|
| Greg's diary |
| September 2009 |
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
This view of the diary is limited to these topics: technology. There may be lack of continuity in the text, and some days may be completely missing. In case of doubt, please enable the complete display.
| Tuesday, 1 September 2009 | Dereel | Images for 1 September 2009 |
| Top of page | ||
| next day | ||
| last day |
| Topic: technology, music, opinion | Link here |
The real issue is knowing what is being played. Like iTunes, it displays the concepts Artist, Album and Track (interestingly not song), and with the help of Gracenote I get displays with values Artist “Trevor Pinnock”, Album “SC 1” and Track “Allegro”. It's not easy to deduce that this is the third movement in a Brandenburg Concerto—the first movement has identical information, as does the first movement of the second concerto. The “track numbers” don't help either—they're allocated sequentially from the beginning of the card, so this was “track” 113.
The real problem I'm seeing now is that the concepts that the industry uses are just not appropriate for the kind of music I'm looking at:
They consider the “Artist” to be the most important piece of information about a work. This might apply to modern music, where the “Artist” is also the composer, but for classical music it's not very useful, as witnessed by entries like “Stephen Gunzenhauser_ Czecho-Slovak Radio Symphony Orchestra”, which refers to “Greig”'s Peer Gynt.
It doesn't understand the concept of a “work”, a collection of tracks that form a complete work, as indicated by an iTunes search for “Harmoniemesse”:
This is one of the few times where “Artist” would be of use, but both people who entered this data entered the composer's name instead.
The lack of the concept “work” means that finding specific information is very much a multiple grep, not helped by the inaccuracy of the database. Thus the search term “bach” brought hits from Johann Christian Bach, unnamed Bach and Johann Nepomuk Hummel:
|
||||||||||||||||||||||
Things aren't made any easier by the truncation which seems to be a way of life in modern GUI software. I thought that “Franz Joseph Ha...” might be easy enough to recognize, but you need to know your classical composers. Haydn? Händel? Hatzenbacher?
I can complain about this all I want; it won't change anything. But I still need to work out how to organize my own collection. I see a number of alternatives: I can try one of the other software packages available on the web. I have so little confidence in this approach that I'm not even going to try. Alternatively, I can do everything manually. At the moment I'm creating a separate directory hierarchy organized by composer and work:
=== grog@dereel (/dev/ttyp9) /home/Music/MP3 220 -> l Bach-JS/
I suspect that this is a better approach, but still not sufficient. At the very least I need a further level with performance (so I'd have BVW147/Harnoncourt and BVW147/Rifkin).
Even this isn't enough, though: the MP3 players only seem to understand a single level of subdirectories. So I need to map them again to a local copy where these directories are in the top level. Maybe I should just give the files numbers and store information in an SQL database.
| Wednesday, 2 September 2009 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: brewing, technology | Link here |
Finally got around to brewing beer again today. It's been over 3 months since the last brew, an indication of how much I hate it. It takes up the whole day, and my current scheme is so far from convenient that I'm always worrying that something catastrophic will happen. Today it almost seemed so: on powering up brewer.lemis.com, my fermentation temperature control machine, it didn't come online. I used this computer almost as a joke: it's about 18 years old, it's running a development version of FreeBSD 5.0 dating back to December 2000, and it's in pretty poor shape:
|
|||||||||||||||||||||||
In particular, the display card no longer works, so all I have to go on is the noises, disk accesses and LED status that happen during boot. On this occasion it seemed to be performing lots of disk accesses. Pressed the Big Red Button, and it came back normally enough. I need to plan to replace this machine.
Apart from that everything went smoothly enough, but it still took all day.
| Topic: technology, gardening | Link here |
More work on porting wview to NetBSD, and finally finished. I never realised what a pain it is that NetBSD stores packages in /usr/pkg and not the more usual /usr/local. Many packages, including parts of radlib, hard code the path name in various configuration files, and I had to go out and fix them. Now I have a complete standard installation of wview; the next part is to incorporate Steve Woodford's patches for my Fine Offset WH1081 weather station.
| Thursday, 3 September 2009 | Dereel | Images for 3 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: photography, technology | Link here |
Issue 19 of c't, which arrived yesterday, contains a test of various panorama software. One package stands out: Autopano Pro 2.0 (and though there's no “normal” Autopano, there is an “Autopano Giga”). It gets much higher marks than any of the others, including hugin and Panorama Maker 5 Pro. Decided to try it out. It doesn't run on PPC Apples, but it is advertised to run on Linux, so downloaded the image and installed it on cvr2.
And then? How do I run it? There are no instructions (it's clearly a “modern” program), but there is a Wiki. And it doesn't tell you how to run the program.
Went looking in /usr/local, but there was nothing there. Finally—find(1) is your friend—found an executable /usr/bin/AutopanoPro and ran it:
=== grog@cvr2 (/dev/pts/3) ~ 1 -> AutopanoPro
Checked for help, but all I got was a contact form with these horrible CAPTCHA, so sent off a message, and then downloaded it and installed it on Microsoft.
