Review: Manjaro Linux 17.0.1 "Gellivara" Xfce

This is the next installment of my series of reviews to determine which Linux distribution I can use to replace my current installation of Linux Mint 13 LTS "Maya" Xfce on my personal laptop. The (not strict) criteria that I am considering are that the distributions should be well-known, which is reflected to some degree in DistroWatch rankings, as this implies that the distribution may have official or strong community support for popular proprietary packages; additionally, the distributions I consider should preferably have MATE or Xfce editions (though I'm open to other DEs as well), and should have a long (more than 3 years from now) support cycle or use a rolling-release support model.

Main Screen + Xfce Whisker Menu

The current distribution I am trying is the latest Xfce edition of Manjaro Linux. It is a rolling-release distribution that was formerly based on Arch Linux, though it still uses the Arch User Repositories (AUR) for many packages that the distribution maintainers do not officially test. It also has an official KDE edition, as well as community-supported editions for other DEs. I tested the 64-bit version (though a 32-bit version is available too) on a live USB made through the "dd" command. (This time, I used a USB stick that I have never used before, to avoid the issues seen in my recent review of openSUSE with my previous aging USB stick that I have been using for reviews for the last 8 years.) Follow the jump to see what it's like.

Featured Comments: Week of 2017 April 2

There was one post from this past week that got one comment, so I'll repost it.

Review: openSUSE Tumbleweed GNOME Snapshot 20170329

Reader msian_tux_lover said, "I think you missed the point and objectives of Tumbleweed - it isn't meant for beginners and more for tinkerers and hobbyists i.e. people with average or higher know-how of Linux like yourself for instance; that is why Tumbleweed is a rolling distro with newest/er versions of packages rather than a more fixed/controlled versions of most packages like LTS versions of your favourite Ubuntu/Mint platform. I have to agree that openSUSE Tumbleweed needs to be more polished (I am thiking of a similar edgy distro like Fedora but then again they aren't a rolling distro, although you 'could' upgrade the distro using fedup - a bit of a hit and miss there), but that is what openSUSE Leap is doing - a hybrid distro with core stuff from SUSE Linux Enterprise (SLE) with community userland apps. Tumbleweed can be rough round the edges - but it is running fine on my low end Acer notebook."

Thanks to that reader for that comment. Now that I'm done with my conference and the pace of work has settled back to normal, I can post more in this blog. This week, I will have another Linux distribution review as well as another book review out. More Linux distribution reviews are to come through this month and next month as I continue to search for a replacement for the distribution currently on my laptop. Anyway, if you like what I write, please continue to subscribe and comment!

Review: openSUSE Tumbleweed GNOME Snapshot 20170329

For the last 5 years, I have been running Linux Mint 13 LTS "Maya" Xfce as my main OS on my laptop. Its support cycle is only 5 years long, so its end-of-life is fast approaching (within a month). This has spurred me to seriously start looking into replacements/upgrades. In the interest of having an open mind, I don't necessarily want to lock myself into sticking with Linux Mint; while I may still be biased toward the Ubuntu/Linux Mint family, the collection of minor issues in the latest releases of Linux Mint may actually make me more open to other alternatives. However, I'm probably not going to go with some small one-person distribution, especially if the community around that distribution is small and it doesn't have officially-supported packages. My criteria for considering a distribution are that it should have either official support for Skype and Google Talk (both of which I use regularly) or it should have a large enough community to make unofficial support viable (through the implication that the distribution will last a long time); related to the last point, I expect the support cycle to be as long as possible, and in particular, to be more than 3 years. Additionally, while I am not completely opposed to KDE, my most recent interactions (in my most recent distribution reviews) with KDE 5 have left me somewhat less than impressed, so when possible, I will try to stick to MATE/Xfce when possible, thanks to my familiarity and confidence with their tools, ability to recognize peripheral devices, et cetera. Additionally, because I'm seriously trying to test aspects of these distributions for my daily use, I'm going to touch upon a few other things beyond what I might write for a typical distribution review.

With that in mind, my first test subject is openSUSE Tumbleweed GNOME. I've tried openSUSE before, but it has been a while since the last time. Additionally, its support cycle is only 3 years, but it does have a rolling-release version called Tumbleweed, so I figured I might try that. I created a live USB of the 64-bit ISO using the "dd" command, as recommended on the website. Follow the jump to see what it's like.

