Older blog entries for vorlon (starting at number 10)

Let's see how many Planet Debian posts in a row we can get about the fact that sarge is released! Yeah! ;)

I've already been personally congratulated a number of times today on the release, and I know the other members of the release team have also been congratulated. This feels awkward to me, because even though I've certainly put in long hours trying to pull this together (and hold it together), this is a community effort: the release team cannot and does not do this alone. There are many other teams involved in achieving a Debian release, and all the release schedule frustrations and eventual triumphs :) happen in a much larger context than that of "the release team".

So when congratulating the people responsible for the release of this latest iteration of Debian's high-quality, free operating system, don't forget to congratulate and thank the FTP team, for their behind-the-scenes work keeping the system running that makes development possible; the documentation team, for their fine work on the sarge installation manual and the (somewhat hurried :) release notes; the CD team and mirror team, for the long hours they put in this weekend to get our official distribution channels in shape; the installation team, for the flexible new Debian-Installer; the porters, some of whom had to fight last minute buildd outages (naturally ;) to get those last few bugs fixed; and some other teams too that aren't unimportant to the process just because I'm in too much of a daze to acknowledge them here.

And thanks to the many maintainers, bug fixers, translators and documentors who have helped to make sarge a distribution worth releasing.

Sarge is once again proof that communities can do great things — even communities of irritable, cantankerous, grudge-holding, flaming Free Software nuts. ;)

Raphaël writes that Debian's decision to hold everything in the Debian main archive to the same standard of freeness (the DFSG) was a mistake, and that it's necessary to fix this mistake with another general resolution.

While it's fine to disagree with the outcome of the previous changes, please don't start a new GR unless you think there's a real reason to believe the outcome will be different. As you can see from the tally sheet for that vote, with 396 developers voting, we had the participation of a sizable percentage of all developers (though, interestingly, less than in most DPL elections), and came nowhere near the 3:1 supermajority needed to rescind the changes. Since everyone's votes on this issue are public, it should be straightforward to canvas the developers who didn't vote or who voted against reverting the change, to get support for this change before putting everyone through the GR process again.

In contrast, I don't think public pronouncements about how bad of an idea this change was are going to get us anywhere. In fact, we've had almost a year of such public pronouncements, from various people, that haven't gotten us anywhere.

No, jdub, the Ubuntu kernel team don't have real ultimate power; ninjas do.

I'm afraid the use of port 445 is not a new and random port choice by Microsoft, Tollef.

$ grep 445 /etc/services 
microsoft-ds    445/tcp                         # Microsoft Naked CIFS
microsoft-ds    445/udp
$

This port assignment has been in place with IANA for several years now, and actively used by Microsoft implementations since at least the release of Windows 2000. If anything, your Uni's policy change seems to come a bit late to the game. But I guess that also explains why they've done badly enough with it that they're dropping packets on the floor...

Needless to say, trying to stop people from implementing new network protocols and asking for new port assignments would be futile. ;)

Jordi expresses his concern that the American public could've voted this chucklehead in for a second four-year term. While I certainly voted for Kerry, and think that the vast majority of Bush backers should be sterilized along with their children, it's important to remember that the best that can be said about Kerry is that he's the lesser of two evils.

Kerry is a career politician, and in that sense is no better than any of the other alternatives out there. The only candidate in the Democratic party's primaries who didn't reek of politics was Kucinich, who stood a snowball's chance in hell of taking the nomination — let alone beating Bush. Voting for Kerry was very much a question of the lesser of two evils; these were votes against Bush, not votes for Kerry.

And the main point of differentiation between Bush and Kerry is that Bush has gotten us into a war we shouldn't be in; on many other issues, as people have pointed out, Kerry is likely to be worse than Bush, not better. When so many Americans — including my father — still buy the administration's line that this war was necessary, getting a sufficient majority of Americans to understand the war was wrong to the point of actually voting the slime out of office is quite a challenge — and one we failed at in the 2004 elections.

The bright side in all of this is that four more years of Bushocracy may be just the wake-up call the American public needs in order to advance some real reform in 2008. Assuming that Bush doesn't attempt to declare martial law at the end of 2008 and try to make himself dictator for life, that is. :P

As long as the US is beholden to a self-sustaining, two-party political system that no one believes can be changed, real reform is impossible. The dollar signs behind Kerry are the same as the ones behind Bush. It doesn't surprise me at all that Nader stayed in the race, in spite of the possibility that his candidacy could negatively affect Kerry.

One tantalizing suggestion from an NPR commentator, though, was that the 2008 elections would be the year we could see a genuine Internet candidate emerge courtesy of the blogosphere. So, who wants to make the first nomination?

Jury duty

I was called up for jury duty this week. An interesting experience. Of the first twenty-four jurors, only one was not passed; pulling in the next candidate in line moved me up to the third seat for consideration as an alternate juror.

The first potential juror knew one of the officers who responded to the incident socially. The second potential juror would fail one of her college classes if the trial lasted longer than two days. So there was a good chance I would have been selected as the alternate juror, had it not been for my views on jury nullification.

