Software

There is a lot of speculation about the April Software Update (ASU). Is it a single image you install, or is it a line of development? Is it a rebranded Qtopia or something new? How's the sounds quality? What Wifi management tools will there be? Is there a package installer? What does it look like?

I thought I'd take some time today to hopefully answer those questions, and more importantly, show lots and lots of screenshots, because everyone likes screenshots. :)

When you first boot the Freerunner, the screen backlight fades in gradually to reveal the familiar orange and black Openmoko start splash. After a few seconds, the screen turns to the black with white scrolling text that just about every Linux user in the world is familiar with.

Then come the boots. As a bit of geek humor, probably chosen by Raster, there is a pair of boots on the boot screen. A green bar slides in and begins loading. A great example of the early stages can be found on this flash demo.

By default, the ASU image is set to a grid system application launcher.

For those uninterested in this view, there is a bit more unique take on it using the slider mode.

Atop the action at all stages of using the ASU is Illume, which essentially functions as the system tray and task bar. Here, you can launch the keyboard, launch the settings panel, enter the Enlightenment settings (note, this is slated for removal and doesn't directly affect the use of ASU), switch to a running application or close an open application.

Let me visually cover those things, in order. First, the keyboard:

The keyboard is a QWERTY like keyboard that has a dictionary lookup built in. To bring the keyboard down you tap the “qwerty” area of the Illume bar. To begun typing, simply hit a letter. To see a more complete list of words meeting your typed characters, click the white triangle on the keyboard. There is a shift key which displays upper case numbers. If you look, you'll see that some keys are missing. If you make a line on the keyboard going up or down the keymap changes to special characters (!,@./) and numbers, making the keyboard usable on a terminal. Though, in fairness, I didn't check to see if there are CTRL or ALT keys which might affect that use.

The settings panel is opened by clicking the wrench icon on the Illume slider. In here there are settings for all kinds of things. I won't go in depth with them all, the most “visible” is probably the option to set the slider or icon grid launcher modes.

The Enlightenment settings, as an E17 user would recognize, are kind of a “left over” from the development. These settings are planned to be removed, since the functionality of them is either irrelevant for Openmoko or will be absorbed into other applications.

One of the best features of the Freerunner, in my personal opinion, is the ability to utilize GPS without a non-free driver. Because of this, a basic GPS application has been included with the ASU. It's really very much a placeholder or prototype now, as the only thing it seems to do is display a map that can be zoomed or moved around and prompts that it fails to get a GPS fix. I'm not exactly sure of the name of this application, in all honesty. I'm used to calling it “Diversity” thought I've also heard the name “Splinter” tossed around and I'm not sure of what the distinction, if any, is.

The second big hardware change for the Freerunner was the inclusion of wifi. In order to connect to wireless networks, there is a program called “Campwifi” that once launched, probes for networks. While not viable in this shot, I have managed to connect to an open wifi access point “in range” of my apartment and been assigned an IP address. I personally have no experience with wifi at ALL aside from what I've gleaned from reading tutorials over the last few years so I'm not sure if some of the encryption and authentication schemes are implemented – with only one unsecured wifi AP within range of my apartment, it's not been an issue for me yet. Hopefully I'll be able to check that out more as I'm able to shift away from the 1973 and towards the Freerunner as my day-to-day phone.

Another nifty feature of the ASU is the Assassin package manager, a GUI front end to opkg. Once you launch Assassin, you're presented with a list of categories to choose from, and some of those categories contain items that can be installed. So far, there aren't too many packages listed. Certainly far less than show with an opkg –list. It leaves my mind to speculate as to how this is important – perhaps Assassin will present users with action based installs (Install US Maps for GPS) or be used for more than just software applications for things like ebooks and ringtones. With an open platform, the mind runs amok.

Once an application is selected, there's one big “Touch here to install” button which is very simple to figure out what to do. Once pressed, you're prompted again to confirm at which case the install begins and completes.

From the screenshots you can see I installed “openmoko-messages2”, an application from the 2007.02 line that was based on GTK. At one point, when the switch to Qtopia was announced, there were a lot of GTK fans upset. It seemed to die down a little bit when it became clear that GTK would still be usable on the phone so I wanted to be sure I got a shot running a GTK application. It looks hideous because the GTK theme wasn't installed, but any GTK theme could be applied and hopefully there will be a matching one in the future. If there was any remaining doubt, GTK is alive and well within the ASU framework.

Finally, I'll show a few “glitches”. Firstly, the video processing on the Freerunner is slower than the 1973 thanks to the Glamo. I found that after a while of using the device (and with over 40 screenshots, it was “a while”) the transitions (Illume slides down, launched applications slide down) began getting choppy. I found reducing my frame rate to 10 fps helped this quite a bit but was still noticeable. Additionally, NONE of the Qtopia apps would launch in slider or grid mode. Setting the display variable I was able to launch Qtopia Media Player which complained about lack of Qcop and then segfaulted as I closed it. But you'll see that there are no screenshots of the Qtopia apps since they wouldn't launch “normally” for me.

You'll see below that occasionally applications crash. I've been shown the “Enlightenment crashed” white box a few times, and recovery happened. Every now and then, a half rendered Enlightenment window would pop up but not show me anything good. I had to use my tiny screwdriver stylus to close that box, my fingers are too big for that. Lastly, while earlier versions of the ASU had scrollbars, this version does not. However, when scrolling in certain areas, a distortion appears in the area where a scrollbar would be. These, luckily, appeared in screenshots.

The ASU, while certainly not usable as a daily phone, is certainly full of promise! I'll be tracking this closely and writing reviews as applications become more complete and the images become more functional.

The improvements in Iceweasel 3 are great. The new styling really fits in well with my Gnome desktop. However, as with the migration to any major “point-oh” release sometimes things break compatibility.

For this reason, I have been hesitant to fully move to Iceweasel 3. The improved password manager, the new visual styling, improved searching (you can not enter the title of a page in the address bar and it pulls it up in your list of auto completions) and improved zooming are all great, but when I can't do things I usually do, it's not an upgrade.

One thing I do is play a game called oGame. To enjoy it a bit more, I use two extensions: FoxGame and GreaseMonkey with a script from Userscripts.org. Both of these extensions broke with the upgrade to Iceweasel 3 which has moved from Debian Experimental to Sid, which means my days of 2.0 are numbered.

Luckily, I complain. Having complained in the Libervis IRC channel, netdaemon offered me a suggestion that it was possible to override the extension compatibility check. I'll be damned, there is! Here's how:

• Type about:config into Iceweasel's address bar and click the "I'll be careful, I promise!" button.
• Right-click anywhere. Choose New>Boolean. Make the name of your new config value extensions.checkCompatibility and set it to false.
• Make another new boolean pair called extensions.checkUpdateSecurity and set the value to false.
• Restart Iceweasel.

Unfortunately, FoxGame actually is incompatible with at least the RC 2 version of Iceweasel. So I had to manually disable that one, but Greasemonkey, which I utilize more frequently, works just fine thus far.

Also, at some point my work computer (also running Debian Sid/Experimental) updated to Iceweasel 3 and lost it's back button. This is annoying, but not incapacitatingly so, since you can always use ALT key commands or use the right click menu.

Anyway, to restore the back button simply go to View->Toolbars->Customize and click “Restore Defaults”. Problem solved. Both of these tips today gave me a mostly functional, feature enhanced version of Iceweasel 3 and I'm happier for it.

For the past several years, I've been a proud supporter of the Free Software Foundation. The ideals of Free Software have always rung true to me, and I've not only adopted Free Software solutions in my home (even my wife runs GNU/Linux) but advocated for others to evaluate what they find important and adopt free software themselves.

So when I pulled my funding last month from the FSF, I was asked "Why?" from some friends. "Do you not care about free software anymore?"

I still care deeply.