Review: Chakra 2015.11 "Fermi"

Main Screen + Kickoff Menu

Not only has it been a while since I've done a Linux distribution review on this blog, but it has been an especially long time (over 2.5 years, in fact) since I've looked at Chakra. I figured that now that KDE 5 (technically incorrect terminology, I know, but please bear with me, as I'm using this for the sake of brevity) is being used in Chakra, it may be time to see how a distribution I've rather liked in the past has evolved. In case you don't remember, Chakra was originally based on Arch Linux, but a few years ago, it branched off into its own independent distribution with its own repositories, though certain tools (like the package manager Pacman) are based on things found in Arch Linux. It focuses exclusively on KDE, and it uses a semi-rolling release model in which core system packages are updated less frequently in order to maintain stability, while front-end applications seen by users most often are updated more frequently to provide a competitive desktop experience.

I tried this on a live USB using the "dd" command; as in my review from over 2.5 years ago, neither UnetBootin nor MultiSystem work anymore for reliably creating Chakra live USB media. This wasn't a terribly difficult thing to do, but in any case, the Chakra wiki contains a little more information for people who might need a little more help with these steps (especially if they are new users from Microsoft Windows who are trying a Linux distribution for the first time). Additionally, note that Chakra can only be used on 64-bit computers. Follow the jump to see what it's like.

Review: KaOS 2014.04

It's been a while since my last review. Now I'm a bit more free because the semester ended over a week ago. At the moment I'd really like to get my hands on the official release of Linux Mint, but that isn't out yet. In the mean time, though, I'm going to check out KaOS.

Main Screen + KDE Homerun Kicker Menu

This distribution caught my eye from a DistroWatch review. That review concludes that it isn't clear exactly what the goal of this distribution is. Looking at the website more, I can't say that it's any clearer to me either. All I can glean is that this distribution aims to please more experienced users with a rolling-release model, maintain a small base of packages so that those will be polished before use, and target newer computers by using KDE and only 64-bit releases. I'll have to try this distribution out to see if there is any more information regarding the target audience of this distribution. I tried KaOS on a live USB made with MultiSystem. Follow the jump to see what it's like.

Review: Linux Mint MATE 201403

Main Screen + Linux Mint Menu

It's been a while since I've done a review. In fact, it's been a while since I've posted in any form, because this semester has turned out to be a lot busier than I anticipated. It likely will remain so until it ends; the only reason why I can post a review right now is because of spring break, and even that has been busy for me. Anyway, I initially wanted to do a review of Frugalware because it looked intriguing, but I couldn't get the live USB to work. I'm reviewing this (which I had planned for later) instead. If you've passed by this blog, you've probably already seen my thoughts on Linux Mint, so I'll skip the introduction. I tried this updated ISO file as a live USB made with MultiSystem. Follow the jump to see what it's like. There isn't too much that has changed since last year, so I will simply link the review from then, point out any changes, and put out any other thoughts that occur to me about this.

Review: Fuduntu 2013.2

I haven't checked out Fuduntu in over a year. I wasn't particularly planning to do so either, because I wasn't exactly expecting huge changes. But then I saw some news that changed my mind.

Welcome Screen + Main Menu

Fuduntu, as regular readers know, is an independent distribution that maintains GNOME 2 essentially as-is and uses the RPM package format, so it can sometimes use third-party packages developed for Fedora. Recently, though, there was a discussion among Fuduntu developers that culminated in the developers announcing a feature freeze for Fuduntu along with support ending by this September, along with the lead developer Andrew Wyatt (also known as FEWT) announcing his official resignation (though he may still unofficially consult with the project from time to time) from the project after support ends. The main reasons for this were that maintaining GNOME 2, keeping Fuduntu independent from Fedora while maintaining support for RPM, and having Andrew Wyatt work way too hard on this were all unsustainable; the remaining developers may decide to turn Fuduntu into something else entirely, in which case it would be once again based on another distribution (a likely candidate is openSUSE, which is interesting because I am not aware of any major distributions based on openSUSE at all), and it would probably need a new DE (perhaps the Consort DE from SolusOS, though that is purely speculation on my part). In any case, I am reviewing Fuduntu because this will certainly be the last such review I can do of Fuduntu in its current incarnation, and may be my last review of it ever.

For this final review, I tested Fuduntu using a live USB made with MultiSystem. Follow the jump to see what it's like and how it will be remembered.

Review: Linux Mint MATE 201303

For those of you who have been waiting for a review, I think I may have said before that my writing would shift more to science-y stuff and away from distribution reviews. However, that does not mean that reviews will stop entirely. I'm on spring break now and have a little more time to do these reviews, so today I am reviewing Linux Mint MATE 201303, which came out earlier this week.

Main Screen + Linux Mint Menu

This is the version of Linux Mint based on Debian rather than Ubuntu. It uses a variant of a rolling-release model, in that while existing users can get the latest and greatest software simply by applying updates as usual, the updates come in large bundles (I almost want to say they are like the Microsoft Windows Service Packs, except that they work) rather than individual package files. This means that the most common packages used on a Debian-based Linux Mint system are tested so that they can be guaranteed to work not only individually but also together, so that the problem of an individual update breaking other dependencies becomes moot. Around the time of releasing a new update pack, a new ISO file snapshot of the distribution is released, as was the case this time around.

