I'm trying to install Trac with NetBSD pkgsrc, but clearsilver is the big wall for me. cleasilver depends on GNU find and gcc, so I need to modifide some codes. GNU find is called from rules.mk, so I installed it and replaced "find" to "gfind". Then, util/neo_err.h uses __PRETTY_FUNCTION__ macro, it is a gcc extension, so sunpro compiler can't handle it. So i replaced it to __LINE__, it is not so good solution but better than nothing.

So you read the documentation for IO::File and see:

open( FILENAME [,MODE [,PERMS]] )
open( FILENAME, IOLAYERS )

so you write:

my $rules = new IO::File('debian/rules','w', 0755);

and wonder why it hasn't changed the permissions from 0666. Stracing confirms it is opened 0666:

open("./debian/rules", O_WRONLY O_CREAT O_TRUNC O_LARGEFILE, 0666) = 4

A bit further reading of the documentation you discover:

If IO::File::open receives a Perl mode string (">", "+<", etc.) or an ANSI C fopen() mode string ("w", "r+", etc.), it uses the basic Perl open operator (but protects any special characters). If IO::File::open is given a numeric mode, it passes that mode and the optional permissions value to the Perl sysopen operator. The permissions default to 0666. If IO::File::open is given a mode that includes the : character, it passes all the three arguments to the three-argument open operator. For convenience, IO::File exports the O_XXX constants from the Fcntl module, if this module is available.

and the correct way to write this is

my $rules = new IO::File('debian/rules',O_WRONLY O_CREAT O_TRUNC O_LARGEFILE, 0755)

Thank you very much perl for ignoring the permission parameter when you feel like it. Update: Steinar, yes sorry, I did have 0755 rather than "0755" originally, but changed it just to check that didn't make a difference and copied the wrong version. I've changed the post to have the right thing.

% strace -eopen perl -MIO::File \
   -e 'my $rules = new IO::File("foo","w", 0755);' 2>&1   grep foo
open("./foo", O_WRONLY O_CREAT O_TRUNC O_LARGEFILE, 0666) = 3
% strace -eopen perl -MIO::File \
   -e 'my $rules = new IO::File("foo","w", "0755");' 2>&1   grep foo
open("./foo", O_WRONLY O_CREAT O_TRUNC O_LARGEFILE, 0666) = 3

Incidentally, 0755 and "0755" are different:

 perl -e 'printf("%d %d\n", 0755, "0755");'
493 755

In an effort to put an end to the endless time sink of hardware problems on the small cooperative ISP I run, I’ve decided to outsource the responsibility of actually taking care of hardware and move most of the operations to a virtual private server. Part of our hosting is actually already on a VPS, but the service there has been lackluster: sometimes disk space is mysteriously unavailable; we don’t always get the promised amount of memory; throughput is slower than we’d like; customer support usually gives us cryptic responses about what was wrong and how they fixed it after a few hours; etc.. (Not to give them too bad a name, but the provider here was SPRY.) We decided to find a new provider. Lots of people had good things to say about RIMU Hosting, and after just a day it’s becoming clear to me that they are an excellent provider. Tech support (even before we signed up) was incredibly responsive; the packages of virtual servers or dedicated hardware seem like an excellent value; and my first impression is that there is none of the VPS flakiness that seemed to permeate our prior provider. It looks and feels like a real machine: top provides an accurate task list that reflects the memory we really have, dmesg gives realistic kernel-like output, etc. In such a short period of time, I can’t genuinely claim that RIMU Rules, but I will say that they rule so far. They also have a Debian developer discount, as well as a discount for hosting multiple machines with them.

So I was sitting at work bored the other night. I, for some reason or another, started playing with my razr. I had installed an ssh client on it a long time ago and never got it to connect. While browsing the web I had noticed that Opera had a mini version for cell phones. Knowing that the current wap browser blew donkey nuts, I decided to give Opera Mini a try. It is truly one of the best ways I have found to browse the wen on a cell phone. It can render normal, non-wap pages. It can even load images. Surprisingly, it is a lot faster then the normal text only wap browser. Opera also featured a option to enable small text which allows more to fill the small screen. That is the biggest seller for me. The next cool thing I found is that it has a page-up and down key via the joystick. I always did hate the browse by selecting the links to go up and down. All in all, I am happy I had added the unlimited internet plan to my service cause I see myself using Opera Mini a lot in the future. Now I can keep up on all my news when I am on the road or at the bar. Probably more the latter then former. Other then that, went to New York City this past weekend. Had a good time. Me and the sister went up to just hang out and relax. All in all was a good trip.

