tonywhitmore.co.uk  

Lunchtime on Sunday was the GPG keysigning in BOF 3. It was reasonably well attended, with not too many people to make the informal style a pain. It was also nice to be plunged into an almost entirely different group of people for a while; a great chance to meet yet another bunch of community members. The keysigning, with different people presenting different IDs from around the world, has lead me to put together my own GPG keysigning policy, which I will put on my wiki when I’m entirely happy with it. Browsing round the exhibition after the official signing session, I exchanged keys with Mark Shuttleworth. Paul Sladen (who remembered me from his visits to HantsLUG) also took my key to sign. Neither of them have signed my key yet, but I’m sure it’s just a matter of time.

I was keen to see Michael Dominik’s talk on DIVA as I’d seen the rather cool videos on the project website and heard the interview on a LUG Radio episode. The talk didn’t really demonstrate anything that new over what was in the online videos but it was good to see it in action, processing some MPEG and OGG files. It looks featureful enough to be useful for our LUG talk videos and again, I was able to ask Michael some specific questions after the talk. That’s one of the cool things about LRL, being able to just chat to the developers of the projects you’re looking at. I’d like to track both DIVA and Pitivi as they both look promising. I’ve got a feeling that DIVA isn’t going to be easy to get up and running, though it has the edge over Pitivi in terms of features at the moment.

The penultimate talk I attended was the “Women in Open Source” talk, given by Kat, Jen and Phated. This talk was easily the most packed lightning talk on the Sunday, with all four large gents and most of the crew in attendance too. I don’t think think this was just because they had pulled Ade up to the front of the room as an example of male chauvinism. It was very noticable how many more women were at LRL 2006 than the previous year’s event, but they were still a minority. Although the talk was not entirely negative, it looked at how women can be treated online, how they can be treated in LUGs and how they can be subjected to uncouth behaviour. Although they made quite sure that the talk was well-controlled, there was sufficient interest in the topic that an impromptu BOF took place in the corridor outside the talk room which saw a crowd of people taking the discussion even further.

Seb Payne was the last lightning talk of the day, a brief but very well presented tour of iFolder, something that I’ve been unsuccessfully trying to get working for several months now. Seb was a confident and amusing speaker and handled even the more tricky questions well. He convinced me to try and compile iFolder from SVN on my Ubuntu Dapper systems. I have promised that once he produces some documentation, I will. I missed most of the goodbye from the gents on the main stage as I was shutting things down in the lightning talks room. I caught the end of it though, and there was a real feeling that the weekend had been a success, that it had strenghtened the community, fostered debate, shown what’s new and given some projects a firmer sense of direction. The gents, and all those who helped them organise the event, should be applauded for this. Afterwards, Adam Sweet told me that Matt wanted to see me. So I toddled off to find him. He thanked me (on behalf of the team) for my help over the weekend and gave me some Linux games from the tuxgames.com people as a thank you. It was really sweet and I was quite touched.

The journey home was even more interesting that the journey up. After a detour or two to avoid the traffic jam on the motorway and to wave at Alan’s brother’s house, discussion soon turned to our thoughts of the weekend. Everyone had enjoyed the weekend a lot, everyone having their own highlights. This in turn lead to a discussion of the “Women in Open Source” talk and the issues it raised with the laddish culture in some parts of the community. The exchange centered around whether this culture puts people off joining firstly the LUG Radio community and secondly the UK LUG community. The former is certainly true but is perhaps less important; There are other podcasts people can listen to if they don’t like LUG Radio’s style. The latter is more contentious and something I’d not given much thought to before.

LUG Radio has always had an attitude of “this is us, take us or leave us,” which personally I feel they can do because they aren’t representing anyone other than themselves. They aren’t bound by a company to present a perfect public presence and they make no pretence at being a professional outfit. However, it was suggested by that LUG Radio might, through it’s name, be perceived by some as representing LUGs or LUG culture, regardless of whether this is the intention of the show itself. I’d not really thought about this before. If that is the case, the point went, do the presenters have a moral responsibility to reflect this in their deportment on air? The argument was that by calling the show LUG Radio it does at least align itself with LUG culture and acknowledge they are at least a part of that culture.

