Posts for December 2009

2009-12-04: backport 1.22

This is a new version of my script to automate some of the work of backporting Debian packages. This version sets the urgency of the changelog of the backport to the highest urgency of the changelogs included in the *.changes file and allows adding an additional entry to the changelog of the backport. It also handles a distribution of etch-backports properly.

I unfortunately had to revert the change in the previous version to use -nc, since unaccountably using -nc to suppress running make clean before building the source package forces a binary package build that doesn't include the source package in the *.changes file. I filed a bug against dpkg about this. In the meantime, you'll continue to need to satisfy the requirements for debian/rules clean outside of any build chroot.

This version also adds some automated backporting logic that I needed for the Shibboleth packages.

You can get the latest version from my miscellaneous scripts page.

2009-12-05: New book haul

It had been quite a while since I'd put together a new book order and there was new, interesting things out by several authors I follow.

Catherine Asaro — The Ruby Dice (sff)
Elizabeth Bear — By the Mountain Bound (sff)
Philip K. Dick — VALIS and Later Novels (sff)
David Langford — Starcombing (non-fiction)
China Miéville — The City & the City (sff)
Carlos Ruiz Zafón — The Shadow of the Wind (mainstream)
Karl Schroeder — The Sunless Countries (sff)
Jo Walton — Lifelode (sff)
Jo Walton — The Prize in the Game (sff)
Ysabeau Wilce — Flora Segunda (sff)

Now to make more time to read. It should help that I'm mostly over my cold and hopefully will stop sleeping twelve hours a day.

2009-12-05: Usenet RFCs finally published

Last month, after a rather extended AUTH48 period with a variety of different annoying holdups, the new Usenet (Netnews) RFCs were finally published, obsoleting RFC 1036. RFC 5536 is the new document describing the Netnews article format, and RFC 5537 is the new document describing the Netnews protocols. These join the RFCs for NNTP (3977, 4642, 4643, and 4644) as finally fairly completely describing a current Netnews implementation.

I was the document editor and author of much of the text in RFC 5537, based on earlier work by Charles Lindsey. Many thanks are due to Charles Lindsey, who was a co-author on both documents and who kept pushing to get them both published despite multiple roadblocks and delays.

There was originally supposed to be a third companion document that described best practices and recommended implementation strategies, but the working group long since ran out of steam for producing more work. Hopefully someone will pick it up, since a best practice document would be very good to have.

I've updated my Netnews article format and protocols web pages for the new RFC publication and to include the Internet-Drafts of son-of-1036, which is in the process of being published as a historical document. I also created a new reference page for Netnews-related standards and included copies of many of the most relevant e-mail standards there, since Netnews is now officially based on MIME and the most current e-mail standards.

It's been well over ten years since the document revision process started. USEFOR wasn't the longest-running IETF working group, but it was close. RFC 5537 holds an interesting distinction: it represents the largest numeric gap between one RFC and a subsequent obsoleting RFC (by a margin of only 1).

2009-12-08: filter-syslog 1.22

This version, in addition to fixing some minor bugs, adds support for a regex matching the program name. If the program name is surrounded by slashes, it's now considered a regex, and the filter regex on that line is applied to all entries from any matching program.

You can get the latest version from the filter-syslog distribution page.

2009-12-10: Things that piss me off

The news has been rather full, recently, of things that are really getting under my skin. I'm going to rant about them here so that hopefully I can stop turning them over in my head and move on to doing something more productive.

The things that are pissing me off:

2009-12-11: Peter Watts attacked by US border guards

Signal boost.

While returning from a trip to the US (in other words, going to Canada) on December 8th, Canadian SF author Peter Watts was attacked, beaten, pepper-sprayed, and imprisoned without access to legal counsel by the US Border Patrol.

See also his post on the subject.

The government's case has obviously not yet been presented. However, given the past behavior of the US Department of Homeland Security, I find Watts's description of the event entirely credible.

This is an outrage. This is the kind of thing that happens in third-world dictatorships.

See also Making Light and Boing Boing.

John Scalzi's post has information on how to donate money to Watts to get through this legal fight. I will be giving him money. I also plan on calling my US Representative and Senators when I'm no longer at work, as this sort of treatment of a Canadian citizen warrants a Congressional investigation and heads rolling, not only at the agent level but at the Homeland Security management level.