Free Software is, at it's most very basic for me, a matter of property rights. To me, it is an affront to property rights to sell or give something someone and enforce conditional restrictions upon them. If Oster sells you a toaster, they have NO right to prevent you from taking that toaster apart, studying it, adapting it and using those adaptations in the marketplace. Free Software then, has ALWAYS been about me holding my right to study that which is mine - and affirms that everything on my computer is in fact MINE.

Due only in part to Free Software, the activist nature within me has been awoken. Even more than with free software, I feel it's important to stand for what I beleive in and make decisions that reaffirm that believe. The newest belief if that government, in some way shape or form, is the cause of most of the day-to-day gripes I have. That isn't the point of this blog entry though...

What has become clear to me is that the Free Software Foundation is not truly comitted to user freedom. Furthermore, they're quite willing to use the guns of government to enforce their "freedom". Freedom is free market freedom. Freedom is, at it's very base, the right to choose. I'm still firmly comitted to the ideals of Free Software, but I stand against the Free Software Foundation, as I stand against anyone, who feels it's morally justifyable to use the guns of government to enforce compliance with ANYTHING.

If free software is better, free software will stand it's own ground, and hundreds of men with military weaponry can't part with it. Bad ideas, however, don't seen the threat of violence to be abandoned, as it makes no sense to continue with it. Free Software stands and fights it's own battles, using only consumer opinion to oppose Microsoft and Apple and Adobe. I'm quite content to leave it there, and in order to do that I found it necessary to pull my funding of the Free Software Foundation.

Viva Libre!

I'm a Debian user. One of the strongest points in the Debian world is the package management system that is both binary and source based. In addition to the software, Debian maintains strict rules regarding the package management process to ensure high-quality packages with the minimum of conflicts.

Sometimes, however, those conflicts occur especially running the "rolling" testing or unstable distros. Frequently the issues come when updating "sets" that contain many packages, such as KDE or Gnome, when one piece didn't build but other chunks are there. Being able to pull down the source code and build it on your own system before it hits your mirror can be a lifesaver.

In addition to building missing packages, Debian's source based repositories are the only "safe" and "easy" way to mix sources - something I've personally had to do to leverage the large number of packages in GetDeb that are designed for Ubuntu but will build and play nice with Debian systems.

I've been meaning for a LONG time to sit and write out a tutorial on how to easily use the tools, but someone else has done basically that so I'll merely link there and call it a day. :)

http://www.ducea.com/2008/03/06/howto-recompile-debian-packages/

Today I've done a new review of the 20 February 2008 snapshot of OpenMoko for the Neo1973.

The full review can be found http://wiki.openmoko.org/wiki/Snapshot_review/2008-02-20

The major news is that today SD support has returned, which means I can recommend all users to upgrade to today's image and kernel. :) Today also saw decent (almost 50 seconds less!) improvement to boot time.

These reviews make me feel good. :) I got thanked by Sean Moss-Pultz (CEO of OpenMoko Inc.) himself, and have had developers give me info on changes, causes and plans for the distribution. It feels awesome to be able to help the project even though my coding skills are sorely lacking.

I'm a bit concerned about where to go from here. For the most part, my reviews cover the major functionality of the device. Phone calls, SMS messages and media capability. I don't touch GPS because it requires non-free.

There are applications that I don't review also because I consider them less than critical (like the fact that there are 4 sudoko games) but at this point I'm thinking that I have time to check the "less than critical" things. I also really want to begin moving into building a realistic, step-by-step wishlist. The audio system is in place now so the phone rings, but most people consider multiple ringtones a basic feature. Profiles are also pretty basic (phone in vibrate only mode, for instance) and don't exist.

I've asked the device-owners list to give input on how to improve the review and where to go from here. :)

Yesterday I forgot my USB cable so my phone died and I was unable to flash the 12 Feb 2008 image. No review. :)

Today's marked the first 2.6.24 kernel in the official OpenMoko snapshot pool. It also broke quite a bit because the rootfs contains the 2.6.22 kernel modules. :)

Phone functions didn't work, SMS didn't work and my 1GB media card didn't work so the ability to play music has been greatly decreased.

General users are urged to stay with the 11 Feb 2008 snapshots while moderate to advanced users are urged to test and report on the new snapshots.

The daily images were not built while the staff of OpenMoko Inc enjoyed the time off for the Chinese New Year. :) Now that the holidays are over, OpenMoko images have resumed.

The full review can be found at http://wiki.openmoko.org/wiki/Snapshot_review/2008-02-11

The biggest change today comes in the form of improved sound. The review hits on this when I mentioned that ambient noise is echoed a LOT less, and that the "air" sound is gone. Another undocumented improvement is that sound volume of the ringtone is now good, meaning I can hear the ring of the phone while in my car with music up loud. :)

Erin Yueh of OpenMoko has also submitted instructions to the OpenMoko device-owners mailing list on how to remove the crappy multi-tap input and replace it with the matchbox keyboard. While neither solution is good (both are considered horrible and where one lacks features that makes it useable, the other has bugs that makes it unusable) I personally prefer the matchbox keyboard.

I've also ammended the review process in regard to GPS. On the Neo1973 GPS requires the use of a non-free driver, and efforts to reverse engineer it are horrifically understaffed. For this reason, GPS doesn't work out of the box. Despite some community members insisting on providing this information in the review, I consider the use of non-free software to be unethical AND contrary to the stated mission of OpenMoko - my reviews now reflect this by adding in three new critera for reviewing of GPS functionality.

For the record, the GTA02 (OpenMoko Freerunner) has corrected this problem, and the GPS on it "speaks" standard protocol so no driver is needed to make it work.

My recommendation is that Neo users upgrade to the 11 Feb 2008 image to reap benefit from the improved audio settings.

I finished my OpenMoko review today. In general, this image is good.

The full review can be read at http://wiki.openmoko.org/wiki/Snapshot_review/2008-01-31

Boot time is back in line.

SMS works.

Calls work (but still require "the fix" for good sound quality on my side).

Media player works for audio, still no video (and it may never happen on GTA01)

Battery life seems decently improved but still not "good". No hard and fast tests done there.

My recommendation is for all Neo users to use this image, it's the best thus far. :)

Today I reviewed the 30 January 2008 rootfs and kernel images.

Overall, this was a giant leap backwards. Stability has suffered which prevents the use of SMS to the fullest and prevents the use of media playback.

In addition, I found and reported a bug where missed calls were reported twice.

Bootup time jumped up by about 40 seconds.

In general, I suggest NOT flashing the 30 January image, preferring the 29th. The kernel is, as far as I can tell from the filenames, unchanged.

I've noticed that many people who own Neo1973 devices simply don't update every day. There are several ways to do this, one being the apt-get reminiscent application called ipkg (or opkg, the official OpenMoko fork). Another way, particularly useful in rolling distros (like Debian or OpenMoko) is to install a new image.

I choose to install fresh Debian images every weekend and I choose to install fresh OpenMoko images every day. Since people are reluctant to flash a non-working image (which would erase all of their contacts, scripts and hacks) frequently I decided I'd begin documenting it when I do.

Today, I published my first review of the OpenMoko snapshots, since it's finally good enough for me to use daily as a phone and SMS device.

http://wiki.openmoko.org/wiki/Snapshot_review/2008-01-29

I'll summarize them here, but the real good stuff is in the wiki. :)

Summary:

Sound quality on the phone sucks because it picks up ambient noise, luckily a fix has been identitifed!

The multi-touch input method sucks, I think.

Video playback is half-there, i.e. Audio only.

Music playback is better and improving, a LOT less skipping versus a week ago. There's a hack to do even more.

Calendar mostly works.

SMS works.

There seems to be some confusion with exactly what OpenMoko is.

The most simple explanation is that OpenMoko is a software framework designed to empower users of mobile devices. OpenMoko is to the Neo1973 what Gnome is to a PC. The analogy is particularly fitting, since Gnome and OpenMoko are both GTK based.