We all agreed that LUGs should be as inclusive as possible and that this should be reflected in the places they meet, the activities they run and the way meetings are conducted. Running a LUG in the same way that LUG Radio runs would certainly put some people off getting involved in their LUG. However, it can’t be denied that some people prefer this sort of LUG meeting, but they are probably already committed to the FLOSS movement, already part of the community. Conversely, it isn’t good advocacy to put new recruits off attending meetings in the first place or scare them off if they do attend through swearing and laddish behaviour.

But does LUG Radio represent LUG culture in the UK? The atmosphere of the show is certainly very different from our HantsLUG meetings, so I would have to say “no”. There was a lot more LUG-related content in the early shows, with phone calls in from LUGs around the country, but this hasn’t happened in the last two series. There’s nothing in the show’s content that allies it with LUGs, other than some discussions being prompted by events in the presenters’ local LUG. But this is perhaps something that regular listeners pick up on, rather than those who are new to the show or to LUGs.

The articles written by the presenters and published on their blogs, in magzines etc. are well-written and intelligent, but this does contrast with the atmosphere on the show. It was pointed out that bad language or crude jokes wouldn’t be acceptable in such articles, but they are deemed to be OK on the radio show, despite the fact they may exclude some (potential) members of the community. Space for freedom of expression is important though, and I can totally sympathise with the presenters wanting to have a outlet for their frustrations. I don’t think I would want to see the show self-censored any more than it is. Personally, I think there is room for talk about advocacy and still enjoy kicking back in a podcast, but conversely I wouldn’t want the show to be perceived to be representing UK LUGs in general. Perhaps a “clean” podcast will prove more popular than LUG Radio at some point in the future which can be better said to represent UK LUG culture, but LUG Radio is clearly doing something right as its download figures and active community demonstrate. Regardless of the lack of conclusion to the debate, it was rewarding talking to people who have clearly thought a lot about the issues.

So, I suppose through these posts, I’ve been trying to say that LRL 2006 was a blast, really good fun, but also much more thought-provoking than I had expected. The efforts of the organisers were noticed and appreciated. They should be proud of their acheivement and the community that has grown around their show. It’s great that people care so much about FLOSS that communities like this form around it, and it’s fantastic that there is room for diversity of opinion and discussion. I enjoyed meeting lots of people I primarily only speak to online: Jono, Matt, Aq, Adam “drinky” Sweet, sward, mrben, neuro, sebpayne, Xalior, ChairmanMeow, Cillian, bilboed, TMM, Phated, pickle, resiak (and anyone else I’ve left off). (Ade too, but I never speak to him online. ) I enjoyed LRL 2006 immensely, as much for the community around it as the talks themselves, and I’m looking forward to being involved next year.

Sunday was the day of the lug.org.uk BOF that I had proposed on the UK LUG Masters mailing list back in April. It was clear from the Saturday sessions that BOF area 3 was much better suited to running a BOF than the spaces in the auditorium, so we arranged to move the lug.org.uk BOF from BOF 2 to BOF 3. I hadn’t been that optimistic about the BOF originally, mainly due to being in the hangover slot – the first session on Sunday morning. However, the gents had sensibly pushed the start of Sunday as late as possible, so it kicked off a little after 11am. It was a surprise, but a pleasant one, to see so many people turn for the session. There were about twenty people there, with about eight or nine different LUGs represented. With that many people crammed in to the bizarrely green room though, we soon started to run out of air!

There was some confusion over the BOF, as two guys from Sussex LUG thought that they were running the session. It turns out that they had volunteered a Linux User Groups BOF and we (UK LUG Masters) had agreed a lug.org.uk BOF. However, only the lug.org.uk one had made it to the final schedule. As it turned out, the two Sussex guys were also responsible for running the practical aspects of their LUG (their Chairman apparently in absentia), so were great contributors to have at the lug.org.uk meeting.

I was really impressed with the discussion we had. The first half we dedicated to practical issues: The problems with the lug.org.uk hardware; how we really need a replacement; how to cover running costs for the hardware (replacement disk drives etc.); the administrative structure that might be needed to accept donations from UK LUG members. The over-riding feedback from those present way to just get on and do it, without getting bogged down in endless debate and trying to please everyone all the time. The second half of the session focussed on what people would like to see happen in the future. Lots of suggestions in this section including: revised way of helping people find their local LUG; a UK-wide Planet; a UK-wide wiki for technical information; information and advice on how to set up and run a LUG; information and advice for newbies on basic LUG concepts like “mailing list”. These were by no means all the suggestions, but others did a better job of taking notes than me.

