running Linux (including #1) and now it’s going for the low end as well. It doesn’t matter if you are a fan of Windows, Linux or Apple, this is good news, because it means that Microsoft and possibly Apple will have to lower prices and compete when Linux gains widespread acceptance from the newbie crowd. In addition, everyone is looking to a future where software runs over the Internet, and for the underlying Operating System to be less and less significant. For that to happen no one company can be allowed to steer the direction by using existing market share. The rise of Apple, the growth of Linux and the innovation of Google are going to see that the future of personal (and mobile) computing is not steered by any one company.

Having used Linux on servers for many years now, I’m always keen to see if it has progressed enough for me to use as my “daily driver” desktop/laptop OS.

The test this time was on a NEC Centrino Laptop (1.6gig, 1 gig of ram 64MB Intel Video.) Could I get Linux installed and use it for both work and home and completely exclude Windows without making myself less productive in the process. The answer sad to say is no. But it was a really close thing and only one area really let the side down, but I’ll get to that in due course. I’ve been a big fan of CENTOS for my servers and since it’s an rpm based distro, I figured I’d stick with a desktop OS that has as much in common with CENTOS/RHEL as possible. To that end I’ve based this on Fedora Core 7.

1. Installation: Awesome, Better than Windows by far. If I had to pick anything wrong with it, it would be the partitioning tool. Except for the really basic stuff, it is very “non obvious”. Mandriva makes a far better GUI frontend tool for partitioning, it’s a shame that they don’t co-operate since it’s all OSS anyway.

2. First boot: Excellent ran me though all the usual for setting up network email etc. No complaints here.

3. General usage: Very nice, everything worked well and as expected. I updated Firefox to the latest. (It annoys me that this is easier in Windows than it is in Linux. I found myself having to edit symlinks.) Openoffice, The Gimp all the usual suspects were there and in fine form. Very impressed up till this point and as a desktop OS, it was looking like I could personally leave Windows XP totally behind and work just as well or better.

At work I used Linux GUI tools to connect to a Microsoft PPTP VPN, take over Windows machines via rdesktop, copy files to and from SMB shares on a domain and basically do all the things I’m required to do, that up to this point had been done on a Windows machine. I was stunned with how much better things had gotten since I last tried. Windows was easier in some respects, (like the VPN) but only because they had dumbed their tools down and limited their abilities. I was however still required to know a fair bit about networking, routing and VPN’s to get usable PPTP connections though. Once setup however, everything just worked, every day without fail. My personal feeling is that setup wizards would help newbies make full use of the tools available.

My biggest complaint is wireless. This NEC is a true Centrino in that it has the lousy Centrino 802.11b Intel IPW2100 wireless card, which while limited to 11mb, worked flawlessly in Windows. My daily travels involve me connecting to at least 3 different Wireless networks. My home network has WPA with hidden SSID, and two at work have WEP with differing settings. No GUI Wireless tool I tried in Linux would consistently work with all of them. If I was prepared to mess around for 10 or 15 minutes each time I could usually get it working, but that wasn’t the point of the exercise, doing so would have made me less productive than Windows. For the record I tried the Redhat Wireless tool, then wifi-radar, then wlassassistant and then I got desperate and tried anything else I could find. The problem seems to be that these tools hand off to other apps for different things, and none of them do it flawlessly. WPA, DHCP etc etc, the result being that I couldn’t go from one AP to the next without issues. On all three of the wireless networks I was required to connect to used DHCP, and Fedora only scored an IP address wirelessly once in every 10 or so attempts, and thats regardless of what I set the timeout to. I plan to change the Intel IPW2100 with a different chipset (non Intel) and try again before I write off Fedora Linux’s wireless ability, but I’ve got to say I’m not hopeful. If however, you just connect to one wireless network, then there really isn’t a problem as the Redhat tool will do that just fine, but I’d really hoped for better. Sad to say it but Microsoft XP SP2 Wireless tool is surprisingly good and better than anything I’ve seen in Linux thus far. Ditto with Intel’s wireless tool that comes with their Windows drivers. (Wouldn’t it be great if they ported it to Linux and released it OSS?)

To get a bit more cutting edge, I then swapped to KDE, installed the kde-redhat groups updated KDE packages, and then headed over and grabbed the latest Compiz-fusion stuff. This laptop had a 64MB 3d video card so while it wouldn’t do much with Vista, it did some amazing stuff in Linux with Compiz-fusion. I was blown away with the 3D eyecandy that this old laptop was capable of. Especially after seeing what is required to get a good Vista 3d performance.

Summary.
Is Linux ready for the desktop? Yes, it is for all the usual everyday stuff, if your needs are more complex, like you need to connect to a Microsoft network via VPN, it can do the job, but don’t expect to get out of it without learning about routing tables and the like.

