I've been thinking about this for the last week or two, and I've decided to run for the Fedora Board elections. I felt a bit reluctant to do this since I'm a Red Hat employee and I'd like to see more non-RHers on the Board and other committees, but in the end I decided it was the best way to move forward with my idea.

Ah, yes, my idea. This will be a single-issue campaign. As I vaguely pondered back in September, I think it would be of value to Fedora to take a good look at the Fedora release process. I'd like to get elected to the Board so I can form a working group to look at the release process and various alternatives for revising it, and report back to the Board on those ideas. I think it's worth taking a close look at whether the Fedora release process, which is essentially the same process Red Hat had in 1999 - we release an operating system every six months - is the best one to aid the goals of the Fedora project as we now see them.

I don't claim to have all the answers here, and I can see a scenario where I go and form the working group and we go away and look at all the angles and conclude that actually, the current process is indeed the best option. I don't think that would be a waste of time, either. But I'd certainly like to take a close look at moving to a longer and more flexible release cycle, or moving to a rolling release model, and see if those options would help us to better achieve Fedora's goals.

So for now I'd like to suggest that if you agree with me, well, Vote For Me! As far as other Board business goes I've set out my stall in the nominations page, but broadly I would try to work with the other Board members to achieve consensus and will support and work to implement the Board's overall view on issues even where I may personally disagree. To put it simply, I'll try to be a 'good Board citizen' as I've always tried to be a good Fedora citizen.

If anyone on the current Board, or any of the other candidates for the open positions, thinks my working group idea is a good one and would like to move forward with it themselves, that would be another good way to move forward, and I'd seriously consider dropping my Board candidacy if that's the case. So if you're that Board member / candidate, please drop me a line by IRC or email and we can discuss it.

Thanks!

(The hot dogs? Why, as a Board member I would of course work to forward the cause of the Glorious Hot Dog in all things.)

So we just got done signing off on the gold images for Fedora 14. I'm amazingly proud of the whole little release management group - development (especially Anaconda team, who were awesome), release engineering, and QA teams: we had an unbelievably smooth ride through the Final validation testing stage. Unprecedented in the annals of Fedora history, we span one publicly-announced Test Compose (TC) build (there were five unannounced ones, but they were just to test small fixes which we needed an image compose to verify) and exactly one Release Candidate (RC) build, which was the build signed off as Gold today. We have never needed just one candidate build to get a release right before.

Both the composes happened right on time (thank you release engineering and development groups!) and then the QA group swung into action for testing. You can see the results of the extensive RC1 validation testing in the installation test matrix and the desktop test matrix, confirming that the RC1 build passes all of our critical validation testing. We have test results not just from Red Hat employees and interns in the Fedora QA department, but from RH employees in other departments (shout out to jreznik for KDE/Plasma Desktop testing!) and, most importantly, from multiple community volunteers: huge thanks to Andre Robatino, James Cassell, Robert Spanton, David Ramsey, John Watzke, Timothy Davis, A.J. Werkman, Sandro Mathys, Garrett Holmstrom, G. Wolfe Woodbury, Juan Pablo Daza P., Masami Ichikawa, "kevmif", and "Adams" (Wiki username is all I have for those two :>) for contributing to the TC and/or RC testing matrices. Special thanks to Andre and Ichikawa-san for the huge amount of testing they both did across TC and RC composes. Also thanks to everyone who posted feedback on the composes to the test mailing list or the Fedora forums. For me personally, one of the coolest things about the Fedora 14 cycle was sitting back and watching the results roll in to the RC1 desktop testing matrix: I didn't run a single RC1 desktop test myself and that was really awesome, after having to do almost all the tests for the first few runs of that matrix myself.

Of course, Murphy will do his best to make sure it turns out we've all somehow overlooked some major issue that the hyenas^H^H^H^H^H^H tenacious reporters of the Linux press will leap on gleefully, but I'm really happy with the way we've taken advantage of the awesome Fedora community to try as hard as possible to catch all the big ones. I'm sure F14 will be a release to be proud of.

Finally, huge thanks to John Poelstra who worked really hard this cycle to try and drive the entire release process, across the entire Fedora community, to make everything happen on time. He's a big reason why everything ran so smoothly.