Even apart from basic questions of human rights, Homeland Security is going to systematically destroy the tourist industry in the United States if this sort of behavior is not stomped on immediately and severely. There are border patrol guards who are probably still on duty today who are common criminals and belong in prison for assault.

ETA: The government's side of the story (now a dead link). This is pretty much what I expected, and with the data available so far, I feel confirmed in my initial reaction. This level of force is completely disproportional and flatly unacceptable short of Watts pulling a weapon on the agents, which they are not claiming that he did and which therefore he clearly didn't do.

I'm highly skeptical of this supposed "choking" of a police officer after they pepper-sprayed him. This reads like the standard cover-your-ass charges filed by the police after they beat someone up. Once they've provoked the struggle by grabbing you and have done something as painful and disorienting as pepper-spraying you, chances are you're going to physically react in ways that they can conveniently misinterpret and use as the basis for charges.

People elsewhere have pointed out that contacting Congressional representatives is probably better done Monday morning since nothing will happen during the weekend.

2009-12-20: Year close

Snow on rock

In honor of the blizzard that hit the east coast, have a snow picture. It's not snowing around here (although that would be fun, it's unlikely), but it's gotten fairly cold and it's expected to be wet this week.

I haven't been posting pictures, regular journal entries, or reviews for a while because, after getting back from vacation and catching up, I had company for a great week of playing video games. Immediately after that I got a cold, and then the rest of the time was spent trying to get as much done as possible before the winter closure.

But now, I'm on vacation. I've already read one book, I've written a couple of book reviews, and I'm sleeping in luxuriously late. I have a huge backlog of non-Stanford on-line "work" that I want to do, mostly around Debian and software maintenance, and I'm hoping to spend the next couple of weeks divided between reading and doing that.

This has been a hell of a year. Having this couple of weeks to gather my thoughts, wind down from the year, think about what I want to focus on next year, and hopefully do a bit of catch-up and prioritization feels very valuable. But no pushing to do anything tomorrow; one more day of pure relaxation seems in order.

Hopefully this will mean that I'll be posting more regularly, and adding more things to my web pages, over the next couple of weeks, since there are some habits I'd like to work on rebuilding.

2009-12-21: Vacation transition

Weeds on grey

Weeds silhouetted against a grey and cloudy sky along the 804 trail in Yachats, Oregon. It's starting to feel like winter, so I wanted a nice winter picture.

It always takes me a few days to kick into vacation mode, sort out how I'm going to typically spend my time, and relax enough that I can "waste" time on taking on larger projects that don't immediately pay off. Today was a great day for all of those things. I didn't end up reading, which I'd intended to, or take a second walk, but I did putter around on Lintian for much of the afternoon. I've written the first draft of a new module that moves the current Tags module into the Lintian::* namespace, documents it, and cleans up the interface to be object-oriented. It requires some API changes elsewhere and I have to figure out how to test it, but the first draft is done. (Strangely enough, I got started on this while trying to figure out the best way of implementing a mode that checks only the ftp-master tags.)

Other than that, I helped my mom make macaroni and cheese, only partly successfully, and showed her a few things on her computer. I crossed three things off my to-do list, which is excellent given that today was not supposed to be an accomplishment day. And I watched football. And now I'm feeling far more relaxed, able to take each day at a time, and thoroughly on vacation.

Goal for the next week: take each day at a time, do plenty of reading, follow whatever feels the most fun, and don't think about anything beyond the horizon of vacation.

2009-12-21: Picture sorting

Gleaming rocks

Sunlight on rocks and sand near the end of the beach outside Lincoln City, at the point north of Road's End.

Today was not horribly productive, but it didn't need to be either. I wrote another review, which means I've been sticking to a pace of one per day. I also finished debugging the Lintian coding work that I did yesterday, so although some additional testing is needed, it's about ready to commit. Then I can go back to working on solving the problem that caused this detour.

I also started sorting through older pictures. I like to have a file of notes accompanying each picture and to make notes of which pictures I want to eventually post on the web, but I've gotten way behind on that. I only did one batch today, but hopefully I'll work up some momentum and do a batch or two a day during vacation and get closer to caught up, as this has been nagging at me for a while.

Expect to see me post a lot of beach pictures as I go through them.

Other than that, I read. I brought a pile of books with me on vacation, and while this won't be a reading vacation to the degree that the Oregon coast trip is, I'm trying not to spend all day in front of the computer (successfully so far). If all goes according to plan, there will be a flow of reviews over the next couple of weeks and possibly a backlog for posting in the new year.

