Episode #15

Direct download links: MP3 & Ogg

Interview with Campbell Barton

Toggle full show notes
A slightly different show this time, as we bring you another fascinating long-format interview. Tune in next time to finally hear how Joe got on with KolibriOS, and for all our usual programming.


Richard Stallman’s Slashdot Q&A
Gartner’s predictions for Android sales 2014
Tech-deficit at the US Supreme Court
Relevant episode of Dan Carlin’s excellent Common Sense podcast


Ubuntu for Android runs out of steam in absence of OEM interest; Canonical attempt to contain PR backlash
Canonical Fixes Critical Bugs Allowing Users to Bypass the Ubuntu 14.04 LTS Lock Screen
Top tip, power users – upgrading Ubuntu may knacker your Linux PC

Heartbleed updates:
Cash, the Core Infrastructure Initiative, and open source projects
How I used Heartbleed to steal a site’s private crypto key
Heartbleed used to uncover data from cyber-criminals
How to Prevent the next Heartbleed
When Porting LibreSSL, Don’t Assume Your OS Is As Sane As OpenBSD

Did Mozilla jump the shark with Firefox 29? For Joe, 24.5.0 ESR to the rescue
Chrome URL shenanigans

This Google Motherboard Means Trouble for Intel

Linux desktop environment LXQt achieves first release (LXQt are interestingly planning to introduce Intents, à la Android, using D-Bus; presentation at FOSDEM this year)

KernelCare: New no-reboot Linux patching system

U.S. military UAVs migrate to Linux

Seen Elsewhere

An interview with George Neville-Neil about the network time protocol and the precision time protocol

Techmoan’s Chromecast Review


Thanks to everyone who mentioned us on Twitter, and to the Ubuntu UK Podcast guys for the shout-out.

A huge thank you to Eric Crampton and Brendan MacWade for the PayPal donations and comments. And thanks to johanv, Mikael Inscius, and an anonymous donor for the Flattrs. It really is appreciated, and lets us know that we’re doing something right :)

Paddy mentioned that O’Reilly has just released the OpenStack Operations Guide, which is a print version of the continuously updated online guide.

SonOfNed offered some thoughts on why Chrome OS is such a compelling proposition for so many users, albeit one with freedom-denying ramifications. He’s right, of course, on both counts. So why can’t full Linux distributions learn the lessons from Chrome OS? Are they really just too busy constantly reinventing the desktop UI, which is the least important part of the puzzle…?

Jason told us that he’d unfortunately bricked an Acer C710 Chromebook whilst attempting to install Crouton; see the show notes last time for more details.

Michael Tatum suggested we take a look at Voyager Live, and Morten Juhl-Johansen Zölde-Fejér pointed us towards another talk by PHK. Michael Albertson put Paddy straight on a particular turn of phrase, whilst Mark Smith told us that we’d prompted his escape from the clutches of KDE, and his subsequent move over to the tranquil waters of Xubuntu.

Last show, Joe asked if any listeners could suggest a decent Linux-ready wireless keyboard and trackpad combo. Ian Barton and Stilvoid both suggested the Logitech K400. Matthew Platte and Richard Kline own the slightly newer K400r, but the consensus of opinion there is that the keys may be a tad too small. Richard now owns an Adesso WKB-3000U, but that’s a trackball model, and the Keysonic ACK-540RF as suggested by Charlie Ogier may also be a little cramped. The hunt goes on.

Danny Knestaut, SonOfNed and Rob Mackenzie responded to Paddy’s question as to whether the balance in our News section is about right or not.

As you’d expect, a fair bit of feedback following our look at Ubuntu 14.04 Unity last time. Unlike Joe, Stephen Martinez is a fan of multiple workspaces and has somehow even configured them on his Windows box. Richard Marsh agreed with us that the Ubuntu Software Centre is a load of old bobbins, and was perplexed by having to ‘buy’ the Steam client for free from Ubuntu One.

Nigel Verity said that whilst we probably all owe Canonical for helping to popularise Linux, he doubts that nowadays we need Ubuntu as much as we think we do, as several other good ‘Debian with drivers’ distros now exist. Ghislain Vaillant agrees with us that the Dash doesn’t exactly speed your workflow, that the distro performs really poorly on older hardware, and that the Software Centre is the weakest point of the entire distro. But he does like the idea of web apps.