I think I've mentioned before that I find the Reg sadly diminished from its heyday, but it still nails things perfectly sometimes. From this story on Street View, featuring another gobsmacking Eric Schmidt quote...

"Earlier this month, Schmidt said that Google's policy is "to get right up to the creepy line and not cross it" — which was just the most prominent example of Schmidt getting right up to the creepy line and leaping across it with abandon."

I have to admit, you get a much better class of sarcastic comment on Boing Boing than you do on most web sites.

My current favourite - "If we don't allow this, prehistoric people will retroactively stop building monuments. Then where will we be?"

This is a bit of a jarring post on a blog that usually just writes about tech geek stuff, I know, and probably not many of the people who'd usually see my stuff are interested in it. But I'm mostly writing it to support the tour and to give Steve something to link to. With that said...:)

I spent most of last Thursday at the opening night of the second Next Music From Tokyo tour, at the Biltmore Cabaret in Vancouver. I went to the first tour too, back in May, and loved it so much I offered the promoter, Steven Tanaka, any help I could give with the next one. So with that in mind, this is more of a fan's recap than a hardened music journo's unbiased review. :) In the end I printed off a few hundred flyers for the event and handed them out at arcades and Sakura Media and stuff, to try and help with attendance, and then on the day I and a couple of others showed up early for the show and bought pizza and water and cookies and opened curtains and tried to help get the crowd going and stuff. But mostly, we were just enjoying the whole event.

The show is a mini-festival of bands mostly from the amazing but heavily underground Tokyo indie scene. A lot of these bands are kids from high school to university age doing 'the band thing' for a while without many professional aspirations, and the scene as a whole has a heavy independent/DIY/non-commercial ethic, so it's pretty rare to see the bands outside of Japan and few of them go on to become professional acts on major labels even in Japan. So this tour is a really rare opportunity to see bands, and a whole music scene, that would normally be pretty inaccessible in North America. Steven runs the tours because he's a huge fan of the scene and thought it would be cool, and does it at a major financial loss, so the whole thing is an incredibly rare enterprise - there's basically no commercially viable way to have five niche bands tour a foreign country playing shows to crowds in the hundreds for $10-$15 a head, so we're incredibly lucky to have the tours at all.

With all that said, the show and the whole experience was just as cool as the first one. Before the show the bands were obviously pretty tired but at the same time excited about the tour and just like on the first show were really happy to chat with fans (despite the language barrier!). I saw a lot of people who came to the shows chatting away with the band members and getting t-shirts and CDs and flyers signed, which made the whole thing have a really nice friendly feel and is one of the things I loved about the first show.

Which just leaves the performances. The night opened up with Susquatch, who are described on the NMFT page as playing "smooth jazzy melodic compositions furnished with aggressive drumming and intricate guitar work to add a forceful edge", which is as good a description as any, I guess. One of the things that the tour really showed is the incredibly high standard of musicianship in the scene. I've been to local indie nights at places like the Railway, where the same sort of Vancouver bands - small groups composed of local kids doing it for the love of music - play, and the comparison in most cases is pretty embarrassing. Most of the bands at those events tend to be somewhere between talented but under-rehearsed and sloppy - flubbing intros, missing notes, losing time - and just flat-out amateurs; one particular highlight I recall is a band composed of a guitarist/vocalist who knew one chord and thought he knew another, and a drummer who owned a snare, a tom and a cymbal and didn't have the slightest idea how to play any of them. From the evidence of the NMFT shows I've seen so far, this ain't how it works in Tokyo. Even the smallest bands on the tours showed an impressive level of professionalism - they all had a really amazing level of musical proficiency and without exception their performances are very tight, they play in key and in time and I don't remember any of them flubbing anything. Susquatch are a compelling case in point; they aren't necessarily very musically original but they're extremely committed and skilful and they have a very clean stage show. The audience was obviously appreciating the set but weren't completely won over until the last section of the set, when the singer and guitarist Oshikiri Kenta stumbled through a weirdly endearing monologue about how they weren't really as serious as his previous monologue had indicated but were in fact a very 'kinky' band. At first the audience figured he'd made a vocabulary mistake but his story about how cute the waitresses at Banana Leaf were made it clear that he meant more or less what he said! After that they launched into Spin The Words, a great number with an incredibly insistent drum part (which must be tiring as hell to play) which had the crowd bouncing and cheering wildly and kept them fiercely on the band's side through the closer. Nakano Maki's incredibly precise drumming was a highlight of the whole set and a great treat for drumming geeks.

