|
| Greg's diary |
| August 2012 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Wednesday, 1 August 2012 | Dereel | Images for 1 August 2012 |
| Top of page | ||
| next day | ||
| last day |
|
Fine time for spraying
|
Topic: gardening | Link here |
The weather isn't any warmer for the new month, but it was more pleasant: sunny and low winds, just the kind of weather I needed for spraying weeds. Got rid of an entire backpack full (16 l) before I got fed up. I should really mow some of this stuff before continuing.
Also made some progress towards the next row of Hebes, between the garage and the greenhouse. This tidying up somehow goes so much against the grain, but with any luck I'll be able to lay the weed mat tomorrow.
|
Repairing laser printers
|
Topic: technology | Link here |
My old Brother HL2700-CN colour laser printer hasn't worked for years: it had extreme paper feed problems. That's a
typical problem for devices without tractors, of course, but getting it repaired
even looked at would have cost me a minimum of $80, and that was more than the price of a
new black-and-white printer, so two years ago I bought one. Now that one has run out of toner, and of course a full toner cartridge
costs more than a new printer, so here I am again. I'd like the flexibility of duplex
colour printing, and the prices are now down to as low as $250 or so.
But what's wrong with my current laser printer? Surely paper feed problems can't be that bad? Clean the rollers and it should be running again in no time.
So I dragged the printer out of the corner where I had boxed it in and fired it up. Press Go to print a test page, right? No, that just goes to Pause. Checked the instructions: “Press Go to print a test page”. Wonderful.
So I looked for an Ethernet cable to connect it to the network. None long enough for where I had put it, of course, so I had to move it again, and it's heavy. And yes, the web interface allows you to print various kinds of test pages, so I tried that, and sure enough, I got a “Jam A1”. Paper had hardly moved, and I didn't hear any sounds suggesting that it had even tried. This jam is the roller at the end of the paper tray:
|
|||||||||||||||||||||||
That's looking up from the paper feed. The roller looked pretty clean, but I gave it a wipe with a moist sponge cloth anyway, which didn't remove any significant amount of dirt. Still paper jam A1. Took a look at the next rollers (which would have been involved in jam B). They looked dustier, but then, the paper never even made it there:
|
|||||||||||||||||||||||
The magenta line appears to be spilt toner.
So: I've seen this before. A sensor somewhere? I didn't see anything there. But maybe a connector? Yes, there's one there, placed in a very inaccessible place. Spent an inordinate amount of time trying to remove as much as possible to get at it, and finally gave up after removing the main logic board and discovering that there was no access behind it:
|
|||||||||||||||||||||||
Put the thing back together, and for the sake of completeness tested it and—feed sounds, then Jam B! That means that at the very least it is now feeding. Removed that and the subsequent Jam D (in the duplex unit) and tried again. It managed its 3 page (2 sheet) test print with no problems. So it looks as if it's sort of functional again, though I didn't stretch my luck by trying any more.
Somehow modern devices fail in different ways from old ones. Light bulbs used to burn out; now it seems that some other circuitry is often involved. Disk drives used to have head crashes; now it seems that it's the electronics or the drive motor that die. Paper jams used to be due to dirty rollers. Now they seem to be due to poor contacts.
| Thursday, 2 August 2012 | Dereel | Images for 2 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Google Translate: not Google Maps
|
Topic: language, opinion, technology | Link here |
Somebody pointed me to an interesting article today. The Malaysian constitution has an interesting clause that gives specific rights only to the Malays, one of many ethnic and religious groups in Malaysia. This has given rise to a certain amount of ethnic and religious tension, and in view of the increasingly radical nature of Islam it's nice to see a more balanced attitude from a Malay: “Remember what Islam has instilled in you, not what Muslims have told you.”
The bio of the author was written in Malay, a language I once thought I understood. I didn't understand it all, so I went to Google translate for help. Between that and what I already understood I was able to make sense of it.
But how does Google translate go if you tell it that it's in Indonesian and not Malay? The languages are almost the same: mainly the loan words are different. And indeed, Google translate did about as good a translation when told it was Indonesian. But how about translating Malay into Indonesian? It seems to be roughly OK, with one important and obvious difference: “di sebuah firma guaman di Kuala Lumpur” becomes “di sebuah firma hukum di Jakarta” (“in a legal firm in where?”). Clearly Google Translate is overstepping its boundaries.
|
Temperature control problems
|
Topic: gardening, food and drink, opinion | Link here |
Sometimes the influence of temperature surprises me. It's becoming clear that my seed germination is still suffering from too low temperatures, despite a heater in the vicinity: I'll have to reinstate thermostatical temperature control. And keeping the house warm is difficult for two different reasons: firstly, the terrible temperature control of these Fujitsu air conditioners, and secondly the increasing cost of energy. The results became clear when I baked a loaf of bread today: I put it in the server cupboard to rise, but after 3 hours, it had hardly changed. The temperature there must be lower than normal. Put it in the oven at 40° for an hour and things improved, but it's clear that here, too, I should do some kind of thermostatical control.
|
More garden progress
|
Topic: gardening | Link here |
I had intended to finish the row of Hebes between garage and greenhouse today, but I wasn't able to start either lawn mower: the push mower was out of (2-stroke) fuel, and the ride-on mower had a flat battery. Mañana. Instead did some rose pruning and planted some cuttings—again. The ones I did in autumn started to develop leaves which then died off again, and now with one exception they're completely dead. Replaced the cuttings in the same pot, in the process noting that the old ones had developed no roots at all. No wonder they died. Hopefully it's not too early to propagate the new ones.
Also moved one of the shelves from the shade area to the other (south) side of the garden shed, where it'll be in the shade almost all day long, with just a little sun in the early morning. Hopefully that'll be a good environment for the cuttings to grow.
| Friday, 3 August 2012 | Dereel → Ballarat → Dereel | Images for 3 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
To the doctors
|
Topic: general, gardening, photography | Link here |
Yvonne brought something unplanned back from Adelaide: a cold. She's not feeling at all well, and she even managed to give me part of it, a throat infection. That's bad news for me: I almost always need antibiotics to get rid of it, and today it seemed I had reached that point. I had to go to town today anyway to have the stitches taken out of my gum after the operation last month, so made an appointment with Dr. Peddi as well.
A little late to Leela, the periodontist, and as a result he let the next person ahead and I had to wait half an hour. The gums seem to be healing nicely. A while back Tom Maynard recommended the use of Waterpik to keep the gums clean, so I asked him what he thought of it. I've seldom seen him so vehement: there is no clinical evidence that it does anything useful, and it's a waste of time. It's true that I found little use in a similar device decades ago, but times could have changed. What I heard from Leela sounds like a version of “For every problem, there is a solution that is simple, elegant and wrong”.
Coming out of the dentist's, found a tree that I hadn't expected:
|
|||||||||||||||||||||||
Aren't toy cameras fun? This was taken under normal situations with flash fill-in, and it still managed considerable camera shake. But more importantly, it's a Viburnum prunifolium, and it's about 4 m tall. I have one at home in a pot, about 30 cm tall. For once I know just the place to plant it.
Then to the Botanical Gardens to take some panoramas, and discovered that in my haste I had forgotten to bring a rail with me, so no real panoramas, just a view of Lake Wendouree taken freehand:
|
|||||||||||||||||||||||
Then on to Bunnings to buy an inspection lamp to heat my seed trays. The normal el-cheapo ones have a 60 W globe, but they also had much better looking ones with 150 W halogen lamps for the same price, so took one of them instead.
Down to the doctors. They're now in a new building in Sebastopol, between ALDI and Dahlsens, much more convenient for us. Building still looks empty, and it turns out that 12:30 is not a good time to make a doctor's appointment: it's the first appointment after lunch, and they all came back 15 minutes late.
|
Continuing planting hebes
|
Topic: gardening | Link here |
Back home and finally got round to mowing the lawn and putting down weed mat between the garage and the greenhouse. And of course I ran over something hard with the (ride-on) mower. Didn't do the mower any obvious harm, but it made a hole in one of the greenhouse panes:
|
|||||||||||||||||||||||
I should have expected that. Now there's the question whether it's worth the trouble to replace it. It's not as if it would make a big difference to the function of the greenhouse.
Got as far as laying the mat and enough mulch to ensure that the mat wouldn't blow away. But at least I'm getting somewhere.
|
Panorama brackets: the other shoe
|
Topic: photography, opinion | Link here |
It's been over 7 months since I got my new panorama bracket. On the whole, after adding a rail, I'm happy with it, but this silly rotator got on my nerves, and I wasn't able to get it off.
And then today I discovered it was really wobbly. I still couldn't get it off, but the presumed epoxy glue had clearly come undone, so out to the garage to drill out the central screw. And when I had done that, it was tight again. There were two smaller screws offset from the centre, and it seems that the vibration of the drill caused them to tighten up again. But that wasn't too difficult to fix, and finally I have the panorama bracket mounted directly on the rail. Here the situation I had in January and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
| Saturday, 4 August 2012 | Dereel | Images for 4 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More insights with house photos
|
Topic: photography | Link here |
On the first Saturday of the month I take even more exterior photos than normal, and today it kept me busy all day long. Nothing of any great interest except for the verandah, where I have always had difficulty with the strong contrasts. Today was bright sunshine, which makes things much worse. There wasn't too much wind, so I tried HDR images made from 3 exposures bracketed at 2 EV apart, but with automatic exposure offset by 2 EV, so that each image was composed of three components exposed at 0 EV, 2 EV and 4 EV from what the camera thought. The result was individual images with an exposure distribution of 7.7 EV (at fixed f/8, times between 0.4 s and 1/500 s). Automatic exposure is supposed to be a no-no, but the latest version of enblend does very well with it, and I'm quite happy with the results. Here last week and this week:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Despite the sunshine today, the shadow detail is much better:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
So there's only so much that DxO Optics “Pro” can do. As long as there's no wind, it seems I should return to multi-exposure techniques.
|
More hebe row work
|
Topic: gardening | Link here |
Didn't find much time for work in the garden, but did manage to complete the mulching around the row of Hebes between greenhouse and garage.
|
|||||||||||||||||||||||
Now there's just the irrigation and the plants themselves to do.
| Sunday, 5 August 2012 | Dereel | Images for 5 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Nemo: champion in obedience
|
Topic: animals, opinion | Link here |
Towards evening I saw Yvonne out in the garden, so went out to join her. Nemo wanted to come too, so I let him out, and of course he ran straight to Yvonne—until he saw a kangaroo. Then there was no stopping him. I did manage to get him to come back, but only briefly, and then he was off again into the lagoon.