That was strange, too. They're a French company, and I've seen a couple of cases on the web site where I'm transferred from English into French. This time the installation was in German, most of the time, at least partially due to my difficulties to get Microsoft to stay in one language.
This version was easy enough to start, but of course I had to spent minutes changing directory. The photos are on dereel, and it insisted on reading the entire Photos directory, over 1600 entries, before coming back to life again. Then it went off and did the “autorecognition” thing which c't had found so reliable:
|
|
||||||||
This directory already contains three panoramas of this collection, which it happily included in the collection.
Off to try to convert things, not helped by the strange icons. Clearly this is not software that you can just use: you need to read the manual first. After a long time waiting for the conversion to finish—not helped by the fact that both the button “Render” (well, “Rendern”) and “Cancel” (“Abbrechen”) were selectable, and no progress bar was present, checked the specified “output” directory, C:/Documents and Settings/groggy/Desktop (that's right, with slashes instead of backslashes), but nothing was there. Gave up for the day and decided to wait until I had time to read the documentation.
| Friday, 4 September 2009 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: photography, technology | Link here |
Reply about my problems with Autopano Pro 2.0 today. I had asked:
I have just downloaded the trial version of Autopano 2.0 for Linux and installed it. Then I went looking for how to run it. You seem to have missed this point in the documentation. Please tell me how to run the program.
The reply:
Under linux, it depends on the installer you use. If you use .deb, the software will be available in the main menu under graphical tool.
The error you have is caused by the fact that you don't have openGL on your linux. The v2 of autopano needs an opengl 2.0 graphic card with proprietary drivers installed.
Clearly you need more than OpenGL; you need their idea of what kind of window manager to use too. None of this is mentioned on the web site. Went off and played around and confirmed that yes, the executable is called /usr/bin/AutopanoPro (something that wasn't addressed in the reply), and it works on :0 on cvr2 (using UNIX domain sockets), so I suspected that it might be related to TCP. And of course this system (default Ubuntu installation) has gdm installed, with the option -nolisten-tcp, so had to go off looking for how to fix that. The answer is in /etc/gdm/gdm.conf:
And yes, it worked fine. So is it really a driver issue, as they claim? I don't want to have to reconfigure my whole fragile X setup for a program that wants something special.
Gave up on that and tried with the Microsoft version—what a letdown! Yesterday's attempts showed that you need instructions to understand Autopano Pro 2.0, so off to look for them. As far as I can tell, there are none! The available documentation is still for release 1.4! Instead, followed the Autopano example workflow document, which had the horrifying recommendation:
In Explorer (or Finder, on MacOSX), locate the images you want to stitch into a panorama. Select them, making sure you're including only images belonging to a single panorama.
Drag and drop the selected images onto the Main Window of APP. A new Image Group will be created. You can repeat steps 2 and 3 as many times you want, creating multiple Image Groups. Each group should contain a single panorama.
Doesn't anybody know how to manipulate file names any more? In a directory with multiple files, all looking similar, this is a Real Pain. What's wrong with specifying the wild card verandah-*? I could scream. And yet all this broken software does this sort of thing, so I can't blame Autopano beyond saying that good software should (and easily can) offer something better.
One alternative is what I found yesterday: let Autopano go and look for all photos. It can handle Olympus raw format, so if I let it loose in my orig subdirectory, it happily includes each photo twice, once as raw and once as JPEG. And with only a few hundred mouse clicks and lots of careful comparison (look at those file names; why have verandah-* when you can have P8081353.JPG P8081353.ORF P8081354.JPG P8081354.ORF P8081355.JPG P8081355.ORF P8081356.JPG P8081356.ORF P8081357.JPG P8081357.ORF?) you can remove the files you don't want.
For a first attempt, that was too much, so I went and removed the superfluous files from yesterday's selections, notably the existing panoramas, which had caused the spectacular mess of the verandah panorama yesterday. Also discovered why the “rendering” window had both buttons selected: there were multiple windows, one per panorama, and they all wanted to be selected. When they were, they went off and did their thing, storing the results in files with particularly emetic names, like [Group 16]-verandah-1_verandah-5-5 images.jpg, which even Microsoft's file name couldn't handle: it put a spurious [ at the beginning.
And the result? Good, better than the others, but not perfect. Here the results from Arcsoft Panorama Maker 5 Pro, hugin and Autopano Pro 2.0 for 22 August 2009:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
These are all “automatic” renderings, and apart from the dramatic failure on the part of Panorama Maker 5 Pro they show that Autopano 2.0 Pro is a little bit ahead of hugin. The edges of the swing are rendered better, but not perfectly (first hugin, then Autopano):
|
|
|||||||||
|
|
|||||||||
For some reason my browser renders the Autopano panorama very badly at full size, but that seems to be a browser incompatibility; as the image details show, the image itself is smooth enough. But clearly this panorama requires manual intervention, no matter what software I use to create it. The rendering speed was nothing breathtaking, though the disk activity light on pain, the Microsoft box, suggests that this is because it only has 512 MB of memory, while Autopano suggest 2 GB. It's interesting that it displays the progress as it renders, gradually bending things into shape.
| Saturday, 5 September 2009 | Dereel | Images for 5 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, photography | Link here |
Photo day again today, not helped by the rapidly changing weather. Finally got it to stay moderately overcast long enough to get my photos.