Susquatch were followed by sgt., a pretty long-standing post-rock band who have worked their way to the kind of status a band like the Cowboy Junkies has in North America - not a celebrity band, but a solid professional act with a strong relationship with a committed fan base. Their 'hook' is the prominent place in the band accorded to the violinist Narui Mikiko, whose parts essentially take the place of vocals, giving the band a pretty unique voice. Post-rock is as post-rock does, and if you're not a fan you're not likely to get into it (if you're not sure what 'post-rock' is, think Sigur Ros, Godspeed You Black Emperor!, and Mono - another prominent Japanese post-rock group), but I found them a really impressive band and thought the prominent placing of the violinist really works. Just like Susquatch, they turned in an incredibly high standard of performance.

Third up were Uhnellys, a bit of a departure for NMFT - they were the first band who really departed from the good old guitar/bass/drums indie template. Uhnellys are a funk/hip-hop duo with Midi on drums and Kim playing bass, baritone guitar (an awesome and massively under-used instrument) and trumpet (yes, trumpet) - he tends to construct a loop at the start of each song using bass or baritone and an array of pedals, and use the trumpet for tricks during the song. It's a pretty cool sound and they sell it really well, with great commitment - obviously most of the crowd didn't have a clue what Kim was singing/saying, but it didn't really matter, the cadence and flow and energy is the key to most hip-hop rather than the lyrics, and Kim comes off as a seasoned performer with impressive technical ability. I'm not the world's biggest hip-hop fan, but I really enjoyed the set, and Midi has a great repertoire of standard hip-hop beats which she peeled off seemingly effortlessly and bang on time.

The final band, and definitely the unofficial headliners, were Mass Of The Fermenting Dregs - note their English website is a bit out of date - who are moderately big news in Japan, having fairly recently signed a major label contract. You'd definitely have to pay more than $10 to see them in Tokyo. They started off as an all-female three piece but the original drummer left a few years back and they now have a male drummer. What they play is basically fairly straight ahead high-energy indie punk rock, but it's damn good indie punk rock and their live set is incredible. It starts with a darkened stage and the standard Japanese count-off 'sei, no...' and then doesn't let up pounding until the close. Halfway through the set Steven successfully initiated a moshpit which didn't let up until the end, and crowdsurfing ensued by the last number. The crowd was won over enough to keep cheering through a fairly lengthy encore break (another first for the NMFT tour, I think) after which the band came back out and killed Rat for an encore. It's hard to explain exactly how the band engages the crowd, but they do - Mishimoto Natsuko plays bass and sings from the middle of the stage, looking like she's in her own world half the time, but also in bare feet, which is a good gimmick. Ishimoto Chiemi plays guitar with her head down most of the time, but has the perfect haircut to look cool while doing it (and also for headbanging), and she scored major points by playing large parts of several songs from the front of the crowd and eventually from the middle with the whole audience surrounding her. Whatever the magic is, it works. I've never heard a fairly small Thursday night crowd sound that loud.

The small crowd is the main reason I'm posting this: Steven was pretty disappointed with the turnout and it's making him wonder whether to carry on running the tours. Given the amount of money he loses on each one, you can hardly blame him. He was also disappointed at the lack of reviews and so on of the first tour, so I'm doing my bit to rectify that for the second tour, even though my blog isn't typically the venue for this kind of thing :). Probably the number of Canadian music geeks reading this isn't huge, but if you are - or you did a search for NMFT and came across this - you should definitely come out and support any future NMFT events, they are amazing and you won't regret it; the music is guaranteed to be good and you get to see the bands in very intimate venues for ridiculously cheap prices, and hang out with an awesome crowd of fans and musicians. I've been to literally hundreds of gigs and festivals, with some huge and incredible live bands, and both NMFT shows still rank right at the top of my best and most memorable concerts ever.

As I write this there's one show left on the current tour, in Montreal tomorrow night (Tuesday 2010-10-19) at Club Lambi at 8:30pm, with local band Devil Eyes joining the NMFT bands. Tickets are $10 on the door. I'm no Montrealer, but I'd bet there isn't a damn thing better to do on a Tuesday night in Montreal for $10, so if you're there, get your ass to the show - you won't regret it. (And tell Steven I sent you :>) After that, Steven has/had the third tour tentatively pencilled in for May next year.

