Greatly exaggerated, no. 81.It was good of Phoronix to clarify their story implying that Nouveau, the open source driver for NVidia graphic cards, was somehow dead or dormant. One thing that had me scratching my head, though — isn’t this open source, guys? I mean, you could just look at the source to see what’s going on. This log sure doesn’t look dead to me! By the way, I have an NVidia video card in my Dell XPS M1330 laptop, and I’m running the pre-F12 version of Rawhide on it. And I can definitely report that the progress between Fedora 11 and Fedora 12 on the Nouveau driver is astonishing! I now have reliable suspend/resume and hibernate/thaw, which was a problem in prior releases.So I know that Fedora 12 “Constantine” stands to be a great release from the perspective of a lot of NVidia users. There’s still a rough spot or two that I know Ben and the other fabulous Nouveau developers are still hard at work polishing in the current feature set — not to mention the work they’re doing to make the driver even better in future releases. The nature of open source is to be constantly changing and improving, and Nouveau is a great example of how that methodology has yielded great results in a relatively short amount of time. This is one dog that is not only still in the hunt, but gaining fast. Nice work, Nouveau guys! |
I hope that other Internet news reporters will take this lesson to heart and check facts underlying their stories. Hard-hitting news is thoroughly appreciated, but needs to be backed up by solid reporting.
UTOSC 2009, Day 2.After this morning’s activities, I spent a little time catching up on emergency email and did a bit of polishing of my slides for my talk later today. At some point I realized my head felt like it was going to explode, which seemed incredibly unfair given the fact that I had maybe a pint and a half at dinner the night before. So Clint and I went to the cafeteria, since he was similarly afflicted, and found some Tylenol, which helped immensely. Over the course of the day, I also played several ultimately unsuccessful rounds of phone tag with people regarding FUDCon matters. I gave my talk on the Fedora/Red Hat relationship and our open source strategic technology betting which was very well attended and well received as well. I got great questions from the audience — which I always like, because it means people are not just listening but also thinking — and had some interesting discussions afterward as well. Very much worth being here! I spent a while talking to someone who runs a business incubation program here at the SLCC, along with Christer Edwards, and learning about the challenges he’s up against in providing services for those customers. I had some “Aha!” moments that I should be able to turn into action items when I meet up with the Community Architecture team in Raleigh next week. Then I went upstairs again for Ian Weller’s talk on measuring community statistics. It also was very well attended and, though relatively brief, there were many insightful questions (and answers) so I know again there is a very rapt audience out there for the work Ian, Michael DeHaan, and others are doing on EKG and other community statistical tools. I spent the next hour working on some personal projects, went back downstairs to the hall and had some more conversations with people around the hall from XMission and Novell about open source education, and then headed back upstairs for a session on open source podcasting, where I’m writing this post. This has been my one subpar experience at the conference, since I’m pretty disappointed the speaker has spent most of his time talking about non-open source solutions. It would be nice to see some movement around this need. Maybe I can recruit some people to help me fill it! Tonight we’ll have the Geek Dinner at a place called Spaghetti Mama’s, which I”m looking forward to now that I didn’t eat anything since breakfast. Then I’ll probably head back to the hotel and try and do a smidge of catch-up before I crash. Overall, another superb day at UTOSC 2009! |
