|
| Greg's diary |
| May 2019 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Wednesday, 1 May 2019 | Dereel | Images for 1 May 2019 |
| Top of page | ||
| next day | ||
| last day |
|
Power failure!
|
Topic: Stones Road house, general, technology, opinion | Link here |
While getting up this morning, heard a “beep” from outside. What was that? Yvonne resetting the RCD for the “non-UPS” circuits at the back of the house. It had tripped again, like it had three months ago.
Why? That's what I asked last time.
Then in to the office to discover eureka, my main computer, dead in the water. Why? That didn't happen last time. Coincidence? A quick look at dischord, the Microsoft box, showed that it was still in sleep mode. All pointed towards eureka shutting down spontaneously.
Recovering eureka was a story in itself, but later I discovered that dischord had been disconnected too—it just recovered automatically, something that I should set in the eureka configuration. The explanation, at least partially: previously I had had my office computers connected via a local UPS to the larger UPS in the shed, but that tended to fluctuate a lot, causing the office UPS to beep annoyingly. So after installing the solar power I had connected to the “non-UPS” power point in the office. And though it's not on the same circuit as the bedrooms, both circuit breakers are served by the same RCD. So the failure was understandable, though some of the details were not.
So what caused the failure? An outside pump sounded a likely culprit, especially as both times it happened in the middle of the night, when we spray the garden. Outside to check: no, the bore pump still had pressure, so it hadn't experienced a power cut while operating. House water pump? No, that had worked as well during the outage. Checked everything that I could think of, but to no avail.
|
Recovering eureka
|
Topic: technology, opinion | Link here |
Rebooting eureka was an experience in itself. It failed to boot with an error I had never seen before, something about an inability to load some lua menu. I didn't think to take a photo: I was supposed to go be going into town for a blood test.
But all the parameters looked right. What happens with just boot? It booted! But at a snail's pace. I couldn't even enter more than about one character per second on the keyboard. And I got lots of messages that I have never seen before, roughly at 10 second intervals:
uhub_reattach_port: giving up port reset - device vanished
The good news was that the journals recovered the big file systems almost instantaneously. But the thing was still so slow, and in the end I shut it down to see if a power-off would help. That, too, took for ever, and it didn't help.
OK, try again and see if
the BIOS UEFI setup could help. No, that all looked
normal, but the system and keyboard response was just as slow as when running FreeBSD, so it wasn't related to FreeBSD. In the end,
dammit, reset UEFI to defaults. Rebooted. And it worked.
What went wrong there? A power failure shouldn't cause that kind of problem. Is the motherboard on the way out? Time to think of a new machine: after all, this one is over 5 years old.
|
Solar power monitoring strangeness
|
Topic: technology, Stones Road house | Link here |
One of the problems that I had from the power outage was the loss of 2½ hours worth of data from the solar power inverter. Not much I can do about that, though it's interesting to note when the database updates finished: 6:40. That's when the office UPS died, of course. Was there anything of interest in the data in the hour before? Not that I could see. It would be really nice if the inverter would log things like mains voltage, which I suspect might play a part.
And how did the ingeconsunmonitor.com site deal with the lack of data? Superbly!
It has complete usage data for the time of the outage, including the spike at 7:00 when I turned on the air conditioner, something that couldn't have been guessed. So somewhere the inverter is storing this information and supplying it when the network connection is restored.
But where? Started up an Wireshark to see what it was doing, but apart from discovering that it sends quite a bit of data, there wasn't much to be found.
So where is it stored? I know the pages http://inverter.lemis.com/ems/last7Days and http://inverter.lemis.com/ems/summary, but of course it's far too modern to give me anything useful at http://inverter.lemis.com/ems/. But clearly there's more to be found for him who searches.
|
The weather breaks
|
Topic: general | Link here |
The first 4 months of this year have been some of the driest on record, but with the new month that changed. We had a total of over 35 mm of rain today, including a quite violent thunderstorm with hail this evening. We can do with much more.
|
More power insights
|
Topic: Stones Road house, general, opinion | Link here |
We were watching TV when the thunderstorm passed through, and we had to raise the volume to hear anything over the noise of the storm. And then the TV turned off. Just the TV, not teevee (the computer that drives it) or the lights. Another RCD trip! Back into the garage to turn it on again, and got a few metres from the switchboard before it failed again. And later on it failed a third time.
One thing that's clear: it's nothing in my office. The UPS was now on the external UPS circuit again, and I've turned off the printer, which only gets used once a month and thus shouldn't be on anyway. On the other hand, it was raining this morning, too, when the power failed. So somehow it confirms my suspicion that it's related to moisture outside the house. But we only have three external power points. Two aren't connected, and the pump on the third is not turned on.
