not all change is progress
June 22, 2014
Direct download links: MP3 & Ogg
Puppy Linux – PCLinuxOS – self-hosted RSS aggregatorsNews
RHEL 7 release buzzword bingo
A big step forward in business Linux: Red Hat Enterprise Linux 7 arrives
Docker 1.0 released
Docker libcontainer unities Linux container powers
Docker container breakout proof-of-concept exploit
Lennart watch continues
Traffic Snarls in Europe as Taxi Drivers Protest Against
Uber registrations ‘increase 850%’ as black cab drivers stage London protest
Taxi medallions have been the best investment in America for years. Now Uber may be changing that.
How does Uber make money?
Indie Labels Face YouTube Block Over Unsigned New Terms For Paid Service
HP starts a memristor-based space program to launch … THE
Dell exec: HP’s ‘Machine OS’ is a ‘laughable’ idea
Alienware Steam Machine now a Windows PC for the living room
Intel will offer a customizable chip to keep data center clients happy
Microsoft Supercharges Bing Search With Programmable Chips
Amazon launch their mobile cash register
All your Android are belong to us; not fixed in the 4.4.4 update
Ubuntu Looking To Bring Click Packages To The Desktop;
Click Packages 101
MATE 1.8 has now fully arrived in Debian
500MB ISO, 76 Window Managers to try out
Thanks for nothing, jerkface
How to mend … a slow computer
Joe took a look at Puppy Linux, and Paddy will be firing up Frugalware next show.
A huge thank you to Jeroen van Rijn and SonOfNed for their PayPal donations, and to an anonymous Flattrer.
We had a large amount of feedback to talk about, most of which can be found BTL of our last show notes. Thanks to everyone for contributing, and our apologies for having to be selective about what we included in the show.
Martin Wimpress told us a bit more about his plans for Ubuntu MATE Remix, and we’re looking forward to having him on the show to talk about that project when it gets little nearer to release. Martin is working with Popey at Canonical, who – as Tsukasa Buddha pointed out – seems to have a slightly schizophrenic view on the benefits of traditional desktops like MATE.
Jeroen van Rijn and Rob Landley had an exchange in the comments about user interface design and change for change’s sake, with specific reference to Unity and systemd; whilst Joel told us why he’s a huge fan of Unity’s HUD.
Will thought that we’ve underplayed the usefulness of clouds, at least in a business environment; and Paddy and Jeroen van Rijn had an email exchange about the costs and benefits of fixed release cycles for corporate Linux users.
Jason, Joe and Jeroen van Rijn had a back and forth about what constitutes ‘a gamer’.
Gregor got in touch asking for advice on what distro might suit a Windows XP refugee. The key thing he was looking for was the ability to run a familiar looking desktop on very weakly powered hardware. We’ll be taking a spin around the sub-LXDE desktop environment field on a future show to see what the state of play really is.
Joel wanted to understand why Paddy thought that Btrfs might not quite be ready for use yet, and Paddy mentioned a presentation Dave Chinner gave back at linux.conf.au 2014.
Tsukasa Buddha asked if we had any ideas about what was coming from the Fedora.next project, and Mark Walton had further wireless keyboard feedback for Joe.
Danny got in touch to explain how the Snowden revelations had changed his approach to computing, and Jason and Warren also pitched in on the same topic.
Andy Jesse gave us an update on his data recovery story and, along with Henrik, endorsed the Testdisk suite of data recovery tools. Both the Arch Wiki and the online Ubuntu documentation provide a reasonable starting place for folks looking to understand what data recovery tools are out there.
Rob Landley, Ian Barton, Campbell Barton and SonOfNed all offered thoughts on the issue of software quality.
Jens asked what language might be a good starting point for a beginner, and Nathan D. Smith suggested Python. Paddy and Ian Barton pointed Jens towards a couple of different free online Python courses; Campbell Barton chipped in to endorse one of them, and Cathryne’s first impressions of one seemed really positive.
Eric and Ian Barton were prompted to get in touch following our piece on the UHH; both were of the view that yes, thankfully, things have improved an awful lot since the book was written.