I was still supervising the live audio in the lightning rooms. At some point during the morning session I began to have my suspicions about the mini-jack connector to the microphone doing the audio recordings in the Beard room. The waveforms in Audacity just didn’t look right, and I was worried we were just recording crackle. It didn’t help that the connector seemed to have become quite loose in the socket. We didn’t really have any other option than to carry on recording and see what the audio was like afterwards.

None of the morning’s talks hugely grabbed me, so after they were up and running, I went to wander the exhibition with Alan. I spent a good 20 minutes talking to Juski from the MythTVtalk.com stand. Some of you may know I’ve been working on my Meeja Box project for some time and, although it works, I keep getting told that having to SSH in from my laptop and issue cryptic commands to watch some TV does not constitute a user-friendly interface. MythTV, along with Freevo, has been on my radar for some time now, but I’ve had quite a few specific questions I wanted to ask and it was an ideal opportunity to collar someone who knows. The outcome of the conversation is that I’m probably going to reinstall the Meeja box with Ubuntu Dapper Server on it, then try and get MythTV working on top of that. There are a few features I’m after that won’t appear until the next release (0.20, I think) but most of what I want is there.

Next stop was Josette on the O’Reilly stand, who recognised Alan from FOSDEM. He clearly bought far too many books there. I was relatively well behaved and only purchased these three books. We had a brief friendly chat with the FSFE guys. I’m interested in the FSFE and supporting their work, but feel the €120 annual subscription to be a bit steep. The guys told me that it was actually minimum donation of €60, which is a little more reasonable. Apparently this is the lowest amount that actually gives them some money to use in their campaigns, once the costs of the membership are covered.

The lightning talks over for the day, it was time for the live recording. Predictably there was a crush in the main auditorium as people piled in to get a good view of the action, particularly as Jono was due to shave his ‘orrible beard off for charity. The upshot of this was that I was stuck at the very back, only able to see one or two presenters round a large pillar. Maybe this impaired my enjoyment somewhat, but I felt the live show was a bit lacking in places. There were some sound issues during the recording, and certainly the live audio wasn’t as high quality as it could have been, sounding very distorted at times. The Open Source Pass the Parcel, piloted on the previous episode, was resurrected with participation from the audience. It was amusing watching a room full of geeks fail to catch a stuffed toy as it was hurled about. To be fair, most of them had a pint in at least one hand. However, expecting every random audience member to have a funny or interesting question ready to ask the presenters might have been a bit ambitious. Most of the people who had the parcel when the music stopped gave their name and sat back down again. The format felt like it needed a little more work, or at least planted questions.

It was a great idea to do a live interview, but rather than one of the more experienced interviewees, they interviewed Juski from the MythTV stand. I had a bad feeling straight away that he would be taken to task over Jono’s run-in with the MythTV developers. This did happen fairly quickly and seemed to take up the majority of the interview.* It didn’t feel very enlightening as a result. It also felt a little unfair as Juski, the interviewee, isn’t a developer on the project but a user and community member. He was repeatedly put in a very awkward situation between acknowledging the possible faults of the developers and not wanting to bad-mouth them. The interview was the source of much debate in the curry house later that evening, trying to decide if running a stand for a project makes you responsible for the actions of the developers. Some people thought that advocating a project gave the impression that you were at least part of it, even if not a developer, and should therefore be prepared to answer questions about the problems within the project. Others felt that advocating a project didn’t mean you should had to like everything about it, just be a happy user. They also felt that as the stand was under the MythTVtalk.com banner, it was detached enough from the main development community. Either way, the interview wasn’t a comfortable experience to watch, so being involved must have felt worse. Also, being interviewed in front of a couple of hundred people is pretty scary, particularly if you’re not used to doing it.

The rest of the live show consisted of the awards ceremony, with Xalior and Neuro getting the Community Hero of the Year award and runner-up respectively. Phated won the furthest travelled award and Bruno picked up an award for his talk, as did Mirco Muller. The climax of the show was Aq and Jono being de-bearded. The former shaved his own beard off, but the honour of cutting Jono’s beard went to Ted Haegar of Novell. Over £700 was raised for Amnesty International, so well worth it.

