Contrary to what the Red Bean book says, when doing an svn switch --relocate, the repository actually is contacted. I don't believe it does much more than confirm that the new location is valid, but the documentation is probably incorrect and is certainly misleading.
Why am I just bitching here instead of submitting this as a bug report? Well, the issue in the docs has been around since 1.0, and I'm lazy. Plus I hate freedom, I know and even sometimes speak French.
Sorry about the "a bunch of new posts just became new" in the RSS feed, I didn't realize it had happened til I poked at my LJ friends page and there they were. They haven't been updated, in case anybody was wondering what wonderful new writing I'd added to previous posts.
I moved the blawg between physical hosts and also changed the base URL from /blawg/ to just /, since that's how all the Cool Kids do it and I wasn't updating anything at all besides the blawg anyway.
I figured while I was breaking things, I might as well go all the way and switch to the RC4 for MovableType 4.2. I made an unpleasant discovery that while 4.1 will backup and restore, 4.2 will only backup. So I had to export my posts from the old installation (fine, wanted to do that anyway) then import them into 4.2, but as a result I lost some settings - my pretty green! - and also some spam comments got accidentally published.
So far, I'm actually not impressed with MT 4.2 - despite its vaunted speed, and the faster CPUs in the new host, it still took 27 minutes to do a full publish run. Then again, it was starting from complete scratch, so maybe next publishes will be faster.
Hopefully this won't break MarsEdit! (It didn't.)
Update:
Changed the base dir back. Damn second thoughts. Plus it broke RSS feeds. I'm too lazy to figure out the rewrite-fu required, plus I was worried about MT stomping on stuff in the base dir. Maybe some day I'll do the reverse: redirect / to /blawg. And I have my green back, but not the BANPC blogroll at the side - I'll fix that later.
I knew I should have saved the post for later; don't feel like editing it now.
I ran into issues with m4 as well, trying to build autotools; this post was useful. As the poster says, grab the m4patches tarball, extract it and mv it to /usr/pkgsrc/devel/m4, then do a bmake makedistinfo ; bmake install. You'll wind up with a /usr/pkg/bin/gm4.
Note that I had to do this before I was able to install cvs, just so that I could update the tree. Maybe it would have been easier to start with sunfreeware's cvs. Since I wanted to get a non-vulnerable mutt, and I don't mind hanging on the edge, I updated to the latest pkgsrc with "cvs up -A".
And a final (I hope) edit: that was still insufficient to get libiconv built (so hardly anything else would go), so I backed up a little and found this old posting which talks about bootstrapping on Solaris. The relevant information is here:
- Now bootstrap pkgsrc (follow this sequence exactly):
1. Go to ..pkgsrc/bootstrap and run ./bootstrap
2. Add the following to /usr/pkg/etc/mk.conf:
CC= /usr/sfw/bin/gcc
3. Copy /usr/sfw/bin/{ggrep,gegrep} to /usr/pkgsrc/bin/{grep,egrep}
4. Temporarily set: LD_LIBRARY_PATH=/opt/sfw/lib
5. a. Go to pkgsrc/devel/libtool-base and do `bmake install'
b. Go to pkgsrc/converters/libiconv and do `bmake install'
c. Go to pkgsrc/devel/gettext-lib and do `bmake install'
d. Go to pkgsrc/devel/gmake and do `bmake install'
e. Go to pkgsrc/lang/gcc3-c and do `bmake install'
6. Go back into /usr/pkg/etc/mk.conf and change CC like this:
CC= /usr/pkg/gcc3/bin/gcc
and add:
GCC_REQD+= 3
I'd bootstrapped already, but the mk.conf + (re)building libtool-base with the LD_LIBRARY_PATH variable set was sufficient to get libiconv. I did not need to copy stuff from /usr/sfw/bin to /usr/pkgsrc/bin though.
The final path I used is
"/usr/pkg/bin:/usr/pkg/sbin:/bin:/sbin:/usr/bin:/usr/sbin:/usr/sfw/bin:/usr/ccs/bin:/usr/local/bin"
And after all *that*, mutt dumps core on the x86 platform. Go figure. Everything else seemed so reasonable too... doesn't inspire me with hope for everything else, but I guess I'll see.