Jezra told us why he doesn’t think that Tizen will gain traction, whilst Henrik, arold, Brian Hall and Jenny offered thoughts around syncing or file transfer from Android devices to desktop machines.
We got a mixed, but mainly positive, response to our suggestion last show that we were looking for additional hosts. Thanks to everyone for giving us your thoughts on the matter – especially to Dhalgren, who offered us some sensible advice. If anyone is interested in putting themselves forward who hasn’t yet got in touch, please send us a few minutes of audio to show@ explaining a bit about your background, and what you think you could bring to the show – thanks!
An independent offering with a strong and vibrant community is a welcome sight in these days of increasing corporate-funded distro dominance. But is it any good? We took a look at the LXDE and MATE versions of PCLinuxOS.
Off the Beaten Path
Thinking about self-hosted RSS aggregators to reduce your exposure to unreliable cloud services? Tiny Tiny RSS, selfoss and KrISS Feed may be good places to start. And if anyone does knock up a good mobile theme for KrISS Feed, at least one of your hosts would like to know about it…
Thought I’d follow up on a few points you raised in Episode 18.
One of you said something like “MATE is fatter XFCE”, which is just about true. I recently conducted memory use comparison for some of the popular desktop environments using Arch Linux. You can find the results below and draw your own conclusions.
MATE 1.8.1 is indeed available in the Wheezy Backports and this weekend I wrote a post explaining how to install a bloat free MATE Desktop on Debian.
Alan Pope (popey) and I (Wimpy) are both regular contributors to Linux Unplugged. Popey did indeed make the remarks you quoted about the reasons why a two panel desktop just wasn’t working for Ubuntu and why they created Unity and if you listen a few seconds longer you’ll hear my come back to what he said about Sun Micrososytems and CDE. The time-line is important here, Popey’s remarks were made prior to my UUPC interview and prior to the Ubuntu MATE Remix project being started. They were also a little tongue in cheek.
Popey wants to see Ubuntu more widely adopted and the MATE team also want to see greater MATE adoption, therefore a Ubuntu MATE Remix serves both our interests. Popey’s involvement with Ubuntu MATE Remix is as a community contributor, not in an official Canonical capacity, and he has already been very helpful in clearing the road ahead for the project. Mark Shuttleworth also briefly acknowledged the Ubuntu MATE Remix project during his recent UOS keynote.
Following my interview on UUPC, Popey took the time to test MATE and his comments in a recent UUPC were “MATE is really rather good” and he was also surprised at the pace of MATE development and the work we’ve done in collaborating with the Sonar and Talking Arch projects. So while Popey may not be the target audience for Ubuntu MATE Remix, I admire that he has seen the benefits that MATE can bring to Ubuntu and invested his own time into to helping cultivate the Ubuntu MATE Remix project. Without Popey’s initial spark of interest the Ubuntu MATE Remix may never have got off the ground.
Hi Martin – thanks for the updates, and particularly your clarification of Popey’s role in the project as a community contributor, rather than Canonical employee. I trust that you can appreciate how it’s sometimes difficult for outsiders to see which hat folks are wearing at a given time when they have several?
As Joe and I have said repeatedly, Ubuntu MATE Remix is a project that we are really enthusiastic about and, if it does gain official Canonical sanction, believe will become incredibly popular. However, we agreed with the listener who pointed us towards that episode of Unplugged that Popey’s comments there seemed to be distinctly at odds with what has happened since. I’m grateful that you have clarified the timeline.
My personal view – and bearing in mind your tentative relationship with Canonical, I wouldn’t expect you to respond – is that Mark Shuttleworth made a fundamental error about the true nature of convergence when he decided that Unity was the way forwards, but is too bloody minded to admit it. Nobody cares about OS convergence; rather it is application and service convergence and ubiquitous availability that people crave, and this is where others in the industry are starting to take us (e.g. Facebook’s push towards becoming a platform; or, on the app front, Apple’s Handoff). Which goes some way to explaining the push-back against Unity; for many, the UI change appears to have offered little benefit in terms of workflow, but resulted in a lot of pain – people still like the familiar desktop paradigm, and the MATE project as a whole has been one of the beneficiaries of this.