Mike assumes that zlib is valgrind clean. In actual fact there are a number of cases where loop unrolling in the zlib library causes the mentioned “Conditional jump or move depends on uninitialised value(s)” warnings. These are safe since the results of the comparison are subsequently ignored but obviously valgrind doesn’t know that. This is discussed briefly in the zlib FAQ. It is safe since the memory being read was allocated by zlib (so it knows the pointer is valid) and the results of the calculations are discarded by bounds checking done after the loop exits. Recent valgrind packages should have rules that suppress the spurious warnings from zlib, though I don’t know if the rules manage to cover all the relevant cases or not and obviously it is still possible to pass uninitialised data into zlib.

Throughout the last couple of years, I've had some bad decisions. I decided to start using my credit to work for me, and to buy things which I couldn't afford. Partly this was because I had a horrible ex, who I thought I should do things for, and partly because of just living outside of my means for a certain period of time. The long and short of it is, I've gotten myself into a bit of credit card debt. I'm not in any big trouble, just needing to pay it down so that I can get on with using that money for other things. There is a couple of web applications that I have in the works which will help me remind myself, but that's neither here nor there at this point. So, I have all of this credit at high interest rates, surely there is someone out there that will look at my credit and finance it at a lower rate, make some money off of me, right? I'm thinking that what I need is a consolidation loan - something to lump together my credit card debt and give me one monthly payment. What I call this, is debt consolidation. Thinking we are in the age of the internet, I type "debt consolidation" into google and give it a try. This is my first mistake. Apparently "debt consolidation" doesn't mean that you are consolidating your debt, it instead means that you are going into what I would call debt negotiation. This means you decide of your own volition that you are going to stop paying your bills, and try to negotiate with the credit card companies to pay off your debt at a fraction of your actual debt. But I don't know this, so I nievely fill out a couple of forms at "debt consolidation" sites. This gets pretty long, so I put in a break. Read on, MacDuff. The Scam The first people who got in contact with me were some people from Credit Solutions. Let me be clear on this: I'm not making broad generalizations about the company here - I can only speak of my own experiences. The person who I got on the phone went through a whole spiel, "explaining" how it works. He was very high-pressure and very fast talking, just from the start I didn't want to buy anything from him because he seemed like a used car salesman who decided that it was better to make money from nothing than from goods. Here's the explanation he gave me: You've probably been paying your cards down already, so you've paid 70-100% of the cost of your original loan. You stopped paying your cards, wait 3 months, and then they start negotiating with their "powerhouse lawyers". Then they negotiate you to pay your debt at 40-60% of the current balance, and the credit card companies get to take the remaining 40% as a tax loss. Now is when the math gets strange. The man on the phone attempted to say that this meant that the credit card company got 140% of the current balance. This doesn't make sense to me, especially because I'm a smart credit card consumer, and most of my debt is on 0% APR promotional rate cards. This means that they have made at most 3% of the balance from me, based on balance transfer fees. The man on the phone tried to dance around the issue, saying that they actually will get the 40% tax loss and that means they'll still get 140%. 140%? This doesn't make sense - 60% plus 40% is 100%, not 140%. Lets add to this that a tax loss is not the same as profit: a tax loss only gets you like, 30% of your loss. At this point, I've decided that I'm being taken for a ride already, but I ask how it works anyway. Here's how this particular company is setup: I pay into a savings account (of my own), automatically deposit around $200 a month, and they take money out. For the first three months, they take the WHOLE amount. After that, they take half the amount, and leave the other half. They then start negotiating with your creditors, and when they make a deal, they tell you "pay them now". In the end, they take 15% of the settled cost of your debt. On $6k settled debt, this is another $900. I really didn't want to do it now, mostly because of the high cost. The man kept coming back to my objections with "you can't borrow your way out of debt". Of course you can't borrow your way out of debt. That's stupid. I wasn't planning on borrowing money and then ending up with less debt. I was planning on borrowing money so that I'm not paying so much in order to have the debt. Apparently he thought I was an idiot, because he kept repeating it to me. He sent me a little packet of stuff, which laid out the whole plan. After that, he gave me a call back. This is when I got really defensive. I asked him point blank: "What is the worst thing that can happen to me