Keeping with this analogy, another "environment" that can be found on the Neo1973 is Trolltech's QTopia which is, like KDE, based on QT.

Most of the hype around the Neo is because of OpenMoko (or perhaps the other way, without OpenMoko the Neo would be ignored) and OpenMoko is what my personal interest is in so I will begin today by explaining just what is "in" an OpenMoko application from the perspective of a user.

Below is an image of the "home" screen, the starting point for all OpenMoko adventures.

The very top of this image is the "taskbar". A closer examination is below.

In the section that displays the words T-Mobile" is the name of the application you're using. On the home page, this displays the name of the GSM network you're connected to. In this case you see "T-Mobile" since that is my cellular carrier. When the phone is first booted (or when GSM is first turned on) this may say “Registering...” as the phone searches for your network.

To the right of the words "T-Mobile" is a kind of dashboard or system tray. The first icon is of a small keyboard, clicking this brings up the matchbox keyboard, used for entering text. If the keyboard is open, clicking this icon will close it.

The icon directly to the right of the keyboard is the Bluetooth Logo. Clicking this icon brings up a menu allowing you to "Power up Bluetooth radio", "Power off Bluetooth radio" or check "Bluetooth Status". This allows you to shut off unneeded Bluetooth capability for added security or to preserve battery life.

To the right of the Bluetooth logo is a USB icon. This icon indicates that a USB cable has been plugged in that is capable of transmitting data. This icon does not appear when “dumb” USB-style cables, such as car chargers, have been plugged in. Tapping on this icon does not present the user with options, it is there only as an indicator.

The next icon represents the GPS menu. As of today this icon does nothing and is not representitive of the GPS's status.

Next to the GPS icon is a GSM signal meter. Most mobile phones have an indicator like this, showing the relative strength of the cellular signal they have. Typically, the more bars you have, the stronger your signal and the more clear and stable your phone conversations would be.

To the right of the signal meter is the battery indicator. This indicator shows roughly how much power your battery has remaining to power your device. A fully filled green battery icon indicates full charge where an icon that is not filled would be almost powerless. This icon can indicate various intermittent states. When connected to a powered USB cable the phone enters “Quick Charge” mode and this icon displays a white zig-zag lightening bolt image.

Below the taskbar is an action panel. This panel varies from application to application but in general it indicates actions a user may take. On the home screen, these actions are, from left-to-right:

Launch the “Dialer” application (Make a call) Launch the “Contacts” application (View your address book) Launch the “Messages” application (Send and read SMS Messages) Launch the “Tasks” application (Set or view important events)

Below the launcher is the main body which contains a digital clock and the date.

In addition to displaying the time and date, this pane will display notification of missed calls below the date. This is the main area to present information to the user – what this section of the screen displays is highly dependent on the application that is currently running.

Tabs for your currently running application appear below this pane. These tabs are different depending on the running application but the concept behind them is uniform to all OpenMoko applications. For instance, the Dialer application has two modes, a keypad mode and a list of all past calls. This section has two buttons in the tabs panel to switch between these modes. I will go in-depth regarding options with each specific application later.

Today is day three with my Neo1973. It is well know by owners of the Neo and members of the OpenMoko development community that the GTA01 suffers many power management issues. Today, I got the first taste of those issues myself when I forgot my USB cable.

The Neo runs on a "basic" mobile battery and it packs a devent punch. A sharp touch screen, GPS, GSM, Bluetooth, speakers and a backlight. In addition, the CPU and memory consume power as they're used. This means, without proper power management systems on the OpenMoko platform, the Neo's battery life is rather short - in my case, about 5 hours.

There's another problem - the power controller firmware doesn't flag "critical" use - what this means is rather than shutting down when the phone's power gets to 5% it will continue to draw power, right until the battery is 100% drained and the system dies - or as SpeedEvil says, "rests" - mercilessly. Why is this so important? Because the Neo has an advantage called "Quick Charge" - when powered on (i.e. running the Linux kernel) it is able to tap the full power it recieves in the USB charging port resulting in a faster charge. However, when the phone is NOT running Linux (dead or running the bootloader, u-boot) it draws about 1/5th of that power resulting in painfully slow charges. So a phone that has "rested" totally need to be slow charged for about an hour before it has enough power to boot into Linux and enable quick charge.

The promising part is the reasons for the total battery death are know and can and will be fixed over time. That's not so much of an important thing.

What IS important to me is that pondering on this caused a paradigm shift. Having been a GNU/Linux user for the past several years (almost a half decade now?) I've seen several Linux kernel release announcements from the exciting release of KVM to things I considered mundane - drive I/O improvements or some such.

One specific area I've always deemed unimportant - power management. Up until recently, it never mattered to me as I'm a full-time desktop user. I hate laptops and have always found them to be novelties. Now, having really looked at how dramatic this is on my Neo, I have a new appreciation for the time and energy that kernel hackers put into these things - and I realize that every patch submittited is someone else's "power management"; their issues that everyone else is oblivious to.

Taking this track along a bit more, I realized just how powerful the Free Software community is. Until today, I've never thought about power on my computer before. The more CPU speed needed, the more power it consumes. The more memory swaps, the more power. More HD? Bright screen? MORE POWER!

My friend Danijel finds a connect between being a geek and "being green" - I do not. Personally, I feel that the concept of Global Warming (implying it's our 'fault' and that we actually have the capacity to screw a planet over in under a hundred years - or more crazily, that we have the capacity to "fix" it if such things were true) to be laughable. I don't, however, think we can ignore issues of pollution, energy consumption and waste - just the opposite in fact. I think these issues are SO important that, like Freedom, they should be addressed on their merits alone, without propaganda scare tactics of an impending global catastrophy to motivate people to do something about it.

That said, until today, my eyes were closed to the sheer about of waste generated by shitty code. The FSF has been pushing that angle for a little while with the Bad Vista campaign, but this isn't an attack on Vista because it applies to us - users of software libre - just as much. How much electricity are we wasting because of crappily optimized code?

Think about this for a moment... There are datacenters FULL of servers on redundant power systems. Each of these servers is running some sort of AMP stack, spinning hard drives and swapping bits around in memory. If Apache itself has a bug that causes just a bit of unneeded movement it has a cascading effect running 24 hours a day on hundreds of machines. How much energy is being consumed because a developer didn't think about being efficient while writing his code?

Free Software users and developers have an added kind of power, the power to consume less. The amount of waste generated by that hypothetical Apache bug is decently large, but imagine now if the bug was in Linux itself - now it's affecting desktop users and laptop users as well as a lot of servers. This isn't merely a case of "bloat" because even on very slim, "minimal" systems it's possible to be doing more than is strictly needed.

We all need to expand our horizons sometimes and I'm glad for having the chance to have done it here with the Neo - it makes me feel powerful. :) Pun intended, unless it's tacky. Then I won't own up to it.

On January 2nd, I purchased a Neo1973 Base from direct.openmoko.com.

My first interaction with FIC, the company that manufactures the Neo, was that of excitement. They have a decently beautiful site that highlights exactly why one would want the Neo – freedom, hackability, being on the cutting edge and having a sexy looking device. I felt no qualms at all with inputting my credit card number and shipping info. Immediately I got a confirmation via e-mail and everything was good.

As I waited excitedly the next day, checking my e-mail every few minutes to see if the confirmation had arrived (did I mention I was excited?) I began to get a bit annoyed by the end of the day. I then checked out the OpenMoko Wiki and found some information regarding the shipping procedure. I found there two things of interest:

Firstly, the price listed on direct.openmoko.com is an estimate. FIC is based in Taiwan, not the USA, so the USD amount I saw was the “target” - they would actually charge my debit card in New Taiwan Dollars in an amount equal to the current exchange rate.

The second thing I learned was that order processing took 3 to 5 days. Okay, “fine” I told myself. I could restrain my excitement.