I reviewed the [32-bit] MATE edition using a live USB made with MultiSystem; I wanted to review the Cinnamon edition too, but it refused to boot, so I will leave my assessment of it at that. I also did an installation of this (which regular readers know is rare), so you will have to follow the jump to see what this is like.

Review: Chakra 2013.02 "Benz"

Main Screen + KDE Kickoff

It's been a while since I looked at Chakra, so I was thinking now might be a good time to do that. Plus, KDE 4.10 just came out with a whole bunch of new features and fixes, so I wanted to check that out too. So this is the subject of today's review.

I've tried Chakra a number of times before. It was originally derived from Arch, but since a couple years ago it has been developed in a fully independent manner. It uses a "semi-rolling" release model, in which applications like Mozilla Firefox and other front-end features like KDE are updated on a rolling basis, while core system components are held to be more stable.

I tried this (64-bit version, as there is no 32-bit edition anymore) on a live USB system made through the "dd" command; due to Chakra now using the GFXBoot tool for booting, neither MultiSystem nor UnetBootin worked, and I didn't have any other data on this USB drive, so I was OK with using that tool. This review almost didn't happen because right after the boot menu, the message "invalid or corrupt kernel image" would come up. I decided to give this one last shot by downloading the ISO file again and trying again, and it worked! Follow the jump to see what it's like.

Revisited: Fuduntu 14.10.1

Main Screen + GNOME Main Menu

It has been almost exactly 6 months since I looked at Fuduntu. Since then, it has had one new minor release (version 14.9 became version 14.10) and an update to that release (version 14.10 became version 14.10.1). You would think that not much would have changed between then and now, so what's making me test this?
Before that, I apologize for the recent dearth of posts here (aside from the very recent one). First, I've been busy, second, I've been sick, and third, I haven't really thought of anything to write about. That's part of the reason why I'm writing this. Also, I can promise you that there will be a new review soon after this, and it will actually be two reviews in one (but not a comparison per se). With that, let's get on with answering that question.

So aside from the fact that I wanted to write about something here, I'm testing this because the Fuduntu lead developer made a pretty big two-in-one announcement yesterday: (1) Fuduntu is becoming a rolling-release distribution, meaning that users don't have to reinstall the latest version of Fuduntu every six months just to stay up-to-date, and (2) Fuduntu will stick with GNOME 2.X for as long as possible. These two really piqued my curiosity, so I gave it a spin.

I tested the live session using a live USB made with UnetBootin. I did not test the installation procedure because I did that before, and I haven't read about any major changes to Anaconda since then. Finally, I don't usually mention details like this, but do note that the ISO file has decreased in size from about 1.1 GB to a bit less than 950 MB; the consequences of this will become evident later. Follow the jump to see what's the same, what's different, and whether I still like Fuduntu as much as I did 6 months ago (and I will make frequent reference to that review).

The Future of Various Linux Mint Editions

Today, I was reading DistroWatch when I saw an interesting tidbit about how Linux Mint's KDE Edition is moving to a Debian base, though the developers originally intended this KDE release to be based on Ubuntu as before. I think the reason why I felt as surprised as I did was because as opposed to the other editions (Xfce, Fluxbox) which were announced as moving to a Debian base without any previous statements regarding them, the developers did say the KDE edition of Linux Mint would be based on Ubuntu, and judging from the forum posts, this decision seems to have been rather abrupt, as opposed to being more carefully planned.

I think this will be a good thing for the developers and for some users. First, the developers can focus even more on the Ubuntu-based main GNOME edition, while the other editions can get the latest software with less maintenance on the part of the developers. Second, the users will be able to get the latest KDE releases thoroughly tested by the developers through the official repositories without having to resort to PPAs, as is almost always the case in Ubuntu-based releases (and even then, sometimes the latest PPAs don't support even slightly older releases of Ubuntu — for example, I couldn't test KDE 4.6 with a PPA on Linux Mint 9 LTS "Isadora", which is based on Ubuntu 10.04 LTS "Lucid Lynx").