He came back some time later, after I had been on a couple of attempts to locate him, and he was unhurt. But I've had enough of this: this is supposed to be a dog who has been well trained in obedience, and he's very much an outdoor dog. But we can't even let him run free in the garden. He's just more trouble than he's worth—to me, at any rate. We should get rid of him before a kangaroo tears his guts out.
|
More meat roasting times
|
Topic: food and drink | Link here |
A rather poor quality Kassler from Standom for dinner tonight. And again the question: how long to cook it? It's pre-processed, which generally means that it should not be cooked as hot, so chose the temperature marked for ham on my ancient meat thermometer, 74°. According to my previous notes, that should have taken somewhat less than 40 minutes; in fact, it took 66 minutes. I wonder if there's something wrong with the oven thermostat.
| Monday, 6 August 2012 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
ABC: redefining language
|
Topic: language, opinion | Link here |
I listen to ABC Classic FM [radio] quite a bit. The programme isn't bad, but one thing that grates is the inability of the announcers to pronounce correctly the many non-English words that they have to use. Why doesn't ABC have a group to help them?
But they do, as Peter Jeremy pointed out to me recently: the SCOSE, a term which sounds like a British dialect, but which apparently stands for the Standing Committee on Spoken English. And they mandate this nonsense. In the interests of uniformity, they bend names like Curaçao to /kjurəˈseɪoʊ/, but don't change “Mao Zedong” to something that might remind you of mayonnaise. They explain that you have to pronounce “Paris” the English way, because that's what our grandfathers did, but they (presumably tacitly) accept changes of name like “Peking” → “Beijing” or “Calcutta” → “Kolkata”.
The last thing that we need in a small country like Australia needs is Yet Another Way to pronounce non-English words. There's only one way to avoid this kind of confusion: pronounce the words like a native speaker, or at least as close as you can get. Of course, when you get to places like Liège or Luik or Lüttich, even that doesn't help.
There's much more to say, but that's enough here. I've written a separate rant on the subject, which I'll probably update from time to time.
|
Stuffed sardines again
|
Topic: food and drink, opinion | Link here |
Sardines again for dinner tonight—Yvonne brought quite a few back with her from Adelaide—and again she filled them. Last time she used ricotta cheese in the filling; this time she used feta. Yes, feta doesn't melt as easily, but it proved to be far too salty. Maybe some other soft cheese next time?
| Tuesday, 7 August 2012 | Dereel → Ballarat → Dereel | Images for 7 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More house inspections
|
Topic: general | Link here |
For some reason that I haven't quite fathomed, Yvonne is thinking about moving house again. Now she has come up with the idea of transportable homes, and she discovered a maker in Ballarat, Bond Homes, so we set off to take a look.
I didn't have much in the way of preconceptions, except that they would be cheaper than conventional houses. They're not. They cost and look pretty much the same, and about the only real difference is that they're designed to be taken apart and transported in segments. That makes sense, I suppose, if you move house (literally!) a lot, but it doesn't seem to offer much if you just want to live in one place.
About the only thing we discovered is that the building regulations have become even more complicated since the bushfires of 2009, and that getting planning permission for either of the sites we had in mind would be pretty difficult. And once again the electricity supply would mean that it wouldn't make financial sense.
|
More ground cover
|
Topic: gardening | Link here |
While in town, dropped in at the Friends of the Ballarat Botanical Gardens and talked to Bruce Holland and Yvonne Curbach about ground cover that would suppress grass and other weeds underneath. Ended up with four Myoporum parvifolium, which grow to a diameter of several metres.
Back home did some investigation and found that they really want dry conditions, so they wouldn't be appropriate for the corner for which I had planned them, to the south-east of the verandah. But by that time I had done what I hope will be the last weeding of half of the area, and covered it with mulch.
|
|||||||||||||||||||||||
Now to decide what kind of ground cover to put on top.
|
Making progress on time-lapse with Hugin
|
Topic: photography | Link here |
The summer before last I tried making some time-lapse photos of some work I did in the garden, with spectacular lack of success: I had taken the first of the three images at a different focal length, and I couldn't find a way to resize it. Today I had another go and made some limited progress by fiddling with the optimizer parameters. But the exposure (all constant and correct) went to hell. Here the two images that do line up (run the mouse over the first image to see the second image) and the corresponding output image I got today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Not only is the exposure wrong: the corners have been cut off. I suppose I should read a lot more about the background of these optimizations.
| Wednesday, 8 August 2012 | Dereel | Images for 8 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More spring preparations
|
Topic: gardening | Link here |
On with the garden work today. After discovering that the Myoporum parvifolium didn't fit where I had planned them, today I changed course and started planting Gladioli. That was more complicated than I had hoped: most are from assorted lots that I bought last year, and which I then tried to categorize after they had finished flowering. The result was that I don't know the colours of some of them. And I'm planning staggered planting to increase the flowering time—particularly sensible considering I'm planting only like-coloured corms together—so it'll be a month or more before I'm done.
Then there's the question of the cormlets that many of them have developed. Checked again when to plant them: about 6 months ago, if you believe some of the information I found on the web. But then, some also claimed that the main corm dies after one year, so you need to be careful. This page sounds reasonable, as does this one. Planted a large number in big pots. We'll see how they do.
|
Alternatives to DxO
|
Topic: photography, technology, opinion | Link here |
I've been using DxO Optics “Pro” for raw image conversion for over 6 months now. It seems to do its job well, but at a completely unacceptable lack of speed—up to 3 minutes per image. Recently Phase One have had a special offer on their “Capture One” software, so I thought I'd try it out.
Like most photo processing software, it seems, there's a free trial—in this case, two months with unrestricted use. I wonder how many people just reinstall every two months. In any case, the installation went smoothly, and once again I was presented with a grey tiled environment. It's not intuitive; I managed to load some images (last week's 239 images, in fact). It insisted on reading them all in and doing some kind of preprocessing, which took over 20 minutes. I'm doing this with Samba and VirtualBox, which could be part of the lack of speed, but I wonder how much faster it would be on a native Microsoft box. It's difficult to say how fast the processing itself is, because I couldn't work out how to do it. Clearly a case for RTFM.
|
Grilling chicken drumsticks
|
Topic: food and drink, opinion | Link here |
Grilled chicken drumsticks for dinner tonight. They're more complicated than they look: it's very easy to undercook them, leaving bloody areas round the bones, and it's not that difficult to overcook them either. Decided today to first bake them under aluminium foil until they were cooked, then allow them to cool a little and grill them.
Set a temperature of 74° with the temperature probes close to the bone. After reaching that temperature, turned off the heat, and the temperature continued to rise to about 80°. Grilled them, which went well, but they were still underdone. I'll aim for about 78° next time.
| Thursday, 9 August 2012 | Dereel | Images for 9 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Firefox: where's the documentation?
|
Topic: technology, opinion, music | Link here |
I've been using firefox for nearly 8 years, and I've been complaining about it for just as long. Why do I even bother? To misquote Jawaharlal Nehru, “firefox is good. I say this because other systems are worse”.
Today I received mail from Volkan Yazici suggesting that I listen to Radio Swiss Classic. OK, select that and get the well-known question:
|
|
||||||||
OK, that's simple: click on “Browse” and type in the name of the helper application. But the latest version of firefox wants to save me the trouble of using the keyboard, and it gives me this window:
|
|
||||||||
This window was never good, but now the defaults are even worse: previously I would get a list of my home directory, as if I would keep “Applications” there. Now the files are just the base names of files in dozens of different directories, with no indication where, though you can run the mouse cursor over the name and get a complete pathname, if for some reason that might interest you. Not a single one is has the x bit set or is in the PATH, and most are PDF or JPEG files. Even a program that doesn't understand Unix permissions can guess that you're not going to run an image or a document as a “helper”. And they've removed the input field for the file name. Why do people do this?
I've seen the “no input field” thing before, and there was some magic key combination that was supposed to bring it back. But I forget. RTFM time. And then it occurred to me that I have never seen a manual for firefox. Went looking, round and round in circles, and I've still never seen an official manual for firefox. It doesn't seem to exist.
I did find a large amount—under the circumstances, not surprising—of third party documentation. This one might be worth reading. But it didn't help me much either. Went through all the Control and Alt key combinations, with no success. Then I tried clicking on the Places links on the left. Search did just what it suggested: started some kind of search. I have 8,870,708 files on this machine (a number surprisingly reminiscent to my last mobile phone number, 0418838708), and I don't need to search them all. The file I'm looking for is called /usr/local/bin/mplayer-playlist. Why can't I just type it in?
Then I tried File System, which just took me to the root. I could then climb down the tree—indeed, that's what Edwin Groothuis considered a solution—but why? I'm not a monkey. But then I saw that this selection had added two unrecognizable icons at the top left. Running the mouse cursor over the left-hand one shows:
|
|
||||||||
And after clicking on that, I finally got the input field that should have been there in the first place, and I was able to type in the name. Isn't firefox easy to use? And why not?
What's the other icon? Not much of an idea. It doesn't have a tooltip popup. When I click on it, the file name part of the window flashes, so maybe it's a refresh button.
Of course, after all that pain, it occurred to me that I don't really want to run mplayer from firefox: it doesn't have a popup window, so I can't pause or stop it. But that's a separate issue.
|
Pruning the ornamental vine
|
Topic: gardening | Link here |
Although spring is ostensibly just round the corner, the weather today was cool and wet, and despite my intentions didn't get much done in the bed I'm currently overhauling. Instead got under cover and started pruning the ornamental vine growing under the roof of the verandah. It's amazing how much I had to remove, and how slow it is; I don't think I'm even half way yet.
| Friday, 10 August 2012 | Dereel | Images for 10 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
A new house after all?
|
Topic: general | Link here |
The big issue with Yvonne's quest for a new house is that we don't really have anywhere to build it. Our intentions had centred round either the north half of our current (8 ha) property or one of Chris Yeardley's plots of land (an entire 8 ha). But for the former we'd need to be able to subdivide the property, and it's been made clear to us that that would be an uphill battle.
For Chris' plot we'd need planning permission, which John of Bond Homes suggested would not be easy. In addition, we'd need to shell out at least $50,000 for the dubious privilege of a Powercor electricity connection. As if that wasn't enough, the current coverage maps of the Radiation Tower suggest that we wouldn't get NBN coverage there. All in all, it doesn't seem to make much sense.
