According to Michael Larabel Thunderbird will become the default choice for Ubuntu’s mail/newsgroup client! That’s a strong prove of progress that Thunderbird has made, and trust me, Ubuntu does not swap the defaults that easily.

Well, it now seems that we’ll have two of our products used as pretty important defaults in this most popular Linux distribution – Firefox and Thunderbird! (although we do have a strong competition for the browser slot).

Congratulations to the team working on Thunderbird!

Jeśli ktoś ma wolną godzinkę i chce pomóc, to zapraszam na Wiki tego tłumaczenia.

So, Linus announced that (most?) of the bkl-removal tree has been merged to kernel for 2.6.27!!!

BKL means Big Kernel Lock, and is the system responsible for several slowdowns in Linux, including several delays in booting process. While from what I understand the first chunk of patches will not solve everthing, it’s a huge step forward!

Congratz to Linux team!

I dream about the day when I’ll be able to move audio streams live between my 5.1 sound system and headphones without restarting apps.

I dream about the day when I’ll be able to use 2 laptops to emulate dolby surround by setting one laptop as a front speakers and another as rare speakers.

I dream about the day when I’ll be able to make browser not play sounds (flash ads!) while Amarok is loud. Until Skype is calling of course, when Amarok goes down to 5%

PulseAudio promises that, and I’m happy my next Ubuntu will go this path.

I must say honestly, I never liked Mandriva (Mandrake) too much. In the times when I was working for MaxWeber, I tried to use it as my main distro for some time and I felt like in a candy shop, with all those plastic colors, plus the overall experience was pretty “old-fashion-linux-like” in terms of feeling that I’m using a combination of pieces glued together. It’s my old and well known accusion against KDE. (Gnome was a bit better, especially in Ubuntu edition)

I hope that the past did not influence my review, thus I decided it’s better to inform readers about it, to stay fair.

I focused during this test course on the things that are specific to Mandriva distribution plus the overall experience during 10-20 minutes test. It’s not sufficient to put any general statements about the quality and (same as with OpenSUSE 10.3 UX review) I’d like people to give it a try before judging. I did not install Mandriva 2008, and this may influence the experience I have had, altough most of the things I focused on are unrelated to what could have change in the installed system. (that statement bases on my knowledge about Linux and installation processes).
The first thing you see after booting from the CD is a nice, blueish theme. It’s still a bit too candy like, but I feel ok with that, and I’m sure many people will really find it attractive.
A little glitch with grub is that you see a (themed) menu but with only one option to choose and 10 delay before auto starting it. Most people won’t notice, and you can’t say it’s a bug, but what for? I can guess that it’s made to allow some people in corner cases modify the launch options, but once more the sin of Linux Distributions is presented. Majority experiences options made for minority.

Boot screen is just like a modern boot theme in Linux distros. I find it less attractive than suse’s one, but that’s probably because I spent so much time with designers and they raped my aesthetics, so in result I love when things are very small. suse’s progress bar is smaller.

The first thing that I found different to most live cd’s is that you have to make some choices before seeing the deskop. You’re choosing languages, time zone, time format, 3d special effects (metisse, compiz fusion or nothing).

I can understand the reason for which they did it. Especially the 3d part makes you aware of what the desktop will be like. There is a clear description of each choice, so even a novice user will understand the consequences of his choice. User who’s already aware of what he’s doing (experienced linux user testing mandriva) will like this idea. I like it. It makes it possible for me to choose in 3 steps how I want the desktop to looks like.

On the other hand I’m sure novice users won’t like it much. The screen is presented before you can see the desktop. It means that the very second UX you have (after the theme igniting some emotions and attitudes) is a choice. For a newbie, the choice is unpleasent element and he may feel unconfortable that in totally impredictible environment, unknown to him, he’s forced to make choices. For example the time (local or global) choice is the best I’ve ever seen since instead of guessing it shows me the result of each (Which time is a right one: 17:34 or 18:34?) – but it’s still a choice. My computer doesn’t know what time is it. And it could easily be autoguessed basing on my partitions, availability of Windows or no, etc. And even if the choice would be wrong, I believe most users of live cd won’t notice, while the blocking choice screen cannot go unnoticed.

What’s the conclusion? Hmm… I think I prefer when Live CD doesn’t ask me at all. I could live with the button on desktop like “Configure the screen” making it possible for me to change those most visible options that will allow me to see what I can get with Mandriva choice.