But for a few reasons, I also feel less enthusiastic about this move than I did with the development of the original Debian-based GNOME edition and the move of the Xfce and Fluxbox editions (the latter of which still hasn't been officially released yet, thanks to the developers' utter devotion to providing quality releases for users regardless of schedules, unlike another certain Linux organization I know whose name starts with a 'C' and ends with an 'anonical'). Why?
First, I and other reviewers have found the Debian-based Linux Mint editions to be a little less stable and slightly rougher around the edges than the main Linux Mint edition. This is supposed to be fixed in the next few weeks or so with the introduction of update packs, which consists of the developers thoroughly testing packages from Debian Testing (to whose repositories Debian-based Linux Mint points by default) and then releasing collections of those packages to users in one month intervals if and only if those packages are stable and don't cause conflicts or breakages. Hopefully the Debian-based KDE edition will be released around the same time as that and not before, so that stability is less of an issue.
Second, I've found that there isn't quite as much software available for Debian as there is for Ubuntu; for example, if I wanted to install the Nautilus Elementary mod, I could just use a PPA in Ubuntu, but I would need to compile the source code and manually install that in Debian. Related to that, I've found that hardware isn't quite as well supported in Debian as in Ubuntu. For example, Skype has been problematic in Debian-based Linux Mint but not at all in Ubuntu-based Linux Mint. That's a deal-breaker for me, and I'm sure it's a deal-breaker for a large number of Skype users on Linux Mint.
Third, as I briefly mentioned above, the Xfce and Fluxbox editions of Linux Mint moved to the Debian base so that the developers wouldn't have to expend as much effort to maintain them and they could focus more on the main Ubuntu-based editions. The move of the KDE edition happened for a totally different reason: the maintainer of that edition was having some small issues trying to get a fully working system (I believe the issues related to the network management program), so the fix was to move to a Debian base. I'm not really sure I get that. Was it that bad? Couldn't the KDE edition use a few more maintainers to get these sorts of things ironed out? Or is there only one maintainer due to the low take rate relative to the other editions? In that case, does it make sense to continue maintaining a separate KDE edition at all/would it be better to create a sort of customized Linux Mint KDE package available to both the Ubuntu- and Debian-based Linux Mint editions?
Or better yet (and please keep in mind that this question is coming from a nontechnical newbie), could the problem have been avoided altogether by starting from Kubuntu instead of Ubuntu? After all, as far as I can tell, the Linux Mint KDE edition is basically little more than Kubuntu with different applications and a different Plasma theme and wallpaper. This is in contrast with the main GNOME edition, which before had a different desktop layout and Linux Mint Menu in addition to the application modifications, and now diverges significantly from the Unity interface included by default in Ubuntu 11.04 "Natty Narwhal". Also, related to this, while Ubuntu continues to stray farther and farther from what users actually expect, Kubuntu is getting more and more refined, and in version 11.10 "Oneiric Ocelot", it's supposed to get a new package manager called Muon to replace the aging KPackageKit (along with, if I remember right, a new update manager too); Linux Mint KDE users would benefit not just by having a wider software selection, better hardware support, and support for Ubuntu PPAs, but would also benefit from these innovations and more guaranteed stability if Kubuntu was used as the base.
Finally, there's a part of me that instinctively associate Linux Mint KDE releases with an Ubuntu base. Part of that stems from the earlier announcements that there would be a Linux Mint 11 "Katya" KDE, and part of that stems from the fact that Linux Mint actually started as a KDE distribution based on Ubuntu (it wasn't really until version 3.0 "Cassandra" that GNOME became favored over KDE, as far as I've read), so this move would constitute Linux Mint abandoning its roots. If there is to be a Debian-based KDE edition, I'd at least like to see a simultaneous effort to release Linux Mint 11 "Katya" KDE, hopefully with more developers on board. Then again, it has been a while since Linux Mint 11 "Katya" GNOME came out, and even longer since Ubuntu 11.04 "Natty Narwhal" came out, and Ubuntu 11.10 "Oneiric Ocelot" isn't that far away, so it's pretty late in the release cycle to release an Ubuntu-based Linux Mint KDE edition, and that's something the developers wanted to do away with entirely by moving the Xfce and Fluxbox editions to the Debian base. Maybe the problems will be ironed out in time for a Linux Mint 12 "L[...]a" KDE release soon after the release of the corresponding GNOME edition.

In short, I'm more ambivalent about a Debian-based Linux Mint KDE edition than about Debian-based Linux Mint GNOME, Xfce, and Fluxbox editions. Plus, I wonder why the LXDE edition is sticking with the Ubuntu base. I think that's because it was originally planned that the LXDE edition, like the KDE edition as originally announced, would stick with the Ubuntu base, and there were fewer problems so a live CD image could be pushed out faster. What do you think?