I am genuinely pleased that you are getting cooperation from Canonical employees, even if only in a non-official capacity at this stage. As you suggest, a successful Ubuntu MATE Remix ought to equally benefit MATE and Ubuntu, which is a great outcome for both projects. We will continue to watch developments with keen interest.
Ubuntu MATE Remix website is published.
I do enjoy Linux Luddites when I get a chance to listen. It was certainly interesting to hear your take on Ubuntu & MATE. :)
There’s not been any announcement of any Canonical-specific involvement in any Ubuntu MATE Remix for good reason. This project is “just” another community contribution to the wider Ubuntu ecosystem as you may see by hundreds or thousands of others on a daily basis. The fact that I happen to be paid by Canonical is largely irrelevant to this particular project. It’s convenient because I know who to talk to if we get problems, but that doesn’t make this any more “official” than any other derivative.
I’ve worked on Ubuntu for about 8 years as a community contributor. I’ve only worked for Canonical for 2.5 years, and I consider Canonical employees to be part of the wider Ubuntu community. Like many Ubuntu community people I work on Ubuntu and Ubuntu-related projects outside my normal 9-5 “day job”.
My initial rationale for helping Martin is because it seemed like a sensible thing to do. He wanted to have an Ubuntu-based derivative built around the desktop and tools he favours. This makes it easier for his users and potential future users to install a well maintained, secure and beautiful (in the eye of the beholder) desktop :)
From a selfish point of view it helps me to get a better understanding of the way that Ubuntu and the official flavours are built while hacking on a derivative. I also have old machines which are not capable of running Unity, and I’m not a great fan of XFCE, KDE or LXDE. Given I used GNOME 2 up until I switched to Unity (before I worked for Canonical ;) ) it seemed sensible to me to continue the blood-line of GNOME-2-like desktops with the same underpinnings of Ubuntu. Sure I could switch to Arch or some other distro which has MATE already packaged up, but I simply don’t want to use another distro.
At the end of the day we’re making this little Ubuntu MATE Remix derivative of Ubuntu, and if people want to use it, great, if they don’t, I’m sure they’ll tell us :) So if anyone wants to get involved or test it, I’m sure we’ll have an ISO image out soon enough for people to play with. For anyone not interested, there’s plenty of other things for them to play with on the Internet. :)
With regards to Patrick’s comment on Unity and convergence. I don’t think he fully understands what we’re doing there. Perhaps that’s poor messaging on our part, I don’t know. Unity is not the be-all end-all of convergence. It’s one part. The goal is to have not just the same shell on all devices, but common platform components (linux kernel, init system, display server), consistent design, and same application code too. No one part is “convergence”, it’s all parts together. We’re not there yet, but on a journey.
While there’s some people who dislike Unity, there’s also people who dislike strawberries. For those people there are other desktops and/or fruit.
For information about Fedora Next (aka 21)… there have been a number of blog posts by the new Fedora Project Leader prior to him becoming the FPL… so check those out. No link provided.
So far as a list of stuff that is in the works for Fedora
21, here’s a list:
It was actually Fedora.next http://fedoraproject.org/wiki/Fedora.next
The snowden revelations does not have anything to do with google. Corporate spying and governmental spying are different things. It is like blaming the post to be searched by the government. There are reasons to dislike google, but they are not the snowdon revelations.
Nothing to hide? Really? As a German hostorican I have to add the thing with the 2WW, the Jews and the data the nazis collected about them. The hadn’t anything to hide too. It is never a good idea for a society, letting the government collect to much data!! Who knows what will come?
I think Mr Godwin would like a word with you.
It’s the easiest example. You can also look at any other persecution of race, gender or religion.
But you are right. Nobody should do nazi comparisons. I’m sorry.
Here’s a more up-to-date rebuttal of the “nothing to hide” argument. In particular the interview with Cory Doctorow starting at ca. 24min:25sec. In summary: Letting yourself being spied upon means helping determine the baseline of “normal” against which people are judged as “abnormal”.