Spent some more time with Autopano Pro 2.0, this time making duplicates of all my weekly panoramas with it. This still isn't a real report on the software—I'm planning that—but more observations:
It does strange things with the EXIF data. Most of it disappears, and I need to copy it back from one of the component photos. Even that isn't without its surprises:
=== grog@dereel (/dev/ttyp3) ~/Photos/20090822 90 -> exifcopy garden-ne-g.jpeg garden-ne-panorama-kolor-raw.jpeg
exifcopy is a little script that invokes exiftool to copy the data across.
It recognizes Olympus ORF raw format images—sort of. The results look like the results I get with ufraw and friends. Here the verandah panorama done with hugin, with Autopano Pro 2.0 on the ORF files, and the same image after being put through the Ashampoo Photo Optimizer:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Clearly this software isn't a contender for replacing other raw format converters for ORF. It is also surprisingly slow, though to be sure I should first try it on a real computer—if I can get it to work there.
Once again it does a better job of rendering the panorama out of the box than hugin does, and once again it's not perfect. I still need to find time to play around with the manual adjustments.
File selection is horrible. Yes, I keep complaining about this, but that's because it keeps annoying the hell out of me. A simple “change directory” can take minutes because it insists on interpreting the files. I've used the same workaround that I did for hugin: create a separate directory into which I link only those images that interest me.
It's not very configurable. My photos today were shown with this irritating date representation 9/5/09, and there's no way to change it short of changing the language. German gives 05/09/09 (note those leading zeroes and the still-truncated year).
It's not overly reliable. It has crashed several times on me, usually while I'm trying to stop it.
| Sunday, 6 September 2009 | Dereel | Images for 6 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology | Link here |
Spent some time today looking at the software for the Fine Offset WH-1081 weather station that Steve Woodford sent me a while back. It was a patch for Wview release 4.0.1, and the current release is 5.5.3. Spent a fair amount of time adapting that, not helped by the fact that I don't know (and really don't want to know) about autoconf. Something went wrong generating the Makefiles, but by that time I was too frustrated and decided to put it off. Maybe I should first install release 4.0.1 to ensure that things work at all.
I installed the NetBSD installation on kimchi from a downloaded ISO, and it didn't give me the option to install X. Went looking for a package, but found none. Went over the web site looking for documentation. Plenty on how to use X, but nothing on how to get it onto the system. Finally found a document and followed that, but ended up with only a base installation—not even xterm was there. And then it occurred to me that this isn't even the standard NetBSD version of X (should be XFree86). This really could do with much better documentation.
| Monday, 7 September 2009 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, general | Link here |
More work on Wview release 4.0.1 today, and got it to work, not without kludges: it depends on sqlite3, but there's nothing in the configuration to include the libraries. How can that work? Tried handing it in via LIBS, but that caused the configuration to fail (“configure:2904: error: C compiler cannot create executables”) because it couldn't find the library at link time.
Probably there's another variable to tweak, but why should I have to? In the end, just put it in manually. Why is this so difficult?
Still, wview looks quite good, and I now have a Dereel weather web site. Spent some time playing around to make things more readable, and got some understanding of what it can do, but it looks as if I should probably postpone most such stuff until I've installed the latest version. I should also port it to FreeBSD: at the moment it's running on kimchi, my test machine, which means that I can't use it for anything else, and I also can't turn it off.
| Topic: technology, music | Link here |
Also finished copying my CDs to MP3. I now have about 9 GB, enough to require some trimming. Spent some time looking for something off the shelf to organize them, without much success. c't has some articles on it, conveniently stored on DVD in page image format, but indexed, it seems, with an application that only runs on Microsoft “Internet Explorer”. More checking required; hopefully I'm wrong on that one.
| Tuesday, 8 September 2009 | Dereel | Images for 8 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, opinion | Link here |
Came into my office this morning to find an eBay login screen on one of my displays. But it looked completely different:
|
||||||||||||||||||||||
|
||||||||||||||||||||||
The one on the right is what I'm used to, and the browser fills in the names automatically. The other one had a different layout, and it hadn't been filled in. Spent some time pondering the cause, and unfortunately closed the first window before looking at the source. On the face of it, it's legitimate, but it brings back to me how difficult it is to detect bogus signin screens. Somehow there should be some way of getting the warm fuzzies that you're really talking to the correct site; public key cryptography maybe?
