Friday June 13, 2008 Bill Sommerfeld's Weblog
Still Under Construction. Watch for falling objects
Friday June 13, 2008 On respecting other people's privacy preferences
So, one of the things I've chosen to opt out of in life are "social networking" sites -- linkedin, orkut, facebook, evite, etc. I don't trust them. It doesn't entirely matter why. I just don't trust them -- please don't waste your time trying to convince me that they're ok.
And if the only way I can get invited to parties is to accept email from evite, I'll happily stay home by myself.
Unfortunately, a fair number of people (typically those who don't know me very well) think that just because they
think a site has an OK privacy policy, everyone agrees that the site is trustworthy and thus it's OK for them to give my email address to the site without checking with me first.
I really don't understand this -- in my mind, it's horribly rude to give my contact information out to a commercial third party without asking me about it first. And I don't really put much stake in official privacy policies - if they don't owe me anything, they have no incentive to prevent my email address from being leaked to spammers, identity thieves, and similar pond scum.
So, if you want to get in touch with me, do so directly using your own email account or phone. Don't indirect through sites like orkut, evite, linkedin, etc.; I generally route mail from such sites directly to /dev/null because it annoys me.
And if you want to be respectful of others, don't assume that they have the same privacy preferences as you -- check with them before you give their email addresses to third parties, no matter how wonderful and saintly you think those services are. They might not trust the service. They might have another email address that they'd rather use with the service. Just take the time to ask individually first!
Friday May 09, 2008 A few good words for OpenOffice from the Instapundit..
(2008-05-09 10:01:11.0) Permalink
Wednesday March 05, 2008 
(That's the last thing Puccini wrote in the 3rd Trombone part for Turandot. Not a lot to do but sit & wait for the trumpet cue...)
You can hear me play this note (and many others) in the Lowell House Opera production of Turandot, starting tonight. I'll be playing tonight, Sunday, Tuesday, and next Wednesday.
Tuesday January 29, 2008 In Memory of Greg McMullan, 1963-2008
I was stunned to get email Sunday morning informing me that Greg McMullan had died in a house fire.
Greg wasn't a close friend -- just someone I crossed paths with now and then, starting when I was a confused and clueless froshling at MIT. He sticks in my memory as a very friendly, approachable and helpful person who could be relied on to give good advice and make me (and many others) feel at home in the high pressure environment of MIT.
I ran into him again perhaps ten or eleven years ago when he was showing off his new toy -- the first generation Palm Pilot. I was sold almost immediately.
And I vaguely recall running into him when I was in a orchestra accompanying a large-scale choral work; he would have been singing, I had my trombone.. but for the life of me I can't remember what groups were involved or what piece we were doing.
Others have written far more about him; his brother's Live Journal page is probably the best starting point.
Monday December 10, 2007 configuring dhcp for almost-seamless use of nwam
When I use my laptop at home at my desk, I usually have it plugged in to a port replicator. I have wireless but I need to plug in for power anyway.
I finally got around to setting up the laptop to use the nwam phase 0 package to provide nearly seamless migration between wireless and wired network. The trick turned out to be to set CLIENT_ID in /etc/default/dhcpagent to the client id used by the wired interface -- 0x01 followed by the six-byte ethernet address. I suppose using the wireless interface's mac address would also work, but it's not soldered to the motherboard and might end up migrating to another laptop later..
When I undock to use the laptop in a different part of the house, nwam brings up the wireless and uses the same client ID with the wireless interface, and the DHCP server gives it the same address. There's a brief period of about 4 seconds when bits don't move, but no manual intervention (aside from the physical act of plugging/unplugging) is necessary.
And when I bring the laptop back to my desk and plug it in, the reverse transition is also automatic.
Very slick!
Saturday December 01, 2007 Pippin opened last night.
We were going along really well when, halfway through the first act, the casting holding the pivot & spring of my F-valve trigger popped off my bass trombone (a Conn 112H, which is an in-line double-valve instrument pitched in Bb/F/Gb/D); I'll note in passing for the benefit of non-trombonists that a Yamaha in-line double-valve bass trombone was the instrument mentioned in the well known but entirely apocryphal tall tale involving a Chilean bass trombonist, a firecracker, a mute, and the "1812" overture. (Now really -- what trombonist would ever use a mute in one of the few places in the symphonic literature where a quadruple forte (ffff) is written??!?!?!)
Anyhow, after it popped off, the linkage and trigger hung down from the valve, occasionally getting tangled in my shirt pocket but fortunately keeping the valve stably "off" the whole time. So I was left with a Bb/Gb instrument. But whoever designed the instrument put in an extra-long tuning slide on the Gb valve so I could actually retune it down to F.