Next part is, as always, welcome screen. Same as in OpenSUSE, after launching the desktop for the first time you see a big welcome screen with a lot of words describing Mandriva and oped-on choice to show it everytime you launch a LiveCD. I may be wrong but my guess is that since it’s LiveCD, my choice to “not show” will not be remembered. So the option is misleading.

The overall idea of welcome screen, borrowed from Microsoft Windows is not clear to me. It may be distrupting, it may be useful. But it’s definitely not the best solution to the problem it’s made for. With today’s technology it would be so much better to provide a “live tutorial” for the user. Imagine – you launch the Live CD and small popup appears – “Would you like to follow the short tutorial on Mandriva 2008?”. You say “no” and it disappears so that you can use the environment as you want. (never block the user!), but much more funny things happens when you click “yes”.

With a set of opacity, animations, and overlays the desktop changes into a tutorial session. You see a small red circle around “Home folder” icon on the desktop and a blue arrow pointing at it, and a tooltip describing in a few words what it is and how you can use it. There are also a few small expanders that you can click to get more info about how does it relates to Windows XP UI, or MacOS one, or gives you some more advanced tips.

Once you click “go” the red circle disappears or moves to “Menu” button in the left-bottom corner and a new arrow+tooltip appears describing this element.

You may be asked to try it and click on it and once you did, the tooltip changes and describes elements of the menu.

Then, you’re led by hand via whole UI of the Mandriva desktop plus you can go for more advanced tutorial where you gets a list of tasks you may want to solve “How to change the default sound card”, “How to change the screen resolution”, “How to change the theme”, “How to run Windows app”, “Where is my Windows Desktop folder” etc.

In each case you have a live tutorial session with a helper that can click on the buttons instead of you, or let you click and presents you the whole workflow you’ve got to do in order to get the task done. It shouldn’t be too fast, and it shouldn’t be “appear/disappear” like. It should be smooth, animated. I imagine the pointer to move to the button you have to click, and once you click it, to move to the place where you have to do the next action. In each case it should be explained with text and possibility to dig deeper for more info.

It would be so much superior over current “text+screenshots” or “video tutorial” sessions! And with XGL/AIGLX, DBUS, HAL we have everything we need to do this. I hope vendors will start using the potential we have.

Once we’re done with closing the welcome screen without reading it, the desktop presents itself with normal KDE bottom toolbar, themed for Mandriva, a set of icons on the Desktop and blue wallpaper.

I could not make a screenshot (possibly because of metisse) because every screenshot was totally black. Minor bug, I doubt a beginner will notice, but blocks me from providing you a screenshot.

The desktop icons are – Home folder, Trash, “Welcome”, “Live Install”, “Removable devices”, “Update to version”, “Register”.

I dislike the amount of it. Register? In LiveCD? Let me install first, huh? Update? In LiveCD? The same…

“Welcome” opens the welcome screen (which should stay. Once you remove the welcome screen opened by default, let the user have ability to click and open it when he wants), but the screen itself is a bit chaotic.

It presents 6 boxes about “what’s cool in Mandriva”. First of them described “more fun”, presents a game screenshot and talks about Transgaming Cedega in PowerPack (payed version). The other 5 are about Compiz Fusion, mp3/ac3/mpeg, migration from Windows, etc.

I don’t like the idea that the first box the new user reads is a trick. What he reads is that he can play games, but what he’ll not understand is that he will not be able to do this for free after installing the system he just launched from LiveCD.

The “virgin” Home folder opening is better than I had with Ubuntu or OpenSUSE. The directory contains empty dirs like “Downloaded”, “Video”, “Music”. and nothing more. I like it. It’s clean and that’s what the user could expect from just opened folder. It’s empty, right? There’s nothing inside until I put something into it. Great!

Konqueror which is file manager is a bit overwhelming with tons of buttons on the left and top, and tens of options in context menu, and I don’t enjoy that I see a big “guest” name (it’s a /home/guest folder name). Actions that I can make with it are not needed for the user at this stage (compress the folder, open in Kaffeine, open terminal here) and I believe it could not be visible by default.

I’d advice Mandriva to switch to Dolphin asap.

Menu is a traditional KDE start menu, with the very same bug as in each linux distro (from the newbie point of view) – tens of apps with totally unknown names. Imagine a WindowsXP user switching to Mandriva, or a user who never have been using a computer, trying to decipher “Akreggator” and “kbtobexclient” – those are the two top apps from the top menu “Network”.

Then, there’s “IRC Client”, “KNode”, “Comjmunicator – Kopete”, “Konqueror”, “Kontact”, “Virtual network connection”… should I read more?