| Wednesday, 9 September 2009 | Dereel | Images for 9 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, opinion | Link here |
We were discussing Apple again today. Somebody mentioned the fact that you could find files with “Finder”. Well, some of them. If you're looking for MP3s, forget it:
|
||||||||||||||||||||||
This is straightforward enough stuff—or is it? The “modern” attitude seems to be that you don't need to understand the directory hierarchies, and yet here they're presented with no less than five levels of hierarchy, far beyond what you want. And to do so, the file names are truncated to a point where you can barely recognize them. Yes, you can go and change the view and see just the contents of the directory:
|
||||||||||||||||||||||
And it's still mutilated. Why? There's plenty of space on the right to stretch it out, but it doesn't use it. Is this really easier to understand than the method we've had for decades? Consider:
=== grog@dereel (/dev/ttypf) /home/Music/iTunes/iTunes-Music/Christopher Hogwood_ Academy Of Ancient Music 7 -> ls -l
With the help of the prompt, this shows exactly where you are in the directory hierarchy. But then there are so many other things you can do: ls on FreeBSD has no less than 38 options controlling the way things are displayed. “Finder” has only three viewing options (the third is icons, to waste even more space). Why do the GUIs restrict you to so few views? And why can't they use the available space? I'm left with the feeling that the product managers want to supply as little functionality as they possibly can.
Is this an Apple issue? Partially. Microsoft can at least work out how wide to make the columns, but apart from that it's pretty much the same:
|
||||||||||||||||||||||
| Topic: technology, general | Link here |
After yesterday's problems with wview, spent some time porting it to FreeBSD. That proved simpler than I had feared: the USB code compiled cleanly, but when I tried to run it, I got a flood of error messages:
That's EAGAIN, of course, and it was returned from a call to usb_interrupt_read (). Why is it returning that? Did some tracing and discovered that it was getting valid input, so the EAGAIN was benign beyond the error messages. Ignoring them worked, but the wviewd process used something like 35% processor time. Put in a little code to delay a bit (10 ms) and retry if it hit that error. That works, but it sounds like a bug in the USB implementation to return EAGAIN on a blocking read. Still, I now have wview working the way I want it, sort of, and I can turn to getting 5.5.3 running.
| Thursday, 10 September 2009 | Dereel | Images for 10 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, general | Link here |
Into the office this morning to find dereel's /home file system full. I know it's getting full—I've added about 60 GB of photos and 10 GB of MP3s in the last few months—but it shouldn't have been quite full yet. Further investigation showed that it was a ktrace.out file, which I promptly removed—and it made no difference. Clearly it was still running—and running kdump against the file would have told me what the file was, but I had just deleted it! Fortunately Peter Jeremy explained to me what the -C option did. I thought it stopped tracing for a specific process, but in fact it's for all processes for which the user can stop it.
My suspicion was that it was the weather software, of course, and checking /var/log/messages confirmed it: it had stopped functioning in the middle of the night, and attempts to restart it were unsuccessful:
That continued despite restarts until I disconnected and reconnected the USB cable. Is this a problem in the FreeBSD USB stack? To be monitored.
Spent some time trying to sign up for various weather reporting systems, notably Wunderground and CWOP, both of which have very difficult to understand instructions. In particular, Wunderground mentions a password, but doesn't give the opportunity to set one. Signed up anyway, got no confirmation, and read instructions telling me that it would take at least a day, that various things could go wrong, and in each case the result would be that nothing happened. Wonderful. And I can't even check until tomorrow.
Wunderground is very specific about the location of the weather stations, though. When setting the location of the station, it specifies the latitude to 13 places of decimals, and the longitude to 14 places:
|
|
||||||||
That corresponds to a resolution of 1-8 mm latitude and about 0.8 × 1-9 mm longitude. I wonder what they're thinking.
| Friday, 11 September 2009 | Dereel | Images for 11 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, general | Link here |
Into the office this morning to discover that the weather station software had hung again in the middle of the night, and that the /home file system was full again.
The full file system was for the same reason as before: a ktrace.out file had filled it up. And again I removed it without checking what was generating it. But it seemed to be related to the weather software, and sure enough, found that I had included a ktrace of the wviewd process in the startup file. So hopefully that's over and done with now.
The hang was different, and from the log messages it was clear that it had happened long after the file system filled up—at about the same time as yesterday, but the flood of log messages had already flushed the previous day's messages. This time I had:
That's really helpful, of course. But yesterday it seemed to have happened a little after 02:00 as well. Is there something in the nightly cron jobs that trips over the USB stack at this time of the morning?
Getting things started again wasn't easy. Various components wouldn't stop, and starting things manually is greatly hampered by the presence of PID files that don't get ignored if the process has died.
| Topic: technology, general | Link here |
More investigation why I didn't show up in Wunderground. It turned out that my guess was right, that it wanted my own password. But that contained a character that wview didn't accept, so it just truncated the password to that point. Tried another one, all letters with a verylongcommentaboutthiskindofstupidity, only to discover that Wunderground won't accept more than 10 characters in a password. sigh.