But there's another possibility: buy a house on a smaller plot of land and leave our horses at Chris' place. Did some investigation and discovered that yes, indeed, there are a number of places which we could buy and probably make $100,000 in the process. So off to a number of different places. About four of them stood out, including one of our neighbours. Two of them are in Ferrers Road, and they're both on display tomorrow, so we'll go and see what we can find there.
|
Pruning the vines
|
Topic: gardening | Link here |
Continued today with pruning the ornamental vine on the verandah. It took me another 2 hours, and produced an amazing number of cuttings, including a couple that had grown in circles. It looks a lot barer now, of course, but hopefully it'll be better for it by late spring.
|
The revenge of firefox
|
Topic: technology | Link here |
Now that I have my printer running, it seemed to make sense to print out the real estate web pages I had been looking at. The result was abysmal:
|
|
||||||||
How can firefox possibly make such a mess? I've seen it before, and it seems to be some setting, but which? I can't find anything obvious, and it's not clear how firefox should have code in it to make this much of a mess in the first place. Once again, where's the documentation?
| Saturday, 11 August 2012 | Dereel | Images for 11 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
House photos: increasing the pain
|
Topic: photography, technology | Link here |
Gradually I've got into a routine with my weekly panoramic photos of the garden (“house photos”), so clearly it's time to change things. My routine involves taking the partial images in raw format, converting them to JPEG with DxO Optics “Pro”, and then stitching them with Hugin. But today it occurred to me that it doesn't make sense to use JPEG for the intermediate steps: both DxO and Hugin understand TIFF, so I should convert my intermediate images to TIFF.
The results remind me of the adage: for every complex problem, there is a solution that is simple, elegant ... and wrong. In any case, a seemingly simple modification brought untold pain with itself. The first, still relatively benign, appears to be that DxO emits invalid TIFF files—or maybe they just use features that other software, notably including libtiff, doesn't understand:
=== grog@eureka (/dev/pts/19) ~/Photos/20120811 44 -> convert 169-Ferrers-road.tif foo.jpeg
Apart from the mess on the screen, this doesn't seem to have done much harm. But other programs, notably xv, got really upset, producing a series of pop-up error messages that followed the cursor and stopped me from getting any work done.
The first panorama (the complicated verandah one) worked relatively well, though it seemed that the raw conversion wasn't as good as with JPEG output. That will have to wait until I have processed all the images. But then my own scripts came to bite me, and I managed to accidentally delete some images several times, not speeding up the processing. While I was doing that, continued with some other panoramas. The preview of the north view panorama was surprising:
|
||||||||||||||||||||||
The first thing to hit me was that the 360° panorama didn't close. That proved to be due to complete lack of control points between the adjacent images at each end. It wasn't until some time later that I discovered that the arch in front of the house (about half right in this image) was missing: the alignment had caused it to be hidden. And the choice of control points between those two images made the problem clear:
|
||||||||||||||||||||||
Despite the presence of lots of common features, all of the control points had been “found” in the amorphous sky above. And it's clear that the control points are in the same place in the image, completely unrelated to the content.
Apart from that, it's clear that processing TIFFs is slower. After all, they're 70 MB each; over the 100 Mb/s link I have between braindeath and eureka, that's nearly 7 seconds per image, and the panoramas required up to 40 images. So it's not surprising that that was about all I had time for today. What a mess!
|
Inspecting houses
|
Topic: general | Link here |
Two of the houses we passed yesterday were open for inspection today, both in Ferrers Road. Before lunch we went to number 169, a mud brick house on 2.02 ha. It didn't look too bad, in fact, though some of the details needed improvement. But like most of the houses in the area, the rest of the property was covered in tall trees, and there was no grass worth mentioning at all, so it would be wasted on the horses. And the sheds were in pretty rough condition, reminiscent of some of the sheds we had when we moved in to Kleins Road. So it wasn't too difficult to decide against it.
After lunch to the second house, number 229. It has one of the nicest gardens I've seen in any house for sale:
|
|||||||||||||||||||||||
It also had lots of good, modern looking sheds, several hundred m² of them. But the house itself was only passable, and once again the property was covered in trees.
|
|||||||||||||||||||||||
It's a pity about that; otherwise it could have been a possibility.
|
Mystery vine
|
Topic: gardening | Link here |
At the first house we visited, found a vine growing and flowering vigorously:
|
|||||||||||||||||||||||
What is it? Took a cutting and planted it. We'll see if anything comes of it.
|
Lamb shanks tagine
|
Topic: food and drink | Link here |
A new experiment today, lamb shank tagine. It was a combination of a number of different recipes, none of which appeared overly authentic, but it didn't come out at all badly. Fortunately I started early enough: I had assumed that 2 hours cooking would be enough, but we definitely needed 3. And of course some of the proportions were completely out: surprisingly, we nearly completed one shank per person (Yvonne, in particular, has a very small appetite), but we only ate half of the sweet potatoes.
| Sunday, 12 August 2012 | Dereel | Images for 12 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More TIFF processing
|
Topic: photography, technology | Link here |
Spent much of today trying to process my panoramas, gaining more insight than success. Yes, the problems I had are clearly associated with processing TIFF images. Nearly all the 360° panoramas were badly broken, but using JPEG copies of the same images they worked fine. Here an example:
|
||||||||||||||||||||||
The control point detectors found plenty of valid control points, but something caused Hugin to turn over some of the images, and I couldn't work out how to get it to accept the orientation:
|
||||||||||||||||||||||
So I converted the images to JPEG, making no other changes, and things worked fine:
|
|||||||||||||||||||||||
So clearly something is wrong with the TIFF image processing. Now I need to find out what it is. It seems to be more than one thing: the example above shows that the images get rotated although the control point detection is successful. But then we had the control points in the sky in yesterday's example. With JPEG, the detector comes up with much better results:
|
||||||||||||||||||||||
|
||||||||||||||||||||||
But what about the reason for doing all this in the first place, the higher image quality? I don't see much difference in the verandah centre image:
|
|||||||||||||||||||||||
The only way to compare these images is to run the mouse cursor over them, and even then it's difficult to see any difference at all. And it's not clear which is better. Admittedly, the base image quality is not as good as last week, so I'll try again next week. But every indication so far is that using TIFF brings me nothing but pain—unless it's the disk usage:
=== grog@eureka (/dev/pts/10) ~/Photos/20120811 27 -> du -sm
23 GB! What a lot of data. Normally it's less than 2 GB. Clearly I will have to be more careful about what files I keep.
|
Garden flowers in late winter
|
Topic: gardening | Link here |
Today was the day for the middle of the month garden flower photos. Things haven't changed much since last month, and compared to last year's photos there's a big difference. But they were taken at the end of the month, so it's clear that some things might not yet be in flower.
| Monday, 13 August 2012 | Dereel | Images for 13 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Another house inspection
|
Topic: general | Link here |
Round to 161 Rokewood Junction Road at midday today to look at the house. It's just round the corner from where we live now, about 350 metres away in direct line, though Google Maps suggests it's more like a kilometre by road. It has a lot going for it, including mainly cleared paddocks, a house well back from the road, and good fencing—they're keeping sheep on the property. And the house has possibilities:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
What's missing is mainly water, but also heating. Only the main room (first photo) has any heating at all, and that's wood-fired. On the one hand, it's becoming clear that with current energy costs, wood-firing (with its concomitant CO2 emissions) is financially very attractive. What strange results the carbon tax has!
The house doesn't have a bore, but it does have two wells. We would need to clarify the situation there. But I really don't feel like moving house Yet Again. And not being able to make up my mind keeps me from doing anything useful in the garden.
|
Seed tray heater
|
Topic: gardening | Link here |
I have decided that the reason why my seed raising efforts last year were due to too low a germinating temperature, and had already installed a thermostat to control the temperature. But I only set it to 18°, which, it seems, isn't enough for many seeds. So now I have bought a new heater (a 150W inspection lamp of more sturdy construction), installed that, and I'll aim for 22°.
|
New pocket calculator
|
Topic: technology, opinion | Link here |
I first came in contact with electronic calculators at university, in about 1970. A huge thing from TI with Nixie tubes, so expensive that we only had one and had to share it. Only a few years later I got an HP 45, much smaller but also very expensive. Times have changed. A week or two ago I bought a small basic pocket calculator on eBay for $2.97, including postage. The days of expensive electronics are mainly gone. It arrived today, and it does the job (calculations in the kitchen). It even has sound for the key presses. But the documentation! The only documentation of any kind was on the box:
|
|
||||||||
Those aren't even instructions, just examples. But who needs instructions? Everybody know how to use a pocket calculator.
| Tuesday, 14 August 2012 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
TIFF panorama problems investigated
|
Topic: photography, technology | Link here |
I was surprised by the poor results I got from using TIFF images for my panoramas on Saturday, and today I did yet more investigation. I had thought that I had had two different problems: one where the control point detector is all up in the sky, and one where they appeared to be OK, but the images were rotated. I concentrated on the latter one and found yes, some of the images were rotated, but I also had this “up in the sky” syndrome on others, where the control points were all in the sky and in the same place on the image, not related to what was represented there:
|
||||||||||||||||||||||
|
||||||||||||||||||||||
So possibly it was the control points in the sky issue that caused problems in both panoramas. Clearly it's a bug, but whose? We have the following choices of processing options:
Convert the raw image to a portable format, in this case either JPEG or TIFF. I have three different programs that can do this: UFRaw, DxO Optics “Pro” and Olympus Viewer 2.
Detect the control points. Again, there are several possibilities. Normally I use panomatic, but there are also least cpfind and align_image_stack, which come as part of Hugin.
In addition, given the error messages I had seen, I considered the possibility that the EXIF data in the files could be the problem, so I tried variants without the EXIF data.
After a large number of experiments, came out with the results: this particular issue only happens with DxO with 16 bit TIFFs and panomatic. cpfind has no difficulty processing the DxO output. align_image_stack fails to find any control points at all in any of the versions I tried. I gave up on align_image_stack some time ago, mainly because of the slow processing (up to 30 minutes on this particular panorama), but this failure suggests some other problem that I don't want to investigate.
But what causes the errors? Tried again with just the two images above with the control points in the sky. Bingo! They are a perfect match:
|
||||||||||||||||||||||
The main page claims both average and maximum error of 0.0 pixels. This suggests to me that panomatic is finding something in the images that is identically placed in each of them. And that suggests to me something like a watermark inserted by DxO.