PS: “Abnormal” obviously being arbitrarily defined by the whims of whoever happens to have access to the knowledge garnered from surveillance/spying. Thus, people who “fit” that definition in the vast majority do so without fault or wrongdoing on their part. One example only being the US No-Fly list. And that situation can simply not be conducive for a society that (presumably) wants to have democracy, rule of (constitutional) law etc.
I had the same observation about ttRSS – “you call this tiny?” But as long as we are discussing non-tiny options…
After Google Reader shut down, I ended up with Feedbin (https://feedbin.com/). Shortly after I started using it, they opened the code under the MIT license (see repo here: https://github.com/feedbin/feedbin). It is a Rails application, so it is not the simple flat-file system desired. However, it has a an API with Android client(s), and the web UI is beautiful. Check it out, at the least.
Thanks, Nathan, I’ll take a look.
Have you thought about using a CLI RSS feed reader instead of a web based one? There is ssh client on every platform so, if you put it on your server, you could access it from anywhere. I generally use Feedly like you, but I also use Newsbeuter/Podbeuter to download some video podcast. I find it very neat. You should take a look… It is something like 500k to download.
Hi Arold – I do very occasionally use an SSH client on my phone, but find it rather impractical (4″ screen coupled with my advancing years ;)
One thing I didn’t properly think through when we recorded the show and Joe suggested using a mobile theme in a browser was how I actually consume my news. I read a lot of feeds every day, but tend to only refresh when on WiFi to cut down on my mobile data usage. That’s why I’ve always used a fat client app, rather than a browser interface. For now, I’m sticking with my Feedly back-end (I use an alternate client, rather than their glitzy one), but I’ve also got Sparse RSS running on my tablet and phone and set to local storage, so should the cloud ever disappear again I’ll have access to all my feeds without an issue. Of course, you don’t get the synchronisation between devices using something like Sparse the way I’ve got it set-up, which is really the only reason I’m still using a cloud-based service like Feedly, and why it would be nice to find a (really) lightweight self-hosted alternative.
Here’s an interview with new Fedora Project Leader Matthew Miller that focuses on Fedora.Next:
To keep up with Fedora these days, the http://fedoramagazine.org/ site is getting better all the time.
Thanks, Steven – I’d actually pegged that, along with the KDE ‘state of the nation’ piece on dot.kde.org, to mention in the Seen Elsewhere section next show. Shame there’s not an equivalent level of detail in the Miller interview, but I guess things will shake out a bit over the next few months.
I think Fedora.Next has gotten a bit too complicated, and while I don’t think that complication will be its undoing, it’s not going to win it any fans.
Once we see what the differences between the Cloud, Server and Workstation products actually are, everything should clear up a bit.
There has been a bit of controversy along the way about Fedora Spins (http://spins.fedoraproject.org/), which include all of the non-GNOME-running ISOs, including the Xfce system I use, plus KDE, Mate, LXDE, etc.
After a bit of pushing and shoving on the mailing lists, I think it’s pretty well established that Spins will continue as before. The Xfce spin is surprisingly good, though for reasons that escape me the team (or individuals) who produce it keep a very, very low profile.
I’m hoping that Fedora.Next will allow for different out-of-the-box configuration that benefits each use case.
At any rate, it’s a welcome bit of newness in the Linux distro world that, this time anyway, doesn’t have its motivation in making money off of privacy-exposing services. And for that we can all be grateful.
When I began using Linux way back in 2006-7, Puppy Linux became my go-to live distro, and often main distro, because I had a lot of old hardware that did very well with what Puppy had to offer.
It was a great way to learn about Linux and free software.
Nowadays I tend to use Debian on anything old, but I recently used Puppy to pull the data off of a Windows XP system that otherwise wouldn’t boot.
Personally, I use a Firefox add-on for my RSS feeds (Brief RSS reader) because I don’t have a smartphone and only look at my feeds from one computer. I like that it’s really sparse and can be accessed through browser (not a standalone application), so it was easy for me to set up good keyboard shortcuts for it (using the Pentadactyl add-on). I switched to it from Google Reader to avoid the issue of third party cloud service unreliability. Everything I like about it could be provided by a self-hosted web service, so I look forward to hearing more about what you test out and settle on.