This is a call for contributions to the validation testing for Fedora 14 Final.

Validation testing is a set of planned tests to try and ensure the release meets the basic release criteria. We have installation and desktop validation tests.

Testing is pretty easy: you look at the results page, there's a table of tests and results. You find a test with no result, do the test (the test page will have instructions), and fill in the result. See, I said it was easy. :) Some of the installation tests require specific hardware, but many of them don't, and none of the desktop testing does.

Right now we're testing the TC1 images - that's Test Compose 1, it's the first pre-release compose which lets us identify major issues ahead of the release candidates. The validation test result pages are here:

Installation validation Desktop validation

and you can get the images here:

DVD / netinst

So, here's some miscellaneous stuff which is popping into my head!

First up, Poulsbo / GMA 500 (I really ought to make a category...) on Fedora 14. Short story - it works, now, just as F11 / F12 / F13. Long story - follow the instructions here (use the 'Rawhide' fusion repos), with the exception that you're going to have to set up an xorg.conf file manually as all the foobar-config-display tools in Fedora 14 are currently broken due to this bug. The driver packages use livna-config-display to set up the X config file, so that doesn't work (same bug exists with all drivers in RPM Fusion at present). If you pull over an xorg.conf from any previous Fedora / psb setup, it ought to work.

I saw The Social Network the other day. I agree with a lot of the geek criticism of the strict accuracy of the movie, but as a movie it's stunning. I thought it was really short when I was walking out of the theatre, then checked my watch and realized two hours had gone by without me noticing. Even Justin Timberlake was good (okay, okay, actually I love Justin, are you happy now?) I'm guessing my reaction to the Harvard bits was atypical though: it was, roughly, 'aww, isn't it cute that they try so hard'. :) Has anyone figured out what distro Zuckerberg's systems run? I'm guessing it could well be Mandrake, as it's KDE 3.5. Toss up between Mandrake and SUSE, probably, if they're going for historical accuracy, given the time frame and the use of KDE rather than GNOME...

I want new monitors for my desktop. The only problem is that the monitors I want basically don't exist. The thing is that my laptops have spoiled me - they have high-resolution panels with DPIs of 221 and 140 (small cookie for anyone who can name them). Now, that kind of display with a sane OS (that is, Linux or OS X) with the DPI setting correct gives seriously lovely font rendering with such a panel. Only problem is, now I want that for my desktop, and it's pretty much impossible. The best DPI you can get in a mainstream consumer display is just over 100, with 1920x1080 / 1920x1200 21.5" panels. There are a few extremely exotic displays with much higher DPI than this, but as far as I can tell, they're all no longer manufactured and cost ridiculous amounts of money when they were. The DPI calculator site lists most of them as 'noteworthy' monitors, but you just can't buy the damn things. All I want is, say, WQXGA (2560x1600) at a 23-24" panel size. Hell, I'd even settle for 1920x1200 at 20", which would at least be 113dpi, which is better than a kick in the head. But nope, there's just nothing. It seems like displays are all built around a 90-100dpi range to deal with broken operating systems (cough, Windows, cough) which are terrible at dealing with higher resolutions. Anyone know of anything I'm missing here? I want something under 25" with a DPI in the 120-200 range which costs in the three figures. Why this is possible in laptops but no-one will do it in a desktop display, I've no idea. There's gotta be at least a niche market for it.

Vaguely a propos of the above, I've switched to using gnome-shell on my desktop and main laptop for a bit, just to see how it goes. It seems like it still doesn't do anything with the window hint which, in GNOME 2 / metacity, makes a window's task list entry pulsate; this is very frustrating because I rely on this to know when someone's messaged me on IRC. Still, I can work around this with xchat-gnome's notification plugin, and get an icon in the notification tray when someone messages me. It's not as noticeable as a pulsating task list entry, but it's better than nothing. Aside from that (and not having a Windows key on my desktop keyboard, heh) I'm quite enjoying it so far. Looking forward to being able to switch to Fedora 15 and get all the latest developments (F14's gnome-shell is kind of frozen a bit back in time). I've also switched my desktop to using the Droid fonts I've been using on my laptop for a while now; I finally decided that they do look a bit nicer than Bitstream / Deja Vu (though I'm not sure they have the character set coverage yet).