Review: Linux Mint Xfce 201104

Main Screen

Just over 2 weeks ago, I wrote about how Linux Mint is moving the Xfce edition to a Debian base. Well, a few days ago, they released the official Linux Mint Xfce 201104.
For those of you who didn't read that post, in short, Linux Mint Xfce is now Debian-based instead of Ubuntu-based. The developers had a few things to say about this: (1) the desktop will be faster and lighter on resources (114 MB of RAM at idle, 177 MB of RAM with Mozilla Firefox, LibreOffice Writer, and LibreOffice Calc open all at the same time), (2) the Xfce edition will now include more mainstream applications like Rhythmbox instead of Exaile, and (3) the Xfce edition, being based on Debian Testing, will be a rolling-release branch. All these things sounded very exciting to me, so I decided to try it out.

Please do note that I have been using Linux Mint for almost 2 years now. I've become quite fond of it, so while I will try to be as critical as possible, don't be surprised if some elements of bias creep into this review.

I tried out the live session on a live USB made with UnetBootin, which is surprising to me because past versions of Linux Mint "Debian" as well as #! (since its switch to a Debian base) haven't worked with UnetBootin. I tried out the installation, just to see if anything has changed since the last version of Linux Mint "Debian" that I've tried, in VirtualBox inside the live USB session with 384 MB of RAM allocated to the guest OS. Follow the jump to see what it's like.

Linux Mint Xfce: Now Based on Debian

In all honesty, I didn't see this coming. Sure, if you browse through this blog, you'll see plenty of posts and prognostications on my part about the future of Linux Mint and Debian, but I didn't think this would happen the way it has happened. I regularly open the Linux Mint homepage to check for updates, so when I saw "Linux Mint Xfce RC" released, I got excited because finally Linux Mint 10 "Julia" Xfce will have been released after the release of the GNOME, KDE, and LXDE editions. Then, I realized that in place of "10 'Julia'" was a date number. That's when I knew. In short, the Xfce edition of Linux Mint has moved to a Debian base.

The reason for my surprise is that the move happened only partly due to reasons that I wrote about; the upstream issues are only briefly mentioned as causes for the move. The biggest reason was that the other editions are released sometimes mere weeks before the release of an RC for the next version of the main edition. I mean, just look at it: it has been 4 months since the release of Linux Mint 10 "Julia" GNOME, yet the KDE and LXDE editions were released just a few weeks ago. That's bad. That's why the developers have moved the Xfce (and, apparently, Fluxbox) edition to the Debian base; there's no more worry about late releases, and the rolling-release schedule makes for easier maintenance. Another thing mentioned is the huge performance boost over the previous Ubuntu-based Xfce edition; the Debian-based Xfce edition uses almost 50 MB less RAM for comparable task combinations.

Of course, the KDE and LXDE editions have already been released with the Ubuntu base, but the developers have said they are looking into possibly shifting them over to the Debian base and make the Ubuntu-based Linux Mint come solely with GNOME. This would make sense because those editions are also released relatively late in the release cycle. As was mentioned earlier, the Fluxbox edition will move to the Debian base when it gets released.
The developers have said they are committed to releasing Linux Mint 11 "Katya" this May and Linux Mint 12 "L[...]a" this November. Their uncertainty about releases beyond that also seems to imply that Linux Mint 12 will be the last Ubuntu-based Linux Mint, and future versions of Linux Mint will solely be based on Debian; that seems to coincide with Canonical's plan to put Wayland in Ubuntu 12.04 LTS, so I guess that's a good place to stop.
The developers also said they had two big plans for Linux Mint 11, but these may have to be put off until Linux Mint 12 or later. Ooooh, the developers have really piqued my curiosity!
Finally, the developers have said that this and future Debian-based editions of Linux Mint are to be referred to as simply "Linux Mint XYDE YEAR-MONTH". The "Debian" moniker will be dropped.

One thing that has annoyed me about the Debian-based Linux Mint edition (now editions) is that there aren't any release notes saying what's new. I guess it makes sense from one angle in that it's a rolling-release distribution, so users will always have the latest packages, but I would still like to know exactly what is included in a particular snapshot. The developers left out release notes for past snapshots of Linux Mint "Debian", and they've done the same for this new Linux Mint Xfce RC. For example, I didn't know what version of Xfce was included until I saw the comments section; I was excitedly hoping to see that the brand-new Xfce 4.8 has been included, only to read that it is Xfce 4.6, though this can be changed with a quick upgrade. Dear Linux Mint developers, could you please start including release notes when new snapshots of Debian-based Linux Mint editions are released?

In conclusion, the Xfce edition of Linux Mint is moving to a Debian base for a better release schedule, more mainstream packages, and better performance.
I'm currently using Linux Mint 9 LTS "Isadora" GNOME, so I was thinking of using that until Linux Mint 13 LTS gets released; now it looks like there may not be a Linux Mint 13 LTS, so I'm looking forward to using a Debian-based Linux Mint full time on my laptop's hard drive!