UTOSC 2009, Day 1.9.Eventful morning already from UTOSC 2009. So far today I’ve had a chance to meet Stormy and say hi, and then traipsed off with her and joined a few other friends from UTOS for a meeting with Salt Lake County Mayor Peter Corroon and some of his staff members. They seemed incredibly interested in what UTOS and the rest of the open source community is doing to encourage collaboration, open learning, job growth, and better ROI for businesses. What a treat! We bounced important points about open source around the circle of enthusiasts and proponents, and fielded questions from the Mayor and his staff with aplomb. Yesterday we had intermittent problems with the networks. Not unexpectedly for an open source conference, the Cisco wireless APs used here at the SLCC Miller Free Enterprise Center couldn’t quite keep up with demand in some areas of the building, since they’re only rated for a certain number of connections. To add to the fun, apparently they also see foreign SSIDs as unwanted visitors, and promptly stage an active DDOS against the interloper, thus flooding the wifi network so people can’t authenticate successfully. We discovered yesterday that this effect is not only caused by people with their own wifi network shares, but also by the mesh networking features of OLPC XOs. Meaning that a properly equipped six-year-old can effectively take down the network here. Both the SLCC and UTOS staffs worked together to diagnose and remedy the problem, and today they have put up extra APs to handle the increased traffic. Which is good, because attendance is expected to grow daily through Saturday! (Shout out to Trevor, for the early morning he put into helping with the networking. You’re a trooper, my friend.) |
UTOSC 2009, Day 1.This blog post will skip the descriptive narrative, and behave in a little more cut and dried fashion. It’s late and I am out of steam after a tremendous day here at UTOSC 2009. Up early this morning, got breakfast, caught a ride with Doran “fozzmoo” Barton, and then managed to barely see him again the rest of the day! Doran, thank you so much for your kindness, it meant a lot. Larry Cafiero and his daughter Mirano were already at the booth and had some swag set up. Helped move some odds and ends around, tidy up a bit, and otherwise be as useful as a bump on a log. Keynote was by not an engineer but a humanities grad student. Mainly concerned history of open source/crowdsourcing, quite interesting at times. Also mentioned something I hadn’t seen for years, Ted Nelson’s “Computer Lib/Dream Machines” book. I think I saw it in the mid-80′s, and by then it was already a decade old and still fascinating. Check it out if you have time and inclination. Went back to the booth and helped for most of the time that Clint, Ian, Larry, and others worked on the FAD for an event splash design, something to help us more effective channel new contributors who are constantly signing up at events where Fedora appears. They’ll talk more about that over the next few days, and months leading up to FUDCon. Our booth was super popular, and we gave away tons of swag with lots more still to come over the next two days. Talked up a blue storm about Fedora, Sugar and SoaS, and how Fedora and Red Hat work together. Had a very nice long chat with Andrew Jorgensen from Novell, who had some very nice compliments to pass on about our Fedora community packagers who work on Mono, and about our community-driven packaging in general. We also talked at length about our wonderful experiences with git, and as a fellow dabbler (thank you, Andrew, for geeking out with me for a bit) I passed on my recommendation of the Git Community Book as a tremendous learning tool. I admit I haven’t read all of it yet, but the 60% or so I have read has been great without exception. I got a chance to speak to several of the vendors, including a couple of people I ran into last year at UTOSC. Many are using Fedora-family systems for building, deploying, and servicing their bits for customers. Nice! In the afternoon I participated as a panelist for a multi-distro roundtable. We each got to talk about the origins, release cycles, and strengths of our projects and communities. Besides me for Fedora (and talking about our relationship to Red Hat Enterprise Linux and its downstream rebuilds), represented also were Novell, FreeBSD, Ubuntu, Debian, and Gentoo. We also gave shouts out to Mandriva and Slackware, not represented but not forgotten. I was very happy that, unlike so many similar panels in the beginning of this decade, this one was very positive-spirited. Good questions from the audience too. One question concerned