Back at the hotel, HP helped Alan fix the MMC card from Laura’s digital camera, which had stopped working after Alan copies some photos off it. After a wash and brush up, we got a taxi in to town for a meal and the LRL party back at the Union. Without wanting to be patronising, it’s brave to travel somewhere where you don’t know anyone. You risk spending the whole event on your own and feeling left out, but HP was great company and we were happy to have his company for the evening. Hugo got to educate him in the ways of the British curry house too.

By the time we’d got in to town, had a beer, had a curry and headed to the Union, it was getting on for ten o’clock. The party was pretty busy, but more with beer drinking, pool and chat than actual dancing. There was, as I had predicted, the inevitable problem of keeping such a diverse group of people happy. Presumably being berated in turn by dance fans, metal fans, rock fans and cheese fans he seemed uncertain what would make the crowd most happy. So he polled the audience over the PA to find out what type of music people wanted to hear. The dance fans then cheered, so on went a dance record and the people who had been dancing to the cheesy music cleared the floor. This was repeated several times and felt a bit dis-jointed at times. Probably unavoidable though, given the audience. Cheese and rock seemed to be the order of the night. The sight of several pillars of the FLOSS community doing the macarena was something quite special. And no, I wasn’t doing it.

I spent most of the evening chatting to Bruno, and after a while, Jono joined us. The three of us had a great in-depth talk about some aspects of the LUG Radio community. Later on we were joined by Matt, and Laura and I had an amsuing conversation with him which revealed that, had life been different, Laura and Matt could have ended up working together in the same job. Matt could have been an HantsLUG member: Who knows, we could have had HantsLUG Radio.

Got back to the hotel about 2am, and pretty much straight to bed.

* Update: Apparently this is something of an exaggeration. The part of the interview I’m referring to took up about 20% of the total interview.

After lunch, I dipped in and out of Danny O’Brien’s talk. I enjoyed Danny’s columns in Linux User & Developer, but they seem to have stopped now. Danny talked about quite a few things, including his work with the EFF, standing up for people or projects prosecuted under technology laws. Unfortunately, both he and Bill Thompson were called in to the Mass Debate after half an hour of their alloted one hour slot. I should have seen this coming as both were ideal candidates for the Mass Debate and in the last episode of LUG Radio before the event the guys gave the impression that they didn’t know who was going to be taking part in the Mass Debate. So, note for next year: Don’t schedule interesting contributors from high-profile organisations at the same time as a debate they would be perfect to be involved in.

I took the chance to have a wander round the exhibition area and check out some of the stands. In some ways it was the usual suspects, although MythTVtalk.com, the FSFE and Novell were new to LRL. The bits of Des Burley’s talk that I heard made me reflect on the wide range of talks that this year’s event included. Whilst it’s great hearing about specific software projects and seeing demonstrations of cool new stuff, the issues around FLOSS are so wide ranging and it was great to see that reflected in the talk schedule; the OpenStreetMap, OpenDocument Fellowship, Des Burley, Belial, Danny O’Brien, MySociety and Pledgebank touched on social and political issues. These were by far and away the most thought provoking talks, and often caused the most heated debate both during the talk and in the bar afterwards. It’s really important not to view FLOSS as being created inside a perfect vacuum, where other, non-software related issues don’t intrude. They do and people in the FLOSS community need to be aware of them.

I made sure I sat through all of Edward Hervey’s talk on Pitivi, as it’s a project I’ve tried many times to use without success. It’s in Dapper universe now but that version is quite out-of-date as the software is under active development. The talk was a good chance to see the new features in the software but also demonstrated that there is a long way to go. (I was able to grab Edward in the hotel lobby the following morning to ask him some of my more specific questions.) Ian, who I’ve met on several occasions, mostly though Schoolforge-UK, talked about OpenDocument, why it’s important to governments, businesses and people. Daniel Carrera, who I’ve spoken to online a few times, also demonstrated an OpenDocument viewer. I was already aware of the project but didn’t really understand why it could be so important until the talk. When Daniel started drawing comparisons with Acrobat Reader and explained that the viewer would be cross-platform I finally got it. If people already have OpenOffice.org installed then it’s not a big deal to use it for viewing ODF documents. But if not, then a small, light viewer application (that is a significantly smaller download that Acrobat Reader) is really crucial to aiding adoption of ODF.