if I go with this plan?" His answer: "You stop paying your bills." On the surface, this is true. The worst thing that can happen is that I stop paying my credit card bills. He neglected to mention the fact that ceasing payment on my credit cards would ruin my credit, cause insane fees for lateness, and possibly send me to forced arbitration with the creditors. Every creditor that I have ever had since 2002 has had an arbitration agreement, which basically would end up in garnishment of wages. The Scum About a week after my first bad experience with a debt "consolidation" company, I got a second call from a completely separate company. They introduced themselves on the phone as "Christian Financial Services" or some such. Basically the same deal, but this was a low-pressure sell instead of a high-pressure sell. He was much more reasonable, and talked me through the whole thing at my own pace, taking his time to lay everything out that I could understand it. It was basically the same deal, 15% and they negotiate for you, but they set up an escrow account instead of you doing it yourself. When I asked the same "worst situation" question, he gave me a realistic outlook, and this company is in a much better standing. If the creditors don't like their negotiation and take you to arbitration, you actually get representation from their company to help you out. I was almost willing to do it. Then I went and looked at their actual company online, and discovered that they are a chamelon of some sorts. They were originally called something similar to "USA Debt Consolidation" or something otherwise patriotic. Apparently they decided that God could be on their side, at least it's good advertising. What good company has to change their name to get religion on their side? A good company would not need to change it's name. The Sweet Some people have gone with these companies, and they have actually negotiated with the creditors in order to pay off your debt at a fraction of the price. If it happens, that's the sweet part of this. Just given an example of $10,000 credit card debt, it is possible to pay around $6,000 and have all of your debt paid off. It's basically saying that you can spend money that you won't have to pay back, ever. There is some hit to your credit rating in the short term, but in the long term, it starts going the other direction. The hard part is finding a company who will do this and who you like. I had an even harder hurdle to surmount - I don't really believe that it is ethical to do this. I spent the money, I should pay it back - that's the deal I made with the company when I signed up for the card. To me, I need to go through with paying them back. Luckily, I came into this with a decent credit rating, and will probably go out with a similarly decent one. Right now, I am a prime target for anyone who wants to give me a large loan at a decent rate - if I can pay less for the money I spent, I'm savvy enough to do it. The Alternative There is one other option which I looked into, and will represent the only company that I will link through this diatribe: Consumer Credit Counseling Services or Money Management International are truly non-profit and have good word-of-mouth from anyone I have talked to. They will not negotiate to settle your debt, but they will work with the creditors to get lower percentage, and help you through your budget. The couple of talks on the phone that I had with these people were very productive, and I got some stuff in the mail to help me out. You pay down the entirety of your debt. The downside is that you get a good size hit on your credit rating. My Choice In my position, I decided that I was on a good track myself already - I'm paying off my cards smartly, taking the highest APR out as quickly as possible. On my current plan, I will be completely debt-free in 3 years, less if I get some windfall of any sort. One of the bonuses of having a good credit rating is that people are willing to loan me money at promotional rates for some amounts of time. When my promos run out, I play financial musical chairs and get the lowest rate on all of my money. That's just for me though, to each his own. The most important part in all of this is that I got control of my budget, and started living within my means. My punishment for living outside of my means for so much time is that now I have to live well within my means. I have to live within the means set upon by having to pay a large chunk of each paycheck to the creditors. Once I'm outside this, it will be much easier for me to have lots of happiness and savings.

So I have now installable packages and tools for the mysterious, wondrous SERefPolicy. You can grab them at http://selinux.alioth.debian.org/experimental Note that I havn't tried using the policy yet. For example, the policy is missing dpkg and apt-get rules. So don't even think of running it on a production system, there is still lots of stuff to do. But I just, for the first time, managed to use "semodule", and actually add and remove modules from the modular policy. Yay! Now we just need to rewrite tons of policy in an even cryptier language, using even more M4 macro hell... ;-) Oh, and then we need to fix that Make bug effecting unstable, so we can actually build the new policy... ;-) We definitely need more people working on SELinux support in Debian.