After adapting to these quirks, things worked, and I appeared in the map. Spent some more time looking at other reporting systems, and set up reporting for CWOP. That seemed even easier, but when I tried to restart wview, things went to hell:
What's that? There's plenty of memory available. Built a debug version of wviewd and tried it out, and established that radCfOpen (did I get the studly caps right?) is part of radlib, and it's designed to read in the configuration file. It got the config file name right (something that it didn't bother to report), and somewhere inside it ran into trouble with the “memory allocation”. It then stopped without any further message and with a 0 completion code. ktrace showed that it read in the configuration file, then:
There were lots of these semops, but all with return value 0 (successful). What's all this about? Is it really a semaphore issue, or is it really trying to allocate ridiculous quantities of memory? And if so, why not report how much? About the only thing that I can conclude is that a library that can report this kind of message is that they're not worth having. Reading a configuration file shouldn't require lots of semaphore operations.
The obvious conclusion was that the problem was due to a configuration change. But after reverting the changes (RCS is your friend), it didn't change anything. Spent about an hour trying to work out what went wrong, and in the end reverted to the NetBSD installation, which happily accepted the same configuration files once I fixed the path names.
So, what's the problem? One is clearly a badly documented and rickety framework (the only documentation I can find for radlib is a API reference). The other is the Tower of Babel attitude to software design. It's probably not worth trying to debug it; I need to migrate to wview release 5.5.3, which doubtless is waiting with other pain, such as configuration files stored in a database. But maybe some of the problems I've seen so far will go away.
| Saturday, 12 September 2009 | Dereel | Images for 12 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, general | Link here |
For a change, no full file system this morning, and the wview software (on NetBSD) hadn't hung either. But the problems running wviewd on FreeBSD continued.
| Topic: technology, general | Link here |
One thing that the power failure “fixed” was the “memory allocation failure” that I was having with wviewd. I strongly suspected that it was something to do with left-over System V semaphores—how I hate the three ugly sisters! This tends to confirm the suspicion. On IRC, Peter Jeremy pointed me to ipcrm, where, apart from a way to remove dead semaphores, I read:
Callum and Edwin are both on the IRC channel as well. And that was at a time when I was mentor for Edwin, so I should have known all about it. Checked the commit logs and found:
My mind must be failing me.
| Sunday, 13 September 2009 | Dereel | Images for 13 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, opinion, music | Link here |
I still haven't found an off-the-shelf solution for keeping track of my MP3 collection, and spent some time looking for possibilities today. Most, of course, are the kind that make iTunes look good, and so I returned to look at the article in c't magazine that I had already found by chance last week. This time I wasn't so lucky, and spent a lot of time looking for it. The index is now only available as an application that only runs on Microsoft “Internet Explorer”. Went looking with Acrobat Reader which was doomed to failure: I found hundreds of hits for “MP3”, and just sifting through them was more pain than I could stand. It would have been easier to look through the paper copies. But in the process, saw yet another example of this “modern” gratuitous truncation of text:
|
|
||||||||
Look at that path name! It's only half as wide as what looks like a progress bar below (it isn't—the bar just swings back and forward), and it truncates the path name with something that is barely shorter (in fact, the ... are one character more than the om they replace). Sometimes I despair. And yes, I still haven't found the article.
| Topic: technology | Link here |
More work on porting wview to FreeBSD, and now have a clean build of release 5.5.3. Now I need to test it without disrupting the reporting too much. It looks as if it was a good choice to migrate to the latest version rather than search for the bugs in the old one: the area where the bug occurred (reading the configuration) has now changed completely, though not obviously for the better: instead of storing it in (multiple) text files, which I can maintain with RCS, it's now in a database. We'll see.
While she was there, she mentioned that David is currently in Batam, not a place I know, though I see it's just off Singapore. I asked if it was really called Batang, the Malay/Indonesian word for “rod”. And then Chris asked if it was related to French baton, which means the same thing.
How do you find that out? I have a book by M. B. Lewis on “Malay Script”, also called Jawi, a modified Farsi script that is no longer in general use; what little I know about Arabic and related scripts comes from that book, which I must have bought 45 years ago. It's interesting in this context because it contains a glossary with some etymological information. Dragged it out, but it had no information on the etymology of batang.
Chris took a look, however, and found some pencilled-in notes in a section on Arabic loan-words. Not my writing, and clearly written by somebody who understood Farsi: the comments are about the corresponding meaning in Farsi. For example, janazah means “royal hearse” in Malay and “dead body” in Farsi. Most underlined words have the same meaning, however.
But who wrote it? And when? I'm sure I was present, but I've had the book for ever. Could it have been Shahram Akhavan? Or Ali Madanipour? I can't find a way to find out. I'm sure they wouldn't remember either.