the RHEL release cycle and its relation to Fedora. Another concerned rebuilding where I happily reported that Fedora was self-hosting, and that we provide every piece of source code, tooling, and build technologies needed to make Fedora, 100% free software that anyone can use as we do. Caught up with Clint after that to get a report on the FAD, and then it was time for Ignite Salt Lake, part of which was devoted to a Lego sprint (funny but not my cup of tea), followed by 5 minute lightning talks that ranged from good to spectacular. What a great way to bust us out of a code-focused day into thoughts about the world around us and its many wonders! It was almost 8pm when we got done, and then it was time to pack, gather people, and grab dinner. Clint and Jennifer gave me a ride back to the hotel, where I wrote this. Now it’s time for shut-eye before another action-packed day! I’ve been terrible about getting to email — network has been tetchy at the site on and off, and besides that I’ve not been able to spend more than a couple minutes at the keyboard at a time. Well worth the backlog though; UTOSC is off to another fine conference this year. |
UTOSC 2009, Day 0.Today I drove up to Washington after lunch to catch a plane to Utah for UTOSC 2009. The drive was pretty uneventful, and only about a quarter mile of stopped traffic near the Prince William Parkway and then a brief pause north of Woodbridge — it was practically paradise as far as DC area traffic goes. I don’t enjoy airports other than the fantastic Richmond airport with its comfortable, airy seating areas and free wifi. I thought I’d be able to cope by using blueman and tethering to my phone, but unfortunately the latest code isn’t up to speed with current PolicyKit or NetworkManager, neither in Rawhide nor in the project trunk itself, so no-go. I’ll reproduce the problem tomorrow and file appropriate bugs. I made it in one piece from Washington DC to SLC — no thanks to the terrible headwinds that assaulted us not only leaving DC but then descending into Utah. Haven’t experienced worse turbulence than that in a long while. Also, the plane was very crowded. I think the last row had one middle seat empty, and other than that, we were packed in there like sardines. If that weren’t enough, it was really warm back where I was sitting! This misery was somewhat balanced out by my first experience with in-air wifi. It’s too slow for any real transfers, but it’s certainly fine for email and IRC. I found everything worked, including tunneling back to my IRC proxy at home over SSH, and a VNC session for good measure. Not bad, but I wouldn’t have paid $12.95 even for a five-hour flight. Thankfully, it was a “first flight free” deal. My flight was pretty late in the day, leaving DC at five, so I got to SLC a little after eight, and took a really long cab ride to the hotel. By the time I got checked in it was a quarter to nine, and I hadn’t had dinner yet. I was pretty famished since that was two hours earlier than my bio-clock. Clint had invited me via IRC, while I was still over Oklahoma, to come by the conference center, but the flight had just been so chock full of humanity that I really needed some solitude and quiet. So once I dropped my bags off, I walked a half mile or so to a nearby salad place and it was pretty decent. I did indulge in a glass of the strawberry lemonade, but I avoided the apple cobbler (and yes, the other desserts too) and felt pretty good about it. Granted, it was a bit easier to do this since the staff started putting the food away while I was still eating my meal… I probably should have thanked them! Anyway, the wifi at the hotel is fantastic — thank you UTOSC 2009 organizers — so I’m picking up my email to do some triage, doing a little work on my slides, and then hitting the hay. The fabulous “fozzmoo” (Doran Barton) is graciously picking me up in the morning, and I’m looking forward to seeing everyone, once I’m rested and replenished with the meager helping of wits I’m allotted in this lifetime. More tomorrow on UTOSC Day 1, including the distro panel and other happenings! |
You’re gonna need a quality shoe.This next couple of months will find me gone practically more often than not. My upcoming schedule looks like this for now:
I think there’s a Fedora release happening in the middle there somewhere, too. |
Odds and ends, no. 55.