Wednesday, July 16 2:07PM (ET) Atlanta Thrashers sign UFA forward Marty Reasoner.
That's Greene and Reasoner gone now, just Horcoff and Pisani left. At least we now know that Horc's going to be around a good long time, and I can't see Pisani blowing town any time soon.
So long Marty, it's been great. I have to say, I felt a little sick to my stomach seeing that. I know he's "just" a fourth-liner, business is business and he's left before via trade (and has also cleared waivers), but damnit, he was *our* fourth liner. Unless the money's stupid - and I can't see that it would be - good signing by Atlanta.
This is as much for my own reference as anything else. I have a pair of Solaris 10 machines upon which I would like to install various utilities (for different reasons). I'm not totally happy with Blastwave or sunfreeware, and I'm really unhappy with rolling my own for everything. I'm used to BSD and used NetBSD as a desktop for a year or so, about 7 years back mind. So I figured I'd give pkgsrc a try. One machine is sparc64 (a SunFire 280R), the other x86_64 (a SunFire X4100) so it should be a good test of both old and new architectures. If things go well, I may try this stuff out on a T2000 we have kicking around too.
I got bootstrap binary packages from here, although it turns out I didn't need them. I grabbed the pkgsrc sourceball from here, although I see there's a 2008Q2 directory dated a couple days later - maybe I should have grabbed that, I'm not up on my releases. Never mind though, new enough.
I'd done full installations on both my machines, so I had to remind myself about /usr/sfw/bin/gcc before I could bootstrap the pkgsrc install. I put everything in /usr/pkgsrc, so
cd /usr/pkgsrc/bootstrap ; ./bootstrap
A PATH of something like
export PATH="/usr/pkg/bin:/usr/pkg/sbin:/bin:/sbin:/usr/bin:/usr/sbin:/usr/sfw/bin"seems to work nicely.
Caveat: Solaris 10 does not have a cvs utility. NetBSD has cvs as part of its base, so it's not in pkgsrc. Now you have a chicken/egg: the first package I tried installing from pkgsrc (mutt) actually has a security vulnerability, so I thought I'd grab the latest pkgsrc via cvs. Guess I build that by hand.
Lots of posts about this, but like any petulant child or overgrown boy, Burke just can't stand it when somebody might get the last word.
I guess that's enough said; way to prove Lowe right on at least a few of his points though.
I have a SunFire 280R that I've been experimenting with, and decided to try Debian on it.
The release notes for 4.0 say:
3.2 Issues with booting on SPARCIt has been reported by several users that the installation CD fails to boot successfully upon the 'boot cdrom' PROM command, displaying the error 'Illegal Instruction'.
The apparent explanation for this problem is that it doesn't work because the machine had previously been rebooted from Solaris. The workaround is to power the machine off fully, and then boot it directly into the installation CD.
The problem was reported by users of various systems (namely, Enterprise 450, Blade 2000, Fire V240, Enterprise 250, Blade 100 and Enterprise 220R at the time of writing), so it is believed to be generic. Please let us know if you observe similar issues with your hardware.
It definitely is an issue on a 280R, but it's not because "the machine had previously been rebooted from Solaris". I had OpenBSD on this machine previously, and I still got the illegal instruction issue. And ran into the "qla2xxx is non-free" issue too. Very annoying, almost makes me wish I'd stuck with Solaris on this box, but sunfreeware isn't quite up to snuff for my purposes (harder to ensure you're running bug-free versions of things) and OpenBSD is a PITA to do mirroring of root partitions.
There's an interesting, if currently brief, discussion about the Verizon patching article going on over at the Security Catalyst Forums. Succinctly, Verizon's study demonstrates that patching isn't the be-all end-all of security measures. (Who'd have thunk it?)
I'm not sure precisely where I stand. I disagree somewhat with Andy Willingham's position, although it may be because I also disagree with his use of the word 0day. No patching will save you from a 0day, since by definition there *are* no patches for those. (Yet.)