End deselected topic h -->| Monday, 14 September 2009 | Dereel | Images for 14 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology | Link here |
Part of my weather software is a script that copies the web pages to the external server every 15 minutes. It uses an ssh tunnel to do so, and I found it littering the system with old ssh-agent processes. With a bit of advice, found a couple of environment variables that allowed me to do trap the process on exit:
The trick is knowing about the environment variables; there's also a SSH_AUTH_SOCK which can be of use under some circumstances. I should probably use it to not start any additional ssh-agent processes, but this works for now.
| Tuesday, 15 September 2009 | Dereel | Images for 15 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology | Link here |
More work on the weather station software today, and found out why the build was so clean: I had included all the code, but I had omitted most of it from the configuration information, so it hadn't been compiled. Normal enough problems once I reattached it, with the exception of the dependencies. Any normal build system has a depend target in the Makefile, but this thing uses GNU autoconf, something about which I have never heard much good. Even 15 years ago, in Porting UNIX software, I pointed out weaknesses; nowadays I'm reminded of a Dijkstra quotation:
If Fortran has been called an infantile disorder, PL/I must be classified as a fatal disease.
Finally found the problem—it seems that the dependencies are built by the configure script, and they base on the variable AC_CONFIG_FILES in configure.in, at least in this case.
| Topic: technology, general | Link here |
Of course nothing had happened with annulling my eBay transaction yesterday. Tried again and got the same results. Connected with live help and was told to clear my browser cache, which greatly annoyed me. But of course I suppose they have to go by the book, since they really don't understand what they're doing. And of course it didn't make any difference, and they told me it had been reported and that I should try again tomorrow. They wouldn't give me a case reference number and just promised to send me email, which they didn't, so all I have is the session ID (2232708).
| Wednesday, 16 September 2009 | Dereel | Images for 16 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology | Link here |
More work on wview today. Made some progress, but it's painful. I've had the idea of storing configuration information in a database before, with the Black Box project a couple of years ago. But that was in conjunction with web pages to update it, and of course it used MySQL. This software uses sqlite3, which I don't know, and which is different enough from MySQL that I can't just jump in; instead I need to learn Yet Another Dialect of SQL. And the configuration scripts are still just that, scripts, and not very clever at that. Maybe the intention is to create a web-based configuration system, but the current status seems to have the worst of both methods, and it can also easily lead to the system using two different database systems: there's a provision for storing weather data in a database (MySQL or PostgreSQL, but not sqlite3), but the configuration must be stored in an sqlite3 database. I'm left wondering how much work I want to do on this software.
| Topic: technology, general | Link here |
As I expected, and contrary to promises, I didn't receive any feedback from eBay about cancelling my transaction, and a further attempt met with the same fate. Tried live help again and finally the consultant did the work for me, successfully. Still no explanation why it didn't work for me: “We did not receive any report from the Technical Department that we have current issue on the site.”
| Friday, 18 September 2009 | Dereel | Images for 18 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, music | Link here |
Back to copying data to the SDHC card for my car radio. How painful this is! I've already deferred the long-term goal of a database of the works I have: there really must be something like that already. Surely I'm not the only person who can see a need for it. But there are still other issues: how do I get the data on the card in a form that the MP3 player will play in sequence? And how do I create a list of stuff so that I don't need to search all 1300 tracks (sorry, songs) to find them?
None of that should be difficult. A couple of shell scripts should do it: one to copy, the other to list the contents of the card in the sequence which the player will play them.
But there's one problem: these HORRIBLE file names full of spaces. They completely break normal conventions of shell scripts and other UNIX tools. The only way I can find to get a list of the files in directory order is with ls -f. Normally you'd write something like:
But for uses spaces as a delimiter, so that doesn't work if there are spaces in the file name: it attempts to use every part of the file name as a separate name. The shell does provide some support for this kind of breakage with the construct "$@" instead of the more normal $*, but I can't see any way of applying this to the `ls -f` construct.
In this case, I found a workaround: use sed:
But what a pain! And then I came to the next case, where I wanted to put a track number next to each file name. How do you do that? Normally, you'd write something like:
But how do you do this with file names with spaces in them? I've already mentioned that this approach doesn't work; but neither does the workaround I found for copying, because I have a sequence number to process. Spent all afternoon trying, getting more and more frustrated in the process. Daniel O'Connor tells me that I'm using the wrong tools; but he's missing the point. The whole idea of UNIX is that the tools fit together. There are some conventions required for that, and one is that some characters are special, and you don't use them in file names. Daniel also didn't find a way (appropriate tools?) to solve this problem. People suggested that my workaround was insufficient if people put other special characters in file names, like " or a carriage return character (\n).