Bruno “kNo” Bord’s lightning talk, “This Talk May Contain Swearing,” was easily the busiest lightning talk of the day. It was standing room only, and all four of the large gents made sure they were in attendance. Bruno had listened to 16 episodes of series 3 of LUG Radio and generated data on who swore in what way throught those episodes. Delivered dead-pan, Bruno explained his method and results as if he were presenting a minor paper on ear surgery at a medical conference. Having explained his method, he then talked through various trends in the episodes: average swear rate per minute, who swore most, which swear-words were used most, which episodes contained least swearing. The latter statistic lead to the definition of the “RTHE” or Reverend Ted Haegar Effect. The talk was superbly well received. I’m not sure what the sober souls who had gone to the Grid Computing talk next door thought of the raucous laughter coming from the adjacent room.

The following morning arrived far too soon. The night had been warm and the hotel room boiling. The breakfast room was full of LRL attendees, mostly #lugradio people. After eating a nice enough breakfast (although the sausages were a bit odd), Alan drove us to the venue. The Students Union is actually part of a big building marked “Wolverhampton University”, which includes the library and a number of shops. Parking was in the almost adjacent underground car park and good value at £10 for the day (split five ways).

We got to the venue about 20 minutes before the doors were due to open and already a small queue had gathered. Luckily for me, Jono ushered me through to the main auditorium and presented me rather unexpectedly with bright yellow crew T-shirt, my punishment for volunteering to help out. After a brief tour of the venue, including the two rooms I would be looking after, I was left to mooch about for a few minutes. It was immediately apparent how much bigger the venue was than last year, with two large lightning talk rooms, exhibition and sales space in addition to the bar and main arena. The gaming and BOF areas 1 & 2 were at the back of the main auditorium, whilst BOF area 3 was a small room squeezed between the gaming area and one of the lightning talk rooms. The main auditorium was dominated by a huge projector screen,

The quality of the signage, banners and programme demonstrated a lot of thought and presumably quite a bit of expense. Even the Low Tech Wiki was a great twist on an old idea. Of course, various chuckleheads spammed it over the course of the weekend. In parallel with hi-tech wikis, this reduced the usefulness of the wiki to almost nothing. Next year I suggest better anti-spam protection, in the form of someone mean-looking with a big stick standing guard.

The venue’s plasma screens were used to display important information to attendees: a “now and next” updating timetable, a news ticker and an area for photos. The idea was that attendees could have their photos included into the cycle. I was amused to see that the default photos shown at the beginning of the day were from the photos I took last year, apparently nicked by Aq. The system looked really impressive but did suffer from technical problems over the weekend, which left the photo section showing static most of the time.

Before long the doors opened and the great unwashed (sometimes literally) entered. It was about this time that it became apparent that one of the two video cameras in use in the lightning talks room didn’t have an internal microphone. Somewhere along the line, it looks as if the video people thought the audio was being recorded direct off the speaker’s mic but this wasn’t the case. (I can safely take no blame for this as I wasn’t involved in the planning stage!) After a period of headless-chicken imitation, a second mic was found which was then plugged in to the mini-jack port on the laptop provided for use by the presenters. Audacity fired up and was used to record each of the talks.

During the first talk in the Beard room, Christian’s talk on Gstreamer, the levels of sound were really low on the recording. Adam Sweet was compereing the room for the morning in another of his tasteless suits, which sadly looks like becoming an LRL tradition, set off to find something better. About ten minutes later he turned up with a roll of sellotape. So yes, we taped the second microphone to the first in true Hendrix style.

I spent the morning diving in and out of the lightning talks: Gervase Markham’s talk on “How to destroy the Free Software Movement,” was busy and well receieved; Tom Steinberg’s last minute stand-in (Matt, I think) talked about lots of interesting projects designed to improve communication between the government and the people; Mirco Muller demonstrated the lowfat document management system, which looks very, very swish and Belial from Hacker Voice Radio talked about ethical hacking, mainly a break-in on a server to shut down a phishing site. I didn’t really get to see any of the talks from beginning to end though; There’s an inherent problem with having one person sorting out sound in two rooms, when both rooms are scheduled to have talks start at the same time!