To be fair, my answers were not as eloquent as I might have wished them to be; and my voice not as forceful in the dry air of the courtroom than I had hoped. I made it clear enough that I knew something, but not how much; and a little knowledge (together with an unwillingness to blindly follow the instruction of the judge, which was the nature of the line of questioning) is a dangerous thing in a juror. Still, it was distressing to hear comments like this first-hand from trial lawyers:

Defense attorney: If the prosecution proves its case beyond a reasonable doubt, of course, it's your job to vote guilty [...]

... job?

Prosecuting attorney: If I prove my case beyond a reasonable doubt, what will your vote be?

The solicited answer is "guilty", of course.

Prosecuting attorney: Obviously the defendant has the right to a fair trial; but the State also has a right to a fair trial. Do you agree to this?

The rights of the State? Since when do states have rights!? And what in the world does the state's right to a fair trial give that the defendant's right to a fair trial doesn't already encompass?

So when asked if I had any concerns about the legal concepts that had been discussed, I had a few comments:

You asked each of the jurors what their vote would be if the prosecution proved its case beyond a reasonable doubt, and they answered "guilty". I believe the defendant has a Constitutional right to trial by a jury with the power to acquit in spite of the law. [...] The court's own training videos talked about "intellect and conscience". I believe that the issue of "conscience" requires a jury that has the power to nullify the law.

In retrospect, had I had more presence of mind, I might have said something like this:

If the prosecution does not prove beyond a reasonable doubt that the defendant has violated the law, I will vote "not guilty". If the prosecution does prove this beyond a reasonable doubt, I will most likely vote "guilty", but I will not give you a promise to that effect: proof beyond a reasonable doubt speaks only to the jury's intellect, and while I'm willing to accept the judge's instructions regarding the facts of the law, a jury can only be the conscience of the community if it has the power to acquit in spite of the law.

Would it have mattered? Perhaps not.

And you and your kind, mglazer, are a threat to freedom and civil rights the world over. Luckily, so long as you're spending your time espousing your ideology of mutual destruction here on advogato, we can be assured that you will be completely ineffective, as I don't think anyone in your audience is enough of an idiot to be swayed by your diatribes.

So please, do proceed with your efforts to enlighten us all.

"Once the bullshit speak is installed, instant denial a la Ghost Not kicks in, and we see the characteristic development of anti-altruism instead of selfishness, and anti-selfishness instead of altruism (where altruism and selfishness are actually the same thing). Hence the M0 dictionary definition of "selfishness" explicitly asserts that helping self is a zero sum game, and actually describes anti-altruism. Meanwhile M0 altruism is converted into anti-selfishness - the denying of benefit to self on the false assumption that reducing benefit to self automatically increases benefit to others (the reverse is true).

"Perhaps the idea that no matter what you do, you are doomed to help increase the complexity of the universe is a little odd sounding. Consider Microsoft. If their fervent practice of anti-altruism had been a little less intense, would there have been so many people stacked up, pissed off, and keen to help make Linux work? That's the hard way to do it. On the other hand, I doubt that Linus Torvalds is ever going to find himself without employment prospects at a high salary if he wants it! By helping others, he has (like a cosmological waste product) got loads of wonderful, complex opportunity falling right into his lap."

Profound stuff that I think many in the Free Software community will recognize the truth of. So many commentators on the "Linux phenomenon" try to sort developers' motives into the appropriate pigeonholes: are they selfish motives, or altruistic ones? Some commentators make it so far as to say that our motives include a mix of altruistic and selfish goals, but even this is only half the truth: many of us have goals that are simultaneouly altruistic and selfish. We know the world we live in is not a zero-sum game, and we are not constrained to choosing between helping others and helping ourselves. We recognize that some courses of action achieve both.

Oh, I can't resist joining in on the fun. Diary entries from the commandline? Yum.

Worked on php_pam a bit last night. This is all dead code, really, on account of the fact that I designed it badly and it works with neither pam_unix nor most web proxies; but there's still a lot of demand for the thing, so I went ahead and updated it to work with PHP 4.0.6+. Probably do a release today, then maybe the emails will stop. :)

Still need to find time to sit down and redesign the whole thing. Trying to cram this stuff into the PHP process is rather counter-productive.

The hard part about implementing a distance calculator based on zip codes, vab, is not the code; the actual code to do that is so easy, it's almost not worth the effort to find someone else's implementation for the borrowing. No, the hard part is getting an up-to-date zip code database that'll map to useful coordinates (such as latitude/longitude) that can be used in distance calculations.

I notice that Zipdy's description says it uses a database "created by the US Census". Does this mean updates are only available for this database once every ten years? The customers I've worked with who were using Zip codes for commercial projects (e.g., on their websites all seem to think that having all Zip codes in the database is important enough to pay a $1000 annual subscription fee to ensure it's updated quarterly. Certainly, though, Zipdy is a nice resource for those working within more limited budgets.