Mozilla Firefox 4, HTML 5, and Rolling Releases

Yesterday, there was news that Mozilla Firefox would be releasing version 4 probably by the end of this month and early next month, and along with this, versions 5-7 would also be released this year (2011); do note that version 4 was scheduled for release in 2010 November. While some people were outraged by this announcement given how long it has taken Mozilla to release Firefox 4, it has since become clear that this is supposed to be the biggest new release of Mozilla Firefox we'll ever see, and all future whole-number releases will follow the model of Google Chrome/Chromium of releasing new versions often (and now it's nearly every month) that are more incremental improvements rather than revolutions.
I'm glad that this is happening, for it means that applications and not just whole OSs are starting to follow a rolling-release-esque model of releasing snapshots periodically but sending packages of updates thereafter. Then again, I wonder if fixed-release distributions will provide these newer versions of Mozilla Firefox when they come out; I'd certainly like to be using Mozilla Firefox 7 on Linux Mint 9 "Isadora". I'd at least like to see them added to a PPA available for users of Ubuntu LTS releases.
Plus, this comes a few weeks after an announcement by the HTML developers that version 5 will be the last explicitly-numbered version; from here on out, HTML will just be called "HTML" and will follow a similar rolling-release-esque system.
But I don't think the reason for Mozilla doing this is only to make up for the long delay of the release of Firefox 4; I think they're also concerned that Mozilla Firefox, having a lower version number, isn't perceived to be as "advanced" as its rivals, like Google Chrome 9. I've read that in the open-source community, quickly advancing version numbers when such quick advances didn't happen before in the project is frowned upon (i.e. 1...2...3......4-5-6-7), but I've seen that even developers of free software that cater to more experienced users fall prey to this as well. The best example of this is Slackware: lead developer Patrick Volkerding, concerned that users were leaving Slackware for other distributions because Slackware's lower version number (4) was lower than competing distributions and hence gave the impression of Slackware being less "advanced" or "mature", decided to skip a couple versions and released version 7 right after version 4.
Well, at the very least, I hope Mozilla Firefox 4 gets released soon. Even better, I hope I can use it!

Preview: GNOME 3

Main Screen + Calendar + Notification Area

Usually, when I review desktop environments, I review KDE, specifically version 4. Why? It's constantly evolving and improving, and it's nice to be able to see such changes occurring on all fronts so quickly. By contrast, GNOME and Xfce (not to mention other WMs like Openbox) have remained relatively the same over the past few releases. Sure, Nautilus got tabbed browsing in version 2.22 (I think) and split-pane viewing in version 2.30. Sure, there may have been a couple other back-end changes. But generally speaking, where KDE 4 has changed pretty noticeably between point releases, GNOME has been quite stable. That's all going to change, because GNOME is about to be released under a whole new number: 3. That's right: the number preceding the decimal point in a GNOME release will no longer be '2'.
There are some pretty big changes in store for GNOME 3, much of which can be seen in the front-end. Because many major distributions are planning to upgrade to GNOME 3 once that gets released (in a few weeks, apparently), it's important that users try GNOME 3 beforehand both to get accustomed to it as well as to find and report lingering bugs. Happily, the good people at Fedora and openSUSE have put together live CD ISO files with vanilla GNOME 3 on them, just for the purpose of trying out GNOME 3. I downloaded both files and intended to make a multiboot live setup using MultiSystem, but unfortunately MultiSystem reacted with error messages to both ISOs. Knowing that openSUSE doesn't play well with UnetBootin, I decided to just try out the Fedora version on a live USB through UnetBootin. Follow the jump to see how it goes.