Steinar H. Gunderson

It doesn’t matter which shell I use for completion. (And this goes to the other people who mailed me to tell me that Zsh rulez!) The problem is that packages of Zsh and Bash both offer several completion routines which are not sufficient for me. (e.g. one missing one is completion for the various bittorrent scripts.) So I need to install my own routines for those. I also wrote custom completion routines for in-house tools, and my own personal scripts in ~/bin/. Since these are not available by default in either the bash package or the zsh package I must install them manaully upon each host I wish to use, in order to have an identical setup. And that manual installation is what my new package makes easy. (Or will when it is complete!)

Security Team Meeting

Security issue meeting on IRC went well, although I left rather abruptly since my bed was calling my name.

Pyroman is now hosted on alioth, and uses Debians subversion server. I just did a new pyroman release, version 0.1.2. This is just an interim beta release, a version 0.2 will follow soon. New in this version is: Detailed error reporting: when a firewall rule is rejected by iptables (e.g. because you specified an invalid port range pyroman didn't detect), it will give you the corresponding filename and line number! Automatic rollback: Pyroman will undo any changes to the firewall if either any rule is rejected by iptables, an exception in pyroman occurs or the user fails to accept the changes within a configurable time limit (e.g. because he just broke his ssh connection...) So pyroman is even cooler now! ;-) On the TODO list: add a no-confirm switch for use at system bootup, code cleanups and a iptables-version test, so you can add rules that need a specific iptables version (such as string matches for bittorrent).

XSLT is an outright ugly language. Thats why I often just hate it. In some (rare) cases it's very elegant, but most often incredibly clumsy. I've been redoing my homepage completely, and I'm using XSLT to render my data XML files to the XHTML output. That works quite well so far, but I hit more and more special cases. First I wanted to add a language chooser. This required finding out which lanuages my document supports (read: blocks with that language attribute exist) and check for uniqueness. Some things where rather easy to solve (with some experience in XSLT), so where extremely ugly. First of all, I want my template file to be a valid XHTML file, I don't want to have any layout parts in XSLT. Then I have a separate sitemap, needed for generating the menu. So I have to process three documents at the same time in XSLT. Ouch. But doable, using variables. Basically I have three modes - in one mode, the template is mostly copied over, unless I've specified some special rule, e.g. to insert the page title, meta information or content. A second mode uses the current chunk of the template document as template for the contents a couple of times to iterate over data chunks from the data source file (read: put text blocks from my source files into the output template format). The third mode does the same for the sitemap. Unfortunately, it started to get even messier... Big problem number one: Make a list of the available languages for the current page. Make this information reuseable throughout the page transformation. Part one: make a list of available languages by finding unique (!) @lang attributes. (usual approach: grab all nodes with @lang attributes, check if they have the same node id like the first occurence with this @lang value for uniquness, then output. Ugly as hell, but string magic sucks as much.) Big issue number two: Assume we now have a variable with en in there, access these nodes to transform them. Unfortunately, the generated nodes are "result nodes", and you cannot iterate over them with for-each. Luckily the exslt node-set extension to solve this. But now came the third big issue along... this is where I give up. I want my pages to have a nice modification date... my data pages are stored in subversion... Subversion has keywords expansion like CVS, which I can use nicely to automatically update the last modification date of the file. Unfortunately, the date will look like this: $Date: 2006-02-15 21:29:10 +0100 (Mi, 15 Feb 2006) $ Whereas in a web page meta information it's recommended to use ISO8601. Now if I only had regular expressions and proper date handling functions... I guess I'll stop writing ugly, hackish XSLT code and chose pretty Python code instead. Maybe use TAL or KID for XMLish templating...

Matthew Garrett was briefly my hero today (taking over from Iain Holmes) for uploading into Dapper xserver-xgl and Compiz. Very nice it is too, but it's a bit too crashy on my laptop for any real work. Shortly afterwards he was replaced by Lapo Calamandrei, who told me he was working on some new Sound Juicer artwork if I was interested. Hell yes, I said, and it turns out he has managed to draw exactly what I wanted the Sound Juicer icon to be, several years ago. He rules. NP: Wish You Were Here, Pink Floyd

The following haiku poem was written some days ago. I wasn't too sure about including the final two paragraphs, but after letting it rest for a while I think they don't fit too bad into the overall view.

