Older blog entries for katzj (starting at number 123)

A request for some simple testing

Another thing that’s been on my list to look at that I’ve finally had time to sit down this week is the new isohybrid support in syslinux. This lets you take an ISO image, post-process it and then be able to either burn the ISO to a CD or write it to a USB stick with dd. Given that we stopped making a disk image form of boot.iso a couple of releases ago to save on duplicated/wasted space, this is obviously kind of cool.

The problem was that the first time I tested it, it looked like it overwrote the checksums we use for the mediacheck functionality in anaconda. It turns out I just wasn’t thinking — we need to implant the checksum *after* we do the isohybrid modification.

So without further ado, I’ve built a test version of the Fedora 11 boot.iso that is usable in this form. Testing of it would be much appreciated!

How to test

  1. Download the test image
  2. Try to burn it to a CD like you normally would. Ensure that it still boots normally. You don’t have to go through the full install, just boot it. Extra points if you can test mediacheck
  3. Find a USB stick that’s at least 256 megs that doesn’t have any data you care about on it. Now try to write the test image to it using dd (dd if=test-isohybrid-boot.iso of=/path/to/device bs=1M). Again, you don’t have to install, just boot into the installer. Note that we won’t automatically find the second stage and you’ll get asked where to find the installer images.
  4. Let me know the results in the comments (including type of machine).

Assuming this works, I’ll get the changes in so that we do this by default with boot.iso and then probably also try to make it so that the loader can automatically find the second stage image on either the CD or the USB stick. I’ll also consider doing similar for the livecds, although there’s more value with liveusb-creator / livecd-iso-to-disk there as you also want to set up persistence in a lot of cases.

Comments

Syndicated 2009-06-18 15:04:24 from Jeremy's Thoughts

Boot tales, woo ooh!

(Take the title in the context of the theme from Duck Tales and maybe it makes sense?)

There was a long and rambling discussion last week about the version of GRUB that’s shipped in Fedora and specifically the fact that the support for ext4 did not land in the version we shipped in Fedora 11. Now, as was said on the thread, this is because the patches weren’t reviewed and ready in time for beta (there are a couple of different ones… so which one is right?) and so we didn’t feel comfortable putting them in after beta, especially as with the way GRUB works, the same filesystem code gets used for ext2, ext3 and ext4 with the patches. A little unfortunate? Yes. Would it have been better if we had gotten them in so that you could do an install of Fedora 11 onto a single partition? Sure. But that’s one of the costs of a time-based release schedule.

In any case, one of the things that came out of the thread was that I gave a history of the version of GRUB in Fedora. For posterity, I’ll repeat that here, with some edits.

So, the gory history for those who might be interested. Eight years ago (!), we decided that the advantage of not having to rerun lilo after changing the config file as you can just read the config file off the filesystem with grub was worthwhile. We had, at that point, been patching lilo for quite a while to have a graphical menu. Therefore, keeping a graphical menu was a branding requirement. Connectiva at the time had a patch to grub that worked. We picked it up, shipped it, and it (mostly) worked. Efforts were made to integrate upstream, but they were largely uninterested. Along the way, significant changes to the graphics patch had to be made as grub evolved and a few other efforts were made to push it upstream. Eventually, the answer was “no, we’ll do something in the next big version of grub after grub 1.0″. Then the main developers went away and we were basically left maintaining a (large at this point) fork. As there is no upstream for grub 0.9x left, we’ve been left in a position of maintaining it and we’ve added some real features that have been needed along the way as grub 2’s progress has been slow at best and we were initially unhappy with some of the direction taken

So, that’s where we are today. We essentially ship a fork of GRUB 0.9x with graphics support, support for a lilo -R type functionality (so you can reboot once into a single kernel), EFI, and some more little things that I’m not thinking of right now.