Nicola pointed out that the volume control noise Joe talked about is very similar to that found in GNOME Shell, so maybe Canonical took their inspiration from there, rather than Apple. James helpfully explained how to get the keyboard hints dialog back once it vanishes, and Thorsen gently chided us for criticising the file manager double-click behaviour. Thorsen also suggested we ought to cut Canonical a little slack, as did Steven Rosenberg, albeit for a different reason.

Cathryne wrote in to point us towards an AntennaPod support thread where anyone who’d like Opus support in that podcatcher can make their views known. She also asked whether crowd-funding and micro-payments are an effective means to fund raise and motivate developers to implement specific features. I know that we have some developers listening, so do get in touch and tell us your experiences.

Larry Bain offered some thoughts on the ‘Unix philosophy’. As we said on the show, if anyone else wants to chime in on this topic, or any other that is also related to The UNIX-Haters Handbook, please do so in the next two to three weeks so that we can include your thoughts in our upcoming discussion on the UHH, 20 years on.

Interview with Campbell Barton

We spoke to Campbell Barton about Blender, one of the pre-eminent successes of the FOSS world. The wide-ranging conversation took in not only what the Blender Foundation has been up to (including Gooseberry, their latest Open Movie project), but also thoughts about UI design and coding tools and practices – and frustrations – within the context of a large software project.

During our conversation, Campbell mentioned the BlenderNation news website and Blender’s Stack Exchange site. He also talked about Cycles, their new render engine; for an idea of what Cycles can do, check out this demo reel. Other examples of how Blender has been used include forensic facial reconstruction, a Coca-Cola advert, making fashion garments, the creation of a craniofacial prosthesis, and several animated films, including this French short.

Thanks again to Campbell for talking to us at such an unholy hour, and we look forward to bringing you more occasional in-depth interviews with really interesting people from around the FOSS world.

Linux LudditesAs ever, we’d welcome your feedback about the show either here on our website, via a mail to show@, or on Twitter @linuxluddites.

Thanks for listening.


  1. Andy Mitchell

    Hi Joe, hi Paddy

    Greetings from a very wet and windy Amsterdam. I’m listening to Podcast 15 as I write this. Its a great show guys. My ears picked up when I heard Joe was having screen blanking problems with his Xubuntu. I use Peppermint OS Four with the panel placed at the top where it should be lol. Peppermint uses LXDE running on top of Xfwm4 instead of Openbox, (very slick indeed). Anyway, I had what appeared to be a random screen blanking problem. I try mucking around with “xset”, the God awful Xfce Power Manager, turned the screen saver off and removed it from autostart and this and that. Nothing worked so I threw my toys out of the pram and put the whole thing down to witchcraft. I then stumbled across this. You and I both rate VLC highly and quite rightly so too! Its the best damn video player I’ve ever used. I had just watched a video on VLC and suddenly after about 10 mins the screen started to blank again. After some more scratching around I found that VLC has the “disable screen” saver ticked by default. You would think that is correct for obvious reasons. However the problem starts if you already have your screen saver disabled. After VLC is turned off, it politely tries to re-enables your screen saver. The result is screen blanking. The solution is: open VLC, go to Tools >Preferences. At the bottom of the left panel select “All”. In the left panel select “Video”. Untick the “Disable Screensaver” and click the “Save” button. Low and behold it wan’t witchcraft after all. My screen blanking issue is now a fading memory. Maybe this is also the same issue you are having. Good luck.

    Thanks for the show guys. I’m already looking forward to the next one.



  2. Rob Mackenzie

    The Interview with Campbell Barton, I will admit to not understanding a lot of what was said, but I had to admire his honesty.

    And can we have a “Joes Rant” feature ;-)

  3. Scott Dowdle

    Firefox != Google Chrome Browser. Ok, so the Firefox 29’s default theme makes it look more like Chrome but that’s about it… the cover on the book. You seem to think that makes the two the same in many other ways. Not really. Here’s a tip… with Firefox as the active window, hit the alt key… and notice your menu comes back. That isn’t really new in Firefox 29 but anyone who wanted the menu can get it back. There are themes for Firefox and from what I understand there is a add-on or theme that makes it look like it did before with tabs on the bottom so install that. I haven’t tried it myself. The killer feature in Firefox that I love is Tab Groups… and the fact that Firefox cares about an open web and their users. Google cares more about ways to work around do-not-track features in all browsers.