It wasn't until January 8th my debit card was actually charged and I received confirmation that my order was processing. Perhaps I'm spoiled by NewEgg's wonderful shipping procedure but I consider a 6 day turn-around to be horrific. In addition to being a software libre geek, I am also a home brewer. In home brewing I deal with specialty supply shops to purchase my hops, grains and yeast. Some of these shops are run by retired military vets with little internet experience and some are decently skilled in the internet but in any case they can still manage to have my order packaged, shipped and at my door in a week – I fully expect a company called FIC (First International Computer) to have their order system automated enough to process orders efficiently. I console myself with the notion that they've got a room full of hackers plugging away to get the GTA02 “mass market ready” who simply can't be bothered to check e-mail and fill out UPS invoices.

Later that night I got confirmation that my package had shipped via UPS and included was a tracking number. I excitedly (again, did I mention how excited I was?) plugged the tracking number in to UPS's site and found that my package would be delivered on... the 14th! Have I really been THAT spoiled by NewEgg? I had debated over getting “expedited” shipping on one of many forms, but I decided against it - “standard” UPS shipping was decently fast, I reasoned. I guess not – this package was going to take almost a week more to get to me from California (for the record, NewEgg ships from Whittier, CA most of the time too). It seems NewEgg's “standard” shipping is NOT UPS's “standard” - lesson learned.

The weekend wait for the Neo was pretty easy, I managed to brew a beer on Sunday using my new Barley Crusher, so time went by rather quickly. Sunday night I decided that I was going to drive my wife to work, and rather than go to work myself immediately, I would drive BACK home and await the Neo's arrival. Luckily, my job offers enough flexibility that I'm able to do that without much issue.

At 11:09 am, my Neo arrived. My decision to return home was a good one – FIC ordered a signature before delivery. I normally HATE when companies do that (With 6 hour long delivery windows I can't afford to take a day off work to sign for a package. I work 55 miles from home so picking up my package from a UPS facility ALSO requires I miss work. Sign for packages = bad). In this case, however, I was a bit happy, since it was a rather expensive shipment – and I was pretty sure insurance wasn't part of UPS's “standard” shipping either.

Above is the box that arrived at my door. It was a pretty plain box, no fancy buzzers or bells (I truly didn't expect them but I have to make this dramatic – did I mentioned I was EXCITED?) and attached to the top was a packaging slip. I opened the box with my keys to find that there were blocks of very thick black foam covering packing peanuts. Buried beneath the peanuts was a bubble-wrap packed white box. This foam was very thick and in general the whole thing was well packed.

Inside the bubble wrap was a clamshell box that I thought was inventive. My wife has been in the printing industry since high-school so printing and packaging are her passions, I suppose it's rubbed off on me.

Once opened the clamshell splits again (once each way) to reveal the “goodies” that I've been so long awaiting.

I won't go over every piece in the box but every piece has been photographed by me, if you're interested in seeing those images please check out the OpenMoko album of my image gallery. The Neo handset itself was enshrined in the familiar anti-static anti-UV bags but it was the first time I'd personally seen self-adhesive anti-static bag – it was kind of neat.

All of the contents of the clamshell unpacked. There's the carrying case, lanyard, headset, spare ear buds, USB cable, stylus, battery, memory card and Neo handset.

While photographing the handset, I noticed that the screen itself seemed very reflective and I often saw myself in my own shots. This worried me, since I figured if it reflected my image it would also have horrible issues with glare when in normal use.

Eager to “get on with it” I held down the POWER button for about 4 seconds and the phone gave the smallest of shudders as the vibrator pulsed for a moment and the screen flickered to life. For anyone who has read articles about OpenMoko on the Neo you've heard “the screen is nice” but it's impossible to take a picture that shows exactly how sharp this screen is. When the backlight came on I was greeted with a black and orange splash screen which quickly gave way to a “typical” Linux boot up – Tux at top and white text on a black background. As small as the text was it was still remarkably clear – just so small it's hard to read.

The Wiki explains this very clearly, so I wasn't alarmed by it, but the very first boot ends in a kernel panic. The “simple” answer is that there's no software other than the Linux kernel and the boot loader (u-boot) itself, so it fails to load anything. End users won't have this happen to them (OpenMoko will be DOA if it's required) but developers will have to flash their own kernel, root file system (rootfs) and (potentially) u-boot image.

At this point, I realized I was “pushing it” by not going to work, so I packed up my Neo and accessories, uploaded my pictures to my wife's Ubuntu system, crashed it and then went over to my Debian system to burn my newly taken Moko images to CD and then went to work.

At work I began my “flashing” after reading up on the Wiki. What was slightly confusing for me was the out-of-date nature of the Wiki. While it's pretty clear once you understand some terminology for someone “new” it can be quite scary – the term “bricking your device” is used enough to scare the hell out of me and I'm fairly comfortable tinkering with “internals”. A quick post to #openmoko on Freenode led me to a how to find daily build images page which kicked me over to buildhost. The recommendation I got was pretty much “Use the newest images” - this makes a TON of sense since the software is rapidly changing.

Flashing the Neo itself was a trivial task and I suppose it will get even easier once the OpenMoko build tools become packaged for popular distros like Debian and Ubuntu. In my case, it meant plugging in the USB cable to my computer and my Neo, downloading dfu-util and compiling it – the whole process took about 30 second so it's a FAST compile as well as simple. I then downloaded the uImage file and the rootfs image for the GTA01 (about 48 MB combined), launched u-boot (hold AUX, power the Neo on) and issued one command to flash the kernel to the Neo. The second command flashed the rootfs to the Neo – a slightly longer process that took about 13 minutes. The downside to this flash is that once the phone enters u-Boot mode it will power down after 60 seconds of inactivity and it doesn't consider USB transfer to be “activity”. This means every 45 seconds or so I pressed the AUX button to keep things “alive”. I'm sure I looked stupid to my office mate, but oh well. (For the record, there IS a solution for this but I was excited and just wanted the thing to work). Once the flashing was complete I restarted the Neo and it booted without problems (no kernel panic!). This time, once the splash screen was done I was greeted with a whimsical sounding chime (audio works!) and the phone proceeded to load my home page.

At this point I hadn't charged the device or installed the SIM or microSD cards, so I powered down, installed them and powered back up. The Neo doesn't do a “quick charge” unless the device is powered on so unless it's totally dead, it needs to be on to charge.

I worked on “work stuff” for a few hours before turning my attention back to my Neo. This time I decided I'd play with GSM and attempt to make a phone call. This entire time the top toolbar was saying “Registering...” which alarmed me – was my SIM card incompatible? There is currently an open bug pertaining to 3G SIM cards (mainly AT&T ones, but possibly others) and I was afraid mine was affected. I pulled opened the GSM quick menu (at the top of the screen it looks like a Triforce-on-a-stick) and clicked “Auto-register with network” which I assumed it would have already done. The phone sat there, still displaying “Registering” while I went into the applications menu and fired up a terminal (ain't it sweet? CLI on my phone!). To see what was happening, I issues a “dmesg” and almost immediately I got a popup (it looks very “Gnome”-like) indicating I was connected to my T-Mobile network.

At this point, I opened up the dialer and punched in what I thought was my phone number (to check my voice mail ) and was a bit shocked... Someone answered! It happened to be my wife (my phone number ends in 7446, hers 7445) and we both chuckled. I noticed two things about the call. Firstly, I can very much hear “air” through the ear piece. It is my understanding that as of this moment there's no noise-canceling software of the phone so the microphone picks up (and sends) everything. The second was that even without noise canceling software, the call quality was comparable to a “regular” phone – at least in the mid-range phones I'm used to dealing with.