We're entering the final stretch for Fedora 14 now, which means lots and lots of blocker bug tracking. The blocker list is looking really quite acceptable right now - that looks quite long right now but all the ones in ON_QA and MODIFIED are (hopefully) basically fixed, the ones to worry about are NEW and ASSIGNED, which is a nice short list. Of course, final desktop and install validation of the TC1 images might lengthen the list again. I've been focussing my efforts on this bug, which is an icky regression between kernel 2.6.33 and 2.6.34 in support for some fairly common Intel graphics chipsetted Dell systems; I'm trying to work with Intel's Jesse Barnes to squeak a fix for this in under the wire, but it'll be tight.

Finally, administrivia: I know you probably can't leave comments right now. The RPX plugin I used to use for OpenID authentication now requires me to have Wordpress's 'anyone can register' setting enabled, which it previously didn't require, and I'm not cool with that: it means I get 50 spam registrations per day which I just Do Not Want. So I tried the 'official' OpenID plugin, but it doesn't seem to work - for me and the one person I got to test it, it seems to work but never actually logs you in. I'll try and figure something out, but for right now, I am just leaving it this way because I really don't want to deal with BS spam registrations / comments. Sorry about that. Please do email me your pearls of wisdom, though.

I did a bit more of that old geek standby today, tinkering with my home network. I've been connecting my desktop to the router via wifi ever since moving my desk to the living room because otherwise I'd have to trail ethernet across the floor (or route it around the ceiling), but I've just never got great speeds out of my wireless setup - even after buying an N router I never managed to get the wireless going very fast. Plus it just seems nicer to use wired connections for static systems. I considered getting a contractor to run ethernet into the living room but it's kind of expensive and overkill for a small apartment. So today I went and got a D-Link powerline ethernet kit, and what can I say, it works. It's pretty simple - take the adapters out of the box, plug one in near your PC and one near the router, run cables to the PC and the router, and you're done, it just works. Goes about 8x faster than the wireless connection did. I'm happy with it, and it's a lot cheaper and less messy than running ethernet through the walls. So...powerline ethernet, thumbs up!

So I just waded through the latest Peter F. Hamilton brick (excellent, by the way) and wanted to move onto the newish Terry Pratchett, I Shall Wear Midnight. I read everything on my ebook reader, these days. So off I went merrily looking for a copy.

Oh, boy. I'm fairly sure this is recent, I'm sure I've bought stuff from WH Smith's and Waterstone's ebook stores while I was in Canada before, but it appears the gigantic, moronic stupidity that is territory restriction has now reared its ugly head in the ebook publishing world.

So I found copies of the book - that's the exact same text in electronic format, it's just a tiny little text file with some formatting metadata and DRM crap - in various stores. A couple from the UK - WH Smith's, Waterstones - and several from the US, including Books On Board and Diesel. Remember, same text in an entirely electronic format. The UK stores both cost more than 50% more than the US stores (which all seem to cost $9.99), which is just ridiculous. It's the same damn book! I don't care if you blame taxes, import duties, unions or cosmic bloody radiation, that's insane.

So I decide, obviously, to buy it from one of the US stores. No dice - every one I've tried refuses to sell the book unless you have a US billing address. Finally I give up and go with the ridiculously pricey British ones - no dice. Need a UK billing address now. (Not that anyone with their head screwed on would willingly choose UK prices if they could avoid them).

So I go out and search explicitly for a Canadian ebook edition of the book. What do you know - there ain't one. It just hasn't been 'published'. Canada's effectively a monopoly when it comes to bookstores, Chapters / Indigo being the only game in town, and they don't have one. I found a Google search link on Indigo which mentions an ebook edition; if you go to the current version of that page, the link has been disappeared; if you go to the Google cached version and click on the link, you get to a deleted page.