The control point detectors weren't the only thing that complained about the DxO output. xv also complained bitterly and at length—about both 8 and 16 bit DxO output, but not about the others. nip2, on the other hand, was able to handle the images correctly with only the usual “Unknown field” messages.
Interestingly, none of the image converters produced TIFF files that libtiff liked. The messages themselves leave something to be desired: in particular, the line number at the end of the line appears to be the line number of the command printing it out, which isn't very useful, and the dir at the beginning is in fact the name of the directory and not of the file being processed:
DxO file:
Olympus file:
ufraw file:
It's not clear where these errors are. Amusingly, in the Readme.txt for the Olympus viewer I read:
That's the same library that's complaining here, and Sam is, of course, one of the FreeBSD developers.
Olympus also has the fewest errors, all of which are shared by DxO, and some of which are shared by ufraw (the others are possibly masked by other errors). I suspect that the Olympus errors come from the source file. The errors peculiar to DxO are:
The 8 bit version generated by DxO also generates many error messages:
They include all the errors generated by the 16 bit version, so the messages are probably irrelevant to the problems.
In summary, though, this looks like some problem in DxO that causes problems for some, but not all other programs I try to use to process it.
|
New tap for the kitchen sink
|
Topic: general | Link here |
The mixer tap on our kitchen sink has been leaking for a long time, and we always wanted one with a pull-out hose. Finally I bought one, and it arrived yesterday. I hadn't been looking forward to mounting it, but got round to it today. Surprisingly, had little difficulty removing the old one, not even water leaking out of the open pipe ends. But then I discovered that the connections are different, and I'll need either new hoses or a new adapter, and will have to find a way to increase the size of the cutout in the sink. In the meantime, back to the old tap.
While I was pondering this, heard a gurgling noise: after about 10 minutes of being disconnected, the hot water suddenly disgorged about 15 l of water, much of which I wasn't able to catch in time. And after mopping up that, had the devil's own job replacing the tap: nothing seemed to fit any more. Finally got it done, but what a pain!
|
Belle de jour revisited
|
Topic: multimedia, opinion | Link here |
Decades ago I saw “Belle de Jour“, a film that greatly impressed me. It was on TV recently, so we recorded it again. What a difference! Much more sordid than I recalled, and it has certainly aged a lot. Stopped in the middle, and though I didn't delete it, I don't think we'll watch the end.
|
Wood firing: carbon neutral?
|
Topic: opinion | Link here |
Message from Lucas James reflecting what others had said on IRC: wood fired heating is carbon dioxide neutral. A lot of people say this. I don't agree. It's confusing two issues:
Burning carbon-based fuels, which creates carbon dioxide. The amount per kJ depends on the fuel, but I don't have any comparison. I suspect it's comparable.
Binding this carbon dioxide, which you can do by growing trees.
So the only argument for wood is that if you then go and plant trees, you can bind the carbon dioxide again. That applies to fossil fuels too: the trees don't care where the carbon dioxide comes from. Still, there are some advantages for wood: it's a renewable resource. But it loses in comparison to, say, hydroelectric energy.
| Wednesday, 15 August 2012 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Browser bugs
|
Topic: technology, opinion | Link here |
Finally got round to investigating the print mutilation bug in firefox that bit me last weekend. As I suspected at the time, it must be some configuration issue: I run multiple instances of firefox, all on the same system and from the same executable, and so far only one of them has shown the issue, though I haven't tried them all.
So: off to look for about:config, a pseudo-URL that firefox doesn't honour when presented as a link, and found dozens of user set variables that I don't recall setting. Tried printing a page: success. The bug has gone into hiding.
While discussing it on IRC, Michael Ralston suggested that I try Chrome (or is that chromium? Or google-chrome? Or chromium-browser? Or even edited zucchini? They seem to have difficulties deciding on a name). I've tried it before over two years ago, when I had problems with the configuration, and again last April, but didn't find enough reason to change. Maybe the insistence on tabs turned me off. But now I've discovered that firefox only runs with any semblance of reliability with tabs, I've had to capitulate. So installed the Google browser, whatever it's called.
And again it works. No time to try it out in more detail. But this infuriating insistence on having only one running instance, on one display!
=== grog@eureka (/dev/pts/8) ~ 2 -> chrome
If there's already a running browser process, it completely ignores the DISPLAY setting and opens a new window on the display of the running process—which might be on another computer! OK, maybe that's what many people expect, but there should be a way around it. Did some digging and found this discussion, which might help. But once again I'm left feeling that this is a browser designed for somebody very different from me.
And printing? So far I haven't found anything that suggests it can print to a printer. There's something on the settings page called “Google Cloud Printer”, which requires me to log in to Google—almost certainly not what I want. More investigation needed.
|
Little work round the house
|
Topic: general, gardening | Link here |
Somehow I've been neglecting work in the garden and house again. The weather has certainly helped: cool, wet and windy. Finally got round to mounting an exterior lamp pointing at the dog's kennel, and screwed down the bottom of the shade cloth at the west end of the north verandah, at least partially. I'm going to need some angle brackets to finish the job.
| Thursday, 16 August 2012 | Dereel → Ballarat → Dereel | Images for 16 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Friends General Meeting
|
Topic: gardening | Link here |
Into town today to a General Meeting of the Friends of the Ballarat Botanical Gardens: the guest speaker, Geoff Crowhust, was to give a talk on a recent visit to Bhutan, in particular describing the Rhododendrons he saw there. But he didn't: instead, he talked about Peonies in China, something that interested me considerably less, in particular in view of the time it takes for them to become established. We should be more careful with the advertising of these talks.
|
Fixing the engraving machine
|
Topic: technology, gardening | Link here |
The Friends of the Ballarat Botanical Gardens have a machine for engraving plant labels such as this one:
|
|||||||||||||||||||||||
It had been sent for repair recently, and when it came back they couldn't get it to work. Came in after the General Meeting and took a look. There's a PC (Microsoft, of course) connected via the parallel port to an enormous interface box, and then by another cable to the engraver itself. Adele Thomas, the Engraver-in-Chief, showed me how to run the thing: basically the program prints to the Microsoft spooler, which talks to the interface box as a generic text printer. Only it didn't. The box sat there saying “waiting for data”, while the spooler came out with a Microsoft-typical “There was an error printing”. Clearly a communication problem. Checked the cables: DB-25 at the computer end (of course) going into a DB-15 on the box. Another cable with a Centronics connector on the interface box end went to the engraving machine. All looked OK. Finally got on to John at the manufacturer and went through the software configuration process. No joy.
Finally he got me to describe the connections. Bingo! Both cables have a DB-25 connector at one end. The one with the Centronics connector goes to the computer, and the one with the DB-15 connector goes to the engraver. Terry and Adele had connected them the other way round. Beginner's error? Not really: I checked a couple of times and didn't realize the problem. And without any form of instructions or labelling, how are we to know? At the very least, though, I should have recognized the ambiguity.
|
Rosellas discover the verandah
|
Topic: animals | Link here |
In the afternoon, noted some broken Pelargonium stems underneath one of the hanging pots. Clumsy possum? But then I saw some Crimson Rosellas which appear to have discovered the place:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The pot there contains some Petunias that appear to be making it through the winter. I'm in a quandary: should I chase them away and save the plants, or put up with the damage? I suppose it depends on how much damage, and possibly I can minimize that by offering them something else to eat. On Yvonne's suggestion, put out a bowl with some horse feed, which contains grains. We'll see how they like that.
|
More weeding and planting
|
Topic: gardening | Link here |
The weather lately has really been inappropriate for gardening, but the weeds are enjoying themselves. Spent 20 minutes trying to remove some sorrel intermingled with the Euphorbia “Diamond Frost”. Also planted another round of Gladioli, not made any easier by the lack of knowledge of what colour some of them will be.
|
Chrome: three times no
|
Topic: technology | Link here |
Spent some time investigating why Chrome (or whatever it's called) didn't see my printers. Conveniently found this message on the xterm from which I had started it:
What does that mean? Still, it's an indication, and went searching for it. Finally came up with this bug report, closed as “will not fix”. Chrome wants a PDF printer and will accept no substitute. Potentially CUPS would do it, but that's another can of worms I don't want to open. So, echoing both some of the opinions expressed in the bug report and also Callum Gibson (“There are too many other things wrong with Chrome for me to use it. That's just another”), I'll forget it again.
| Friday, 17 August 2012 | Dereel | Images for 17 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Estate agent visits
|
Topic: general | Link here |
Two estate agents came along to look at the house today: Kate Nolan, the agent selling the house round the corner, and Bram Gunn, the bloke who sold us the house five years ago. Kate wasn't overly well prepared: she arrived with details of the house across the road (admittedly made no easier by the amazing inaccuracy of the authorities). Bram was: he even had a printout of my 5 years later page. The good news from both of them is that we can probably get quite a good price for this house, so we could consider moving. The bad news is that now isn't the time: when I reach pension age next spring I will be eligible for exemption from stamp duty, which would amount to about $18,000. If we're not in a hurry, it doesn't make sense to buy a new house before then.
And again the spectre of building a new house arose. Bram has a plot of land for sale around the corner for $85,000, with electricity. It looks as if we could build a house on that for less than what we would pay for a used house. Went around in pouring rain to see what it looks like. This one isn't for us for a number of reasons, but it's quite possible that others would be.
|
Scanner access: now easy!
|
Topic: technology, opinion | Link here |
Bram wrote a number of notes while he was here, conveniently on a document he couldn't leave with me, so I decided to scan it in. No problem, right? Wrong. It's on pain, my Microsoft laptop, and on starting the scanner program I got:
|
|
||||||||
This happened on attempting to start the scanner program. What does it mean? Where did it come from? I had used the scanner only this morning to scan the house plan, and since then the machine had only been suspended. What's the problem? I didn't find out beyond establishing that it was repeatable, and that it only happened when the scanner was connected. Did the scanner die? It's possible, but I suspect this Epson software, which seems to be particularly fragile and with which I've had lots of trouble.