On the ride home from work, I excitedly showed my wife the Neo. She's super supportive of me and my obsessions with brewing, computers and freedom, so she was poking at it inquisitively rather than finding “broken” things to criticize like some people would (especially after what I paid for it). While I had focused on the phone, the interface and the hardware the first thing she did was find the games menu. I have to hand it to the interface designers on this one – either I've been clueless or they did their research. While I have to poke and prod to find a “rhythm” to menus my wife was able to find and launch some games so fast I was amazed – and even remarked on it.

She played several games like the Mastermind clone, a game called “Maps” which seems like fun where you fill in oddly-shaped tiles so that no two touching tiles are the same color. Her initial impressions were very good, actually. Her complaint was that some of the games are horribly optimized for the device. Sudoku, for example, is almost useless without the stylus (Input options are 0-9, a keypad would work MUCH better than a matchbox keyboard).

Once home I hit my computer with the Neo and began “playing” with it. I pulled up my address book from my old phone and began inputting my contacts and phone numbers. This is where I found the most “buggy” functionality. I then moved to trying to send an SMS message to a friend which didn't fail, but didn't work either.

Having those two things fail (after some prodding, too) I decided to see what this thing could do in terms of media. I transfered a song, via netcat, to the memory card and launched the Media Player app. Playing the song failed at first, so I rebooted the device (certainly not needed, like any GNU/Linux system I could have restarted the sound system or investigated WHY it wasn't playing but this was a learning exercise more than a troubleshooting one) and relaunched the media player, this time “Turn On, Switch Off” played from my Neo. I attached the headphones and the sound switched with a 1 second delay. The headphones crackle. I believe this is a manufacturing issue and NOT a software one, it sounds as if my cord is frayed and introducing signal-noise. I recall reading another complaint of this. I very much plan on replacing the headphones this week for that reason but since it doesn't “crackle” unless I move I can use them for learning over the next few days.

Applications began crashing at that point and my brain's “troubleshooting” mode kicked in. About a minute later I found that I was using 100% disk space – somehow my microSD card WASN'T detected or mounted and I'd copied a 13 MB file to a space-limited filesystem. Oops.

I deleted my music file and proceeded to set up USB networking (well documented in the Wiki). I immediately installed nano, my text editor of choice, and then SSHed into the Neo from my Debian system. This is much easier than typing on the on screen keyboard, but mainly because it's more familiar. Then, using the Neo handset I launched the web browser to see how well it worked.

The browser is WebKit based which is pretty cool. It renders my site pretty well and renders MOST sites pretty well. Some sites, however, look a little smooshed. In time people will be forced to design sites for "the mobile web" as well - and standards compliance makes this easier - but for the most part, the internet will still be usable on the Neo's browser. I haven't figured out how to do it yet but tilting the phone into landcape mode may help a lot. This should be done automatically on the Freerunner, kind of like the iPhone does, with a simple tilt of the device. The GTA01, however, doesn't have accelerometers so switching orientation is a manual process.

Over all, my first day with my Neo was fantastic. The phone works at least as good as my old Motorola RAZR did after a year and it's powered by Free Software – it's already a winner. Over time, this system will improve and I plan on watching it while it does, sticking my hand in where possible. The excitement hasn't left me yet, but my desire to “make it my own” is growing now and I've already found some “issues” I can investigate. :)

I'll have plenty of happy hacking ahead of me. During this time, I'll review OpenMoko software more completely, breaking down individual applications at a time. I'll hopefully be able to write some "official" documentation as well as fix simple things in the source as my skill with programming improves (my second reason for buying the Neo).

Having "played" with the OpenMoko development images in QEMU for a while, I finally decided to "pull the trigger" on a first-generation, "developer only" Neo1973. What makes this purchase so exciting to me is that I'm finally able to support a group of people working for Freedom and openness that try to meet that goal with a tangible product. The OpenMoko community is very supportive of Free Software and FIC itself, the company behind the Neo phone, considers, to it's core, Freedom to be crucial to a sane mobile world.

The Neo isn't ready for "everyone" yet and as of today there's no real idea when it will be ready. The next generation phone being "ready" is determined quite a bit by how well the software is polished on existing Neo phones. The plan is to release the GTA02 and the first "stable" OpenMoko software together: a respectable goal but one that won't come to fruitation until the GTA01 gets enough polish and I hope I can be of assistance to that.

You can certainly expect me to do reviews and provide information about the Neo1973 and OpenMoko itself as I get the device, use it and hack on it. I'm expecting problems with it, and gaps in the feature-set. "It's not ready" they say. Hrm... Where have I heard THAT one before?

Matt Hartley wrote an article about why he believes Ubuntu beats Debian. My responce is below.

Firstly, let me apologize if I am addressing the wrong Matt Hartley. There is no easy way to contact the author on Juniper Media's sites so I had to resort to Google.

Secondly, I'd like to give my response as a long time GNU/Linux user, developer and Free Software advocate.

You begin you article with a phrase, and a basic assumption, here:

"Like Fedora or OpenSuSe, it was developed with all users in mind. Unfortunately, what many of the geekier users to this day fail to wrap their minds around is that this "Windows thinking" they complain about is more or less about ease of use, versus "choice.""

This is incorrect. You make the assumption that the goal of all GNU/Linux distros is to be "the most popular" and this can't be further from the truth. From Fedora's front page (http://fedoraproject.org) you can find: "Fedora is a Linux-based operating system that showcases the latest in free and open source software. Fedora is always free for anyone to use, modify, and distribute."

This does not imply it's built "for anyone" or "for everyone" - it's built to showcase Free Software; it is incidentally AVAILABLE to everyone.

The OpenSUSE project says this "Promoting the use of Linux everywhere, openSUSE.org provides free, easy access to the world's most usable Linux distribution, OpenSUSE. The OpenSUSE project gives Linux developers and enthusiasts everything they need to get started with Linux." (http://en.opensuse.org/Project_Overview) I will say, however, that (politics aside) SUSE Linux (the non-free version of GNU/Linux packaged and sold by Novell - please note the distinction between that and OpenSUSE) actually manages to hit it's target audience ("average" corporate users) pretty well.

We now come to the Debian project: "The Debian Project is an association of individuals who have made common cause to create a free operating system." (http://www.us.debian.org/intro/about) What can't be seen in this plain text message is that the word "free" links to a page (http://www.us.debian.org/intro/free) explaining about Freedom, not price.

As a side note I'd like to mention a non-Linux based operating system here: OpenBSD. As you may or may not know, OpenBSD releases every version coupled with a song used to address issues in the Free Software community, and often with OpenBSD specifically. The 4.2 release song debuted this week and features a verse:

"Slow and steady wins they say but this is not a race It's not about who takes a prize for first or second place

Imaginary rings of brass Were traded for real goals The vision and the mission lost For those with corporate souls"

That line there has actually made me rethink the use of GNU/Linux in favor of OpenBSD. It seems so many people are focused on GNU/Linux adoption that they've forgotten why GNU/Linux was something people should adopt in the first place.

"The idea that the average new user really ought to consider becoming more educated with Linux in order to merely use it is just plain silly."

We (in America) require that all drivers be licensed - they must demonstrate an understanding of traffic laws and safe operation of the vehicle. This isn't to make their cars "more difficult" to use, these are in place because failure to know and follow these things puts others at risk. I'm not in favor of an "internet license" but I sincerely believe that a basic understanding of computers, how they function and how they interact with OTHER COMPUTERS is crucial to the ability to use them. Failure to understand and attempt to prevent botnets and spam bots is NOT an excuse for allowing your computer to be used by them. Drivers have accidents and even the best sysadmin will see a compromised system in his or her lifetime but when you are horribly lax from a disinterest in learning, you are just as responsible fro the damages you cause others. Education is MY attempt to mitigate that damage.

"I hope that most people remain clueless as it does help make sure that our favorite companies giving away their distributions can still successfully provide tech support as a business for their product. Even for generally basic tasks."

I find it somewhat ironic that Linux and GNU were created (and purposely opened) to prevent such a thing. Microsoft, Apple, Adobe and others stake their fortunes on keeping users restricted. GNU/Linux's sole strength is that it empowers them.