For lunch we wandered in to town and found a Subway. It looked big enough to old about eight people, fortunately most the other attendees had gone elsewhere or were too slow off the mark. We had hooked up with Hein-Pieter van Braam, known as HP to his friends and TMM on IRC and were sitting munching in the rather cellar-like down-stairs eating area with another guy who was playing with a Nintendo DS whilst munching on his sandwich. He showed Alan how he had Linux running on it, could run a few commands and could start X. Alan has one of these things (or something similar, I’m not very in to handhelds) so was interested to hear how much RAM and ROM was used in running Linux. HP asked at one point, “Is there a Linux port for the DS already?” “Yeah,” the guy replied, “I wrote it.” Turns out this guy was Malcolm “pepsiman” Parsons who was due to give a talk on Linux on the DS, during the Hour of Power the following day. This is one of the things I love about the FLOSS community: You really never know who you are rubbing shoulders with and find yourself getting in to conversations with people who turn out to be someone whose project you’ve read about or have used.

I had arranged to travel up to Wolverhampton with a bunch of reprobates from HantsLUG, the co-ordination of which involved two cars, two drivers, two hotels, three pizzas and one GPS system. Setting off after our respective work days ended, we converged on Farnborough, the staging point for our journey. We fuelled up on pizza in Alan’s garden (with the lovely and ever-patient Clare) before setting off. Alan was driving, I was in the passenger seat on the pretext of operating the TomTom.

The two and a half hour journey from Alan’s house was consumed with the Geek Car Number Plate Game. Usually players work separately, in different vehicles, over a period of weeks. This rather special game was given extra tension by all the players being in the same car. Comeptition for the available plates was fierce. Alan and I had a strategic advantage, being sat in the front of the car, although my game suffered because I was keeping score too. Laura and Adrian both got a few, but not enough. Adrian got the short straw by sitting in middle of the back seat of the car. Alan did very well at first, getting most of easier two letter acronyms. But Hugo displayed his knowledge of computing minutae by getting lots of very obscure three letter acronyms, allowing him to overtake Alan’s score with fewer “hits”. I should probably go through Hugo’s entries to make sure he wasn’t just making them up.

I stayed at the Quality Inn last year. It’s nothing special, but serves the purpose of somewhere to crash between Linux-y activities. The public areas of the Quality Inn are perfectly acceptable and the staff were friendly. The hotel was busy at the weekend, and we were put in the Business Centre, a short walk from the main hotel. The room wasn’t up to the same standard as the rest of the hotel though. Nothing broken, but it was in need of redecoration. The bathroom fittings were old and discoloured. It took two minutes to find the bathroom, which was hidden in the end compartment of the wardrobe. Really. Ruth Watson would have something to say about that.

On getting to the hotel, we bumped into Christian Schaller of Fluendo at reception. I was sat next to Christian at the curry after last year’s LRL and it was nice be recognised! In the bar at the hotel was Seb Payne, busily uploading photos from his Mac over the hotel’s wireless connection. Poor lad, he doesn’t look anywhere near old enough to get in to pubs, so was stuck back at the hotel whilst most other people were off in town.

Last year I travelled up to Wolverhampton on the morning of the event, so I missed the Friday night frivolities at the Hog’s Head pub. This year though we had arranged the schedule to be there in time for a drink or two before the pub shut. A taxi ride from the Quality cost about £4 and meant that Alan could drink. Having promised Clare he’d have his mobile on him all weekend, he went and left it in the hotel, so left a very sheepish message for her using Adrian’s mobile. The LUG Radio community had arranged to meet in the Beer Garden at the rear of the pub so we dived through the bar to the doors at the back. The garden wasn’t really a garden, more of a yard. It had four walls and something approaching a roof. Still, it is a city centre pub so perhaps that’s the closest to a garden they could manage. The yard was doing a good job of penning in about 60 geeks in differing states of inebriation. We met up with Hugo and Adrian, who were staying at a different (closer) hotel. They had already found Phil Clarke and Andy Smith in amongst the crowd. Phil I’d met briefly at a HantsLUG meeting a year or so ago. Although I’ve talked to Andy a fair bit on IRC, we’ve never met, despite us both being at FOSDEM in Feburary 2005. Both really nice guys, and both staying at the Quality.