With that in mind, I sat down and spent some time with a current snapshot of grub2. Overall, it’s made a lot of progress in the time since I last looked at it (a year ago? maybe a little more?). It was actually able to successfully boot for me in KVM and there’s equivalent graphics support to what we’re carrying in our grub 0.9x package. That said, there’s still quite a bit of things to verify exist before we can switch. And just in my look, there are a number of small things that would need work, especially around the way the config file gets created and updated. And with the very short runway for Fedora 12, I don’t think there’s really time to get it into shape in time. But I do think that it makes sense to look at for Fedora 13. So I’ve started a feature page to track as some of the things get tested and worked on. Then hopefully we can make the switch pretty painlessly early in the Fedora 13 cycle.

Comments

Syndicated 2009-06-17 18:19:26 from Jeremy's Thoughts

Night at the Boston Pops

One of the SDM 09s sent out a note to everyone mentioning that MIT was putting on a small conference to celebrate the forty years since Apollo 11 landed on the moon. The closing little event was a concert put on by the Boston Pops performing Holst’s The Planets with a narration by Buzz Aldrin. It seemed like the sort of opportunity not to be passed up, so I got tickets for Kara and I.

We showed up at Symphony Hall and I was expecting a program that would basically just be all of The Planets. So I was quite surprised and pleased to look at the actual program. As I mentioned to Kara leaving, it really appealed to my geek-ness on a few levels: music geek, space geek and sci-fi geek. The selections were the following.

  • Also sprach Zarathustra — what a great way to start off a concert. It does a really good job of pulling everyone in
  • Blue Danube Waltz — continuing on the 2001 theme :)
  • Selections from The Planets (notably Mars, Venus, Uranus and Jupiter). This was accompanied by a short little film and the narration by Buzz Aldrin. Very very well done. The little films were neat and provided a good backdrop to the music.
  • Theme from Close Encounters of the Third Kind — now we really get into the “Pops” part I guess. I really need to get to one of the Pops concerts with John Williams actually conducting.
  • Premiere of a short film to commemorate the 40th anniversary of the moon landing that was commissioned by MIT and accompanied by a John Williams piece I didn’t know. The little film was cool and hopefully will end up online somewhere.
  • Theme from the original Star Trek followed by the music used for the closing credits of the new movie. This was a nice touch as the original series went off the air just before the Apollo 11 mission and the new movie is right at forty years later. And apparently, Keith Lockhart was (also) a big fan of the new movie.
  • Theme from Star Wars. A piece which needed no introduction.
  • And what was the last thing in the program, in fine Boston Pops tradition, a sing-along. In this case, various moon-themed or moon-based songs. Always fun, impressive how many of the things chosen weren’t well known. I didn’t know half of them.
  • Not in the program was a performance of John Lennon’s Imagine accompanied by the Boston Children’s Choir. This was a good touch and would have been a perfectly good way to end the concert
  • But they finished off with the Stars and Stripes Forever. Which was also a good way to end the night

As always after going to things like this, I had the thought I should do things like this more often. It was a great performance and we had a great time. There’s a reason why the Boston Pops are as world-reknowned as they are — they put on a great show and appear to have fun in doing so.

Comments

Syndicated 2009-06-12 14:38:25 from Jeremy's Thoughts

Spring Semester Wrap-up

My original intent was to do a post for each course from the spring, but since summer classes have now started, I don’t see that happening. So I’ll try to do a quicker post with all of them. I ended up taking 3 classes for credit in the spring and one as a listener. Unlike previously, most of these weren’t with my original SDM cohort — only one was a required class and it was one that many people took last spring. I did get to meet and work with some of the SDM 09s, though. I also had a Sloan course and a Kennedy School course.

System Optimization

One of the required courses in the SDM curriculum (although actually it might be one of the options with the new curricula), on the covers this sounded like an interesting course. Unfortunately, it ended up being largely about using Excel and its built-in Solver to do what boil down to linear programming problems. Certainly important for certain fields and positions, but not so much for me. Luckily, it was just half a semester :-)

Organizing for Innovative Product Development

The second half of the semester, I took Tom Allen’s course on organizing for innovative product development (15.980). The class was organized basically as a discussion around some slides and research that Dr Allen has done in his lengthy time with Sloan. It was a fun class overall with some interesting insights about how organizations can be set up to succeed both in terms of the hierarchy, hiring process, etc as well as things like focusing your usage of space.