If users had the understanding of "basic tasks" the software economy wouldn't collapse. I'd argue that it may actually make it stronger. You have to know that in my belief of this I was raised by my grandmother who is a technophobe. When cellular phones came out they were "useless" and then "too expensive to be worth it" and then "too confusing" and now it's "impossible to live without". There are a ton of things today that we take for granted, like managing a few gigs worth of music (iPod-like devices), having online video conversations (Both real time via Ekiga and in cut-time with YouTube responses) and filling out a form to have a complete, attractive website generated with no understanding of code (CMS). These things didn't pop up because they're simple, they popped up because people who understood the basics took it a step further. EVERYONE has a point when they say "I don't know" and need to hire someone to meet a need. If we raise the bar a little bit it just means that the support people will be cranking out new innovative products to meet needs.

"If you place most people into the "command line" world while expecting success, you will instead hear the pitter-patter of their feet as they run back to their previous OS."

They'd also be a bit freaked out if you placed them in the operating room during surgery... Which is exactly why we keep them out of the operating room. We don't argue that operating rooms are "failures" because they're messy. The same applies for operating systems. the average user has no real reason to USE a command line and can operate a system without access to it. But for the digital "surgeons" who use the command line to keep servers up, to develop the next great application or to backup, sort and compress THEIR music collection faster than the GUI allows seeing the command line is a GREAT thing, not a failure. Ubuntu allows users to have a CLI-less system - Debian, being not designed for the "average user" includes is and expects you to use it. The only thing I see out of place with that whole situation is people who load Debian expecting an all GUI system. As out of place, I think, as those who grab Windows Vista expecting to do some hardcore NT kernel hacking - a hard thing to do with no source access.

"Who is primarily being driven to Ubuntu over Debian? It appears to be a mix between those who need consistency with their desktop; meaning not being asked which desktop they prefer to use, etc, and those who simply wish to have the bleeding edge thrown at them with their Linux installations."

That may be the group going to Ubuntu, I won't refute that. However, Ubuntu doesn't offer a choke hold on that market. Ubuntu rebases on Debian every 6 months and then goes into "freeze" until a new release hits. This means that users of Debian Sid and users of a "just released today" Ubuntu version aren't too far off and users of that same system in five-and-a-half months will actually have a more bleeding-edge system with Debian than with Ubuntu. The users who install the betas and release candidates would then fall out of the "average" realm into the "tester" realm, which I consider to be a critical part of development.

As to the consistency argument I'll agree and disagree. A default install of Debian is a Gnome install, just as Fedora and Ubuntu are. When you start adding non-default applications (Amarok, k3b) your "consistency" begins decreasing. This is a problem with the divide between Gnome and KDE, not with any strength or weakness in Ubuntu or Debian. It should be noted, however, that making the choice between Ubuntu, Xubuntu, Kubuntu, Edubuntu is not something "average" users have enough information to know without at least SOME basic understanding of the differences. A "Desktop Installation" of Debian, however, is Gnome. You have to do some "extra" work to install KDE or XFCE later.

"Until Ubuntu came along, you had to hold up a stick to see Debian development move."

Actually, I'll agree here both as a Debian server admin and as a Debian desktop user. I recall trying the "new" Warty releases of Ubuntu thinking "This is Debian's stuff just... snazzier". It wasn't newer, it was just flashier. Debian saw a decline in user base and took some notes from Ubuntu without compromising it's original objective - a Free Software operating system.

You seem to hold that up as a reason Ubuntu "beats" Debian right after a blurb about how Ubuntu depends on Debian. I feel this best explains why Debian and Ubuntu COMPLIMENT each other rather than explaining why one wins in COMPETITION with another.

"Yes, that was mean -- get over it,"

Interestingly, because I believe that comment was made in all seriousness AND is factually valid, it's the LEAST "mean" comment in the entire article. :)

"Let me simply go on record as saying that Ubuntu, while a fantastic distro, has little interest in seeing Debian become anything more than yet another cog in the greater Ubuntu machine."

Interesting that half-way through both of our views switch: as a Debian user I feel the need to defend Ubuntu here. :)

Mark Shuttleworth IS Ubuntu in most senses that matter. He manages large things like the project management policy but also weighs in (and sometimes rules) on small things like the default desktop wallpaper (Ubuntu Edgy Knot 3 had the sexiest startup and shutdown sounds in ANY OS, ever, but was pulled by Mark for being too flashy. Incidentally, you can download those sounds from my site: http://www.foreverdean.info/kevin/downloads/Sounds.tar.gz). Mark Shuttleworth was also a Debian developer and was committed to the ideals of the Debian project - namely Free Software. This is evident in the Ubuntu Philosophy (http://www.ubuntu.com/community/ubuntustory/philosophy) and by official and non-official initives by Mark (http://www.markshuttleworth.com/archives/130 and http://www.markshuttleworth.com/archives/131).

Mark started Ubuntu mainly because he felt that as good as Debian was, by not focusing on the desktop SPECIFICALLY Debian wouldn't ever been the premier desktop operating system. He contemplated running (as all Developers can do) for Debian Project Leader but ultimately decided that since being the most popular desktop wasn't Debian's goal, he's create a fork that did aspire to that. Any reluctance on EITHER side seems best explained by a conflict in project aims (untested real-time audio patches aren't very important in a stable server install) rather than a desire to have Debian in it's role as a "cog". Mark more than anyone recognizes that as soon as Ubuntu has cogs and not a community it is already a failure.

"Ask any Mac user and they will tell you a thousand reasons why they enjoy their Mac: it just works, the software, etc."

I would dump GNU/Linux in a heartbeat if Mac OS X were liberated today. Not because of the consistent interface, easy to use tools or "it just works" but because it's a damn powerful mash-up of hacker-friendly tools and user-friendly applications. Then again, if Apple released the Cocoa interface as Free Software GNU/Linux would gain that as well and also be build on totally Free Software. Hrm... Perhaps Apple, not Microsoft, holds the "missing key" in making GNU/Linux "adoptable".

"In the end, it comes down to whom the distribution is trying to target. Debian is a fantastic purist distro. And Ubuntu is one of a few great options for complete newbies."

I couldn't agree more.

"However, it's when these two worlds cross within the Linux universe that things get a bit sticky."

I am under the assumption that you make a living as a journalist with geeky passions - that said I understand the point of these tantalizing articles is to grab a reader and keep them reading. However, I appeal right now to the geek in you. When those two goals are mistaken as the same goal it creates tension and misunderstanding. As long as journalists, bloggers and end-users see Debian's "failure" riding on the numbers of users then both Ubuntu and Debian (and in fact, all of GNU/Linux) have been attacked. I mean, sure, Ubuntu has more users than Debian, but Windows has more users than all GNU/Linux and BSD and Mac OS X users combined. Obviously, Windows isn't a measure of ideological, ethical or technical merit but by placing the burden of popularity on a specific distro you imply that having more users makes it a success - a comparison every distro will always fail at.

It would be a beautiful day when users of GNU/Linux stop competing with each other to gain users and start competing with Windows and Mac OS X on the values that are important to them. Ubuntu is more usable than Windows (I'd also argue than Mac) and Debian is more Free than Windows (and Mac). Almost every Debian user and developer would rather see someone use Ubuntu over Windows - in this case both projects gain by working together rather than bashing one another.

We all fail when we get caught up chasing imaginary goals rather than attaining the ones we can.

Something OpenBSD recognizes quite a bit.

-Kevin Dean

I came across an article today that caught my interest. It can be found here [OSweekly.com].

Below is my reply, bolded text is quotation from his article.

Matt,

Firstly, thank you for the mainly neutral article on Libre vs restricted software. Granted, I don't expect anyone to be fully neutral but I do expect that legitimate, professional writers address topics in such a way.