Having learned the show using a bunch of alternate positions involving the use of the Gb valve, I played the rest of the show alternating the 2nd valve tuning. Definitely a higher-stress experience than I was hoping for, but it was an (unexpected) benefit of the in-line double-valve configuration that I hope I never need to rely on again...
Aside from that, opening night went really well.
And it's fixed already -- I'm extremely lucky that I live less than a mile from Osmun Brass Music; they fixed it in about 15 minutes for a very reasonable charge. The tech pointed out several other problems with excessive play in the valve linkages and recommended a return trip to the shop when convenient.
Wednesday November 28, 2007 Best rehearsal quote in quite some time..
During a dress rehearsal of the battle scene in Pippin last night, the music director told the cast:
"Just because you're dead, that doesn't mean you don't have to sing!"
(I'm playing bass trombone in the orchestra of MTG's production. Opens this Friday.)
(2007-11-28 10:36:46.0) Permalink
Monday November 26, 2007 If you think it's too good to be true, it is.
So, I signed up for FiOS TV a week ago. They promised an install date of today, which I thought was a little fast.
Turns out I was right.
The installer shows up, looks around, and says that some outside wiring prep work that should have happened didn't. What's worse, the outside wiring folk apparently closed their install ticket saying "all set" when there is no evidence that there actually is an available tap in the cable anywhere near my house (the installers spent a bunch of time staring up at the poles and wiring before concluding that "construction" needed to be called to make a tap available).
Thursday November 15, 2007 I'm in the midst of building our own flavor of labelled IPsec for Trusted Extensions, and took a look at what the "competition" (specifically, SELinux) is doing. I was surprised to notice that (at least if the ipsec-tools-0.7 source is to be believed) they've grabbed a codepoint assigned to RFC 3168 (Explicit Congestion Negotiation) rather than actually asking for one to be assigned via the normal IANA processes, or using the long-defined but rarely used capabilities of ikev1 to carry a sensitivity label.
It looks like racoon2 gets this right (but doesn't have the SElinux security context support).
I can't be the first person to notice this, can I?
Wednesday November 07, 2007 I'm surprised the complaint didn't include "failed to repair godzilla crush damage"
and "causes vertigo in blind people" (no, really!). I hope we get an injunction barring Gehry from making sandwiches.
(MIT sues Gehry over Stata Center)
(2007-11-07 18:52:00.0) Permalink
Tuesday November 06, 2007 Three strikes, you're out (but it's only the first inning of a preseason game)
The indiana prototype was released with the name "opensolaris developer preview". Having installed it and (briefly) attempted to get work done on it, I'm even more convinced the name was vastly premature.
opensolaris. nope. it's the prototype output of a small implementation team and hasn't been through the rigor and wringer of the development process. It's a fascinating and impressive start, but.. it's not opensolaris yet, and I'm skeptical of some of the design choices (admittedly, I always am..)
developer. I'm a developer so it should be for me to use? I installed it. I tried to run bugster to file a bug. Nope, no java. Tried to build some simple C programs. Nope, No C compiler, no make. Tried to add the binary nvidia driver so the display didn't look crappy -- got that working but it was an adventure -- pkgadd(1m) is busted so I had to pkgadd -R on a different system, rsync the bits over, and re-run the package's postinstall script a few times before it "took".
If you want to play with the new packaging system, make sure you have another system running SXCE handy to actually do builds. And if you see something unexpected, please file bugs early and often. If you actually want to get work done, you're better off with SXDE/SXCE until further notice.
preview. When I hear "preview" I think "movie sneak preview" .. getting a look at the almost-final production before they come out. SXCE & SXDE are previews. The Indiana prototype is like the raw footage viewed by the production team the day after it's shot before the bloopers have been edited out.
Tim Foster's critique misses the point of the engineering objections. Attaching this name to a pre-alpha snapshot demo which is simply not even close to done is a mistake. It might make sense when Indiana is a little better baked, but IMHO, the current name attached to the current bits hurts the "brand". IMHO the name should have been held in reserve until it was ready for community endorsement.
Friday November 02, 2007 Accusations of "stop energy" are "stop energy"
So, there's this hot new newage-y (I think that rhymes with sewage-y) concept being bantered about by certain people called "Stop Energy". It seems that it's been used primarily as a non-rebuttal rebuttal to project review comments that the reviewee would rather not deal with.
Many desireable properties of large systems are not localized to any one part of the system. Security is a key example, but performance and availability are others. Senior engineers in organizations which produce such systems often have an obsessive-compulsive streak -- because they have to. In order to preserve or enhance that property, you need to get all the details right, and this often results in long list of "stuff you got wrong" coming from reviewers. It's easy to misread this as a message to just give up.