Listened to a few episode when on the road so the comment is not be specific to this episode. Here goes:
It is a common theme in Linux podcast commenting that GPG is not commonly used.
Is using GPG really that difficult?
To a Window user, yes.
To a Linux user who goes through the trouble of locating and download an iso, learn to use md5/sha, learn to use dd/unetbootin/imagewriter, plan the partitioning scheme for multiboot, install and hang and reinstall, install and doesn’t work 100% to your liking and goes looking for answers to perfect the setup of your choice, the answer is a definite no.
So why is GPG so uncommon?
Simply because of a chicken and egg problem.
There is no real use case for the common user. If the rest of the world is not using, it does not make sense for me to have GPG setup.
How are you and the community help?
Create the use case or the need to have GPG simply by encrypting your podcast.
Instead of downloading and mp3 or ogg, encrypt it with GPG for every first podcast of the month or better every podcast.
Have a little howto on GPG.
This will generate awareness and need the to use.
Can you imagine if all the Linux podcast where to do
All Linux podcast listeners will be using GPG.
Great isn’t it?
So instead of just talking and saying KDE is crap, start doing something really useful.
Since I am ranting, might as well get this off my
If KDE have so many users, it cannot be all wrong.
I am using fluxbox but do appreciate KDE’s tweakability, it’s like comparing the ability to tweak in Linux to Windows.
Anyway, we Linux users should learn to understand that
not all humans are wired the same way.
You prefer apple, she refers orange, they prefer lychee while I prefer durian (or maybe you’ve not even tasted the fruit before :D )
Saying your distro/de/religion is crap and mine is better will not build a better Linux/world/community.
To end, let’s get GPG going.
Between Google I/O and World Cup news eating up my free time, just now getting to a post for episode 18.
Google’s announcements about cross-Google-platform ‘Material’ UI design and Android Apps that will run on ChromeOS make me suspect that Google is taking a shot at following Canonical’s desktop/mobile convergence strategy.
re: Paddy’s comments on some characteristics of the C language enabling whole classes of severe bugs. Apparently C language masters no less exalted then Ken Thompson and Rob Pike shared some of Paddy’s concerns and created Go (a.k.a. GoLang) a few years back at Google. Go is extremely similar to C, and while it supports memory pointers, pointer arithmetic is not allowed. It also supports memory garbage collection. (https://en.wikipedia.org/wiki/Go_programming_language)
It’s great that a 21st Century version of C has emerged, and some notable development is being done with it, but it would be hard to imagine that the millions upon millions of lines of code in the core of the existing *nixes would be reimplemented in it anytime soon.
Appreciated the update on systemd, I had not paid attention for awhile.
I ran across a Phoronix summary of Lennart Poettering’s May 2014 keynote talk _’Perspective for systemd, what has been Achieved, what Lies Ahead’_.
[ Phoronix summary: http://www.phoronix.com/scan.php?page=news_item&px=MTczNDk
[ Slides from the talk: http://0pointer.de/public/gnomeasia2014.pdf ]
It really just struck me how breath-takingly enormous the expansion in scope and ambition of the systemd project has become. I’ve never been inclined personally to use the label GNU/Linux when referencing distros, but if Lennart’s vision comes to fruition, I could see myself starting to say ‘systemd/Linux’ when referring to them.
Assuming that systemd doesn’t devolve into a disaster in terms of reliability or backward compatibility, it seems unclear to me at this point whether any Linux other then a systemd/Linux will be viable once all the dust settles. Like it or not, the next few years look very ‘interesting’ for Linux distros.
Hi SoN – I saw this when Lennart posted it, and was tempted to pop a link in the show we recorded this weekend. But I didn’t, as I was hanging fire to see if some video surfaced (although Lennart doesn’t think there is any).
You’re absolutely right though – it’s becoming a monster. And, with Debian having now folded, it is an unstoppable force.