I've very firmly in the "purist" category. Many years ago, while running Windows XP I got fed up with constant issues and sought out an alternative. I ended up installing an old (and admittedly pretty bad version) of Debian, perhaps Potato. It didn't suit me, and I ended up installing a few other things. I managed to settle for a while on SimplyMepis, a distro known for blending Libre and restricted software for "ease of use" and "hardware compatibility."

However, I began noticing many of the same issues as I did on Windows, crashes, lockups and glitches. Granted, it was much better than XP, but still annoying enough to hamper my productivity.

As I learned more and more about my system, I came to understand that while GNU/Linux was technically superior, it wasn't something inherent in GNU/Linux itself, but stemmed directly from the freedom protected by the GPL and other Free Software licenses.

I now run 100% software libre on my own computers, including my system at work. I wasn't born some software nut, or raised in a nut case family distrustful of everything, I learned to value my freedom, and see it all as an ethical issue, from HAVING it.

Knowing a bit of my progression, I'd like to address a few things from your article.

"I suspect that many purists are concerned about possible contamination of open source as a whole and fear the potential for quality being lost in the shuffle. In short, who knows what their deal is?"

At this point, it's no longer about technical superiority for me. That's why I moved to GNU/Linux. However, I've come to believe that "Freedom" is the greatest for of superiority. If an operating system supports one hard drive, one video card, one processor and respects my freedom, it's superior to a system that supports ten thousand devices and trods on my freedom. My personal fear when discussing "Closed source" software is that people will adopt GNU/Linux without understanding WHY it's better, and because of this, they won't learn to value their freedom. As more restricted software is "standard" on GNU/Linux, quality will gradully decline - ultimately GNU/Linux will be just as buggy as Windows and just as restrictive. Everyone looses in this case.

"To those who still choose to harbor harsh feelings toward proprietary software, I think I would point out that like it or not, some distros have been and will continue to include it. Now, I’m neither downing this fact or supporting it, I’m merely pointing out that this is a fact of life and we might as well make the best of it."

I don't harbor harsh feelings towards non-free software; I think the vast majority of so-called "purists" don't either. We've come to realize that restricting users is unethical; it violates the concepts of community. Information such as art and science should be shared freely: without benefiting humanity these things are useless endeavours. Instead, our issue is with the mentality that restricting users is okay, and even more, that restricting users should be "common". If Microsoft or Adobe has a program today that is restrictively licensed and tomorrow releases it unchanged under the GPL the unacceptable actions have been removed. The code is the same, but it's now Libre!

"That and to back off when criticizing the distributions that have finally met the needs of the common user through straightforward simplicity."

I'm a bit confused by this statement, truthfully. Both because I'm unsure how you intended it, and because I know I already have strong views about what it implies.

I see nothing wrong with making an operating system simple. I'd not use a system that wasn't simple. This statement was tacked on behind another statement involving "purist" distros which makes me believe you're implying that "Imposing license restrictions on users makes things simpler, more straightforward and better."

This is where I disagree, if that is what the statement meant. Back to my previous point of GNU/Linux getting it's technical superiority as a RESULT of freedom. But to add more, I can't find a single instance where I'd feel it's beneficial to be restricted to do something with my computer.

If a wifi card doesn't have libre drivers then that card is a paperweight, it holds no more value than the materials it is composed of. The problem is that users buy hardware designed for Windows and feel irritated that it doesn't work with GNU/Linux. Having had the opposite happen, I know the irritation. I've been building systems for GNU/Linux for years, and I've got no complaints. On the flipside, I don't expect Windows to install flawlessly on that same system (for the record, i've tried and I have yet to have a sound card work without "tweaking").

"Another factor that is helping in noticing the rise in proprietary software inclusion with specific distributions is the complete lack of open source alternatives."

I disagree here. The issue isn't so much that this software is being added, but that users believe it's needed. The example you proceeded to give was a movie editing application.

While many people will immediatly react to this sentiment, I'll say it. If you can't do it with libre tools the thing shouldn't be done. I have told that to a screenwriter and CG artist who depends on many restrictive tools. I have told that to employeers right before quitting jobs.

"I believe another simple fact of life is that people are going to have to get used to the issue of Windows migrants bringing their need for specific applications along with them. This would certainly explain the explosion is WINE popularity for sure."

I believe it's sad if this is where anyone's settled. I don't feel antipathy to those that use restrictive software, I feel sadness, actually. Nobody likes feeling powerless. America is dependant on "foreign oil" and it makes us feel restricted. We're all on a budget, we have a limited suppy of money, and it rankles everyone at some point. Depending on ONE tools so much that it paralyzes you, and prevents you from making certain decisions is sad and a bit scary to me. Imagine the state of the world if mechanics couldn't choose to use one brand of tools over another, or if doctors were required to use one brand of drug irrespective of it's effectiveness or the side-effects. The same thing applies to software, perhaps even more so considering that alternatives can be created with freely shared knowledge and time.

"So, if we are so quick to allow proprietary applications in WINE, why the resistance and even downright resentment with proprietary applications then? Simple, the rules of the GPL disallow it. This may not seem fair, but regardless, this is simply how it is."

Many people use Wine to run games or restrictive Windows applications. I once asked "If GNU/Linux is a libre OS, why make a tool to run restrictive applications?" I asked the same thing about the existance of the ReactOS project (http://www.reactos.org/).

Wine and ReactOS are essentially bug fixes to the problem of restrictions. In a world with several mature, Libre operating systems, it is understandable to try to introduce people to freedom gradually. Let people try GNU/Linux and bring over that application or two they are still dependant on: in time they'll learn to do without it.

The problem Wine and ReactOS case is that people look at it from the other side: even GNU/Linux has non-free things, so it's not so bad. As more and more people come to GNU/Linux because they CAN bring their non-free apllications the messages, ideas and concepts of Freedom are dilluted more and more.

I firmly believe that GNU/Linux does only ONE thing better than Windows, and that is be a Libre OS. When you sacrifice that, all is lost, because now you have a restrictive, buggy system and it doesn't support "all" of the hardware around.

Thanks for taking the time to read that. :)

-Kevin Dean

It's been a long time since I've actually seen a Gnome vs. KDE battle end with more than "Use what works best for you". It seems that not discussing the Gnome vs. KDE thing is the "political correctness".

But Linus, being his typical loud-mouthed, arrogant and attention grabbing (these things are true, and are not bad in and of themselves. If you dispute this Linus, e-mail me, everyone else can shut up. :) ) self has fired off patches to Gnome in hopes of improving things as he sees fit. The debate, it seems, has stirred up again and is visible on Linux.com and Linuxnews.com.

My rant today will be about why both KDE and Gnome suck. I'm not even going to say "E17 is better than both of them" or "Afterstep does this." I'm simply going to comment on why both KDE and Gnome are not perfect solutions, and why you should be critical of everyone's opinions, and form your own before deciding.

For the sake of this article, I will use DEFAULT settings for KDE 3.5 (Nothing major has changed in any of those versions) and Gnome 2.14. It should be safe to apply these opinions to anything in the same version family.

Let's begin this with the biggest part of the desktop environment, the File Manager. It's a part of the desktop that a LOT of people ignore, but it's something that will play a deciding factor.

Nautilus vs. Konqueror

By default, Gnome's Nautilus File Manager uses what some people call a "spatial view". Clicking a folder opens another window while keeping the old one open. This behaviour is perhaps the most complained about feature I've ever come across while discussing the matter. For the record, there are two ways around that. Double middle-click will open the selected folder in Gnome's "Browser" mode. By selecting Edit -> Preferences -> Behaviour and choosing "Always open in Browser Window".

Konqueror, on the other hand, uses this "browser" mode by default. In fact, the File Manager IS the browser, much in the same way that Windows Explorer is (and Internet Explorer before that). In this case, however, it doesn't make the browser and the file manager suck. :D

On the matter of network transparency, Nautilus might be a bit more intuitive, but it is quite a bit less powerful. For instance, from a default Nautilus install, the option to view Samba shares is immediate, nothing needing configuration. The fact that icons are visible from the beginning make it quite a bit more intuitive than Konqueror.