But it seems that the new thing is to instead accuse the reviewer of applying Stop Energy to the project.
Based on what appears to be the canonical definition, it occurs to me that these accusations of "Stop Energy" are .. an exertion of Stop Energy against the reviewer. The reviewer actually is trying to help, it's just that there's a breakdown of communication such that constructive criticism is interpreted as an attempt at stonewalling. So, the frustrated reviewee counter-stonewalls, perhaps with this accusation.
A more constructive response is to honestly ask "okay, so what should I do?". And then listen, and change your proposal accordingly. Maybe the requirement you just learned about overlaps with your requirements in such a way to produce a null solution set, so maybe you need to go back and adjust your requirements.
UPDATE: Perhaps I should have been clearer. My non-snarky view is that "Stop Energy", if it exists, only exists in the mind of the person who stops in response to criticism. In a large project there isn't a single frame of reference in which you can declare some action unambiguously as "forward progress". Reviewers often point out things where, in certain frames of reference, a proposed change is a big (or small) step backwards. In the large, those reviewers are themselves charged with (say) improving the overall security of the system; and in that scope, one or more proposals that introduce more insecurity form a barrier to progress. Reviewees often hear the "no, don't do it that way" part of the message and then tune out and fail to get the message about the requirement they overlooked and cause the subsequent conversation about alternatives to fail. And as a result a message intended by its speaker as "do it a little differently" is received as "don't do it at all".
Thursday November 01, 2007 Looking good, save for the name.
Ran into a few bugs installing the Indiana prototype.
1) the installer got confused when I attempted to add the user "sommerfeld". (a 8-character username limit is a figment of useradd's imagination). I had to reboot and try again.
2) the lack of the nvidia binary driver in the distribution meant that it didn't cope with a 1920x1200 display.
but otherwise it installed with a zfs root in almost no time flat from CD (system refused to boot from a USB key).
It still needs a name change, though..
So, a preview of the new packaging & install technology produced by Project Indiana was just released. I'm shortly going to be installing it on a spare system in my office just to give it a shot.
Unfortunately, it's being called the "OpenSolaris Developer Preview" and is being portrayed as a distinctly special binary distribution on the opensolaris home page. The name is unfortunate for a number of reasons:
- The vast majority of the changes have not yet received the typical design and architecture review received by Solaris components
- There is not yet community consensus that OpenSolaris should have a reference binary distribution
- There is certainly not yet consensus that the Indiana technology is the right tool for the job.
I hope the folks who chose this name despite ample warning that it would cause trouble quickly reconsider. And I hope that the poor choice of name doesn't deter people from giving it a try. But the choice of names is forcing something of a constitutional crisis within opensolaris.
(2007-11-01 10:29:50.0) Permalink Comments [1]
Wednesday February 07, 2007 When a favorite restaurant closes
Valerie asks what she can do about a favorite restaurant which has lost its lease and will most likely need to move.
Don't Panic.
A while ago (must be over a decade ago by now), the canonical Chinese restaurant at the MIT end of Cambridge, Mary Chung's, lost its lease and was shut down for about a year before they found new space on the other side of Massachusetts Avenue. Mary's was open every day but Tuesday, though she took an annual one-week summer vacation (which was known as "the week of Tuesdays" to some of her regular patrons).
The Year of Tuesdays was painful for some but they came back from it stronger than ever in a better, larger space. Recently they were even one of the five Boston-area restaurants featured in an episode of The Hungry Detective on the Food Network.
There's not a heck of a lot you can do unless you've got connections in the commercial real estate arena, but there are a few things which come to mind:
- Keep patronizing them until the bitter end.
- Stay in touch with the proprieter during the shutdown period (easier with FdM than it was with Mary's since they have the secondary location).
- Most likely there will be some amount of town-level zoning/licensing involved in the move. Generally the only people who comment on such matters are concerned abutters; statements in support of the applicants from satisfied customers will typically make a big impression on the licensing authority.
Calendar
| « June 2008 | ||||||
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 |
8 | 9 | 10 | 11 | 12 | 14 | |
15 | 16 | 17 | 18 | 19 | 20 | 21 |
22 | 23 | 24 | 25 | 26 | 27 | 28 |
29 | 30 | |||||
| Today | ||||||
RSS Feeds
All
/General
/IETF
/IPsec
/Music
/OpenSolaris
/Solaris
Search
Links
- Team IPsec
- Bill Sommerfeld
- Dan McDonald