    Regarding your Stallman comments… saying that he is out of touch with reality because he denies the smart phone revolution (I paraphrase)… that doesn’t change the fact that he is right. The smart phone IS Stalin’s dream device (or whatever the exact quote was)… and the best answer is to completely avoid them. Ok, in the real world… billions of people have them. Yes, billions of people can be wrong. What’s the solution? Getting rid of the cell phone providers and having some sort of (perhaps magical) semi-global wifi network… and abandoning the cell phone data providers completely… and running free software that doesn’t think that privacy is an antiquated concept. Yes, I don’t have a cell phone much less a smart phone. Does that make me crazy? Maybe but happy crazy. :)

    • Patrick

      Hi Scott – it’s been a while, so nice to hear from you again.

      I’ll leave Joe to comment on Firefox generally, but must pull you up on your assertion that Mozilla is so different from Google. Remember how Mozilla wanted to shove ads into your browser? Remember how they only pulled back after a huge user backlash? And now we find out that they’re planning to do something similar in the future anyway[1].

      No, I’m not bashing Mozilla for the sake of bashing them. It’s just that – like most every organisation over a certain size – you have seriously well-intentioned foot soldiers, and a senior management whose goals so often don’t mesh with those further down the food chain. That’s simply life; and it’s as true in tech organisations/foundations as it is (sadly) in NGOs, charitable organisations etc. throughout the world.

      WRT Stallman. As I said on the show, I do have sympathy with his position on mobiles. But. You don’t win friends or influence people simply by telling them they’re stupid for using technology that they find useful. The Q&A offered an opportunity – as he was pressed on the mobile question – to issue a rallying cry to folks to come work on and/or proselytize truly freedom-enhancing ways of using this technology.

      Several times in the past on the show I’ve said that mesh and P2P technologies are what really excite me, and that they hold out the hope for a more free future. These are the sort of technologies that could make a huge difference, but we don’t even have a practical FOSS alternative to BTSync yet (there are a couple of projects[2] that I know are working on one, but they are nowhere near ready). So here was a chance for Stallman to tell the world about how the FSF is pushing forwards in this area, and helping to bring these decentralised services to the masses. But he didn’t.

      People need help to ‘do the right thing’. The vast majority of the public don’t have the technical know-how to be able to help themselves in this space – even if they wanted to. However, the readers of Slashdot are exactly the sort of folks who are capable of making some of these visions a reality, if only they were given a nudge in the right direction. Stallman didn’t. What we got instead was the isual negativity, whereas what we need is some positive leadership.

      [1] http://bit.ly/1jmGiEz
      [2] http://bit.ly/1jmIiwz

      • Scott Dowdle

        Ok, Mozilla might have some commercial suggestions for new Firefox installs… where there isn’t any pre-existing data for frequently visited sites. If that’s all, more power to them. Any previous Firefox user wouldn’t notice because there would already be data to go by. Would I prefer Mozilla not do that? Probably. I think both Firefox and Chrome currently have ad tiles there… but for their own stuff. In reality, the vast majority of sites on the Internet (no idea what percentage) have ads on them. I think Microsoft’s Internet Explorer opens up to MSN by default and it has a lot of ads on the front page.

        In any event, I see Firefox’s potential future for subtle ads for new users a stark contrast to the company that profits from running the dominant online ad networks. Don’t get me wrong, I don’t hate Google… but I do get nervous with all of the “free” services that they offer that are fairly dominant but paid for mostly with privacy. I know… they do (try to) provide anonymity to their users… but even if Google does exercise reasonable ethics (and you already addressed how are that is even among GMOs much less commercial, publicly traded entities) that doesn’t mean that someone won’t/can’t steal their users data from them. People in the US especially, but elsewhere in the world… are quite familiar with the NSA… and yet most don’t seem to mind too much all of the global companies like Google and Facebook collect on them. I’m not a privacy freak but I don’t think privacy is an antiquated concept.