Once configured properly, however, Konqueror excels, making the transition from local file to network file almost seamless. Nautilus is unable to CHMOD files via FTP, for instance, while Konqueror does it with the same look and feel that one would be used to while editing local files.

The final differentiating feature of K vs. N is the file movement dialogues. Most KDE users are familiar with his option, which appears by dragging and dropping a file from one location to another. Once the mouse button is released, the user has the option to cp, mv or ln the files. Nautilus merely moves the file.

At first, this annoyed me quite a bit. However, as I began using the GUI for more and more things (Yes, that's right, I'm a Linux user who had to LEARN the GUI... CLI is quite intuitive and works the same on ALL DE's) I found that this was less of an annoyance and more of a time saver.

Having evaluated MY personal use, I tend not to move files at all. I download everything to the Desktop, use it and then remove it. If it's a wallpaper, I move it to /home/kevin/content/theme/wallpaper. If it's a source tarball I'll FTP it to my server and delete it, or compile it and remove the sources. Everything OFF of my desktop is rather immutable. When I alter something in one of those folders, it's usually for backup, not use. I like COPYING files I'm backing up.

With those things in mind, both Konqueror and Nautilus are etrememly stable. Both will copy files and show files without many issues. Because of the few things they do differently, and my personal use, I declare Konqueror the winner in the file manager battle.

One other thing that nautilus and Konqueror both do are draw the desktop itself. They control the wallpaper on the screen, the icon placement all that jazz. As of Gnome 2.8, Gnome was UNABLE to set different wallpapers for different desktops. This is something I rather like about KDE. I have a dual monitor setup. The idea that I must have a repeatable wallpaper to be consistant is bad. The idea that my wallpaper must be unbroken in bad. Until my desktop environment can eliminate the physical space between the edges of the two monitors, they should make the break as smooth as possible. With KDE, and the different wallpaper, this is handled by treating the monitors differently, which is much less jarring to me that having a wallpaper broken mid-way.

Kwin vs. Metacity

An often overlooked part of a desktop environment is the window manager. Of course, anyone familiar with Compiz or Beryl will instantly refute this, in the 2D world, nobody really cares. :D

There is only one real difference between Kwin and Metacity in terms of usability in my mind. Kwin has "Advanced Window Placement" options. One thing that is CRITICAL for me is that my windows open where I want them. Kwin allows me to place each window, resize each window, pick a desktop and have these options saved and applied automatically every time.

I always open Kopete on desktop 6. I like that when I open Kopete while VIEWING iceweasel I don't have to move it over to 6 from 1, which I'm currently on.

Metacity does not provide many placement options, and only one in terms of which virtual desktop applications should go to. For this, it's quite easy for me to declare a winner for WM between "Gnome" and "KDE". Kwin for the win.

Kicker vs Gnome Panel

Kicker and Gnome Panel are the little applications that run at the bottom of the screen (Gnome Panel is at the top AND bottom) and are a container for your application launchers, contains your taskbar and your system tray.

Let me begin by pointing out something that has ALWAYS irked me. Most window managers put their window controls in the top-right corner of the window. To close a window, top-right. Most DE panels put the useful controls at the bottom. To open an application and then co close it, you have to move your mouse across a full screen (or two, in some cases) to close it. Keyboard shortcuts can eliminate this, but putting the controls at the TOP of the screen can save our little mouse a lot of "walking".

For this reason, I have to give some points to Gnome Panel. On a default install, the program menu is up top, just like Metacity's window controls.

Here's where it gets clunky though... Gnome splits it up! Damn you!

Kicker defaults to the bottom. However, it's a single panel, which can be dragged to the top in about as much time as it takes to drag the mouse up top to close the window.

Kicker has some nice animations and informational bubbles associated with many different things. While this isn't exactly a plus for usability, it is nice. Eye candy is ALWAYS good. :D

As a sub-set of Kicker/Gnome-Panel, I'll address the pager applet for both. Gnome's desktop switcher is VERY bland, in my opinion. Bland being that it is just a set of boring boxes, where KDE's pager pulls the icon of the application and places it in the little desktop. This is useful when you accidentally drag a window off the desktop, or just want an overview of what you've got.

However, I must add in now that I think both KDE and Gnome kicker/panel are lacking in comparison to the general setup of E17's bars. Anyone who's used a dual monitor setup for over a week will appreciate it. However, KDE and Gnome both tend to treat the monitors as extensions of each other. For instance, set Gnome or KDE to have 6 desktops on a single monitor system. Then, add a second monitor. It shows (and treats) each desktop as a double-wide monitor. This is a behaviour that I absolutely hate.

E17 allows each monitor to have virtual desktops. I can keep my browser open on one page, move the mouse over to the other monitor switch to my system monitor, pop over to my mail client and then switch back to the terminals that's open, all without moving the broser off the other monitor. I can still drag applications across to the other monitors, of course.

The final difference between KDE and Gnome I've noticed is contant and nagging; the taskbar. Gnome-panel tends to handle KDE application events poorly, where Kicker handles Gnome events pretty well. It is quite annoying to have a never ending Kopete bar flashing away at me.

GTK vs QT

One thing that is important to touch on briefly is the toolkits used to build the applications. The use of certain widgets and toolkits give applications a consistant look and feel. This is why Gaim is seen as a "Gnome app" even though it's NOT released by the Gnome Project.

Both, when used exclusivly, are pretty good. I don't really know much about the underlying technical aspects, but between GTK and QT, I prefer the GPL of QT better than the LGPL of GTK.

It's when used in comparison to each other, or with each other, that they display their relative weaknesses, which I think epitomize KDE and Gnome.

Gnome seems to accept themes MUCH more easily than KDE. A KDE theme touches many, but not all applications, even applications distributed as part of KDE. Kopete for instance, accepts PARTS of the KDE theme, but other parts of it are ignored.

GTK themes are applied more smoothly, affecting all GTK applications evenly.

I'll end my review today with "KDE" vs. "Gnome" without getting into the specific applications that also define my opinions. However, I think it important to note that I use applications from BOTH KDE and Gnome every day. One problem I notice that affects both KDE and Gnome is their ability to work together.

Running Kopete in Gnome looks out of place. Just as running Iceweasel in KDE looks like the proverbial sore thumb.

This is something many people have dismissed as inevitable, or insignifigant, but it's something that has always reseted on my list of flaws with the Free Software pool. I like my applications to all look consistant. I don't care if that's KDE style QT consistant or GTK style Gnome consistant.

The Portland project attempts to address how underlying things work between desktop environments, but it won't touch how they LOOK. There have been attempts in the past to do something with it. the gtk2-qt engine, for example, will do it's best to render GTK apps like QT ones, making Gnome programs look more like KDE ones. For the most part, it does a good job, except in one very noticable case... Mozilla stuff. This is kind of big for me, but given the advantage of a uniform look, I'll cope.

There is, however, nothing like this to convert QT applications to GTK. If there is, it remains elusive. The best option I've seen on the Gnome side is MetaTheme, which died on version 0.0.6, and that was several years ago. What this did was caught QT, GTK and Java UI widgets and rewrote them in a common theme. It could use on of several pre-made themes or import MSSTYLE themes, making everything look the same. I really like the idea there, but the lack of momentum meant the application had little flexibility.

Since neither KDE or Gnome can theme applications from the "other side", I consider BOTH of them flawed. This late in the game, KDE should have developed a way to integrate non-KDE applications better. And if Gnome is focused on on "sane defaults", there should really be some method of enforcing these defaults. After all, a User Interface that isn't consistant is really bad.

Both KDE and Gnome suck. End of the debate (not really, but I hope readers realize that they BOTH need work. :)