You'd think that I would gradually come to terms with the fact that people want to put spaces into file names. On the contrary: the more I have to do with them, the more they annoy me.
| Topic: technology | Link here |
As if that wasn't enough frustration, my keyboard once again generated a c-a-bs key combination, the one that X servers traditionally use to shut down. I have already disabled this combination. Did the X server care? No: it shut down anyway. I've checked the key maps, and they definitely show no server shutdown; but I also checked the combination, and yes, it shuts down the server anyway. I'm not sure what to do about that. I suppose I should go in and fix the server, but why should that all be necessary?
| Saturday, 19 September 2009 | Dereel | Images for 19 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, music | Link here |
More work on the scripts for copying MP3s today. I'm using ls -f and ls -t to list the file names in the sequence I want them, something that I can't do with find (which has options to handle file names with spaces in them). But ls has an option or two (-b and -B in the BSD versions) to handle unprintable characters. GNU ls only has -b, which seems to be the same as -B in BSD. Both of these options have nothing to do with spaces; but the idea sounded good, and I started off thinking of printing something that would escape delimiter characters as well. But how? It would be nice to be able to convert back again too, so things the octal escapes used in the -[bB] options or like HTML escapes weren't appropriate. In the end decided on printing the names in hexadecimal. That worked straightforwardly enough, but strangely there's no standard program that converts back again, so wrote one of those too. The resultant code for listing the files looks like:
That works, but it's still not very satisfying. Is the -X option interesting enough to put in the already overloaded list of options to ls? I still need to think about that one.
| Sunday, 20 September 2009 | Dereel | Images for 20 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, music | Link here |
A few messages today, from Mads Martin Jørgensen, Patrick Hess and Michael Hughes, all pointing out that I can modify the treatment of output of programs in backquotes (``) with the aid of the shell IFS variable, which describes which characters delimit arguments. By default the characters are space, tab (\t) and newline (\n). By setting the value just to \n, you can work around spaces in file names—if the program in question returns \n between arguments. Fortunately, that's the case with ls. Here an example of when it works and when it doesn't:
So it only works if the program returns values separated by \n. echo doesn't, with the result that all the text gets lumped together as one parameter—the opposite of the previous problem. Still, it solved my particular problem (that one, anyway), for which I'm grateful.
Unfortunately, the problem didn't stop there. As I've mentioned, the naming of the tracks is so variable that it's very difficult to get them back into the original sequence—if that's even what I want. A couple of examples:
I have two CDs of Bach's 6 sonatas for solo cello. For reasons probably related to the length of the works, the odd-numbered sonatas are on one CD, and the even-numbered ones are on the other CD. Clearly this makes no sense when they're all copied to the same medium. How do I get them in the correct sequence? I can't go by the name, because it starts with the CD number (important, eh?). Here the first movements of four of the sonatas, in the sequence shown by ls -l:
In the end, used the fact that they all have a hash sign (#) before the number, so I was able to use this to get the correct sequence:
But that brought me back to the issue with the broken names. xargs has the -0 option, but it only really works in conjunction with find. Did some playing around with the output and got most of them to work, with the exception of two containing the name “Bourrée” (the ' is a separate character; clearly it should have been Bourrée, but this is what it had):
I still haven't worked out why, but none of my attempts to match those names worked. Instead touch created a new file of zero length with a name containing a subtly different name. After much messing around, got it to work, and was no longer able to reproduce the old behaviour. This must be something to do with multibyte character sets, but it's not clear what.
I have a number of Dvořák symphonies, all completely mixed up:
Apart from the impossible sequence (ls -tr is your friend), I have two copies of the 9th symphony. Which is which? They have ID3 tags, right? Spent some time installing various ID3 utilities, id3tool and id3.el (an Emacs macro set). Neither of them could recognize the ID3 in the MP3s. They're there, since the player finds them. What's the issue here?
| Topic: brewing, technology | Link here |
brewer.lemis.com is no more. It went down yesterday and wouldn't come back up again. I suspect the disk has had it, but it's difficult to say without a display. In any case, it's seen better days:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
From recollection the components must date to about 1991, and it's only a 16 MB Intel 80486 (50 MHz, I think). The real issue is that I can't find any disks to repair it with, and I'd need an ISA display card as well. So time to start building a new one; and that will be somewhat hampered by the fact that I'm currently using an ISA adapter with a 802.11 card to connect to the network, something that you don't easily find on more recent machines.
| Topic: technology, general | Link here |
If I were to believe my weather station today, we're having pretty extreme weather: it reported a low temperature of -1840.3 °C, enough to cause the HTML generator to crash. Clearly more work needed. Took a look at the code, but without better documentation it's really not clear what the best solution is. There are clearly two issues: one is ridiculous temperatures like this one, and the other is sudden changes. How quickly can temperature change? On 7 February 2009 the temperature dropped 15° in 30 minutes; that's presumably about as fast as you'd ever see it.
I joined up the wview mailing list a couple of days ago, and, after finding a way which Google groups didn't reject, replied to a thread about access to the repository (there is no access). As I've already observed, lack of access to the revision history has made things complicated, and I said so. The response (to a message sent with texts completely out of order)?
And that was all. No mention of my concern about access to the revision history. This confirms my opinion about people who can't read beyond the first couple of lines, and one of the reasons I hate reverse chronological documents.