One of the really cool things in the course was Gunter Henn’s discussion of how he used some of the work he had done with Dr Allen in building a BMW design center in Germany. The big idea being that you want everyone to be somewhat focused on the same things and so the center of the building is all glass windows and looking down on an area where prototypes of the vehicles are built. It also tends to lead to congregation of people around the vehicle when trying to work through an issue.

The downside is that some of the findings here are hard to apply — while it’s true that communication happens best in sort of flat (physical) structures and when you’re close by, how do you do that when you’re in an already existing building with a structure that you can’t change? Another question, at least to me, is how to apply some of it in the highly disconnected area of open source development. Clearly one of the takeaways is that conferences where people get face time are still very, very important. Something to think about as we change the structure of FUDcons and FADs.

Evolution to Web 3.0 and the emergence of Management 3.0

How’s that for a buzzword worthy title? The course was taught by Professor Stuart Madnick as a Sloan course and this was the first time it was taught. My primary reason for taking it was to get some exposure to the more “typical” MBAs from a technical background. I was also somewhat interested to see what was being proposed as “Web 3.0″. It turns out that the main thing being proposed was the move towards the Semantic Web. Even after the course, I’m a bit skeptical :-)

Overall, though, the course was interesting. I switched to listener (audit) status, but still did most of the readings/work/etc. The first half to two thirds of the course I think actually could make a very compelling course if the content were fleshed out a bit more — in it, we covered a lot of sort of emerging trends in the web / online services. The biggest problem, perhaps, was the fact that a few of us within the class had a much deeper knowledge of the area than the professor and we didn’t really get into the interesting bits.

As we got into the last third of the course, we switched to talking almost exclusively about semantic web technologies. And this is where things got pretty weird as we started to get deeper into details of XML and RDF schema. The lack of consistency in the level of technical depth had to have been frustrating for some of the class, although some of us took it in stride.

All that said, I’m still glad I took it as a listener. Good to see what other people are thinking about in that sort of area. Also, met some cool people some of whom are going on to do cool things as they after graduation. And I was able to wave the “open standards are important” flag a few times.

Game Theory at HKS

I covered this more in-depth already, so no need to repeat myself

Comments

Syndicated 2009-06-10 00:08:04 from Jeremy's Thoughts

Lake Auburn Road Race 2009 Recap

Yesterday, I raced in the Cat 4 men’s field of the Lake Auburn Road Race. Unlike last year, the weather was much better and I had a much better result.

Instead of going up the day before, I woke up extra early to drive up. Picked Kate up and got on the road basically on time. The roads were empty and as it got light, there was quite a bit of clouds. Right as we crossed into Maine, there was a little bit of drizzle and I was worried that we were going to have a repeat of the rain from last year. But it let up after about five minutes and then the sun came out and the cloud cover burned off.

We got to the course start with lots of time to spare and actually had the time to pre-ride a lap of the route. Even more shockingly, the rest of the team (minus one person) had made it there with plenty of time to spare as well. So we headed off as a team to recon the course and jog our memories from last year. It was good to do as little things like “landmark for the turn before the hill” doesn’t stick with you for a year… but half an hour before the race start it certainly does.

The race itself was to be three laps of an 11.5 mile circuit. It started with a quick downhill followed by a little bump and then a steeper little climb. Then a few turns and a mile or so with a slight downward grade to the back half of the course which was about six miles and pretty much flat. This was thus a bit on the fast side. The course then took another turn and began going back up towards the finish with one steeper and then one longer and more shallow section to a bit of false flats for the final kilometer or so. Total of about 600 feet of climbing a lap. Pretty much good pavement for the entirety of it, well marshalled, etc. Honestly, it’s a great course and I was looking forward to coming back and doing better.