The very first option that may be familiar to the user (but it’s not sure) is Mozilla Firefox, which is on position 10th. I don’t believe that anyone will read 10 unknown, misterious names one by one just to get to Firefox, but he may scan the menu list and find a recognizible icon earlier.

Overall, it’s a disaster, as always. LiveCD presents me 85 options, most of which are black magic names like “Envy24control”, “RME digicontrol”, “Kbluelock”, “KMon” etc.

Rarly there’s a description of the tool, but who the hell assumes that Joe Average knows what IRC and it’s client is?

Once more the majority will have to experience UI parts made for minority. Instead of “Document writer” we have “OpenOffice.org writer” (and 6 other pieces of OpenOffice.org named at the beginning of each line). Experienced user will recognize the package by the icon. Joe Average doesn’t give a shit what the app name is, but what he can do with it. Thank you.

The mixture of unknown names, huge amount of tools by default layed out in a flat menu (Firefox is as visible as kbtobextblue but the amount of Firefox kind apps is 1, while kbtoextblue like apps are 84) and the UI law that the amount of focus is splitted equally by the amount of data presented, so the “importancy” of each button is 1/”amount of buttons” makes it a deadly mixture. Most users will stop reading after third or fourth element, because what they look for when they first launch is the answer to question “What’s there”. And after reading first options they answer “Some shit” and close it. They became feeling lost, aware that they’re in unfriendly, unknown environment and how much they want to get back to something they know – usually Windows XP.

Advice – the menu should have much more levels. 90% of the apps should be hidden, especially those small things for “viewing file sizes”, “mapping keyboard” etc. The user should see “Web browser”, “Document writer”, “Spreadsheet editor”, “Instant messager”, “Music player”, “Video player”, “Mail client”, “Notes” and maybe “CD burner”.

When you use Metisse, on the right, above clock, on the desktop you see a cube made of 9 pieces that represents virtual desktops. It’s something new to me, and probably to most of users, but one can safely click on any piece and he won’t get lost. It’s nice.

The magic which makes the new user not get lost is that first of all when you click to different virtual space, it animates the windows on this space to slide away so that you can feel that the windows are now “over” your dekstop, or “on the left”. Secondly, the selected space pointer also animates in the preview cube, wich is consitent and recognizible. In the end, if you had any window opened on dekstop space 1, you see the icon in the preview, so you can easily get back to it once you want. I’m rather confident even my mum would understand what’s going on (she’s a beginner Ubuntu user not familiar with any GUI before).

Wait. Did I said NINE DESKTOP SPACES? Yes. And you cannot forget it because workspace switcher in the bottom toolbar takes 70% of the width presenting each workspace side by side and leaving minimum space for window title list. Hello!? Dear Mandriva authors! It looks horrible. UX of this is bad, there’s no animation, and when you switch workspaces using it, it jumps in one frame, without animation that makes the brain feel safe.

Advice: remove it.

Minor tech glitches are that Mandriva didn’t recognize my sound card properly (chose NVidia onboard instead of Audigy2. Same as Ubuntu. Why? Isn’t it obvious that if I have separate sound card out of the board, than probably I want to use it by default?), but also made something worse. Probably autoselecting some option in Audigy like Digital switch made the horrible buzzer sound like an old radio. UX drops down in one second. No GUI to change that, user is fucked.

I know I can use asoundctrl to change it, but that’s a bad default UX, right? Very bad.

I pefered what OpenSUSE did with their “no sound” idea

Also, as a desktop user, for the first time I saw the “cable powered” icon in the task try. It made me wonder if by removing cable it’ll switch to something before shutting down? To what if dekstops have no battery? Was it to hard to check that with HAL and not present me the icon?

Metisse itself, which I used for the first time, is quite nice. The above described workspace switcher is nice, much more usable than the cube (but cube is more eye-candy and better for markting effect called jawdropping).

I like that by default windows that I’m drag&dropping gets translucent while moving them around. I’m used to wobbly windows, so as always the old approach looks “sharp” to me, but that’s not what newbie will feel.

When you click on the window border or a corner, you see a ring menu (mr. Fitt is happy!) that allwos you not only resize the window, but also Z rotate, Y rotate and X rotate. Right now, I’m writing it using all my environment imagination because the browser window is flipped to the right, rotated by 20 degree and flipped to the top. Totally useless if you ask me, and lower Wow factor than the cube.

I didn’t find anything more in Metisse, so I assume most users won’t in the first-use session, as I was dragging windows, clicking everywhere and that’s not what users do until they’re crazy.

Summarize. Mandriva is one of the core pieces of Linux distro healthy environment for me.