For the fun of it, I connected it to eureka, my new amd64 box. To my surprise, I had already installed SANE, with which I had also had insurmountable problems, to the extent that I expanded the acronym as “Scanner Access Not Easy”. The last time I tried, the USB system had trouble with the scanner, and since then I haven't had the courage to try it. The results were surprising:
|
|
||||||||
OK, I can live with that. Played around a bit and found that yes, indeed, it now finds the scanner and can scan things. Some of the default settings are particularly baroque, such as the 50 dpi resolution, but that can be fixed. So what's the difference? Clearly the problem with Microsoft was the software, not the scanner—I've seen something similar before—but why couldn't I talk to the scanner before from FreeBSD? My guess is based on a suspicion I've had for a long time, that a lot of my USB pain is caused by problems in the FreeBSD USB stack. With the new system, running FreeBSD version 9, there have been significant changes in the USB stack.
Things aren't over, though: I scanned as root, a “don't do that!” action. Logged in as myself I saw a very different picture:
|
|
||||||||
This is SANE's way of saying “permission denied”, possibly excusable because it can't determine if a device is a scanner until it has opened it (read-write, by the way). Sure, I can change the permissions, once I've found out what the device is. But there should be a way to do that automatically, in /etc/devfs.conf. But no. From the man page:
It does not work for devices plugged in and out after the system is up and running, e.g. USB devices. See devfs.rules(5) for setting ownership and permissions for all device nodes, and devd.conf(5) for actions to be taken when devices are attached or detached.
So I took a look in there, which seems to be a real can of worms. You'd think there would be some way to set wild cards, but the documentation doesn't mention it. Maybe the easiest was is an explicit chmod in /etc/devd.conf. To be investigated.
|
Julian Assange: exposing dishonesty?
|
Topic: opinion | Link here |
One of the purposes of Wikileaks was to expose the dishonesty of Western governments, something that it did admirably. Admittedly, it was foolhardy of Julian Assange to do so, but what has happened to him in the last 2 years appears to show further evidence of dishonest behaviour on the part of respected governments.
Our own Australian government offers all sorts of diplomatic assistance to the scum of Australian society, such as drug dealers being caught in Indonesia. But Assange? No help from the government at all. Quite the contrary: our Prime Minster claimed that his actions were illegal and that they should consider revoking his passport. Clearly our government considers scrupulously honest people to be worse than drug dealers.
And who can believe this story about rape in Sweden? The women involved have pressed no charges. He's just “wanted for questioning”, something that the Swedes could do just as easily in England, and with much less effort. But no, they've turned that offer down. And finally he has been offered Asylum by Ecuador—and the British have threatened to violate the diplomatic immunity of the Ecuadorian embassy!
I certainly think that Assange was foolish to expose so much so quickly: the world's just not ready for it. But the treatment he has had suggests that many Western powers are just looking for an excuse to get him to the USA—exactly what he has been claiming. I'm disgusted.
| Saturday, 18 August 2012 | Dereel | Images for 18 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Weeding and pottering
|
Topic: gardening | Link here |
The weather was too bad for photos today, and the forecast for tomorrow was good, so I put things off until then. Time for gardening? Yes, a little. The Hibbertia scandens that I planted at the end of last month now has leaves. And then it occurred to me that Hibbertias are evergreen, and that the leaves didn't look anything like Hibbertias. They're almost certainly the birches that I had lost: I had forgotten a label, and thus mistaken them. Clearly they wouldn't be any good 30 cm from the house wall.
Fortunately found the real Hibbertias and planted them in place of the birches. They don't look very happy for 21 month old cuttings; hopefully they'll survive.
Also more weeding. Managed to remove the remainder of the sorrel from under the Euphorbia “Diamond Frost” (at least, all I could see) and also amazing amounts of grass round the Pandorea on the Eiffel Tower. I wonder if I'll manage to keep ahead of the weeds this year.
|
More scanner fun
|
Topic: technology | Link here |
More playing around with the scanner today, trying to get devd to recognize it correctly. Somehow I'm missing something here. In principle, this entry in /etc/devd.conf should do the trick:
But what I get is the output for the nomatch rule:
I haven't found any documentation for the regular expression for the device name, but I've tried various methods, and none seem to match, even if I omit the device-name specification. What's wrong here? There's also the question of why I need to set the permissions: I have 13 other ugen devices, and they all have 666 permissions. But this one gets 600. Is that because it wasn't recognized? I suppose I should take another look at the devfs rules.
|
X bug: another lead?
|
Topic: technology, opinion | Link here |
Now that USB seems to be working better on eureka, tried inserting the card reader that never used to work, with an SDHC card in it. It does get recognized now, but X froze. I thought it had taken down the machine, but it proved to be another case of the dreaded mouse freeze. I wonder if it has something to do with the USB bus (the mouse is, of course, also connected via USB).
| Sunday, 19 August 2012 | Dereel | Images for 19 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Ideal weather for house photos
|
Topic: photography | Link here |
As promised, this morning was ideal weather for the house photos. And, of course, once again I forgot some checks, taking a number of panoramas with a focal length of 10.25 mm instead of 8.8 mm (the real short end of the Zuiko Digital ED 9-18mm F4.0-5.6). Fortunately, most images had enough overlap for this not to be a problem.
What remains a problem is the dynamic range. DxO Optics “Pro” improves things a little, but not enough. The dynamics in the “garden centre” panorama differ by over 8 EV between the sky and the back of the verandah. DxO might manage 2 or 3 EVs, but not more. By exposing each image automatically, I get another 4 stops of dynamics in the following image, exposed between 1/50 at f/8 (11.6 EV) and 1/800 s at f/8 (15.6 EV), and the sensor itself has about 12 EV dynamic range, for a total of 18 EVs. But that's still not enough, as is clear from the mess in the trees to the right of the centre:
|
|||||||||||||||||||||||
Since there was almost no wind, I did an HDR series with -0.7, +1.3 and +3.3 EV. This time the recorded shutter speeds (at f/8) were between 1/6 s (EV 8.6) and 1/160 s (EV 14.0) for the longest exposure of each set. Given the 0.7 EV that I had offered to the heavens, that gives me a total additional dynamic range of 3.3 + 5.4 = 8.7. With DxO, the sensor and the underexposure for the sky, that's a total of 21.4 EVs. The results are much better. Running the mouse over the image shows the difference, though the skies jump out.
|
|||||||||||||||||||||||
So: DxO helps a little, but nothing can substitute for correct exposure of the entire scene.
That wasn't the only surprise of the day. Hugin crashed on me stitching the “verandah centre” panorama. No log file, nothing. I wish Hugin would write its log file somewhere where it doesn't go away if it crashes. But normally this suggests that one of the images is corrupt, so went looking for that and found one image was empty.
OK, not a problem. This was an HDR image as well, so I tried recreating it from its components. No go:
Something wrong with those images? Well, yes, but not what you'd expect:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's the camera's fault: in the middle of a 5 shot burst, it decides that one the images is oriented differently. Rotating it worked around the problem, but it's strange that it could have happened in the first place.
|
Dam full again
|
Topic: gardening | Link here |
Last week's rain has had its effect. Here's what the dam looked like last week, and then this week:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It's difficult to be exact, but it looks as if it's 40 cm deeper this week than last week. There's not much difference in appearance between now and the fullest it has ever been, two years ago. Here then and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
New blood pressure monitor
|
Topic: general | Link here |
One of the more convenient things about the new Community Postal Service in Napoleons is that it's in the General Store, which is open every day. So Yvonne was able to pick up a parcel for me, my new blood pressure monitor. After my last failure with an el-cheapo monitor, decided to pick a better-known manufacturer. My doctors use Omron, so I chose an “M2 Basic”, something that the web site doesn't really want to know about, though it has documentation. It's similar to the M2. The difference appears to be the lack of memory. It was barely more expensive than the cheapest models—only $25.70 including postage from an eBay seller located in St. Helier, Jersey—not a typical location. And it arrived within a week. By contrast, the cheapest I could find here are in the order of $80.
The difference on opening the box was immediate: instead of the usual sheet of folded A4 paper, there was a 272 page handbook, which proved to be 34 pages each in 8 different languages. It contains the inevitable safety instructions:
If battery fluid should get on your skin or clothing, immediately rinse with plenty of clean water.
Use only four "AAA" alkaline batteries with this unit. Do not use other types of batteries.
Sometimes I think there should be fines for writing inappropriate safety directions. But the second one is interesting: is it stupidity for “don't try to force wrong-sized batteries into this device”, or is it a circumlocution for “this machine is too lame to work on NiMH batteries”? Experiments with NiMH and NiZn batteries suggest the latter.
After inserting the supplied batteries, I compared with my old monitor, and confirmed what I had suspected: the old one shows systolic readings that are roughly correct, maybe a little on the low side, but the diastolic values are significantly too low, presumably because it lets down the pressure too quickly. It also confirmed another suspicion, that my blood pressure is too high. Back on medication, I fear.
|
More house inspections
|
Topic: general | Link here |
Off rather late this afternoon to Ballarat to look at some display homes. It's amazing how big houses have become—some of the display homes were so large that they didn't know what to do with the space, up to 350 m². And we confirmed my expectation that we could get quite a nice new brick veneer house for $180,000 odd. Add that to, say, $90,000 for a plot of land and $50,000 for incidentals, and we're still $10,000 less than the house in Rokewood Junction Road. The only issue is where we live while the house is being built. But we're gradually coming to the conclusion that we should put off the exercise by a year.
| Monday, 20 August 2012 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Power supplies: cheaper costs more?
|
Topic: technology, opinion | Link here |
The power supply fan on eureka.lemis.com (normal ATX) is getting noisy, and I'm considering replacing the power supply itself, which is now nearly 4 years old. Discussion on IRC: what to buy? eureka typically uses about 200 W, and I can get power supplies that deliver that (well, “460 W”) starting at $29.
I can also spend up to $300. Why should I ever want to do that? Reliability might be one issue, but even the cheap ones last several years. If the $29 one lasts for 3 years, the one for $300 would have to last 30 years, by which time it would presumably be completely obsolete.
But Peter Jeremy came up with a different viewpoint: efficiency. He quoted efficiencies of 78% for a cheap one and 90% for a more expensive one. Doesn't sound like much difference, but assuming that I draw 200 W using the cheap one, the system is really only getting 156 W. And I can get that out of a more expensive one for a power input of 173 W. Saving of 27 W. 650 W per day, 237 kWh per year. At the current prices of $0.25 per kWh, that's $60 per annum. Put that into the price of a power supply, and you break even after 12 months.