The #lugradio crew had arranged to wear Jono-style beards in the pub that evening, mainly because Jono had promised to shave off his beard if £400 was raised for Amnesty during the weekend. The idea was that the presenters who had been preparing the venue would arrive at the pub to an assortment of dodgy looking beards, rather like Jono’s. Although we were too late to the pub for this part of the evening, I had ordered five beards from Joker’s Masquerade and we duly donned them:

I went over to say hello to the #lugradio people, including Neuro, mrben, Adam “drinky” Sweet as well as the presenters. Jono loved the beard (there are some photos that will surface in due course, I’m sure). Aq was very amused that these beards were sold as Arab’s beards, especially given that Jono was called Ali Baba by a waiter, as reported during a recent GUADEC show.

For my money Adam “drinky” Sweet was the most hammered. I had a bizarre conversation with him that went something like this:

Me: Hi Adam, I’m Tony, we were at the paintballing together last year.
Adam: Oh, hi, how are you doing?
Me: Not too bad, I’m helping out with the audio tomorrow.
Adam: Nice beard.
Me: Yeah, I figured if I put this on and tell people I’m Jono’s cousin I stand a better chance of pulling.
Adam: So, are you in to Linux at all then? Or has Jono just asked you to help out?
Me: (quietly) I’m not really Jono’s cousin.

After this Matthew Revell came over to say hello, in a rather fetching pink and brown rugby shirt. Matt was really friendly and it was great to meet him again. Ade came and joined our table for a chat later, even sampling one of our beards.

We shared a taxi back to the hotel with Andy and Phil. We all stopped for another drink at the hotel, where at one point I got in to conversation with some very drunk hairy hacker types who obviously didn’t think I looked geeky. (“There’s a big open source conference going on this weekend,” I was told. “I know,” I replied.) To be honest, I’m quite glad I don’t seem to appear geeky. Got to bed not long before 3am, having had a great evening and looking forward to the following day.

OK, so this isn’t the promised review of LRL either. I’m still knackered, and spent 2 hours getting home from work (normally 15 minutes). I’ve spent most of the evening responding to people who signed my GPG key at LRL and issuing them with the relevant challenge to respond to. Oh, and I have iFolder and a card reader to get working too…

I was at LUG Radio Live at the weekend, and it rocked. There’s so much to write about, but I’ve not had time to write anything this evening. (In fact, I wanted to be in bed by 9pm to catch up on my missed sleep – hah!) I will write about the weekend tomorrow, I hope, but in the absence of any actual words, please check out my gallery a.k.a. evidence for the prosecution. Some of these photos are also on the LRL flickr page thing.

PC World adverts are generally very annoying. But their current advert includes the bland sales drone flogging an HP laptop with the phrase, “It has a huge one gigabyte RAM memory.” Just another thoughtless tautology from an ill-informed writer, I suppose. And another reason not to shop there.

My OpenGPG card arrived, looking very much as it did on the website:

These cards are basically the same as the ones issued to FSFE members, but have a different design printed on them. As far as I can tell, they are produced by the same company that produces the FSFE ones, although the specification for the cards is an open one that anyone can implement. The card was shipped with a printed copy of the GnuPG Smartcard HOWTO:

The printed copy includes instructions like “download the latest version and apply this patch.” Given that underlined words on a sheet of paper are not clickable hyperlinks, this presents a problem for people trying to follow the instructions. Fortunately most people won’t have to worry about compiling their own verison of GnuPG as most modern distros should have a new enough version with the appropriate support.

So, I plugged my new card into the reader, and the card responded to the ATR command from opensc-tool. However, the basic test to check that GPG is talking to the card wouldn’t work: gpg --card-status just returned errors. More googling showed up posts on various mailing lists that showed that GnuPG and OpenCT didn’t work together. Some posts suggested this was due to the OpenCT API changing rapidly and suggested that it might be implemented in the future. I can’t tell whether this situation has changed since the posts were made, but it seemed safer to assume it hadn’t. So although I had the correct card, a supported card reader and some software to query the card, the application I want to actually use with the card wouldn’t work with the software I was using. So I turned my attention to PC/SC.

Whilst looking into PC/SC in more detail, I started to get how the various bits of smartcard software available interacted. PC/SC is a standard originally developed by Microsoft, but now maintained by an external group of interested parties. It defines a standard API for applications to communicate with smartcard readers. Any card reader supporting the PC/SC can be addressed by applications that support PC/SC. At least, that’s the theory. On Linux, PC/SC is implemented by the PCSCLite project, from the MUSCLE (Movement for the Use of SmartCards in a Linux Environment) project.