As I stated many times, I believe that there are two kinds of distro that needs to exist for the market to survive and one that is good to have, but not required.

First are core distros, made by communities for communities, by hackers for hackers. Debian, Gentoo, LFS, Slackware etc. We need them to keep Linux being linux. To keep our most important ground freedom rights and options. To compile our software on our own IF we want. To combine pieces that we want carefully choosing the amount of libraries we have on our desktop. To have competing pieces like KDE/Gnome, plus less “marketing interesting” options like wonderful XFCE. We need them and we need them to stay healthy. It’s extremely important. If everything goes wrong, Goerge Bush, Steve Ballmer, Google and an unknown group of jews and cyclists will nuke all the servers, buy all the hackers and change the law somehow to make most Linux distros illegal we should do everything possible to make sure those “ground” distros survive. They won’t lead us to rule the market share, but they let us survive and stay “Linux”.

Then we have “user-oriented, vendor based” distros like OpenSUSE, Fedora, Ubuntu and Mandriva. They are exploring the ways to make UX feel better, to lower the learning curve and switch costs. They remind us all about who we work for – the users, and what the goals are. They pay the money to people who does the things that volunteers will not want to do cause those tasks have near to 0 “fun factor”. Documentation, debugging, QA-ing, Examples like Liberation fonts, AIGLX/XGL are pretty good to picture that. They also show teach us how the marketing looks like, how the mature development cycle looks like. All the things that group of hackers is unaware of. In terms of browsers you can think of it as a switch from the “Mozilla” project in 2000 into “Mozilla” project in 2007. In desktop world the same difference is between Slackware and OpenSUSE.

Third part are various companies that tries to sell Linux in some options, exploring the business models, like Linspire, Xandros, SLED, SLES, RHEL, etc.

It’s good to have them. It makes our environment better and richer. It improves the overall “Linux” thing. But I don’t see it crucial for survival (beside of that lack of it would probably strongly limit the second group, which would be bad indeed, so indirectly we may need them ).

Fourth group is an evolutional ground of small, experimental distros made for specific niche or brave new approaches. There’s a lot of them, they’re not popular, but huge. Long tail theory in real life case

So Mandriva is, in my opinion, important. Mandriva 2008 is nice, it’s usable, it’s way better than any previous approach I tried. It has nice theme, pretty good configuration panel, interesting metisse approach, clean home folder. But my user experience was worse than with OpenSUSE or Ubuntu. It could be because of LiveCD vs. installation partially, but not in all aspects. “I use combination of separated elements” is the worst part that I feel stronger here than in other recently release distros. The “tons of useless apps” experience is the worst I found since some old Knoppix and some minor problems with hardware detection exists.

Same as with OpenSUSE. The technology is here. If the UX people in Novell and Mandriva could spend a month of full time work on the first 10 minutes experience and less detailed “first month of use” experience, such polished distros would be so much better.

I can’t resist the feeling that those are all low hanging fruits. No sky rocket technology required to remove those problems. Heads up guys Mandriva 2009 may without a problem remove all those glitches

The initial feeling after login is… fresh. It feels fresh. I like the wallpaper, sound theme. The start menu is as always extremely cool and usable. Those three aspects make OpenSUSE first experience really nice. My gf spent about 10 minutes and had absolutely no problem to find herself in the new environment. She knew it’s different from Windows, but she liked they way how different it is.

The first unpleasant experience came after updating system with suse system update. Sound card went down. After some debugging it came out that all devices in /dev/snd were in group audio, and I was not a member of this group (nor any other system user was). I just added it to /etc/group, but that’s totally a blocker that makes the system useless. Imagine Joe Average on my place with his first, brave try of this Linux thing, when he can hear cute login sound on the first time, then he clicks on “update” icon to update his system and then, after reboot – no more sound Joe!

Second unpleasant experience came few minutes later during short user-testing session on my gf. She really liked the wallpaper set, she found the start menu very nice and intuitive, she clicked through apps/docs/places, read right column, wowed over search menu, but then said “It seems there’s not too many apps here” – user used to Windows like start menu with zillion apps feels a bit “empty”. She did not find the “Other apps” icon, but this may be due to the nature of this experiment with me standing over her head.

But the problem appeared when I pointed her this button. She clicked on it, and it took over 30 seconds for the new menu to appear. The new menu is a window, almost full-screen with huge amount of apps in a flat layout. That’s different to Windows, but nothing “bad”. Until she started reading the descriptions. “DMA channels”, “OpenGL”, “PCI”, “Partitions”, “SCSI”, “Samba status”, “Processor”, “X Server”… thank you! Stop!