        Regarding the Stallman thing again… I don’t think he is a saint nor a fantastic spokesman… but he did do certain things (like starting the GNU Project and the FSF and devoting much of his life to the cause). His main problem is that he has a number of personality quirks that seem to plague many in the technology industry. While it would be great if RMS were more like Steve Jobs was with regards to presentation and hype (not content) there isn’t much in his message I find fault with. The fact that he says the same things now that he said 25 years ago is a feature and not necessarily a flaw. The GPL did get updated from 2 to 3 in an effort to address the changing landscape (and that created quite an uproar)… but yes, much like the legal system, free software does seem to lag behind quite a bit trying to figure out what is meaningful and what is meaningless… in the longer run.

        Thanks for the discussion.

    • Joe

      I tried to customise Firefox 29 to make it like it used to be and although I got it reasonably close, it wasn’t quite right.

    • Rob Landley

      The thing to realize about Stallman’s FSF is that it started as a conservative reactionary movement trying to recapture the status quo of the 1970’s. RMS was objecting to the rise of proprietary software (due to the 1983 Apple vs Franklin decision extending copyright to cover binaries, previously “just a number”), his free software foundation wasn’t working towards an envisioned _future_, it was idealizing the then-recent _past_. (The unix tradition of “everybody has the source code” wasn’t invented by Stallman. BSD wasn’t created in a vacuum. Until apple v franklin, that was _normal_. Stallman was just moving over to Unix because his ITS ecosystem died with the PDP-10 hardware platform it was tied to, when DEC cancelled project Jupiter. Unix was 15 years old by that point, unix culture was already quite well-established.)

      The main thing RMS has really been good at all along is self-promotion. He didn’t invent emacs (even James Gosling maintained his “gosmacs” before then, and Jamie Zawinski’s xemacs was contemperaneous). Unix was independently cloned multiple times: ignoring BSD (and pcc) which proved in court they had replaced all of AT&T’s code, Coherent from the Mark Williams company had its own kernel, userspace, and C compiler written from scratch (AT&T sent Dennis Ritchie to confirm this, and he did), and Andrew Tanenbaum’s Minix had its own implementation of all that as well, from scratch, started when AT&T changed the licensing terms so he couldn’t use Unix source to teach OS design classes anymore. When Linus created Linux he was explicitly cloning _minix_. (Remember the original announcement on comp.os.minix? The one that only mentioned the GNU project by saying this was NOT like that, with a smiley because GNU was already vaporware? Yes, by 1991 the GNU project was stalled enough to be the Duke Nukem Forever of its day, because coherent took about 3 years to ship a commercial release, minix shipped in 1987 (3 years after the 1984 AT&T breakup and Tanenbaum wrote a _textbook_ to go along with it in that time), and Linux had its 1.0 release in 1994, about 3 years after his 1991 announcement. But gnu was still vaporware at 8 years.)

      Linux development took off because Linus hijacked the minix development community, which could cheaply get source code to the OS but could never get their own patches integrated upstream. (There were standard patch stacks you’d track down and apply to the base OS to convert it to 32 bit, add virtual terminals, and so on. Linus took patches so the next release had it integrated, so the developers switched en masse to his codebase. That existing developer base is why Linux leapfrogged BSD (which was coming down from minicomputer hardware rather than up from DOS, so the hardware you needed to run it priced it out of the hobbyist market until sometime after Linux 2.0 had already shipped).

      Each of those Unix clones took about 3 years to reach “usable”. RMS’s gnu project is nominally 30 years old and you still can’t really use it for anything nontrivial. (Did you know the “Cathedral” in the 1997 usenix paper the Cathedral and the Bazaar was the FSF? Not proprietary for-profit software, it was the guy who used to maintain the GNU Emacs Lisp extension library trying to figure out why Linux development was so much more effective than any of the FSF’s projects had ever been.)

      Yes, Linus used the FSF’s compiler because the minix one was 16 bit and he wanted 32 bit output. Fabrice Bellard wrote a new C compiler from scratch (tinycc) and got it to build the linux kernel (tccboot) in about 2 years before losing interest and moving on to his next project (qemu). The Linux guys got fed up with the FSF’s slow pace of development and forked the compiler (EGCS) and the C library (libc5). Then a company called cygnus (later bought by red hat) hired full-time developers and smoothed things over so the FSF could take those projects over again. (A bit like the ubuntu/debian relationship, they thought it was bad PR for their nominal upstream to die off.)