2009-12-23: Plod and poke

Overleaf under clouds

A picture of the Overleaf Lodge in Yachats, Oregon on an overcast day. I love the lighting effect, although it made the foreground just a little too dark and playing with brightness and contrast wasn't helping much (so this is entirely undoctored). But it's only just too dark; I wouldn't want the foreground to be clear.

I'm concluding that as exhausted as I thought I was, I possibly was even more exhausted. I'm also starting to understand why I was getting nothing done the last few weeks before vacation. Four days into vacation and I'm still sleeping over ten hours a day (although that's down from the eleven or twelve at the start). I'm also only slowly getting into the rhythm of working through projects.

This was a hell of a year. Now if I can just figure out what to change internally to help me handle these sorts of years better, since with a few exceptions the external causes are unlikely to change tremendously.

But, I have finished setting up log parsing for my personal domain, partly out of curiosity and partly so that I understand how awstats works so that I can do similar things for work. That took rather longer than the one hour my to-do list claimed it was going to take. I've also made reservations or started the process for my vacation in February, which always requires some pushing for me to actually make the reservations. (I love the vacation, but I hate traveling and I hate committing to spending money, both irrationally compared to how much actual problem either causes, so I tend to have to talk myself through it.) I also read some more and walked twice instead of only once today, which I've been wanting to do for a couple of days. All in all, not bad for a vacation day.

No more picture sorting today, and no more Lintian work. Maybe tomorrow I'll feel like making a push on Lintian, but I'm playing it by ear.

2009-12-23: Lintian work committed

Layered bank

This is one of the many dirt and rock banks between the 804 trail and the rocky beach in Yachats, Oregon. My parents and I spent some time picking agates out of this bank. There were dozens of them, right on the surface, although you had to see them from the right angle.

I committed the Lintian work that I'd been poking at the past few days and pushed it all to Alioth. The old Tags module is no more, replaced with a new Lintian::Tags module that, while it still needs work, is at least better-documented and better-structured. I also added a flag to Lintian to test only the tags that result in automated rejects from the Debian upload queue, which is the problem that started me working on the Tags module in the first place. One more thing crossed off the to-do list.

Other than that, I didn't get much done today on the computer. I helped my mother decorate Christmas cookies, finished a book, helped my dad with a few things, and wrote another review. Projects are still moving a bit more slowly than they feel like they should, but I seem to have gotten to the point in the vacation where I at least know where all the hours are going to. I also didn't sleep ten hours last night, so maybe I'm finally caught up on my sleep debt.

No picture sorting today either. Maybe tomorrow.

2009-12-26: Lintian 2.3.0

Continuing a Saturday after Christmas tradition (well, two years in a row, at least), have a new Lintian release. This release welcomes Raphael Geissert as a member of the Lintian maintenance team. He's been providing good patches for quite some time, so this was overdue.

Releases have been infrequent lately, so there was a rather large bug backlog built up. I also took some time to rewrite the tag handling module and move its namespace around, which is largely why the larger version bump.

This version has a few new command-line switches. --suppress-tags and --suppress-tags-from-file allows disabling specific tags that one never wants to see, added primarily for Emdebian to be able to turn off particular tags but possibly of use in other situations. -F or --ftp-master-rejects shows only those tags that might result in an automated reject, although currently doesn't distinguish between ones that can be overridden and ones that can't.

I'm going to update lintian.d.o in a few moments and start an archive-wide run, since this release fixes a bunch of bugs and should hopefully fix handling of multiple versions of the same package in the archive.

There are a lot of code changes in this release, so it may not be completely stable. I'm planning on making another release next Saturday.

2009-12-27: Halfway mark

Driftwood whorl

Driftwood along the 804 trail in Yachats, Oregon. I still haven't been doing more picture sorting, but thankfully I'd already identified quite a few pictures I wanted to post.

It's about halfway through Christmas vacation, so a good point to take stock of how I'm using it, what's been working, and what hasn't. I also am using this vacation to clear my head and refocus my life, so I'm now through the unwinding phase and moving into the phase where I start working out goals for after I return home, and more generally for next year.

One thing that's worked well is catching up. I haven't been putting large chunks of energy into it the way that I was hoping that I would, but I've been fairly successful anyway. My non-work to-do list has shrunk considerably, particularly after a bout of productivity on Saturday, and I'm completely caught up with Lintian for the first time in quite some time (last May at least). That's taken a lot of psychic pressure off.