So I just want to buy the damn book, but I am not allowed to buy it from the US or the UK, and I can't buy it from Canada because they haven't 'published' it here. Again, it's a fricking text file. Why does this have to be hedged around with cartel-y nonsense? What do you, the publishers, honestly expect me to do here? Wait, and then pay whatever you decide to charge the captive Canadian market, never mind it being available cheaper to anyone who happens to have an American credit card? Buy a paper copy? Or get it from Bittorrent? I'm not telling you what I'll do, but I'll give you one guess what most people will do. For christ's sake, at least try and pretend you're running a business here. I just want to buy your product!

Apologies for the lateness of this wrap-up - I didn't want to interfere with the Graphics Test Week posts on my blog! A Graphics Test Week wrap-up should be coming tomorrow.

Indeed like many others at RH and in the Fedora community, I was at FUDCon Zurich last month. First I'd like to say thanks again to Sandro and Marcus for their amazing job of organizing the event - everything went off smoothly and they even contrived to have enough money left over in the budget to beat out the legendary Czech alcohol tolerance at FUDPub. Incredible!

Note that all the pictures in this post are courtesy of Máirín, as I stupidly forgot my camera. Thanks Mo!

I arrived early the day before the conference, got myself set up at the hotel (which was excellent, and had a scarily fast internet connection), and did some Beta release process work. After a while, I heard that Jared, Jesse, Spot and Mo were at the hotel, so we got together and went out for fondue. Really very nice fondue. Bert happened by the restaurant while we were eating and joined us, and we chatted very seriously and exclusively about important, work-related topics. There were definitely no anecdotes about NASA. (That was for the benefit of Jared's expense report).

After the fondue, we went to the venue and got in the way of - I mean, assisted in - setting up the hall for FROSCamp. I started off plugging in routers but my QA genes took over and I wound up testing connectivity by running round the hall and plugging my laptop into each one in turn...

After that we headed back to the hotel. The others made it successfully back to their rooms, but I was snared in the hotel bar by Jeroen and Peter Robinson (and also Xavier and a few others, but once you have Jeroen, me and Peter at a table, no-one else gets a word in edgeways for four hours). Peter being Australian and me being an adopted Canadian, whenever we meet up we always wind up having a beer or eight and sticking the knife into the whinging Poms (that's the English, for my American readers...), and Jeroen's always pleased to have an audience, so I didn't manage to stagger up to my room to write my talk for the next morning until several hours and several beers later.

This of course had absolutely no impact on my stunningly professional and captivating delivery of my talk the next morning. Actually, I didn't change the talk much from the last time I gave it, at FUDCon Toronto; I should really give it a more thorough rewrite and make it look snazzier, I was a bit disappointed at how many things I remembered during the talk I should be covering and wasn't. But it seemed like most people watching managed to stay awake, I kept it short, and some people signed up as proven testers after it was finished, so I don't think it was a complete disaster! The talk is a general overview of the Fedora QA team's tasks and processes and how to get involved in them, I certainly encourage anyone who's involved in QA and going to an event to propose a similar talk, and please use any material from my slides that might be useful.

I read that some people felt the talks were a bit disappointing, but I was pretty happy with the ones I went to. On the first day I dropped in on Spot's packaging workshop, which I thought he did really well, and which seemed popular. I also sat in on the release engineering SOP workshop, and helped Jesse out with a few things, but mostly sat there and worked on Beta testing.

On the second day I went to Kamil's AutoQA overview talk, which was impressively detailed and smoothly run. It sparked some passionate audience debate and feedback, too, which always means you're on the right track! Kamil did a great job of explaining both how the AutoQA project is structured and being developed, and the value it will provide for Fedora. I also went to Jesse's talk on the new git-based RCS for Fedora packages, which covered a lot of ground I'd more or less come to terms with already but also gave me a few useful new tips, and must have been invaluable for anyone who hadn't figured out the new system yet. I also saw Spot's talk on legal issues surrounding Fedora, especially Fedora packaging, which was informative and amusing in the way only Spot can manage. I highly recommend anyone involved in packaging go to one of Spot's presentations, or just corner Spot and let him tell you legal anecdotes for a few hours; you'll come out a lot better informed and a lot more cautious about licensing, patent and trademark issues. I also went to Jared's talk on VoIP, but wound up working through most of it! I spent most of the weekend juggling conference stuff with the F14 Beta validation process, you'll probably have noticed by now.