Lazyness — a pain?
It helps you to optimize...
or do just nothing. Some say it is art —
a required skill for some,
e.g. syadmins. To notice repeats
and work on scripts to ease them.
It truly helps there. But then the downside:
when you don't manage anything
you are lazy stunt. Not able to do
what you are expected to
and just waste big time. I strive for former
but fall into latter group
I am but sorry. The expectations
that I built — to some degree
all by myself — hurt. I have no idea
why people believe in me,
I don't do myself.

John Williams, of GNOME fame, recently blogged on window placement as part of his "things about gnome that suck" series. Well, I basically agree with him - but he doesn't propose a solution. Devil's Pie isn't a solution, because it's not configurable by the average user. Or even by me, I failed to have it tell apart my terminal windows and use a sane placement for them. Just remembering the last window position doesn't work either - I'm annoyed by this "feature" by nautilus, for example. I tend to have a default position for my windows, by I occasionally have to move them away, and I do not want to have them reappear at this different position ever. My personal approach basically is to run all windows (except my terminals and popups such as IM windows) fullscreened, on dedicated desktops. Even the IM windows are usually on my "IM & Chat" desktop. This works really well, especially since I keep my terminal windows open all the time, and the default placement to minimize overlap works fairly well. Still I couldn't specify a general rule how any window manager could learn where to place my windows... so just complaining that Gnome doesn't do this good isn't all - we need to find a way to make this work just fine without the user noticing... I don't know about OSX - how does this work there? In windows it definitely is horrible. When you watch windows users, you'll often see them stating their Internet Explorer, then clicking on the maximize button. All the time.

Usually I try not to get involved in the mind-numbingly pointless desktop wars, but Benjamin produces one of the most flawed arguments I can remember having seen.

Fair enough that you want to use the keyboard. I'm much quicker on the keyboard than on the mouse, and presumably so are you. After a long time using computers, and CLIs in particular, this is no surprise.

Most people are not this way. I can assure you that my mother is not this way, and prefers the comfort of the mouse whne navigating around. I don't see what your assertion that 'stating that a typical Gnome user does not want keyboard is simply false' is grounded in. Have you done studies? Do you know the exact breakdown of geeks vs normal people using GNOME? Remember that it's not just for the cool kids who can write their own code anymore -- I know plenty of 'normal' people using Linux now, typically with a GNOME-based desktop.

But, it's 'but don't say it is for users sake' that really throws me. Why do you think they do it? For fun? To see how many people bitch on blogs about it when they do it? Because they hate people and this is the best practical way to take out their anger on the world? They've done studies on their userbase (e.g. the Sun usability study), and they're doing what they believe is best. This is true of every desktop: they simply do what they believe will produce the best desktop for their userbase. Sometimes this results in decisions which people think are bad. Fine then, so be it. This is what's happened here, as opposed to your suggestion of a deep-seated hate for all their users, individually and collectively, or something.

Your claim to be a normal user is simply false. I can't claim the same thing. A good rule of thumb is that if you're capable of writing the application you're using, then you're not a normal user. In some cases, this means you're not its target userbase (not always: this holds more true for, say, Evolution, than mutt).

If you have truly constructive suggestions (not using the words 'crap' and 'simply false'), then provide feedback to the developers through the appropriate channels. If you don't think GNOME is for you, try some of the many excellent other desktops (KDE, XFce, GNUstep, whatever else is around). If you like GNOME otherwise and would like to make a concrete improvement, then submit patches.

Daniel Burrows's point was simple: you are probably not GNOME's target audience (this is not specific to benj; if you are reading my blog, then odds are you aren't GNOME's target audience). So, assertions that they've got it all wrong and they should listen to you on matters of interface design aren't really going to be helpful, because you just aren't the primary target market.

PS: Just because you write it in Latin, doesn't make someone else's argument a logical fallacy.
PPS: If you're still having trouble wrestling with this entry, imagine instead someone bitching about Debian being crap because they don't include the proprietary Flash plugin, a bunch of closed-source games, VMware, etc. That's not what Debian's about. Make sense now?