Fresh OS, Rolling Releases, and Debian as of 2011 January 5

It's been quite a while since I've done one of these massive combined posts, but I feel like it's necessary today.
I mentioned in a post a couple days ago that, along with releasing an update to UberBang, I would continue working on a new build of Fresh OS. Well, those plans (regarding Fresh OS) have hit a snag (actually, a couple of snags). For everyone's information, I've been starting with Linux Mint "Debian" and using Hadret's Debian PPA for all the requisite Elementary-fication. The thing is, there are two separate Hadret repositories ("unstable" and "experimental"), and some packages from "unstable" have been moved to "experimental" (for reasons that I still don't get); these include pretty important ones like Nautilus Elementary (a modification of Nautilus that makes it a lot less cluttered and a lot easier to work with). What this means is that these packages, as I found out the hard way, are more likely to break certain things in the system (though I would think that Nautilus Elementary would be essentially unchanged). Anyway, installing Nautilus Elementary caused the desktop wallpaper to be replaced by a plain blue background, though the wallpaper switcher claims that the wallpaper I selected is still in effect. This is combined with Gloobus Preview and CoverGloobus breaking some other packages and preventing updates from running smoothly.
I looked for solutions to this in the Linux Mint forums and found that it probably is because of Hadret's repository not playing nicely with Linux Mint "Debian" and its rolling-release model. I also saw another interesting tidbit about Jupiter OS, a similar project to Fresh OS, being abandoned due to the developer having difficulties with porting over the applications and other things as well as feeling competition with Pinguy OS. While I don't feel any competition with Pinguy OS (hey, we're all in this together), I can certainly see where the difficulties come in (as it seems like the Jupiter OS developer also used Hadret's repositories).
That brings me to my general beefs with Linux Mint "Debian". I've said most of them already, but just to sum it up, it seems like Linux Mint "Debian" is having a lot more trouble with stability and working packages than standard Linux Mint. Granted, the developers themselves warned of this, but for supposedly keeping very close to the standard edition, it certainly is unstable. Maybe I just need to give it a few more months for it to work out problems with stability associated with a rolling-release model.
Finally, though this is somewhat unrelated, Debian 6 "Squeeze" should be coming out within this month! I say this only because some developers have said that the current number of known bugs is about the same as in version 5 "Lenny" when it was released. Also, there's been a bit of complaint over the new artwork. I too am not a fan of it, but it comes from the codename, which comes from the Squeeze alien toys from the Toy Story series.
Given the issues I've been having with Fresh OS based on Linux Mint "Debian", I'm going to try to do it now with plain Debian 6 "Squeeze". I'll try that and write more later about how that goes.

Featured Comments: Week of 2010 November 21

I want to apologize for not having posted this yesterday as usual. I was traveling back to college (more on that in an upcoming post) and didn't have computer access all day.
There were quite a few comments on posts this week, so I won't repost all of them.

Adafruit Bears Fruit for Microsoft

In response to my question about why Microsoft seemed so defensive, an anonymous reader had this to say: "Because Microsoft stir hackers' defiance whenever they say they have protected their products. For Microsoft, it was just saying "I challenge you to hack my ultra-securized device", and some hackers successfully took in the challenge. the Microsoft PR guys are just brilliant, they just took advantage of the company's reputation and the situation actually did beget creativity, the sort of creativity that will eventually benefit the Redmond-based behemoth."
Another anonymous commenter counters this: "Why do you think the developer at Microsoft who claimed that it would be easy to hack really telling the truth? I think it is just a post-construction when they realized that it was impossible to stop. And then everything is back to normal again, MS is and will remain evil :-)"
Another anonymous reader thinks it's because of sheer ignorance on Microsoft's part: "The data format was not "ultrasecurized" at all. They didn't know what was going on they just heard "kinect hacking" and gave a generic response which applied to a physical type hacking, with soldering and all. This was not a physical hack but a reverse engineering of the data format."

Ubuntu to Become a Rolling Release Distribution

In response to the update about the news being not-quite-true, reader T_Beermonster wrote, "I think that's a shame that they back-pedalled. Since I went rolling release with aptosid I don't think I'd be willing to go back to a step-change release model. I can see why it may be easier to sell support contracts for a step-release model but I don't think it actually offers any benefit to a desktop user."

Linux Mint: Good for Low-Requirement and Paranoid Users

Reader Arjun Krishna had this to say about it: "Windows is one of the worst OSes I have ever had. Linux Mint 10 "Julia" is definitely one of the most stable and user-friendly Operating systems in the world! Open SUSE is also a good alternative to Linux Mint, in case the system is older, and has less RAM. In any case, any Unix based OS would be much better to work with than a Windows based OS."
Also, commenter herbalfroot wrote, " Everyone for whom I have installed *buntu and mint have nothing but praise for the desktop they now have. These include non-technical users. I roll my eyes to the sky whenever I hear 'Linux is too difficult for the average user'."

Thanks to everyone who commented this week. Unfortunately, for the next two and a half weeks, I'll be quite busy, so don't expect to see a whole lot of new posts. In any case, as always, if you like what I write, please subscribe!

Ubuntu to Become a Rolling Release Distribution

This just in, folks: Ubuntu is about to become (Susan Linton, OStatic) a rolling release distribution! SWEET!
Of course, this means no more weird numbering system, and no more silly "[adjective]-[animal]" names...or does it? Mark Shuttleworth does say that like any other rolling release distribution, Ubuntu will release ISOs periodically for people who are installing for the first time as well as for people who need to reinstall Ubuntu for whatever reason.
I'm not too happy about the move to Unity, and I'm cautiously optimistic about the move to Wayland, but I can say for sure that I'm ecstatic about this news. I really do agree that in an Internet-oriented world (reflecting Ubuntu's new/revised goals as well), rolling-release is the way to go.
Of course, this leads me to the question: what about Linux Mint? One of the reasons Linux Mint made a straight-up Debian-based edition was to take advantage of the rolling release model in the "Testing" distribution. Now that Ubuntu does that too, does this mean that Linux Mint will follow suit whenever that happens and drop the "Debian" edition altogether? I'm excited to see what's in store for the future!
(UPDATE: As it turns out, Ubuntu isn't actually going on the rolling-release route. All it's doing is essentially integrating the PPA functionality into the main system to allow people to get the latest versions of third-party software like Mozilla Firefox. I remember some Ubuntu developers mentioning this before (specifically regarding Mozilla Firefox), so this doesn't come as a huge surprise. That said, I'm a little disappointed that it's not what I thought it was.)