      In 1998 the famous “212% growth” of LInux was all the Java developers moving over to Linux en masse when Netscape released their code and made Linux a tier 1 platform. Those guys didn’t know the history of Linux, and the Linux guys had their hands full bringing them _technically_ up to speed (without time for proper enculturation), so Stallman fed them a revisionist history of the FSF instead of the truth, claiming credit for other people’s work.

      (Sheesh, he even claims credit for the GPL. You seldom hear about the actual _lawyer_, Eben Moglen, who actually wrote most of the thing.)

      The farther the FSF gets from “recapturing the glory of 1979”, the less connected to reality Stallman is. I drove up to Boston to interview him for a computer history project in 2001 and asked him why the FSF wasn’t doing anything with software patents, and he said it wasn’t his fight. The EFF was already trying to do something about them, but the FSF simply wasn’t _interested_. And people think he’s a visionary? He has trouble seeing the _present_.

      Sorry for the rant. I did a whole talk about this last year, and am preparing an updated version for Texas LinuxFest:

      • Scott Dowdle

        Thanks for the history. I lived through all of it… but I wasn’t necessarily conscious of it all because I wasn’t in the industry until much later. I have read quite a few books on history with the primary one being 1994’s “A Quarter Century of UNIX” by Peter H. Salus. I do disagree your conclusions on a number of points… but going into them all would be so tedious and no one would care to read it anyway. I don’t think the FSF was the Cathedral. While it has been a number of years since I read the book, my recollection is that the UNIX and / or BSD was the Cathedral (not the FSF) because they had a limited number of people with commit access… whereas Linux was the Wild West.

        I do agree with you that RMS has gotten a lot of credit for things that he doesn’t necessarily deserve but I don’t think that is because he has been personally claiming credit in the speeches that he gives although there might be a few minor exceptions. Yes, the reality is that the FSF is a fairly small organization with little money, almost no power, with many of the projects they sheppard having being donated by outsiders with a waning ability to keep it well maintained much less modern. No, the GNU Hurd kernel / OS hasn’t arrived (in all the years) and that’s why it is “GNU slash Linux”…but your characterization that the GNU Project is mostly a failure (maybe you didn’t say that) denies the fact that a lot of people everywhere use a lot of their programs and libraries… and will continue to do so for the foreseeable future. Sure Eben Moglen deserves credit… after all, he is a lawyer and RMS is not… and a license needs to be as legally sound as possible.

        This kind of reminds me of the whole Stan Lee and Jack Kirby debate. I’m definitely a Jack Kirby fan myself but Stan Lee sure seems to have gotten all of the fame and cash. :(

        • Scott Dowdle

          Rob, before you correct me, I’ll correct myself. Looking at the Wikipedia page for tCatB to refresh my memory… I do see that two examples of Cathedral were Emacs (which as you pointed out has a number of forks) and gcc. I don’t think the FSF itself, but yes two of its projects were examples. I think that is still true for Emacs which has been moving very slowly over the years… but certainly isn’t true for gcc anymore. As you mentioned Cygwin made a copy of the gcc ball and ran with it… and a few years later a particular Red Hat developer was seen as the ass-in-charge of gcc who was overly resistive to change. gcc has since had a control revolution. So far as OSes go, I do think that the BSDs are more of a Cathedral than Linux is… which is extremely obvious… so if the book got updated I think the BSDs should be the new example… but then again, some people see that as a feature and not flaw. There certainly has been a lot of “Internet time” since 1998… and the new ideas in tCatB are now the ancient wisdom handed down.

          • Rob Landley

            Eric Raymond personally told me that CAtB’s Cathedral was the FSF during in the 3 months I crashed on his couch in 2003 editing The Art of Unix Programming. (See paragraph 2 of http://www.catb.org/esr/writings/taoup/html/pr01s06.html)

            Peter Salus is fun, he used to live here in Austin. (I drove to california to meet him the first time, at the last Atlanta Linux Showcase in california, and it turned out he lived like 4 miles from my house at the time. Last time I bumped into him, at Penguicon, he was living in Canada.)