Who the hell came with this idea? Let me guess… no one. No one actually took care to do this very freakin simple user action flow – login, click on Computer, click on Other Apps, read the first line, compare it with what the user expects to see. Hello!?

Clicking on “Home folder” icon on the desktop to see a window with “bin”, “public_html” directories and one file named “nautilus-debug-log.txt” is also something that should be considered as a suicide.

That’s it for now. OpenSUSE looks really good, it makes great first impression, more familiar for Windows users than Ubuntu, it has way more complicated installation process and lacks the LiveCD+Installer idea from Ubuntu which is so obvious once you get used to it.

Anyway, it’s really frustrating that with each and every linux distro I launch, it takes up to 3 minutes to find a first UX bug. Always. And I’m not talking about super-duper-heavy-LFS like distros. I’m talking about Ubuntu, Mandriva, Fedora, OpenSUSE…

We overcame huge limitations of X system, we developed eye candy to the extend never seen before, we have very, very secure systems, we have huge variety of distributions that create healthy ecosystem of competing solutions and ideas. We developed the best ever seen system of application management and installation, we made it all. Few years ago the blockers for Linux adoption were technical. Sound cards didn’t work. Monitor resolution was badly autodetected. Printer was unavailable. CDROM was not detected. Today, we overcame ALL of those. Virtually ALL! Last three or four hardware pieces I bought were much better recognized by my Ubuntu than Windows! Including USB headphones which are recognized as USB hotplugged sound card! No problem for Ubuntu! Huge problem with Windows.

But in the end, it’s depressing that we still fail to provide the UX without very visible, simple to avoid, flaws. In Ubuntu, you have great chance to see something like “/dev/sda2″ on the very first desktop you open after logging in. In SUSE you hit “nautilus-debug-log.txt” in your virgin home folder and “DMA Channels” as an example of “other apps”.

I know that users will learn this. After one day, such problems disappear and new patterns are memorized, I know that people with motivation (and the motivation is easily raised by the blue screen of death) will switch and will be happy in the end. But all those “mistakes” looks like ignorance. Like if no one did actually install his own distro on an empty drive and SEE how it works. For 5 minutes. To make those 5 minutes perfect.

Tomorrow I’m going to try Fedora 8.

I was also updating my flash player basing on the latest releases from Adobe, and realized that this company, and many others will have to exploit packaging model of modern distributions in order to achieve what they want.

See. Modern, user-oriented linux distros like OpenSuSe, Fedora, Ubuntu, are preparing set of packages for the release, and then “half-freezing”. When I use Ubuntu 7.04, I use a set of software that was ready by April 2007. There are two exceptions – security patches, and community contributed backports of the newer packages , but for the latter, I have to manually select that I want to get them.

It means that being a company, that wants to upgrade users browser, mail client, game, or office package, I should claim  that it’s a security release. It’s not an issue right now, since linux is not popular enough to be on most product managers radar, and the releases happen pretty often (half a year in case of Ubuntu), but as Linux will become more popular, I’m more than sure that it will start to happen. All companies I was working for would like their latest versions to be deployed for all users soon after the release. Not half year later. Also, what about users who will not upgrade?

Look at the browsers. Browser X ver 3.5 has been released on Sep 2010. The Ubuntu (by the time used by 35%  of end users on Earth!) release 10.10 uploads it and uses in their release. Users are happy, confetti is everywhere and We Are The Champions can be heard in the background. Win-Win.

Then, the vendor of Browser X prepares release 4.0, and they’re ready on Apr 2011. Unfortunately  the release cycle of Ubuntu says that Ubuntu is already freezed and will not use ver 4.0 in Ubuntu 11.04. So this release is delivered to the users with Ubuntu 11.10, 6 months after original release!

At this point, many can say “Yea, Mark is calling for synchronizing releases”, but that’s not a solution. What if 11.04 is so great, that people don’t want to migrate to 11.10? If it’s a LTS and majority will want to stick to it for a loong time? (see XP-Vista migration rates) or if it’s simply not good for some reason, and journalists advice to stick to 11.04 and wait for 12.04?

I think that the only proper solution is a vendor controlled backporting highway. A process that would allow a vendor (or vendor licensed volunteers) to backport apps (usually the more front-end user oriented ones) and deliver them with the updates to all users of a release.  Otherwise, vendors will start pretending that such a release is fixing some Scary and Serious Security Vulnerabilities  that might kill your cat or grandpa.

Business is business… :/