Map of the Race

Map of the Race

Course Profile

Course Profile

In the Men’s Cat 4 field, we had six people in the field out of a total of somewhere between fifty and sixty — myself, Jim Gomez, Charles Wescott, Nessim Mezrahi, Kenton Eash and Andy Tucker. My personal goal was to finish with whatever the main pack ended up being and from a team perspective, we were hoping to get someone at least in the top ten. The race started on time and it started out pretty quick. I was at the front and was able to maintain my position through the fast descent in a nice improvement from last year. The first lap continued pretty quickly; I know I saw an average speed of above 25 mph at one point on the back stretch. When we hit the big hill up to the finish, that dropped a bit. We had definitely started to drop some riders off the back, though.

The second lap was much of the same and I realized that hanging on was really about all I was going to be good for. I did get in some good work with moving up in the pack and raising my comfort in doing so. It helped that the peloton for the field was pretty smooth overall. The exception was that for every corner, the speed dropped somewhat dramatically and then people accelerated like hell on the other side of the corner only to let up after 100-200 meters. A little annoying, but I kept with it. Andy and Nessim spent some time during the lap attacking and trying to weaken some of the stronger riders in the field. I was content to just sit in and let things happen.

By the third lap, I realized that the only Quaddies who were left were Nessim, Andy and myself. I talked briefly with Andy and he said that Nessim was going to try to set him up with a lead-out. I didn’t really have anything to add to the effort, so just was going to keep my head down. I also noticed at this point (not far into the lap really), that there was a rider a little ways up the road and that the pace car seemed a little further away. Not that I was going to be able to do anything about it. As we started up the hill for the finish, though, the gap dropped — by the time we were cresting the second hill up to the finishing flats, we passed the guy who had jumped off the front and this was when people really cranked it up a notch.

With a little more than a 1km ago, someone decided to start pushing for the sprint and I decided I had done enough to accomplish my goal. So I sat up and got passed by 6 or 8 people coming across the line about 26th although I did make it look like I was sprinting for something. Andy ended up with 7th and Nessim was somewhere in the pack between 15th and 20th. All in all, a respectable day by the Quaddies.

A sprint for the photo at least

A sprint for the photo at least (photo courtesy Charles Wescott)

By the numbers:

  • 34.7 miles, overall average of 24 mph
  • Second lap was the slowest by a small margin, first and then the third was the fastest
  • Pretty usual race heart rate for me averaging 170. Max was only 193, though, which is a little lower than usual for me in races
  • Cadence only averaged 80, although maxed out at 132. Low average is probably as I got to do some decent coasting sitting in the pack and the Garmin averages in those zeros
  • 1750-ish feet of climbing and some of that was serious grades. Not long climbs though

So overall, a very satisfying result from my point of view. I finally feel like I’m getting back the right level of fitness for racing. Also, a pretty good team result with the seventh place plus we also had two women in the Cat 4 women’s field (Nancy Labbe-Giguere and Kate Leppanen) who finished fifth and sixth out of a field that was probably about twenty deep.

Comments

Syndicated 2009-06-08 03:16:02 from Jeremy's Thoughts

May in Maps

It’s been a good month of riding — over 800 miles on geared bikes1 recorded on the Edge and another on the order of 150 on the fixed gear. No racing, but I’m feeling a bit more confident there and looking forward to the Lake Auburn Road Race next weekend. Rather than saying any more, I’ll just sum it up with a fairly representative set of GPS tracks from the month

Usual Bobby Mac loop out through Carlisle
Ride out to Westford and Groton
Century for the month

Out to Westford from Home
Home from Westford via Sudbury
Ride out through Stow, Bolton, Harvard

Hilly ride out to Harvard and returning via Bolton and Sudbury

1. Almost entirely on the new CAAD9 as the Merlin’s shifter needs rebuilding which I noticed on the one ride I did on it this month.