            Computer history is a hobby of mine (http://landley.net/history/mirror) and it’s easy for me to talk your ear off if we get started on that, I’m _trying_ not to blather. I covered some of it in the above linked talk. The FSF annoys computer historians because in the effort to put the spotlight on themselves, they push revisionist history that Just Ain’t So:


            Most of the second generation computer scientists are still alive, or were recently, and not in such high demand that you can’t phone ’em up and arrange to meet them for lunch or something if you’re willing to travel across the country. I mentioned interviewing Stallman in Boston once. I spoke to Dennis Ritchie on the phone and exchanged email with him. I was on an email exchange with Ken Thompson when he was reviewing The Art of Unix Programming (although that was more me being cc’d than really participating). Ohio LinuxFest tends to invite Unix bigwigs as speakers: I met Kirk McKusick there last year and Doug McIllroy the previous time I went there. Plus you can track down all sorts of fun stuff on youtube, here’s Grace Hopper on David Letterman:


            The FSF’s initial popularity stemmed mainly from 2 things:

            1) before the National Science Foundation changed the internet backbone’s “acceptable use policy” to allow for-profit ISPs (loosened in 1990, ended in 1994), getting an FTP server was _really_hard_. It required a valid educational or government purpose. MIT gave the FSF a server with broadband internet, so for example Larry Wall signed over his “patch” program (despite supporting perl on Windows years later) because not because of idealism, but because doing so got him distribution. But Linux got 3 non-FSF mirrors in 1992, the same year sunsite was founded:


            And by the time geocities launched in 1995 the FSF’s distribution function had been rendered completely irrelevant. People stopped signing over their code to the FSF because they no longer needed to do that to get it distributed.

            2) In 1987 Sun’s new head of marketing Ed Zander decided to “unbundle” standard OS components and sell them as add-ons for an extra fee. So Solaris stopped coming with a compiler built-in. The two original gcc targets were VAX and Sun3, so people looking around for a compiler for their Sun workstations found gcc, which was _crap_ at the time, and those Sun users pushed to gcc by Ed Zander’s short-sighted greed provided the critical mass of bugfixes and upgrades to turn gcc into a reasonable compiler. (Note the first gcc release _not_ support the i386, which first shipped in 1985 and was in the Compaq Desqpro in 1987, because Stallman was not good at predicting the present, let alone the future.)

            Also, Cygnus != Cygwin, the second was a product of the first. Cygnus wasn’t anti-gnu, it was in fact founded in 1989 by a total Gnu fanboy (somebody so impressed by the 1987 gcc release he founded a company to support and upgrade it). After Red Hat bought it about half the staff re-coalesced under under Code Sourcery (which then got bought by Mentor Graphics).

            A 1999 snapshot of Cygnus, written by its founder, is here:

            The FSF did indeed do some interesting things back in the 1980’s, but A) not as big as it claims, B) not as unique as it claims (they had nothing to do with Project Gutenberg, Creative Commons, BSD Unix, and their claims about Linux are just ridiculous), C) that decade ended 24 years ago.


          • SonOfNed

            Rob and Scott, I have really enjoyed reading your comments on the history of Unix and the FSF. Thanks much for sharing your knowledge.

    • Patrick

      Thanks – they are aware of it, and looking to resolve.

      Podtrac is normally pretty reliable, and they are used by some huge shows (e.g. Leo Laporte’s), so we’re hopeful they’ll sort things out promptly. If not, we’ll look at alternative ways of doing things.

      I’ll update after I’ve spoken to Joe. Meantime, our RSS feeds are working fine, and I’d recommend any regular listeners to pop one of those into their podcatcher of choice, rather than using the download links from the Show Notes.

      Again, thanks for the heads-up.

  4. Steven Rosenberg

    I love the idea of “intents.” This is one of Android’s best features, and I was a bit crushed when I got an iOS device to find that Apple doesn’t match this feature. (I’m also crushed because iOS doesn’t allow me to download an MP3 file, like it’s their business, though I guess it IS their business).

    In any event, a Linux desktop with “intents” is something I am very interested in.

    Swinging over to Canonical, I’m in a bit of fresh hell right now as I upgrade a Lubuntu 12.04 system to 14.04. Lubuntu 12.04 is not an LTS, so there wasn’t a direct upgrade to 14.04. Instead you are offered 12.10 as an upgrade.

    I tried to upgrade directly to 14.04 via the CD, but the disc was squaking about my system not having a swap partition, although it clearly does. I tried to remedy this but wasn’t terribly confident that I did.

    Then I tried to back out of the upgrade.

    Unforunately the installer had already seen fit to trash enought of my root partition to render the installation unusable. I should have gone straight away to a 14.04 install, but I decided to reinstall 12.04, and go in steps.