The Ideal Linux Distribution, As I See It

Before I begin this post, if you're wondering how I manage to churn out these posts and still claim to be busy, I will say that I write almost all of these posts on the weekends and then schedule them to automatically publish during the workweek. I don't usually have time to write posts on weekdays.

As I review more distributions, I'm continually fleshing out exactly what I want to see in a distribution. There are a few things that I would like to see in Linux distributions, none of which should be especially hard to do, as some of these features have already been implemented. Here's what I'd like to see:

• Mozilla Firefox (Rekonq and Arora are acceptable substitutes in KDE; Opera and Google Chrome are also acceptable substitutes anywhere)
• OpenOffice.org (AbiWord and Gnumeric are acceptable substitutes if the distribution is more lightweight)
• A graphical package manager
• Most proprietary codecs either included out-of-the-box or installable by clicking on a highly visible link
• Support for various peripherals out-of-the-box (especially mice, webcams (and external mics), and printers)
• Stability and security
• Rolling releases (just so that installation only needs to be done once)
• Minimal visible bloat

The ideal candidates for this are PCLinuxOS and Linux Mint "Debian"; both are rolling-release distributions but test their packages extensively to ensure the stability and high quality of the packages. Both include Mozilla Firefox, but only Linux Mint "Debian" offers OpenOffice.org. Both include graphical package managers and most proprietary codecs (out-of-the-box). Finally, both have excellent support for mice, webcams, and printers. As PCLinuxOS doesn't have OpenOffice.org, I'm going to continue with just Linux Mint "Debian".
I would say that the best-looking GNOME theme present today is the Elementary theme, and that Linux Mint "Debian" with Elementary (and things like the Nautilus Elementary mod) would look positively stunning and will age well with the distribution (which doesn't make sense considering the distribution will always have the latest packages). (On a side note, there are some rumblings in various blogs about how Linux Mint may use the Equinox Faenza icon set in version 10 "Julia".) I've already done this with my FreshOS respin of Linux Mint "Debian". So what else would I like to see? Not a whole lot, except for one other thing:
I want to see an installer (and Linux Mint "Debian"'s installer, while adequate, isn't quite up to the level of polish of the Ubuntu-based releases' Ubiquity installer) that gives options for different application categories. What I mean is that if someone is big on multimedia, applications like F-Spot will be replaced by digiKam and OpenShot during the installation. Similarly, if someone needs programming tools, programs like Emacs, Eclipse/Netbeans, and other similar programming utilities could be installed. This way, while the user can always go to the package manager to install and remove packages of his/her choice, there are options in terms of what default applications are present after installation, nicely grouped into different categories.
How does that sound?

200th Post: Preview: ArchBang 2010.09 "apeiro"

Happy 200th post Das U-Blog! I hope you readers have enjoyed reading these posts as much as I have enjoyed writing them! And, as always, if you have not already done so, please take a moment to click any one of the various subscription buttons to get updated on new Das U-Blog posts as they come out (either by email or in your favorite RSS/Atom reader)

Main Screen

1. This is not a typographical error.
2. This is not "CrunchBang ('#!')".
3. THIS. IS. ARCHBANG! [insert dramatic music here]

Of course, the real story is slightly more nuanced (did I use that word right?) than that. ArchBang, while not a badly-spelled version of #!, is actually inspired by (but not derived) from #!; it aims to be to Arch Linux what #! is to Debian (and was to Ubuntu before version 10 "Statler"). That's right: ArchBang is an Openbox derivative of Arch Linux. I was actually looking for other reviews of Linux Mint "Debian" when I found this article whose comments section linked to an article on #! and ArchBang. Naturally, I was intrigued upon seeing Arch's response to #!, so I went ahead and downloaded the ISO image to try it. Please do note that to better reflect this distribution's advertised capabilities as a solution for old computers, I have decreased VirtualBox's RAM allocation for this distribution to 192 MB (keeping video memory at 12 MB). Technically, the minimum requirements are 128 MB of RAM, but I think 192 MB is a reasonable stand-in for an old computer these days. Follow the jump to see how it fares in these harsher conditions.