Is Linux (fedora at least) ready for the laptop? No, I don’t believe it is, GUI Wireless is frankly far behind Windows and I also had several other issues that may have been specific to this laptop, but googling shows not. I was required to install 855resolution in order to get my screens native resolution. (a bug of Intels not OSS, but Fedora could have detected the problem and loaded 855 resolution for me.) In addition, I had a massive key bounce problem with the touchpad that I did eventually solve, but not without much gnashing of teeth.

Fedora annoyances:
1. No centralized control center. I’ve been spoiled by Mandriva in this respect I suspect. Redhat hasn’t had a centralized unified control center since Linuxconf went away. It isn’t a huge thing for me, but I think it would be to those still learning the ropes, and I must admit I did find it annoying when I didn’t know what tool I needed and had to google to find out rather than looking in a central control.

2. This one is more about KDE/Gnome than Fedora. Why can’t they work together? Neither has a total “better” experience, both do different things better than the other. But since they compete rather than co-operating, you get two lesser experiences as a result. Sad really since we’re essentially on the same side. I’d love to see some sort of system that allows Gnome apps to fully integrate into KDE when run from within it, and vice versa with KDE apps in Gnome. That way we could actually get some consistency in form and function and allow the best of each to compliment the other for the betterment of all.

3. This one is touchy from a security perspective, but on a desktop machine, I hate having to enter the root password to start networking tools, to start wireless tools and other stuff of that nature. I know why it is the way it is, but on a desktop it isn’t good enough. I know there are ways around it, like sudo for example (which I did make some use of) but none of them were setup that way out of the box, and thats annoying because real people like to burn CD’s and join wireless networks and most of them won’t want to have to start them as root to do so. I am not suggesting we get as lax as Windows, but we need to find some middle ground here because in an average morning on my Linux desktop, I’d have entered my root password at least a dozen times. (And since I’m paranoid, my root password is traditionally quite long and convoluted. By the end of the second day, I was considering making my root password a couple of characters long to speed things up and that is even more unacceptable.)

4. Wireless, for people with a wireless lifestyle, Linux GUI is not gonna blow their socks off. It’s on again off again performance was a surprise and a disappointment as it was a deal breaker for me at this stage. If I can’t consistently connect as I need to in Linux, I have to use Windows, simple as that. Before I give up, I’m going to change the MiniPCI from IPW2100 to something non Intel to see if it makes any difference, but as I said earlier, I’m not overly confident that it will change anything. One of the big Linux guns like Redhat or Novel really needs to take Wireless in hand and come up with something as good as or better than XP’s tool, because the current setup is a mess. I actually find myself missing UnitedLinux for the idea if not the result. WLassisstant and Wifi-radar are both better in look and usability than Redhat’s tool, but their lack of functional consistency lets the side down badly. We need to help these guys build on what they have and get it right.

5. Wired networking. While I had no problem at all with Ethernet networking, I must say I was surprised that there isn’t some sort of hotswap system in place with it. Everytime I unplugged/re-plugged the Ethernet, I had to restart networking or do the old ifdown/ifup eth0. In Windows it detects/recreates the connection it all by itself. I was surprised that something so simple hadn’t been addressed yet. It isn’t a big thing, but on a battery powered computer, it’s definitely nice to not have to open a root terminal and re-establish your network connection

There is no doubt at all that Linux has come leaps and bounds in the last couple of years. But anyone that tells you it’s all the way there yet is deluding themselves. In many cases it can be a useful and productive replacement for Windows on the desktop but in just as many cases, it can’t. I know the *nix ideal of each tool doing one job really well, and I agree on servers, but it just doesn’t work on the desktop unless all the parts seamlessly work together to create the whole. Being a huge supporter of Linux and OSS, I really wanted to only have good things to say here today. But failing that, honestly is the only way the issues will get addressed and so I’ve not held anything back. I haven’t gone entirely back to Windows on that machine, but I am dual booting it now.

Anyone who has suggestions of other Linux wireless tools to try should comment below and I’ll give it a shot.

Apple has announced that they are releasing a version of the Safari web browser for Windows. Apple say that Safari for Windows is twice as fast as Internet Explorer (is that saying much?) and that a beta release is available for XP and Vista.

It will be interesting to see if the Windows and Mac versions render the same (unlike MS IE on Windows and Mac). Also interesting is the fact that Safari’s rendering engine is in part based on KHTML the Linux/*nix GUI, which will hopefully help raise the status of Open Source code even more.

Many people (myself included) were not terribly impressed with the layout of Windows Itunes, hopefully Safari will be a more intuitive experience. If we (as web developers) are really lucky, Safari will end up with 25% market share (who knows, perhaps the Iphone will drive that since it comes with Safari.) and the rest is divided between Firefox, IE and Opera. That way it will be impossible for any one browser to drive developers away from W3C standards ever again.

See Apple/Safari for more details.