As I’ve repeatedly said on the show, my concerns aren’t that it won’t all work well; I think that Red Hat has too much invested in this as a play to totally dominate the landscape for that to happen. Rather, they are both exactly that – who will need any other distro when they eventually all turn into Red Hat clones, plus the fact that we’re going to end up with something very un-Unixy. It may be a terrific product, but simple, loosely coupled etc. it won’t be.
As you suggest, the next few years are going to be interesting to watch; and in 10 years’ time, I guess some bright spark will turn around and say “You know what we really need? A FOSS Unix lookalike…”. At least that time around we may all get a head start from the rubble that systemd is now leaving in its wake…
Dell swapping a Linux device for a Windows device is a chronic Dell issue caused entirely by their internal politics. It has nothing to do with any actual qualities of SteamOS except “not being windows”. Dell has never understood “not wintel”, it’s the core business they built the company on and central to their corporate identity and self-image.
There’s a similar issue with HP, but with a different backstory. When Compaq bought the corpse of DEC, this threw together a windows shop and a unix shop with hugely different corporate cultures, which fought each other to the death. The windows guys were where the money was coming from, so they won the political battles and drove out the Unix guys.
About the time that fight wrapped up, Carly Fiorina decided HP should acquire Compaq. But Compaq’s Windows culture had been weaponized into vereran unix-killers, and immediately went for the throat of HP’s existing unix culture (HP-UX and such). The unix guys got sucker punched and taken out in preemptive strikes; the political fight was over before they knew there was a problem. A flood of existing HP engineers left the company rather than fight these insanely confrontational Windows developers who wouldn’t leave them alone, blaming Fiorina for killing the culture of HP.
(My uncle Alan was a 35 year HP veteran who took early retirement to get out of that mess, which is how I heard about it. See also David Packard’s fight to stop Fiorina’s destruction of the company, ala http://www.politico.com/news/stories/0310/34086.html )
Driving out the unix guys killed the entire existing HP R&D effort (except printers, which the Windows guys took for their own). Fiorina used the exodus as an excuse to cut the R&D budget, a classic “I meant to do that” move that deprived HP of new product lines once their old ones were exhausted. This is probably why a decade later they’re betting the company on unproven technology in a desperate attempt to climb out of the pit Compaq’s old niche turned into.
Recently got a whole lot of Lenovo s10e laptops that were headed for ewaste.
By combining bits from multiple machines, I Have been able to cobble together close to 60 machines which I shall donate to the local community (freecycle is my charity of choce) running mint xfce of course.
Problem is most of of the motherboards have bios passwords and I haven’t been able to reflash the bios and successfully remove the passwords
If you know of a method to remove the password I would love to hear from you.
Hi oldnerd – according to the Lenovo Hardware Maintenance Manual, it’s the usual route of taking to pieces and temporarily removing the backup battery, I’m afraid :(
See pages 34-36:
Thankyou thank you. I couldn’t for the life of me find this manual and read somewhere that being a lenovo you needed to reflash the bios as removing the battery wouldn’t work.
Your find says otherwise – that’s what I get for trusting the interwebs (who would of thought that some of the information on them isn’t 100% gospel truth).
If this works then I owe you a beer (or other beverage of your choosing).
More details to follow (when I get chance to try it)
Also really looking forward to the round up of lightweight XP replacement distros as requested by Gregor in ep 19 (which I haven’t listened to yet)
Well i took the plunge today and dis-assembeled a unit and disconnected the battery (a harder process than it needs to be).
After 1/2 an hour with the battery unplugged, the bios is STILL requesting a password. Grrrrr
Don’t suppose anyone out there knows of a good bios passwrod reader / re-setter for these units?
Hi oldnerd – since nobody has got back to
you… I realise that it’s an IdeaPad that
you’re talking about, but this post – and
the comments under it – about ThinkPads
suggest to me that if the battery removal
suggested by Lenovo doesn’t do the trick
you might be in a somewhat sticky
Comments are now closed.
The content of this website, and that of the podcasts produced by the website owners, is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.