These items both lead me to believe that there’s no automated substitute for Knowing What You’re Doing. |
Into the future.Because we’re trying to stomp out a handful of nasty bugs — some of which appear to be thoroughly smooshed, and some of which we’re still attacking with the Rolled Up Newspaper of Free Code Wrangling — the Fedora 12 Beta will be pushed back one week. We expect right now to release it on Tuesday, October 20th. Of course the schedule has already been updated to reflect the latest schedule news. We’re still working out what this portends for the final release. Generally the QA team’s cumulative wisdom is that changes like this need to echo down the schedule, so that we don’t compress periods of public testing. Many people in the community don’t keep up with daily Rawhide, and depend on a DVD, CD, or Live image release to do their testing. (More on this in a moment.) We want to make sure that the Beta, which is the last release before final Fedora 12, has the chance to be tested by as many people as possible. This used to be the position and purpose of the so-called “Preview Release” that happened several weeks before GA. However, in an attempt to make our test releases less confusing, we have gone to the industry-standard practice of making Beta the release that is meant to be code-frozen, and ready for wide public testing. The only changes that are supposed to go into Fedora during this period are fixes for problems detected that would make it unsuitable for release. In which a wrinkle enters the fabric.To make that testing period as long as originally intended, we’d have to make the final Fedora 12 release a week later as well. There’s a thorny problem in that plan, however. Part of the Fedora infrastructure will soon be undergoing a relocation from one physical facility to another, and that’s scheduled to begin some time around the 18th of November. This is happening toward the end of a longer process that is supposed to be fully complete by no later than November 30th. Currently our final Fedora 12 release is scheduled for November 10, but if we were to echo this Beta slip down the rest of the schedule, that would mean a release on November 17. Moving infrastructure the day after a major release? Our community Infrastructure team is the very definition of awesome, but… Yikes! So, moving pieces of infrastructure around the day after release doesn’t seem like a great plan. And neither does shortening the period for public testing. We’re looking into whether our infrastructure relocation can be postponed, but of course the week of November 24th has a major US holiday, making rescheduling difficult. In which you, Dear Reader, play the starring role.Well, for one thing, by testing the currently available bits for what will be Fedora 12 Beta. The few changes still landing are mainly correcting problems found and noted in filed bugs, but the critical stuff in Rawhide, our development branch, is frozen. When we spin those packages into a release of DVD, CD, and Live media, it takes several days to prepare all of it and get it shipped to mirrors in time for the release date. And during the days even before that, leading up to the spinning, you can test the vast majority of what will be in the Beta. Here’s how:
This method may take some additional time, depending on the speed of your connection to the mirror you’re using, but the bits are the same ones that Fedora experts are using every day for installation, updates, and testing of the latest software that will go into Fedora 12. And most importantly, if you find problems, FILE A BUG! Remember, we’d rather hear about a problem twice than not at all. We have a Bugzilla primer that will tell you what you need to know to file a more clear and useful bug. It’s vital that we get great testing during Beta phases so that we can make the final as good as possible. All software has bugs, but what makes free software improve faster than alternatives is that we can all help stomp them out through an open, collaborative process. You can be a part of that process! |
Tightening the belt, no. 32.No, this blog isn’t about my new gym routine, although I have lost a few pounds already (not that you can tell). Nope, it’s about making way in our budget for FUDCon Toronto 2009 so that we can subsidize additional travel for community members. Originally we planned to get some sort of lunch brought in on Saturday (the technical session day), Sunday, and Monday (the hackfest days). However, provisions aren’t that useful at a community conference unless you have people to share them with. Things are really shaping up for a marvelous event in Toronto. We’re still two months away from the event and we have almost 70 people pre-registered to attend already! If you plan to attend, please visit the pre-registration area on the FUDCon Toronto 2009 wiki page and add your name. And again, if you’re in the Boston or Westford, Massachusetts areas (or can get to either one, because you live there or it’s cheaper), you’re invited to ride the bus. There’s a separate sign up for the bus, so you’ll need to both pre-register and put your name on the bus riders list. We also have a growing number of conference sessions and hackfests at FUDCon. Remember, the list you see on the wiki doesn’t represent everything that will happen there. Part of our conference sessions on Saturday will be BarCamp or “unconference” style, meaning much of that day’s schedule will be plumped up in the morning with proposals made the morning of the event! There will be plenty of sessions and opportunities to learn and work on fun and meaningful challenges, regardless of your interests or skill level. Everyone is welcome at FUDCon, and remember that by pre-registering, you can secure a space at FUDPub and score some other cool free stuff. |
Unleash the hounds!Greg, So it was written, so shall it be done. I contacted both of Virginia’s Senators, Mark Warner and Jim Webb, to show my support for the Open College Textbook Act of 2009. It did, indeed, take under five minutes to exercise some of my power as a voting constituent beyond showing up at the polls every couple of years. What about the rest of you USA readers? Honestly, it couldn’t be easier — visit this site, choose your state and Senator, and fill in their “Contact” form (always featured prominently on their web page) to send them a quick note. Check out Greg’s post for a sample; staffers and Senators do not mind receiving the same note many times, because they still know the issue was important enough to you to bother contacting them at all. |