Mark Teel is the principal author of wview, and also the first person I've ever seen to ask me to write messages upside-down. Clearly this is not a list in which I will participate. Still, maybe this is not such a bad thing; there are so many details I don't like about wview that the lack of requirement to feed back my data might turn out to be a relief.
| Monday, 21 September 2009 | Dereel | Images for 21 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, brewing | Link here |
More thinking about brewing computers today. Put together a much newer machine, only about 4 years old, and found a FreeBSD disk to go with it. And now? I had included the special hardware in the case of the computer, but it seems to make more sense to have a separate box. I now have a few dead UPSs that have power connectors on the back, which makes them a good choice. Now to find enough energy to gut the cases and put in the hardware.
Also started working out the network cabling to the garage. I put some in quite a long time ago, and there is a switch in Yvonne's office cupboard—6 cables and 5 ports. Found an 8 port switch and put that in there; all I need to do now is put a connector on the cable in the garage.
| Tuesday, 22 September 2009 | Dereel | Images for 22 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: brewing, technology | Link here |
A little more work on converting the UPS housing to a housing for the temperature control hardware. It has three power sockets on the back which I can use for connecting hardware. I only need two, but the third could come in handy, especially as I have a total of 8 relays. It's interesting to note the “surge protection” on the back:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
If there's any “surge protection” at all for the left hand socket, it must be the fuse. But now I'm left wondering how to connect things to the box. There's much more space there than I need, but where do I put the connectors? There's not much external surface except for the cover. More thought required.
| Topic: technology | Link here |
In the afternoon Yvonne called me in to tell me that her machine had hung. It hadn't: it had paniced, with a double fault. While I was looking at the dump, it happened again, and a further investigation showed that it had happened in the morning and—somehow—Yvonne hadn't noticed.
The kernel was ancient, and I had neither sources nor a debug kernel, so started building a new kernel. That failed with the message “I/O error writing to Make.log”. Make.log is where I write the output of the build, and like the source tree, it was NFS mounted. So is Yvonne's mail inbox folder, and the last two panics occurred while she was writing back to it. And yesterday I changed the switch connecting the two machines. Are we getting some kind of data corruption? First I need a kernel, and building that took the rest of the day.
Peter Jeremy reminded me of the typical example of the stupidity of “top posting”:
| Wednesday, 23 September 2009 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology | Link here |
Last night I had rebooted Yvonne's computer with the new kernel, but hadn't done the installworld yet. Came in this morning to find it not running properly; the root file system was full, and the network was down.
OK, I had just built a new kernel, and I had collected a few dumps during the day, so moved the objects to the /home file system, after which I had barely enough space. The network issue was more interesting: the name of the interface has changed, from nve0 to nfe0! I don't know why BSD needs to have different names for each interface; I think this is one area where Linux does better. What I saw in the dmesg output was (before, then after):
That's easy enough to fix, but it needs fixing. And that explained the real reason why the root file system was full: the nightly dump, which goes via NFS, had written itself to the root file system. What a pain.
While tidying up, came across another strangeness:
=== root@lagoon (/dev/ttyp2) /var/crash 3 -> l
=== root@lagoon (/dev/ttyp2) /var/crash 4 -> rm vmcore.* info.*
=== root@lagoon (/dev/ttyp2) /var/crash 5 -> df .
So I had removed 6 GB of dumps from a file system that wasn't quite full, and at the end I had only 4223 MB free? How did that happen? It occurs to me that I should have looked to see whether there were any symbolic links there, but locate didn't find anything else. Strange. Are we writing dumps with holes in them nowadays?
Another recent occurrence may be the result of the recent bad weather: on Monday the Internode data centre in Adelaide had a power outage, and the old AUUG web server didn't come back up. Today David Newall got in contact with them to find out what had happened: the (only) disk had failed. Stephen Rothwell intends to put up a replacement machine soon, but currently it's off the net. So not only is AUUG dead, the web server is too. It's something like the end of an era.
Peter Jeremy is finding some nice silly mail exchanges. Here's another:
The reference to USENET shows how things have changed since we were all able to communicate with each other.
| Monday, 28 September 2009 | Dereel | Images for 28 September 2009 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
| Topic: technology, opinion | Link here |
A few months ago I signed up for Twitter, for reasons that didn't even make sense at the time, and only a couple of weeks later they changed the password rules, so that when I tried to log in, I was told that my password—previously accepted—was invalid, and I had to change it.
I didn't use Twitter again, though I note with amusement that the announcers on ABC Classic FM now refer to the users as “Twits”. Today I got a message with the subject line “Deporte6am wants to keep up with you on Twitter”. That shouldn't be difficult, given the speed with which I use it. Still, tried to log in again, and this time I didn't get the message that my password was wrong or invalid. I just got a new login screen, repeatedly.
Doubtless this is an adaptation to the preferred clientele, but clearly it means “your password was not accepted”. Sent off a message to reset my password, and didn't get the promised mail message: it had been eaten by SpamAssassin:
Reset my password again, and got the message:
|
|
||||||||
Clearly the message is wrong: the password did contain spaces, but they don't accept it. So they're continually weakening their passwords. Never mind; I haven't found anything of interest on Twitter, so I'll leave it the way it is. No idea who Deporte6am is, but it wouldn't help anyway.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