I've been doing well with reading, although my pace of reading has now dropped off. I've been doing reasonably well with reviewing. I still have a two-book backlog, but I feel like I've gotten that sorted out and am getting back into a rhythm. For some reason, if I go for too long without writing a review, writing reviews starts feeling intimidating and I have a harder time doing it. Writing them fairly regularly seems to reduce the pressure.

Pictures are somewhat in the not working category. I haven't taken any on vacation so far, and I haven't done anywhere near the sorting of older pictures that I was hoping that I would. Maybe something to focus on for next week.

General relaxing is going very well, though. I'm feeling much calmer, much more centered, and much more emotionally grounded. I've shed the constant feeling of pressure and need to push myself, and I think I'm getting into a fairly good mindset for planning for next year. Tuesday or Wednesday, I'll sit down for a while and do some serious thinking about that.

My main goal for the rest of the week is to spend some time reading each day, write a book review each day, and cross a couple of things off my to-do list each day. That's fairly aggressive, but if I do that, I'll clear a ton off my plate, be completely caught up with reviews, and take a big step towards re-establishing some habits that I want to have going into the new year.

I hope everyone who celebrates it had a wonderful Christmas, and everyone who doesn't celebrate it had a wonderful and hopefully long weekend.

2009-12-27: AFS SRV record draft

As part of my work on OpenAFS, I wrote an IETF Internet-Draft specifying how to use DNS SRV records for service location with AFS. This is intended to replace the experimental AFSDB DNS RR that was specified in RFC 1183. I've now put the draft up on my AFS notes page, and of course you can get it via the IETF I-D tools pages.

Current status: I have approached the application area directors to ask the correct path forward for this document, and one of the ADs is currently reviewing it to see if he wants to sponsor it as a standards-track document. It's already been reviewed by the AFS community and the text should be reasonably close to final, although there are some typos and grammatical errors that are fixed in my copy but not in the latest published draft. (Thanks to Julien √ČLIE for the proofreading!)

2009-12-28: podlators 2.3.0

This release is rather overdue. The last release was in January of last year, and the first changes that went into it were made in February.

As of this release, anchor text in L<> tags with URLs is now supported and displayed as "text <URL>". This preserves the anchor text so that people can use anchors for nicer output from POD HTML modules.

Thanks to David Hull, the Pod::Text subclasses now correctly format short item tags that have embedded hidden formatting. Previously, the length of the tag, used to determine whether it would fit in the margin of the paragraph, mistakenly included the invisible formatting codes.

Pod::Man has multiple, varied fixes in this release. Expansion of all formatting codes in the NAME section is now suppressed, keeping the NAME section clean for tools like lexgrog. The only markup used is regular double-quotes around C<> text. Handling of backslashes in X<> formatting codes is fixed. Pod::Man also no longer uses POSIX to make it easier to use as part of bootstrapping processes.

The Pod::ParseLink module now also handles anchor text for URLs and anchor text of "0", which previously was treated the same as missing anchor text.

Because of the changes around use of anchor text in L<> formatting codes, podlators now requires Pod::Simple 3.06 or later.

You can get the latest release from the podlators distribution page.

2009-12-29: Staying caught up

804 trail

Near the southern end of the 804 trail. We're generally in Yachats around Halloween, and the local community apparently does something or other. This year, they were marking trails with streamers, and you can see one in the middle of this picture.

So far, so good on my daily goals. I spent some time reading, wrote another book review, and crossed two things off my list (actually, more like four). Productivity today was working on podlators and finally getting another release out, which took care of two active goals and a couple of other future work goals.

I'm feeling much more on top of things. Catching up on Lintian helps a lot. I will also need to catch up on Debian Policy work to feel entirely on top of things, and that's more challenging, but I can feel the difference since the start of vacation. The hard part is sustaining this, or at least not falling so far behind that I put things on my to-do list to work on and then don't touch them for nine months. As with reviews, a lot of things seem to get harder to do if I let them sit for too long. I suspect that if I can get better at identifying which things are like that, distinguish them from things that don't get harder if they sit, and then prioritize accordingly, I'll feel positive more frequently.

There's just no substitute for a good vacation to clear the decks, though.

2009-12-29: control-archive 1.2.0

control-archive is the code that drives the ftp.isc.org archive of control messages and the active and newsgroups files published there, as well as the source for the control.ctl file distributed there and used as the default in INN.