What's wrong with this picture? The efficiency ratings. Where do I find them? Nobody publishes them. And there are documents such as this one that suggest that power supplies are between 94% and 96% efficient. At that rate, the cheaper one wins.
|
Finishing the row of Hebes
|
Topic: gardening | Link here |
I started planting a row of Hebes between the garage and the greenhouse weeks ago, but other things got in the way. Today finally got round to removing a whole basketful of Hebes from the Japanese Garden, and planted 8 of them in the prepared bed. Also replaced 3 of the Hebes that I had planted in the east of the east garden last month: it seems that kangaroos had nibbled on them, in one case down to the ground. We have a whole lot more Hebes waiting, but hopefully I won't have to replace many more.
Removing the Hebes from the Japanese Garden also goes a long way towards removing the weeds. Maybe I'll get that done soon too.
| Tuesday, 21 August 2012 | Dereel → Ballarat → Dereel | Images for 21 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Identifying plants
|
Topic: gardening, general | Link here |
Into town today for a blood test, and then to the Botanical Gardens with a number of plants for identification. I thought that most of them were weeds, and Yvonne Curbach confirmed this, unfortunately without any further insight. Also had two others: the “Cathedral Tree” and this mystery plant that we bought on Australia Day:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's almost certainly a Mimulus aurantiacus. The image in Wikipedia is much darker in colour, but it's not typical of the photos I have seen:
|
Minor garden work
|
Topic: gardening | Link here |
The weather was milder today, but it didn't seem so: once again there were strong winds, and I didn't do very much. Finally spread the mulch over the Hebes, and cut a few flowers. Time for spring to finally come.
|
A harrowing experience
|
Topic: general, animals | Link here |
Yvonne has been waiting for some time to harrow the paddocks, which she does with the car (Holden Commodore) and an old gate. Today wasn't a particularly good day to do so, but she decided she had waited long enough, so off she went.
She was back fairly quickly: as I've noticed, it has rained a lot lately, and the paddocks were waterlogged. The Commodore is a conventional rear-wheel drive car, and she got bogged down. Not something that we're used to here. Went out with my Mitsubishi Magna and confirmed that the ground was very soggy, and I couldn't get enough traction to pull her out. We'll have to wait for Chris Yeardley to bring her Toyota Landcruiser.
|
More thoughts on power supplies
|
Topic: technology, opinion | Link here |
Peter Jeremy commented on yesterday's thoughts on power supplies. It seems that the 94%-96% efficiency applies only to the power factor correction. He pointed me at a certification site for power supplies with efficiencies over 80%. It's a bit difficult to navigate, and the legend for the tables is missing (or maybe just obscured by the markup errors): the last three columns only get displayed some of the time. But it appears to display the efficiencies at various loads, notably at 20% load. And the most efficient one is not quite 94% efficient—at 20% load.
There's a problem here, of course: all the el-cheapo power supplies are missing. Maybe none of them make 80%, but that's not clear. In addition, of course, there's the question of the load. The winner at 20% load is only 90% efficient at 10% load. So to get the best efficiency, I need to do quite a bit of research—probably more than I want to.
| Wednesday, 22 August 2012 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Yet more weeding
|
Topic: gardening | Link here |
The weather's gradually becoming milder, though the wind continues, and I had to force myself to do any work. Most of it was weeding, of which I need to do much more; but in some areas I get the feeling that I can see a difference.
|
“Open source” rants: not just me
|
Topic: technology, opinion | Link here |
In the past couple of days I've had my attention drawn to a couple of rants about free software that I almost could have written myself. One of them was from an acknowledged Microsoft-centric person, and a number of people I talked to dismissed the article because of it. But despite his perspective he has a number of valid points. In his case, he was complaining that a specific Apache module, mod_rewrite. And its behaviour has changed between releases 2.2 and 2.4. The documentation doesn't tell him so, and the release notes are very vague about how he should recover from the problem. All in all a good parallel with my pain with PHP and MySQL last month. I doubt his claims that commercial software is better in this regard, though. Both documentation and upgrade paths seem at least as rocky.
The other rant is much more interesting, because it comes from exactly my background: Poul-Henning Kamp (phk), one of the longest-standing FreeBSD developers, and one with whom I have frequently openly disagreed. In this case, though, it's clear that he has some good points. He takes the historical perspective, looking at free software from the perspective of the Dot com boom and The Cathedral and the Bazaar, somewhere half way between when Richard Stallman launched the GNU Project and now.
There are many ways to interpret phk's article, and follow-up comments show many of them. He describes the pain of updating a FreeBSD machine, as I have done vociferously over the last couple of months (and no, the upgrade to 64 bits that I started over a year ago and finally cut over last month is still not complete). He complains—rightly—that it takes too long (“a full day” for his spartan work environment—and spartan it is). But the rest appears to be a condemnation of the FreeBSD Ports Collection, blaming it on Eric Raymond (esr) and the sentiments he expressed in his book.
Yes, installing ports is a mess. He quotes firefox as being dependent on libtiff:
Here is one example of an ironic piece of waste: Sam Leffler's graphics/libtiff is one of the 122 packages on the road to www/firefox, yet the resulting Firefox browser does not render TIFF images.
But whose fault is that? I can think of two possibilities: either it's not really needed and it's a bug in the FreeBSD port, or some part of libtiff is really needed in firefox. I have libtiff on cvr2, a Linux box, so it's probably the latter. Should firefox have extracted this part and distributed it separately? One of the things that phk complains about is the lack of code reuse. That would go against it. So possibly the inclusion of libtiff is the Right Thing.
He goes on to describe the hideousness that is libtool, and finally:
That is the sorry reality of the bazaar Raymond praised in his book: a pile of old festering hacks, endlessly copied and pasted by a clueless generation of IT "professionals" who wouldn't recognize sound IT architecture if you hit them over the head with it.
What he forgets here is that he too uses these programs. Why doesn't he use one which is built better? I certainly would; I complain enough about firefox too. But the problem is: there is no useful alternative. It's not the bazaar or the cathedral; it's the bazaar or nothing. And that is the message that I think esr tried to get across all those years ago. We now have (phk's count) 22,198 ports, many of them quite unsavoury when seen from the inside. 20 years ago I created a CD of ported free software for Novell's USG—I managed to get 150 of them together, and at the time the FreeBSD Ports Collection wouldn't have been any bigger. The difference comes from the fact that people can write their own software and distribute it—very much the bazaar. I don't like the quality either, but what choice do you have? Even phk (the originator of “shut up and code!”) doesn't contemplate writing his own version.
In summary, then, I'm reminded of Jawaharlal Nehru's comment “Democracy is good. I say this because other systems are worse”. And on searching, I discover that I have quoted this only two weeks ago.
|
Installing conkeror
|
Topic: technology, opinion | Link here |
Jashank Jeremy asked me today if I had used conkeror, something I had never heard of. It proves to be an attempt to create a web browser with Emacs bindings. It's in the FreeBSD Ports Collection, so tried to install it. Not easy. First of all I need to find it. You'd expect it to be called www/conkeror, but no, for some reason it's called www/xpi-conkeror. Building it starts with the rather confusing message:
Doesn't that seem the wrong way round? Anyway, continued and saw:
That's ancient! What does it need that for? The general opinion was to extract libxul—almost exactly the issue that phk was complaining about in regard to libtiff in his rant. OK, download, and then? It started building libxul and got as far as:
I suppose that means something to somebody. Tried a few obvious workarounds without success. I didn't feel like fixing Yet Another Broken Port, so I tried downloading the binary package:
=== root@eureka (/dev/pts/14) /usr/ports/www/xpi-conkeror 71 -> pkg_add -r xpi-conkeror
=== root@eureka (/dev/pts/14) /usr/ports/www/xpi-conkeror 72 -> pkg_add -r conkeror
perl-5.14.2_2? I have perl installed already, of course. Do I really need to install a new one? No. I don't really need conkeror either. That's enough pain for one day. phk is right.
|
Bauhn 12MP Video camera
|
Topic: photography, technology, opinion | Link here |
Yvonne takes a number of video clips with her Canon IXY 200F. They're nothing special, and only 640×480, so I was interested when this week's ALDI specials included a “12 Mega Pixel High Definition Video Camera” for only $70. The great thing about ALDI is that I can return things if I don't like them, so I got Yvonne to pick one up.
ALDI's technology things are a bit of a mixed bag. Some are good, some are not so good. Some are quite bad. But nothing comes close to the uselessness of this device. The lens looks as if it has a cover which closes when not in use, but it doesn't. And it doesn't have a lens cap. There's no way to protect it. Inserting the battery and SD card was made much more difficult by lack of instructions about which way round to insert it, and the cover is really difficult to close.
But what about the image quality? Terrible! Here are some rough comparisons of still photo quality from the Bauhn, the Canon, and my Olympus E-30 in that sequence:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The highlights (trees in the background) are burnt out, the shadows are too dark, and there's extreme unsharpness and distortion in the corners. Here the top left corner of the same photos:
|
|
|||||||||
|
|
|||||||||
|
|
|||||||||
Clearly the Canon can't compete with the Olympus, but the Bauhn is an order of magnitude worse still.
But that's nothing! The thing doesn't have a flash, of course, but its light sensitivity is completely unacceptable. Here photos taken inside:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
There should be no difficulty getting correct exposure for a still shot: just leave the shutter open longer (the Olympus shot was 1 second, hand-held, and still recognizable), while the Canon jacked up the sensitivity to 800/30° ISO.
Still, this is supposed to be a video camera. The camera function is relatively unimportant: what are the videos like? Tried the same scenes in video, this time without the Olympus (which doesn't do video). The first problem was getting them in a format that I could display. I couldn't download the images via USB:
So I took out the SD card and read it directly. And then I couldn't display it with mplayer with either the vdpau or xv output drivers. With vdpau I get:
The corresponding video display is alternately acceptable and completely broken, maybe every alternate frame. xv gives the same swscaler message along with a broken display:
|
||||||||||||||||||||||
Finally got it to display with x11, in the process discovering that the image format is only 1280×720. Why? Clearly it has a 12 MP sensor.
Uploading to YouTube worked, but I got a strange message:
|
||||||||||||||||||||||
What is there to fix? This appears to be another YouTube markup problem. At one point I found a fleeting message that disappeared again:
We detected that your video may have bad lighting. Would you like us to fix that?
Yes, I would like. Unfortunately, YouTube also has difficulties with the images. Here (finally) is the outside:
The green banding that appears when playing the video appear to be YouTube's attempt to recover the broken image. They're even more obvious in the inside, where the image is really just about completely black:
There's more. The camera writes invalid EXIF data to its still images:
The shutter speed and aperture appear to be invariant, and of course there's no flash. I set the date when I got it, but it forgot it at some point. There is no flash, so that's incorrect. And the exposure is always given as the same.