My own experience in the last year or so has been that *every* compromise has been for reasons other than no patches. I will stipulate in advance that we've actually had very few recently, at least in my group, but none would have been prevented by patching, since there is no patch for human stupidity. All of them have been weak passwords on system accounts, so what would have helped us would have been a strong password and auditing policy. To counter that, the first time I really let Nessus loose on our network, I discovered (and pwned) a Windows machine that had not been patched in over a year. That machine was somewhat protected by a host-based firewall, but anybody on our networks could have seen it and done the same thing I did. (Yes, it's getting patched regularly now, and yes, I'm fairly sure nobody else *did* do what I did.)
I lean more towards Randy Armknecht's position: patching is great as part of a bigger security strategy, although I wouldn't be so dismissive as to say, "Nothing new here." If it's worth saying something once, it should be worth repeating it, and Verizon, at least, have done a fairly comprehensive study to back up the repetition.
This seems a very Windows-centric discussion. The unfortunate part of desktop Linux installations is it's difficult to keep all your machines patched, or to ensure that patches are installed at appropriate times. It took me about 3 weeks to track down and cause to be patched all our Debian-based boxes with vulnerable SSH host keys. Granted the time could have been drastically reduced if I was actually allowed to concentrate more time on security issues (it only took five or six hours, counting time updating old machines and notifying clients, then beating a few up into actually updating), but the time could also have been drastically reduced if the vendor made that sort of thing at all easy.
Furthermore, as a colleague at the UofT points out, sometimes patching is worse than the alternative. How many times have you applied the latest patch, only to have to apply a patch for the patch? Hands up, everybody who's lost time over that. What about when a machine is rendered unusable as a result of a patch? There was a particular MacOS 10.4 patch that checked for sufficient free disk space before installing, but got the figure wrong - so if your free disk space fell into a certain range, the patch would half-apply then fail and you got to figure out how to make your machine go again.
At any rate, I wonder how much time it's worth spending on ensuring a decent patch strategy, and how much time one should spend on other issues.
Tangentially related to a paper I wrote for my cognitive science seminar, I've been collecting references to stories on CBC, CTV, and other outlets that allow commenting. What I'm interested in is the popular reaction to matters related to what can loosely be described as justice. For instance, I've bookmarks to stories on Robert Pickton, Thomas Svekla, street racers who've killed people, Gregory Despres's trial, some child pornography cases, and so on.
Besides the obvious effect of the stories themselves - litanies of the indignities which we visit upon one another with depressing regularity - I've found a couple of other things. First is an extension of that thought; it's similarly depressing how people who are presumably largely Canadian citizens, my peers, presumably fairly well-educated and in the top 10th or 20th percentile worldwide, folks who are well-enough off to have both the free time and access to read and comment on these stories online, are able to so casually dismiss and pass judgement on other human beings with a minimum of information. Second is now every time I see a story that's likely to get such comments, I mentally grin and rub my hands in anticipation. Then I feel guilty, because each one of those stories means something horrible has happened to one or more human beings, and chances are something horrible is going to happen to at least one more person.
I have a rough idea where I'd like to go with this stuff, and I even have a rough idea about the direction in which I'd like to see our society headed. But given that we live in a democracy, it's fairly unlikely we'd even come close to what I'd like to see, at least in my lifetime.
Terry Jones doesn't apparently think much of Horcoff. After pointing out that Gilbert, Visnovsky, Hemsky, Penner, and Souray are all signed to 2012+, with Nilsson, Staios, and Moreau to 2011, he says "What they really need is the main man in the middle of the first line, scoring 40-plus goals and 50-plus points."
Earlier in the column he'd talked about trading Pitkanen *and* Horcoff and mentioned getting a scorer to play with Hemsky. Now, while I'll grant that Horcoff is not a pure scorer, he's not a dog's breakfast either, and he knows on which side his bread is buttered when it comes to coming back too. He's sort of like a reverse Mike Modano, progressing from a checking centre into a scorer. I think giving up on Horcoff now would be a huge mistake; he's got a lot to offer the team, and I don't think they need... well, I'm not sure who Jones has in mind that would be both available and a major upgrade to Horcoff at the first line centre spot. I can't think of anybody off the top of my head.
Perhaps Jones really did mean to package up Pitkanen with Horc. Trading Horc right now though, would leave the club awfully inexperienced up the middle; anybody as good as or better than he would be just as valuable to the presumed trading partner team as Horcoff is to this one.