PCSCLite runs as a daemon process (pcscd) in the background. It detects USB readers as they are connected to the PC, but requires manual configuration in /etc/reader.conf for serial devices. (On Debian and Ubuntu systems, the /etc/reader.conf file is managed by the system using entries in /etc/reader.conf.d, but the principal is the same.) The daemon is contained in the pcscd package, and the userspace tools in the pcsc-tools package.

So, this was roughly the revelation point for me. PCSCLite and OpenCT are roughly equivalent pieces of software. They both provide drivers for smartcard readers and an API through which other applications can talk to the smartcards. Readers supported by one may not be supported by the other, although there seems to be quite a few that are supported by both. The principle difference seems to be that PCSCLite provides a PC/SC interface, whilst OpenCT is designed around a PKCS interface. (Some webpages seems to suggest that OpenCT can be made to provide a PC/SC interface for applications to use, however I couldn’t find any concrete documentation on this.) Another difference is that openct has the drivers for supported readers built into it, whereas pcscd requires an external driver to be loaded for each reader.

I couldn’t use GPG to communicate with my OpenGPG card through OpenCT, because GPG itself couldn’t “talk” to OpenCT. OpenSC, however, can talk to OpenCT. The GnuPG application itself supports OpenGPG cards directly, so doesn’t need an interpreter to talk to them, only the ability to address the card through the reader. GnuPG supports PC/SC, not OpenCT.

As mentioned earlier, OpenSC supports three different backends, including the project’s own backend, OpenCT and the wider-spread PC/SC. The third is CT-API, which seems to be popular in German projects, but hasn’t really featured in my investigations. OpenSC supports PKCS#15 cards (a Public Key Cryptography “standard”), including OpenGPG cards. This means that with a working PC/SC backend, I should be able to use opensc-tool to query the card, and applications written to use the OpenSC interface should be able to do the same.

With the pcscd and pcsc-tools packages installed, I just needed to ensure there was a driver for my card reader available to pcscd. Some poking around showed that there are relatively few drivers available in Debian and Ubuntu, most of which are in their own package. There are quite a few smartcard reader drivers available from the MUSCLE people, including the CCID driver, which seems to be a generic driver for quite a few smartcard readers. libccid doesn’t support the ACR30, but there is a driver for the ACR30 available for download from the manufacturer. I downloaded the USB version, although there is also one for serial devices and a driver for what are labelled as propriatory drivers.

The driver Linux driver as downloaded contains a README crediting David Corcoran from linuxnet.com as the author and pointing people to the LICENCE file for a copy of the licence. This is a method commonly used by people distributing software under the GPL although the LICENCE file is missing from the zip file. The headers of some of the files point to a COPYING file, also missing. None of them include references to the GPL as is usual with GPL software and recommended by the GNU project. This is possibly why the driver hasn’t been redistributed by distros. I’ve e-mailed David Corcoran to see if he can help clear it up.

Fortunately compiling and installing the driver was very simple. I already have a build environment installed on my PC (the build-essential package) so I unpacked the zip file, changed into the newly created directory containing the source code and ran make. I then copied the driver to the correct location (which differs from that in the Makefile), /usr/lib/pcsc/drivers. See the wiki page for more details on the compilation, such as they are.

In order to test the device, I stopped the running pcscd instance with an /etc/init.d/pcscd stop command. I then ran pcscd from a terminal using pcscd --apdu --foreground, which keeps the process active on the terminal and shows lots of debugging information. The output on the console (again, details on the wiki page) showed pcscd starting, the card reader being detected when it was connected and the card ATR being shown when the card was inserted. The ATR of the card matched the ATR of the OpenGPG card type listed in /usr/lib/pcsc/smartcard_list.txt. I was be able to unplug and replug the reader as often as I like, and remove and reinsert the card repeatedly. pcscd coped fine.

So, for the acid test. The command gpg --card-status is used to show basic details about an OpenGPG card and is a recommended way of testing whether GnuPG can talk to your OpenGPG smartcard. I ran it and the output matched the sample output in the HOWTO. Time to finally start doing some GPG stuff with this card!

To be continued…