All in all, then, an amazingly bad camera. Why did they ever try to put it on the market? But, in passing, it tends to confirm my comments on the quality of commercial software as compared to free software.
| Thursday, 23 August 2012 | Dereel | Images for 23 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Diganosing network problems
|
Topic: technology | Link here |
I had a number of disconnections of my HSPA network connection in the course of the day. I have a statistics page which is showing significant signs of neglect: the graphs no longer work, and the thought of debugging gnuplot scares me, so I've left it.
It's not as if it helps much anyway: the amount of status information I get from this Huawei dongle is minimal, and I don't really detect disconnects well. As I write this, I have had the following reconnects:
=== root@nerd-gw (/dev/pts/3) ~ 16 -> grep "IPCP: myaddr" /var/log/ppp.log
My scripts didn't detect any of these disconnects, though they did detect some other problems. Should I fix them? Hopefully the radiation tower will be here in a few months, and then I can forget this HSPA stuff altogether. So I'll just let it collect what stats it can until then.
Called up Internode Support and spoke to Michael, who went through his script: what OS are you running (and no, he didn't ask what FreeBSD is; I'm not sure if it was because he knows or didn't want to ask), whether I was using the software supplied with the stick, whether I had tried it in other computers, whether I had deinstalled and reinstalled it. At least I was able to convince him that deinstallation and reinstallation was just plain not possible.
The background of this effort seems to be that Optus, the operator of the wireless network, require it. And the next step is to swap the modem. It's quite possible that it is defective, though it's not the first place I'd look. The fact that every single reconnect was to a different IP address suggests some issue in the network. Normally the reconnect would be to the same part of the network, and then I would get the same IP address (or at the very least the same /24), but in every single case the reconnect was to a different /24. Still, that's their (slow and clumsy) procedure, and they have to go through it before they can even report a problem to Optus. So they're sending me a replacement dongle, which will probably not make it here until Monday. Another 5 days of rocky connectivity. Roll on the radiation tower.
| Friday, 24 August 2012 | Dereel | Images for 24 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Investigating the network disconnects
|
Topic: technology | Link here |
A couple more network disconnects today, becoming stable round midday. Probably things will be back to normal before the replacement modem from Internode arrives. But we discussed things on IRC, and with help I managed to find a little more information: the command AT+CREG=2 tells the modem to return status information when there is a change in registration:
When <n>=2, and the cell information changes, the following will be reported:
+CGREG: <stat>[,<lac>,<ci>]
...
<stat>:
0 Not registered. The MS is not searching the new operators to be registered.
1 Local network is registered
2 Not registered. But the MS is searching the new operators to be registered.
3 Registration rejected
4 Unknown reasons
5 Roaming network is registered
<lac>: Position code information, composed of four characters and expressed in
hexadecimal. (Example: "00C3"= "195" in decimal)
<ci>: Cell information, composed of four characters and expressed in hexadecimal.
Extended Information: according 3GPP Rel7, four characters are requested, but if
before Rel7, for example the currently network is Rel6 mostly, only the last two
characters is valid, the other characters is invalid and should be ignored. For example,
if the <CI> return 3B3DE1C, only DE1C is valid and could be used as DE1C is the last
two characters.
It's a little difficult to digest, but in practice I got messages like:
It looks as if we're hopping between two different cells. It's interesting how often it happens. As time went on, the hopping became less frequent and more asymmetric.
For some reason it never stays long with the tower ID 8FC8F2E. There are also occasional losses of registration:
This appears to be short enough that the ppp process doesn't notice. It continues running and reports no errors. My statistics also don't show anything obvious:
That middle line indicates complete loss of connectivity.
So: what's going on here? Under normal circumstances there should be no continual jumping from cell to cell. Is this related to the problem? Whom do I ask? I suspect that neither the Internode help desk nor their contact people at Optus have much idea: they're more a “how many bars does your software show” kind of person.
|
Attracting rosellas
|
Topic: animals, general | Link here |
The Crimson Rosellas have made quite a mess of my hanging baskets. The pelargoniums will probably survive, but I'm not so sure about the petunias. So we've bought some bird seed and put in one of the now-dead hanging baskets, beyond the reach of a very interested Piccola:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
Remote flash pains
|
Topic: photography, opinion | Link here |
In the good old days, electronic flash was simple: you plugged it in to the camera via a cable. Then people invented hot shoes, and you didn't even need the cable. So they got rid of the now-redundant sockets for the cable.
Problem: what if you don't want the flash unit in the hot shoe? Easy: wireless remote control, implemented differently by every manufacturer. My Olympus E-30 does in fact have a connector for a flash cable, but my Mecablitz 58 AF-1 O digital flash gun doesn't. Still, it understands the amazingly complex and poorly documented Olympus protocol.
Today I had the issue of taking the photos of the rosellas. I was using the Zuiko Digital ED 70-300mm F4.0-5.6, so I really needed flash (something that doesn't seem to worry birds at all). And I wanted it outside the house so that it wouldn't reflect in the window panes. A clear case for remote control. But what a pain it is to use! You need to set channels, groups and kinds of flash, and then mount the flash where it can see the camera flash unit. My old el-cheapo tripod came in handy there, and the first test shot worked well. But once the rosellas arrived, it didn't work. Instead, the camera flash fired at full intensity. Why?
It seems that there must be some response from the slave flash back to the master, and if the master doesn't get that response, it assumes (in this case correctly) that the slave didn't respond, and does what it can to make the correct exposure, which in this case wasn't enough. And that problem was caused by a slight repositioning of the slave flash. Next time I should be more careful.
| Saturday, 25 August 2012 | Dereel | Images for 25 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Network problems: more insights
|
Topic: technology | Link here |
My network connectivity is getting better again, but it's clearly not perfect. The last two reconnects were:
That's well over 24 hours, but more importantly, the addresses are located 1000 km apart:
=== grog@eureka (/dev/pts/10) ~ 138 -> host 118.209.122.167
=== grog@eureka (/dev/pts/10) ~ 139 -> host 121.44.12.163
It's not difficult to guess that mel is short for Melbourne, and syd is short for Sydney. There's nothing obvious about the local link to explain that: the cell hopping continues unabated, and the only real connectivity issue was a little earlier:
The ppp process reports:
And the statistics log reports:
The important column here is the 5th, which reports RSSI: it's 8 all the way, so there's no issue with the signal strength. At this point, irritatingly, the log file stops. I'm not sure what went wrong, but probably a bug in my code. The really interesting part, though, is in /var/log/messages:
I've removed irrelevant entries to make the size more manageable. What is very clear here is that the device has gone away. So Internode's guess is probably correct: this horrible flaky Huawei modem is disconnecting and reconnecting. The messages from the device should give more information, but ^STIN is not described in my documentation. ^SIMST is a USIM status change (to 1: “Valid USIM card state”). That could be normal on initialization; it's a pity there's no documentation for the ^STIN message.
The other thing that's interesting is that, apart from this information, everything was pointing to a problem in Optus' network. I'm convinced that this continual hopping and changes in IP addresses does indicate a problem, but it's not the one that's biting here. I wonder how long we need to establish that whether the modem is defective or just typically flaky.
|
Catching up in the garden
|
Topic: gardening | Link here |
Somehow this month has brought particularly unpleasant weather. It's not clear why. Looking back at the weather for the last 3 Augusts I find:
So two years ago was considerably cooler than this year. But, looking back, I complained then too. Maybe we just forget.
Today finally got round to repairing the broken pane in the greenhouse. I don't know whether it was a result of the broken pane and the wind, but some panes on the other side of the greenhouse had loosened and slid down next to another, leaving an even larger gap at the top. Fortunately they weren't damaged. But if we buy a new house, we won't take the greenhouse with us; it's just too much pain.
When do you prune roses that bloom all year round? The “Gruß an Aachen” and “Monsieur Tillier” still had buds on them, but it's clearly time to do something for the spring, so cut off some very ordinary looking buds and pruned the “Monsieur Tillier”. There are too many snowflakes growing round the “Gruß an Aachen” for me to prune it. Some other time.
Also planted the next round of gladioli, somewhat overdue, and removed two roses from the north bed, including the one that I had planted there less than a year ago: it has pink flowers, and Yvonne doesn't like that, so Chris Yeardley will get them.
| Sunday, 26 August 2012 | Dereel | Images for 26 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Back to the veggie patch
|
Topic: gardening | Link here |
I've severely neglected my veggie patch over the last several months, but it's time to plant potatoes. A few weeks ago I sprayed the weeds, and today I finally got round to tidying up the remains. A surprising number of potatoes in one corner, all apparently Kipfler: I dug up a bucketful of them from about 1.5 m². Some are edible, and others will serve as seed potatoes. Now to wait for Yvonne to decide which she wants.
|
Pruning disaster
|
Topic: gardening | Link here |
Also spent some time catching up with pruning the ornamental grapevine on the verandah—with disastrous results. Found one stem going in to the jumble of Lonicera and Jasmine, where clearly it couldn't do very much. So I cut it off:
|
|||||||||||||||||||||||
Then I discovered that it came out a few centimetres later and formed the back part of the framework, visible here from the middle beam and back:
|
|||||||||||||||||||||||
So now I only have half the coverage I had this morning. Hopefully it'll grow back quickly.
|
All the wrong citrus
|
Topic: gardening | Link here |
We now have five citrus trees, none of them really bearing satisfactory fruit. The grapefruits are too dry, though I still have hope that they will ripen further, the Tahitian lime has fruit that are too hard, and the Meyer “Lotsa lemons” has had exactly one fruit in the 3 years we have had it. But the Kaffir lime tree bears lots and lots of fruit:
|
|||||||||||||||||||||||
What a pity you can't do anything with them.
|
Pollo en mole verde
|
Topic: food and drink, opinion | Link here |
My chiles poblanos didn't exactly bear lots of fruit, but I did get a number:
|
|||||||||||||||||||||||
What to do with them? Went looking and found a recipe for pollo en mole verde, a Mexican dish from Puebla (like the chiles), which specifically used them. The recipe looks complicated: the various parts of the sauce need to be chopped separately, and it seems you have to skin the poblanos before cooking them. This seems just to be intended to make the resultant sauce smoother, not to add flavour. It doesn't seem necessary with modern kitchen machines.