What I'm dreading is Horcoff getting the Smythian brush-off. "Thanks for all the help, we'll trade you for some prospects now that your contract is expiring, we can't afford you now." I honestly don't see Horcoff asking for more than 4.5-5 million, and unless Lowe knows something about the player's shoulder that we don't, I think he's likely worth it. The team should be able to fit that in under the cap.
Is Lowe waiting to see what happens with his UFA offers before he sees how much is left for Horcoff? The prospect of seeing a pair of genuinely viable power plays featuring Hemsky, Hossa, Penner, Gagner, Visnovsky, and Souray is, no doubt about it, very attractive. But I'm not sure the cost of losing a guy who can play with literally anybody else on the ice and look good, a fellow who did nothing but help lead the team to the Stanley Cup Finals a couple of years ago, who has improved his points per game total from day one, a guy who has paid his dues at every level and is probably one of the smartest players on that team is worth it. If somebody told me "you're the Oilers GM today and you have to make a choice: sign Hossa and trade Horcoff, or keep Horcoff, what will it be?" I'm not sure I could let Horc go. Let's hope that Lowe finds a way to fit Horcoff in, whether or not Hossa takes that huge offer.
Media outlets, both mainstream and idiots like me, are abuzz with the news that Henry Morgentaler may be named to the Order of Canada.
Love him or hate him, like the news or not, it seems pretty clear that the various commenters on the issue are dead wrong. The G&M story quotes the executive director of the Catholic Civil Rights League as saying, "[w]ith this choice, the one thing that everybody really agrees on about Morgentaler is that he is a very divisive figure." A Liberal MP is noted as saying "if he is admitted to the order, it will polarize Canadians." (That's the Globe and Mail's wording, it's not a direct quotation.)
Canadians are already divided and polarized around the issue of abortion. Morgentaler's appointment would merely bring that polarization back out into the light; that division is there whether or not we choose to acknowledge it. Poll 100, 1000, a million, or thirty million Canadians; I'm sure you'd see that most Canadians already have an opinion on abortion and would argue their point of view quite vociferously. Morgentaler's appointment can't cause what's already done, so in that respect the anti-abortionists are either deluding themselves, or trying to delude those few still on the fence.
Edit an hour later: he's admitted. Bring on the polarization.
First, the Other Team news. Sounds like Pittsburgh's signed Dupuis to a very reasonable 1.4mm/3 year deal. I've been a fan of Dupuis for a while, I always thought he was a suitable Dvorak supplement/replacement, and at 1.4 a year, the Penguins have him signed for a very reasonable number. Too bad, although I guess it's understandable he'd show the Pens some loyalty (and maybe leave some money on the table).
Torres for Brule? It's hard to know what to make of this. Salary dump, getting rid of a problem child, who knows? I do know his even-strength is good, and that's hard to find in a guy who can pot 15-25 a year while steamrolling players. This is almost like a mid-90s Oilers trade. I don't much like it, unless it's going to be matched with a signing or a trade later - and I don't believe that it let the Oilers trade for Cole, that was going to happen anyway. If he was causing problems on the team, then sure, ditch him, but without knowing that sort of inside information, this trade stinks. The team was already soft up front, Glencross is not certain to return and Moreau's brittle enough the last couple of years that even glass looks tough. Combine that with losing Greene on the back end while adding Vish, and this is a team ripe for the Regehrs and Prongers of the world. Brule's addition has got to have Schremp looking over his shoulder though - poor guy can't buy a break, he should just get traded already because it's pretty clear he's going nowhere fast in this organisation.
Pitkanen for Cole is understandable. I love the Finn's style of play, when he's healthy. And therein lies the rub. He missed an awful lot of 1-2 game streaks this past season, and it didn't sound like he had the coach's confidence by the end of the season. With Torres out, the Oilers needed some size on the wing, and Cole could fit the bill. I'm not as concerned about his previously-broken neck as some commenters are; Gary Roberts had a broken neck too, and that was about 600 regular season and 177 goals ago. Cole's broken neck didn't prevent him from playing 73 games last year and 71 the year before, or from winning a Stanley Cup against the Oilers. 29+22 goals and 61+51 points is doing pretty well. Cole will be a good Oiler, and Pitkanen was obviously the odd man out with two other $5m+/year blueliners on the team.