    The reinstall went smoothly, then I started the in-place upgrades via the Update/Software manager. Luckily after 12.10 I was able to “skip” to 13.10, and I’m just about done with 14.04.

    We’ll see if this works after the final upgrade.

        • Campbell Barton

          re: Oracle v. Google, recommend episode 0x44, http://faif.us, interesting discussion on the matter, even if you’re not so much a fan of the FSF.

          @Scott Dowdle – another non-phone-owner here, I like to keep attention seeking gadgets confined to my office. (don’t have a laptop for similar reasons).

      • Steven Rosenberg

        It did turn out eventually. Once I got to Lubuntu 14.04, the desktop for one of my user accounts was broken, so I had to create a new account and chown all the files anyway.

        I thought I would save time and trouble by upgrading with all of my apps installed, but I had to reinstall them anyway.

        Yep, it would have been easier to do a full reinstall of 14.04 from the beginning.

  5. Stephen

    Another great show – though I had to look up the definition for FORTNIGHT – I had an idea of what it meant, more or less, but I wanted to make sure. It’s just not a word that you hear here in the States – Well at least in this neck of the woods, Los Angeles.

    As others commented, I enjoyed the interview regarding BLENDER. Also like the questions that were asked about coding.

    Well look forward in a fortnight to listening to the next podcast.

    Hasta later,

    Stephen Martinez

  6. Esteban

    The linux action show also loaded lxqt and interviewed a couple of their developers. But yeah, it does load better in xubuntu because of the qt. As for firefox vs chrome – firefox now = chrome. I still stick with firefox cause chrome often just hangs.

  7. Jack Dennahower

    I am not going to get into the above quagmire. I have just downloaded, and am starting to read, the Linux Haters’ Handbook. So far I am laughing to myself as I was a user of HP-UX back in the 80s through 90s. The Canadian Met. Service was using 23 inch CRT screens with HP boxes with HP-UX for continuous satellite and RADAR loops, dissemination of surface and upper air charts, and (for a couple of years) as data entry and transmission of observations. We loved them as it did away with mufax machines with chemical saturated paper and wire helices that burned the charts on said paper. When the internet and intranets took over we went gradually to RHEL for graphics. I got to play with some of the x window scripts and was fascinated with the stuff.

    My last workstation was 2 23 inch screens with unix boxes and 2 23 inch screens hooked to a Win-XP box. The keyboard tray looked like a pipe organ keyboard.

    Love the show as always.



  8. Andrew Precht

    It would be nice to try to listen to you. To bad you have no rss that works with carcast. a simple rss url works for all other podcasts. not here….

  9. Bruno Miranda

    Greetings from a hot blued sky Portugal!

    About the volume control noise on Ubuntu that sounds like the Apple one, I have a friend that was studying a degree on music and composition, and I remember him mentioning me a couple of years ago that in contemporary composition it seems that there was a trend about creating what they called popcorn sounds.

    Since I am a Linux Mint user and never used Ubuntu (ahah!) and very seldom used a Mac (I have a full brain and an empty wallet), I really can’t compare the sounds. But popcorn trends or wannabe copycats, who knows?


  10. Richard

    Regarding Joe’s Skype problem – I just had a hell of a time trying to get sound working on Xubuntu. I’ve never had a problem before, but recently re-installed from scratch from Ubuntu 12.04, switching from 32 to 64-bit. In the end, I uninstalled pulseaudio. On Xubuntu it has no dependencies that end up uninstalling the world, and Skype seems to work fine now. It reminded me of Ubuntu 9.10 (IIRC) where this sort of pulseaudio-induced sound screw up was all the rage.

    • Steven Rosenberg

      Skype in Fedora seems to have continual audio problems that I’m sure are due to PulseAudio.

      Sometimes I get no audio, sometimes I get constant noise through the speakers. Both problems seem to be solvable by quitting, then restarting Skype.

      When I do run Skype, PulseAudio grabs way too much CPU for a program I’m not currently using to do any actual talking. I generally use it for IM, and it’s got way, way too much overhead for that.

  11. Oskar

    Thank you again for a great interview. Too often the interviews are bit hollow and only the obvious are repeated. But you seem to be able to do interviews where great opinions and insight into things are revealed. It was nice to listen when Campbell when talked about coding and work with Blender.