FUDPub was the evening of the second day, and it went very well - we went to a local pizza restaurant, but didn't get pizza. Instead the vegetarians got a sort of cheesy-vegetabley-potato-ey bake-y thing, which sounded worryingly vague but turned out to be delicious. I sat with most of the North American RHers, Peter, and Xavier, and we had a good time reminiscing. Or rather, the others reminisced and I pretended to have been around at the time...

On the final workshop day, I ran a workshop to do some desktop validation testing on the Beta - RC2 at the time. The turnout wasn't huge, but those who showed up - especially the awesome Christoph Wickert - really helped check out the state of the non-GNOME desktops. We managed to turn up some significant bugs in the LXDE and Xfce desktops which we eventually resolved in time for the Beta release, which was a great result.

Of the other workshops, I went to Peter's workshop on Fedora Mini and Sugar, which was very productive in terms of getting a bunch of people with an interest in Sugar, Moblin/Meego and the importance of getting Fedora onto non-boring-Intel-desktop-y-machines in general together and with a vaguely cohesive sense of purpose. We also renamed Fedora Mini to Fedora Mobility during the workshop, which was cool. I'm hoping to be able to help with making sure Sugar is in great condition for Fedora 14, and getting Meego in shape - hopefully we'll have something at least usable in Fedora 14, but we're mainly targeting Fedora 15 given the timeframes involved. Personally I think Meego in particular is much more significant than a lot of people realize yet, and I really hope we get it to the state where we can have a 'showcase platform' for Fedora 15 - some kind of handset or tablet - where you can stick in a Fedora USB stick (or SD card or whatever) and have it boot up into a fully-working, shiny, Fedora-ified Meego environment. That would be a great thing to be able to show people. It would also be good to let people have the Meego netbook UX right out of the box with Fedora.

I also caught half of Máirín's incredible Awesome Art Stuff workshop, which is basically two hours of Máirín demonstrating how to do insanely cool stuff in the GIMP (and other free art tools). If you've never been to one of these, you really really really should. No matter who you are. If you've ever dabbled in the GIMP even vaguely, or even wanted to but been too scared, you will learn a ton of stuff (and have even more stuff whiz by too fast to take in yet). Even if you've never touched the GIMP and never want to, it functions as awesome entertainment just to watch a badly-lit mugshot turn into a pristine hackergotchi, or a shiny shiny pony emerge fully-formed from the ether, or a scary-yet-cute dragon guy be dreamed up Live On Stage. It's just great.

Finally I went to Jared's Getting Connected To Fedora Talk workshop, which went off exactly as advertised. Actually I'm already connected to Fedora Talk, but we got everyone else there connected too, and did some test calls. We did find out that Fedora Talk can't really manage video well yet, as there's no protocol that the clients we tried and the Asterisk-powered server end can agree on. That's a shame. It'd be nice to fix it.

So all the formal events that were part of the conference were great, but as usual, probably even more valuable was just being able to meet people and work in real time. Having Hans de Goede and Jesse around in meatspace certainly helped resolve some issues and get all the F14 Beta blockers fixed in time for the Beta to ship, and having Christoph around let us get the desktop spins fixed up properly. Meeting Jeroen and Bert and Xavier and Sandro and dozens of others was great, and just being able to chat face-to-face about various little issues really helps to smooth them out. And I spent most of the event hanging around with Peter - as well as a deep mutual interest in drinking beer and complaining about the English, we talked a lot about Sugar and Meego and I came out with a strong sense of the importance of those projects and an enthusiasm for integrating them into the QA process. I'm hoping to get Sugar into the desktop validation process for Fedora 14 final and Meego integrated for Fedora 15, and I'd be really happy if we could.

I also wound up sharing my hotel room with Dave Crossland, who does font stuff for Google and volunteers on font and design stuff for Fedora, and survives on less sleep than any other human being I've ever met - I was doing eight hours of conference and twelve hours of F14 Beta validation per day, but he still managed to sleep eight hours after me every day and wake up enthusiastic in the morning. (No, don't do the math on that, please). He's also cheery and optimistic and has grand, entrepreneurial plans - it's like he's not really English at all.

The Commonwealth - Peter, me, and Dave in the hotel lobby

So this was another enjoyable and valuable FUDCon. I'll be back at Tempe, hopefully with an all-new remixed QA Introduction 2.0 talk. And a unicorn. Probably just the talk, though.