And the result? Quite nice, but not more. The recipes I've seen suggest that it's something special, but I don't find it so. Is this one of these cases where the food is made with common ingredients like green tomatoes, chile poblano and pepitas, much like cassoulet is also made with simple ingredients like smoked sausages and confit de canard, all of which are really difficult to find outside the area where the dish comes from?
| Monday, 27 August 2012 | Dereel | Images for 27 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Interpreting the network logs
|
Topic: technology, opinion | Link here |
My network disconnect problems seem to have got better over the last few days. Since Friday I have only had a single disconnect:
At the same time, the number of cell hops diminished greatly. In 18 hours I had only the following reports:
I still don't understand the link-level disconnects, but they're practically invisible. More interesting is that between Aug 26 15:19:18 and Aug 27 09:13:55 cell ID 8FC8F2E didn't show up at all. Is there a connection?
It didn't stay that way. In the course of the day the number of hops increased, and shortly before midnight there was another disconnect. In this case, though, the device didn't disappear. Instead, I find:
Unless I'm very much mistaken, this is a disconnect explicitly requested by the other end.
So it seems we have at least reasons for disconnects: hardware problems with the dongle, where it goes away and comes back, and explicit disconnect requests from the other end. There's also the question of cell hopping and occasional very short link-level disconnects, which may or may not be related.
|
Transplanting the Tahitian Lime
|
Topic: gardening | Link here |
Four years ago we bought a couple of citrus trees and planted them to the north of the house:
|
|||||||||||||||||||||||
The one in the front is a Tahitian Lime, the one to the right is a grapefruit, and behind the trees, a row of Hebes is barely visible.
How times have changed! We had had Hebes before in Wantadilla, and they grew to about 60 cm. These have grown to more than double that size, engulfing the poor citrus trees in the process. Now things look more like this:
|
|||||||||||||||||||||||
Clearly the citrus trees need to be moved elsewhere, and today I tackled the lime tree. The photo above shows the hole where it came from, and the grapefruit is barely visible to the left of the gate.
I have spent some time agonizing over where best to replant it, and finally decided to put it in the north-eastern part of the garden, just north of the row of Buddlejas. That's not far from where the other (Kaffir) lime tree is, and it is doing well. But I had been concerned about the area because other plants had not thrived there, but on digging the hole I understood why: there was a thin layer of packed gravel about 15 cm from the surface. It was relatively easy to remove, and hopefully the tree will be happy—if it survives the root damage. I'll keep an eye on it, and if it does well, there's space for the grapefruit next to it:
|
|||||||||||||||||||||||
The leaves are turned to the left because the trunk ended up growing away from the Hebes. Presumably that will soon right itself.
| Tuesday, 28 August 2012 | Dereel → Sebastopol → Dereel | Images for 28 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Into town again
|
Topic: general | Link here |
To Sebastopol today to get the results of my blood tests (good; all in normal range, with the exception of some marginal haemoglobin values that suggest I should be taking folic acid), and then to have my hair cut.
|
Gardening: too windy
|
Topic: gardening | Link here |
Finally it's getting warmer and drier—so of course it had to be windy too. Did some more weeding in the south-west part of the south-east bed. It's full of underground runners of what I think must be couch grass, and I'm not very convinced that I will get rid of it easily.
|
Transplanting tomatoes
|
Topic: gardening | Link here |
The tomato seeds I planted two months ago are finally big enough for planting into bigger pots. But that means taking them out of the warm box in the bathroom and putting them in the greenhouse, where at least it's light, but where the temperatures can drop well below 10°. Which is preferable? Hedged my bets and planted some of them. We'll see which do better.
| Wednesday, 29 August 2012 | Dereel | Images for 29 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More tree mutilation
|
Topic: gardening, general | Link here |
More tree cutters from (presumably) the Vemco Group are in the area, and of course they have entered my property without permission. Not much harm done, though—until this morning, when they came to “trim” the Schinus molle (“pepper tree”). Not much I could do about it, since it was too close to the power line. They did leave some of the tree behind, though the pile on the right shows how much they cut off:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
About the only positive thing is that this view is from the swamp, not from the house. And compared to four years ago, when I took the last of my photos from the swamp, it's not the only drastic change:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I suppose the tree will recover relatively quickly, but they've made quite a mess of things, and I'll have to trim off some more.
|
Two months of marriage
|
Topic: technology, opinion | Link here |
Looked in on facebook again today, for the first time in a long while. The things that happen while you're away:
|
||||||||||||||||||||||
Not only that, but a number of people who should know better “like” the marriage (in fact, I've been married for nearly 30 years). Showed it to Yvonne, who explained that she had been updating her profile and entered that she was married (presumably to me)—and the software took that entry as the day of her wedding. How much is wrong about this, particularly the fact that I have been implicated by somebody else's entries! And no, we're not even “friends” on facebook, and her name isn't in my (minimal) profile. And I can't find a way to enter date of marriage either.
|
Network disconnects: more dropouts
|
Topic: technology | Link here |
Part of chasing last week's network problems was for Internode support to send me a replacement modem, by express (usually overnight) post. It arrived today, six days later—a ZTE MF652:
That's not what they said they would send: they were going to send a Huawei E1762, like I have now. But it was probably worth trying anyway—until I discovered it had a different antenna connector. So there was nothing I could do with it. Set to to write a page summarizing the disconnects I had had so far, and came to some interesting conclusions. The disconnects last Thursday were really all accompanied by the device going away and coming back:
The device goes away and comes back again 5 seconds later. So a device error? Probably a firmware reset or some such. Why that should happen is beyond me, except that it seems to be in keeping with this device's reputation for reliability. But it's not the only scenario. Of the 15 disconnects I documented, two were physical disconnects (removing the modem from the slot), and there were further 8 disconnect messages. Three were like this:
There were no terminal messages involved here, and it looks fairly clear that in these cases the other end caused the disconnect. It's also interesting that they were all in the middle of the night, an hour either side of midnight.
And then there were two of these:
That looks like something else in the system had shot down the ppp process. What? Did I do it? I don't think so. I'll have to keep an eye on it. Wouldn't it be nice to be able to log this kind of signal?
All in all, though, it looks as if I can only blame Optus for three of the 15 disconnects, and all happened after I reported the problem. I suppose I should just grin and bear it, and hope that the radiation tower will be built soon.
|
Cleaning out potato beds
|
Topic: gardening, opinion | Link here |
Continued removing weeds in the veggie patch today—there's sorrel and some creeping grass, both of which leave long runners. The sorrel is particularly tenacious, with runners growing down to about 15 cm. Cleaned up a few m² and got out another bucket full of potatoes. How can you rotate potato beds? It's almost impossible to remove them all. We have some in the small succulent garden to the north of the verandah. They've been there for over 5 years, and despite attempts to remove them, they still come back.
| Thursday, 30 August 2012 | Dereel | Images for 30 August 2012 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Inactive day
|
Topic: general | Link here |
Some days I do so much that I spend half the following day describing them in this diary. Today was not such a day. The weather didn't help: yesterday seemed almost like spring, with a top temperature of nearly 21°. Today the maximum was 12°, it was windy, and in general miserable, so I didn't even get to do any work in the garden.
| Friday, 31 August 2012 | Dereel | Images for 31 August 2012 |
| Top of page | ||
| previous day |
|
Finally: the mixer tap
|
Topic: general | Link here |
I've been dragging my heels about installing the new mixer tap that I got two weeks ago: I was really concerned about my ability to widen the hole sufficiently. Finally, yesterday, I called Bryan Jackson to help, and he showed up at 9:00 this morning. It was clearly a good decision: he brought a pair of metal snippers with him, with which he was able to do quite a good job of creating a larger hole. Also discovered that the new hose connections weren't necessary.
|
|||||||||||||||||||||||
I was somewhat comforted by the fact that Bryan also had problems with the screws. He had bought a long box spanner to attach the nut on the threads, but the new one had a smaller nut (9 mm instead of 10, I think). Still, it's done, and the total cost was less than I spent on the bits and pieces to do it myself. Time to accept that I don't have to do everything myself.
|
Garden at end of winter
|
Topic: gardening | Link here |
Since this time last year I have moved my monthly flower photos from the end of the month to the middle of the month. This month the difference was noticeable, so today, the last day of winter, I went around and took another look. It's certainly different from last year, but it's not clear why. The roses are overlapping. Here a „Gruß an Aachen“ bud from last year's growth and a “Lilli Marleen” from this year:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Apart from that, there's not too much to note. The second shoot on the Cymbidium is starting to flower:
|
|||||||||||||||||||||||
That's a big difference from last year, where it didn't flower until late October. The other issue that is still with is is that animals mutilate the plants. I suspect that this is due to a kangaroo:
|
|||||||||||||||||||||||
And the rosellas are still eating seed on the verandah, but they have also caused severe damage to the pelargoniums in the hanging pots. I'm sure they'll recover, and it's clear that with the increase in shade from the vine, they don't belong there anyway.
The daffodils have been blooming for over two months, and they're gradually finishing. Or that's what I thought: now a new variety is coming into flower:
|
|||||||||||||||||||||||
|
Fixing the rain gauge
|
Topic: gardening | Link here |
The rain gauge in the weather station has been non-functional for months: the mechanism seems to have jammed up. Today I finally replaced it with a spare. It shouldn't have been much work, but I chose a bad day: cool and windy. I was glad when it was over.
|
New business cards
|
Topic: general, technology, opinion | Link here |
Since retiring, it's clear that I don't have “business” cards, but from time to time I want to give people something similar, more like what people used to call “visiting cards”. And from time to time I get a web advertisement with this offer from Vistaprint: 250 business cards, free.
They're something like a free lunch. You only pay the postage. Letter postage for up to 500 g should run to about $3 and take between 1 to 4 business days to reach its destination. But this “postage” costs $7.85 and takes 21 days. You can have it faster, of course, but then it costs (much) more. So: TANSTAAFL.
Still, the price isn't bad, I can wait 3 weeks, and they had a marginally usable online configuration tool, so I chose a pattern and tried to sign up. “Invalid mail address”. OK, it was a one-off that I had created specially for them, but that shouldn't matter. No idea what the problem was, so eventually I called up their support line, which also didn't want to believe that the email address was mine. But they finally created one for me, and then I had difficulty setting a password. My guess is they're making assumptions about both the format of an email address and that of the password. The latter appears not to allow special characters; maybe the former wants digits, like most people seem to have in their email address nowadays.
Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