  12. Mark from Pompey

    I have Ubuntu 14.04 LTS but i’m not getting the volume control noise Joe talked about.

  13. MikeF

    Great podcast as always, now in my Top Three :-).
    Regarding Firefox, Joe, I feel your pain. The new UI is fugly and less customizable. Who demanded black / flat icons?
    Fortunately I discovered Pale Moon, a FF fork that strips out some cruft and seems intent on keeping the same UI. If they do that and maintain add-on compatibility, they could get a large following and hopefully a sponsor. So far the Linux port is not official, just a tarball that you ‘untar’ into the /opt folder.

  14. SonOfNed

    Really enjoyed the David Wheeler article/blog on software testing strategies (How to Prevent the Next Heartbleed) as well as the most excellent interview with Campbell Barton. One of my favorite parts of the interview was Campbell’s comments about the ‘real world’ trade-offs that are necessary when dealing with ideals in s/w development such as the testing ideals outlined in David Wheeler’s article.

    Continuing the ‘Canonical Watch’ theme (with an OpenStack element), Mark Shuttleworth revealed a new aspect of Canonical’s business plan with his announcements at the OpenStack Summit: https://tinyurl.com/pkvvtpy

    It appears to me that Canonical is going full steam ahead in their enterprise web server and services business.

    Joe commented in the podcast that he would prefer that Canonical drop Desktop Ubuntu to focus resources on Mobile Ubuntu rather then do an inadequate job with them both. That suggested an interesting thought experiment to me. What would the Linux Desktop landscape look like if Canonical we to drop Desktop Ubuntu the near future?

  15. Cathryne


    My question about how crowdfunding (or voting with money on) particular features in Free Software projects is viewed around here seems to have not been very interesting. OK, one more try. Even an explicit “not interesting” would be an OK reply ;-)

    I consider myself a power user (10y Win, 5y Ubuntu, 3y Mac OS X & Android), but never yet managed to jump the hurdle of deserving the term programmer. I can test software from the user’s perspective and write a proper bug report or usability summary, tweak software via its config files and apply a list of thoroughly commented terminal commands. Not code anything myself (yet), though, sadly.

    Therefore, supporting FLOSS projects monetarily is one of the few options I have (besides translation and testing). Services like BountySource, ADN’s Backer IMHO take this a step further by focussing discussions on particular features and on how much users are willing to spend on them. What I am interested in, is how FLOSS developers see these options? Annoying chatter by, or useful input from, users?

    Thanks for any reply and many greetings!


    • Patrick

      Hi Cathryne,

      /I/ thought that your question was an interesting one, and was a little surprised that nobody responded to it. I’ll be very happy to bring it up again next show.


  16. Campbell Barton

    Hi, I didn’t see your previous post, but I find the topic quite interesting and have some experience from both sides.

    I found it can work well, but theres a bunch of issues…

    First is that a lot of the low hanging fruit in a project has often already been picked, if there is some feature a lot of people are passionate about which is missing, theres probably some good reason existing developers didn’t add it already. (though that depends on the projects age).

    Maybe some internals need to be rewritten to support the new feature, or its a can of worms. or its simply a really big project.

    So typically the people you want to do these kinds of projects are experienced developers, which brings me to the second point…

    In my experience most of the best developers already have jobs, weather they already work for the organization which makes the software or they have an unrelated day job, it makes the incentive of money less interesting.

    As a developer I see these kinds of projects as a liability, If its simple – I rather do on a weekend for no pay. And if its complicated, then its very hard to guess at how much time the project even takes.
    (estimating software dev time is notoriously difficult).

    The added paperwork/red-tape on-top of doing the actual work also isn’t so attractive.

    Also theres just a lot of communication with users and making sure theres no room for confusion before starting. (getting details ironed out, managing expectations… etc)

    All this overhead means for projects less then a month or so, I just wouldn’t bother to attempt to get funding. And any single feature which takes longer then that ends up being a risk/unknown for both users and devs.

    After saying all this, there are projects I think it can work for so I don’t mean to paint such a negative view on the matter, but in general I think its tricky to do well.

    • Cathryne

      Hm-hm, OK, thanks for listing these counter-points. I guess the conclusion is to learn programming myself :-/

Leave a Reply

Your email address will not be published. Required fields are marked *