Comments

Syndicated 2009-06-01 02:25:17 from Jeremy's Thoughts

Garmin 705: Definite Upgrade from the 305

As I’m intending to eventually get a power meter1 for my new bike, I decided it was also probably a good time to upgrade from the Garmin Edge 305 I got last year to the newer, nicer Edge 705. A month into using it, it was a very worthwhile upgrade.

The 305 was really great in terms of tracking my rides. I was able to both upload rides to MotionBased and also keep the GPX files around. This made having a training log relatively easy for the rides I used the GPS on. I could then also share the routes with others, go back and see what I’d ridden, etc. This was awesome.

The 705 does all of that. And as a plus, it does it with a color screen and longer battery life. Those probably wouldn’t be worth the upgrade. The ability to talk to power meters probably is. But the other thing is that its routing functionality is way better. I’ve now taken a route that I’ve done in the past and loaded it back onto the Garmin as a saved route and then followed the route again with audible prompts for each turn. And then for today’s ride, I created a route with MapMyRide and loaded it onto the GPS the same way and followed it. We did a 75 mile ride, the first half of which was all on roads that I’ve never seen before and we had lots of very nice roads and only missed the one turn that I didn’t listen to the GPS for :)

There’s only one set of caveats with the Edge devices in general — the software stack kind of sucks. There are a few things to keep in mind/tricks I’ve figured out

  • Unlike the Edge 305, the 705 just shows up as a USB mass storage device. This means you don’t have to use garmin-sync under Linux and can just mount and copy files off. This is an improvement!
  • The shipped software is kind of crappy for a lot of things. I’m finding you’re better off using various web sites.
  • To load a route on the 705, first you want to create a route as a GPX file. Then I’ve being going to GPSies and converting it to a GPX track and then I can just copy it to the Garmin/GPX directory and it shows up as a saved route.
  • Garmin purchased the company behind MotionBased and has put them to work on Garmin Connect which is intended to replace MotionBased and is what you have to use with the new devices. Unfortunately, the new site is lighter on functionality, slower, and I’m generally less than happy with it. If you’re using another site (TrainingPeaks.com, MapMyRide or something else) for keeping your riding log and like it, I’d love to hear what you’re using.

Even with the above, though, I’m very very happy with the 705 and would recommend it to anyone who rides a lot and likes to over-analyze data afterwards. Or for anyone who likes to ride in places they’ve never been and see new routes without getting lost or following a paper map.

1. I’d also love to hear opinions on a Quarq vs the more ubiquitous PowerTap as a power meter option

Comments

Syndicated 2009-05-26 02:43:09 from Jeremy's Thoughts

Semester Wrap-up: Game Theory at HKS

My game theory class at HKS wrapped up with a final exam a little more than a week ago after a two week reading period. Taking a class at HKS was pretty different from an MIT class.

For one thing, it was a fairly different set of people. While all the SDM students have an engineering background and even a fair number of Sloanies come from some sort of quasi-technical background, very very few of the people at HKS seem to have that sort of background. So they tended to approach problems a little bit differently and think about things from a different angle. This policy angle also really started to become apparent in the presentations that we had to deliver at the end of the course — many of the games proposed were based around political conflicts or things of that sort. Very different from what you’d get in an MIT class!

Overall, the class was quite good. The professor knew what she was talking about and did a good job of keeping the class moving along in most cases. The examples were a pretty wide mix drawn from a variety of different cases rather than just being based in one subject area. There were four problem sets over the course of the semester and they did a good job of preparing me for the final. Compared to an MIT game theory class, I suspect there was a bit less of the math details behind the theory but with what I was looking to get out of it, that was perfect.

So I enjoyed the experience and the course and am glad I took it. As far as applicability, a good chunk of why I took it is that I’ve done some reading on game theory in the past and so wanted to learn a bit more about it. I don’t think I’ll be sitting down and drawing out game trees for things, but it’s a nice framework to simply sit down and think about how people might respond. And most importantly, I had fun in the class.