There are a few minor tweaks to the reporting code in this release and some better debugging, but mostly this is a data update, including all of the data changes since January of 2009. I was meaning to do these releases more often.

You can get the latest version from the control-archive distribution page.

2009-12-30: Fun with Git

Wall climber

I've been posting a lot of grey photographs lately, so here's something with some color. I'm not sure what variety of ornamental climbing vine this is, although I suspect something related to Boston Ivy is more likely than true ivy.

Mission accomplished again for today: I read some, wrote a review, and crossed two things off my to-do list (and did a software release that wasn't on my to-do list at all). I even went for an extra walk.

Much of the productive part of today was spent fixing the Debian packaging repository for OpenAFS. This started out as a CVS repository, was converted to Subversion, used svk for a while, and now is a Git repository. In its days as a Subversion repository, we checked each upstream tarball release directly into the repository, as well as maintaining a vendor branch of the contents of it, so that one could build the Debian packages directly from a checkout of the repository. Now, with Git and pristine-tar, this isn't necessary and means the repository takes up tons of space and takes forever to clone over small links.

Git has a tool for rewriting repositories, git filter-branch. This was the first time that I used it, but it works surprisingly well. I wrote a little shell script that moves the tarballs out of the repository and, if the source in the repository was in an openafs subdirectory, moves it all up one level and deletes the subdirectory. Then I ran that as a tree filter, used cat as the tag-name filter, and added a mini Perl script to remove the git-svn-url bits from the commit messages that I'd mistakenly added when I initially converted the repository. I applied that to all revisions, let it run for a while, and it worked like a charm. I then committed the tarballs with pristine-tar, using the relevant upstream tag as reference, and cloned the new repository to a fresh one.

It reduced the size of the repository from over 300MB to under 40MB. Totally worth it for an afternoon of work, and now I know a lot more about manipulating Git repositories.

2009-12-31: Goodbye to 2009

Lifting fog

Fog lifting from the beach near Chinook Winds in Lincoln City, Oregon. It seemed like a good picture for the turn of the year.

This is the last journal post of 2009. It's been a hell of a year. While it was productive in places, and parts of it were extremely good, it was also even more stressful than 2008. I'm happy to see the end of it.

The year started with a serious illness in the family that cast a bit of a shadow over much of the year. I was also sick a couple of times, once rather miserably, and felt overly stressed for much of the year. Partly as a result, this was my worst year for reading and reviewing since I started, and I struggled with a few other personal goals.

Much of my work year was spent working on virtualization, which was ultimately successful but very frustrating along the way. Work also saw strange management prioritization decisions and a few rushed projects and ended with proposals for reorganization that were distracting and promised more hard overhead work to come that will take time away from projects. One major upgrade project is still looming.

But, that said, I'm still in the job I want to be in, doing the work that I want to be doing. I also got a lot of volunteer work done last year, particularly early in the year, with numerous Lintian releases and two Policy releases for Debian. The Shibboleth packages for Debian are in use and working. And the upcoming work reorganization, while it will drain energy in the beginning of the year, has some potential to give me more focus and let me shed some of the annoying work I don't want to do (such as more virtualization).

I've spent the last couple of weeks clearing my head and putting the year behind me, and I spent today doing a comprehensive review, cleaning out my to-do list, and starting to write down the things I want to accomplish and work on for next year. The combination of extended vacation and the feeling of list-making and time change in the air always makes that process easier. Right now, I'm feeling quite positive towards 2010, and know I will be in even better shape once I get over a few early hurdles in the year.

I won't be posting any lists of new-year resolutions or goals. I think goals are better managed privately, to avoid creating artificial constraints and pressure against rethinking and reconsidering them. I only have one that I want to mention publicly: In this next year, I want to waste less energy on meaningless arguments. Before entering an argument, I want to think carefully about whether the argument will serve any purpose: a decision that matters, a useful exchange of views, a broadening of my understanding or thoughts. If it won't, I don't need to participate. I'll waste energy entering the argument, and I'll waste even more energy and emotional calm trying to exit the argument. It's okay to sit on messages for a while until I can decide whether I really want to send them, and it's okay to let meaningless arguments slide away without a reply.

I hope everyone who reads this has a very good new year, and if you also do mental brush-clearing and realignment this time of year, I hope you've reached a thorough and rewarding ordering of your thoughts.

See you in 2010!

Last modified and spun 2014-10-22