Comments

Syndicated 2009-05-26 02:17:50 from Jeremy's Thoughts

New race bike

As I offhand mentioned a few weeks ago, I recently built up a new bike. Although my Merlin is awesome, I’ve been wanting to build up something that’s set up a little bit more aggressive and a little bit more geared towards racing as opposed to the long rides the Merlin is for. After a fair bit of looking around, the fact that Quad was going to be becoming a Cannondale dealer for this year led me to settle on the CAAD-9 as a good choice of frame. And since the MIT cycling team is sponsored by SRAM, I figured I’d give SRAM a try — the Merlin has Campagnolo Chorus, but the continuing increase in cost for Campy meant I was interested in trying something a little cheaper. So a SRAM Force group was purchased.

Everything finally arrived and I had the bike built up. Odds and ends include an older, but awesome pair of Velomax Ascent wheels, a Fi’zi:k Antares saddle (as I was happy with it on the Merlin), white bar tape, Speedplay Zeros (another new thing to try compared to the Keos on the Merlin). I think it turned out pretty well — it looks sweet and more importantly, is a lot of fun to ride. The setup is definitely more aggressive and it just feels like it wants to go fast. Stiff as rails too. So I’m pretty happy with it.

No white garage door, but for an obligatory picture

Full Bike

Now, three weeks later, I’ve got about 500 miles on the bike and am really enjoying it. I raced it the first weekend I had it and have also gotten in some other, longer efforts including today’s century. It is meeting all the criteria I set for it. Still to be considered — power meter options and if I want to get a pair of flashier wheels.

Comments

Syndicated 2009-05-17 02:14:02 from Jeremy's Thoughts

PPC and Fedora: What’s Next?

Thanks to everyone for the feedback on PPC usage.

At this week’s (public IRC) Fedora Board meeting, Seth had kindly put the issue up for discussion and since it was a public meeting, I took part as well. Now the first question that might come to mind is “Why does the Board care / make the decision?”. And the reason is that it was the Board that originally voted to keep PPC as a primary arch until there was another functional secondary arch. Now that we’re two years later and there’s still not another functional secondary arch, I think that the topic was worth revisiting.

The discussion ended up being somewhat lively, but to me there were just a few key points. On the side of requiring PPC kept as primary, the argument seemed to mostly be that by going to secondary, PPC as an arch in Fedora would not be successful. One stated reason was the timeframe. So there was a proposal for keeping the requirement in place until either a secondary arch is up and running or some time limit (six months? a year?) was hit. The other side is that it’s been two years, what difference is a known six month time horizon going to make?

In the end, the Board voted and decided that from the Board level, PPC is no longer required to be a primary arch. That does not mean that PPC is now automatically a secondary arch.

So, what’s next? The next step is that I am proposing to FESCo that they consider a proposal to have PPC become a secondary arch for Fedora 12. And in doing so, put out a call for someone to lead a PPC secondary arch group. That would entail the work in keeping builders running, creating test releases, getting testing, etc. And then also being the person responsible for getting the release out and calling it done.

By making PPC a secondary arch, there are a few tangible benefits.

  • Likely less build time for packages, rawhide, etc so that hopefully development can move a little faster
  • Less last-minute scrambles to fix the PPC-specific distribution issue (whether it be installing on some platform or fitting on the DVD)
  • More people caring about the secondary architecture process and thus hopefully helping it accelerate.

There are, though, also some risks and downsides. I’ll list them as well, but with my refutations :-)

  • It’s possible we could not have everything in place for a Fedora 12 PPC release. While I freely admit that this is a concern, I think that this is a concern no matter when we move PPC to be a secondary arch and I actually see the argument here getting stronger over time as the number of PPC users decrease as old Macs die.
  • We may lose some portability testing in builds. Also, true. Hopefully the build infrastructure for secondary arches is far enough along that those can be reported still.

Thoughts, comments, etc appreciated as always

Comments

Syndicated 2009-05-08 15:56:31 from Jeremy's Thoughts

114 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!