How are these external power points constructed? You'd hope a DPST switch. Maybe we should consider connecting them to a separate RCD.
| Thursday, 2 May 2019 | Dereel → Ballarat → Dereel | Images for 2 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Blood test again
|
Topic: health | Link here |
Into Ballarat this morning for another blood test. For once, I wasn't attended to immediately: I had to wait 10 minutes. I had thought that this might be due to the relatively late hour (10:30), but it seems that they were short-staffed. And after all, 10 minutes is typical to good at other places. And somehow I couldn't think of anything else to do, so back home again.
|
Android navigation revisited
|
Topic: technology, general, opinion | Link here |
My GPS navigator has expired: it thinks that the date is some time in 1999, and there seems to be no way to adjust it. OK, it's over 5 years old, so it's probably time to replace it anyway.
But are standalone GPS navigators still the way to go? Nearly every mobile telephone, even my el-cheapo Nokia 3, has not only a GPS receiver, but also accelerometers and a compass. Why not use that instead?
An obvious reason would be my previous experience, which has brought nothing but frustration. Still, practice makes perfect. Finding my way to the Dorevitch collection centre in Victoria Street is not easy, since it's in a side road only accessible from the Bakery Hill roundabout. A good chance to try out Google Maps again. OK, Google, take me to 53 Victoria Street.
No reaction. When does this voice recognition work, when doesn't it? It took me 4 stops and a lot of messing around, and finally it found a route. One of the issues is that I normally have mobile data turned off, and it didn't like that. Did it say so? No, these things have voice recognition and only very limited voice output. When I stopped to look, I discovered a confusing message on the screen that I finally construed to mean “please turn on mobile data”. And of course, once it found the route, it said... nothing.
And the map was strange. It's convenient to put the phone horizontally in front of the speedometer, but then it only gave me half a display. Here the same place horizontally and vertically:
|
|||||||||||||||||||||||
|
|
|||||||||
Why is the left half of the map missing? Looks like a software bug to me. I considered stopping and restarting the app, but who knows how much work it would then be to get back to where I was?
While waiting for the blood test, did some more playing around, and finally found something like a settings menu, embedded far too deeply in the strange and non-uniform menu structure. Not much to set, but at least I got it to set North to up (since I can't find any way to get it to display the direction in which I'm going).
On the way home, tried a few more things, noting that it had a very different view of the route from my GPS navigator. Apart from choosing inappropriate routes, it had a strange view of “straight ahead”, like this display, taken while I tried in vain to find the settings menu again:
|
|
|||||||||
Why turn right? The road goes straight ahead! Ah, but I'm on the side of the road, and I need to get into the traffic again:
|
|||||||||||||||||||||||
So what's the real problem here?
I don't understand how the app is structured. The previous images show the problem: what choices do I have? The four icons at top right seem to have some meaning. The tulip is a microphone. But what do I do with it? It seems that I need to touch it to enable it. If that's correct, it somehow detracts from the utility. And then there's a magnifying glass. What does it do? I forget, and I can't find out again because I can't reproduce the map display. I would have expected it to magnify the display, but I think it does something unexpected like search for places around the map area. This seems at least to be a problem with the lack of transparency of the app structure. But though I had plenty to grumble about with the standalone GPS navigators, the menu structure was easy enough to understand.
This seems to be a general issue with Android. Is it just that I don't understand? My feeling is that it is the result of applying too many ad-hoc patches to address issues that inexperienced users have. The result is that frequently used functions are (marginally) easier to use, and the more complicated ones become significantly more difficult to access. After all, how difficult would it be to have a single, clearly structured menu and a clearly visible icon on the screen at all times?
As noted, the magnifying glass doesn't magnify the display. For that you need a two-finger gesture (we used them when I was a boy to express disapproval). And that invariably rotates the display, even if set to “North is up”. Grrr!
There are more icons on the right. The third seems a model of clarity: turn loudspeaker on or off. And the fourth? From recollection, report problems along the way. Now isn't that much more important than access to a main menu?
And part of the issue seems to be the glacial reaction times of the system. That's possibly related to the low-end phone—after all, it has only a quad core 1.3 GHz processor. How does that compare to the Control Data 7600, my benchmark for 1970s supercomputers? My guess would be about 10 to 50 times the speed, but it seems to be becoming more and more difficult to compare CPUs. One thing is almost certain: the mobile phone CPU is probably an order of magnitude faster than the CPU in the GPS navigator, which hasn't caused this kind of problem.
|
How much power am I using?
|
Topic: Stones Road house, technology, opinion | Link here |
I'm still puzzled about the RCD trip that I had yesterday, so spent some time investigating. The first thing that occurred to me was the display from the inverter:
|
|||||||||||||||||||||||
|
|
|||||||||
OK, I know about a number of bugs in this display, including the fact that the day are the wrong way round in the second image. The first entry, marked “Fr” is today. It seems that this always has tomorrow's day of the week. And the following ones, in correct weekday order, are the previous days. But look at the entry for “Th”, which must be Friday, 26 April 2019. Nearly 70 kWh! And “We” (Saturday, 27 April 2019) isn't much less. What does the external web site say?
|
|
|||||||||
A couple of things: firstly, no, nowhere near 70 kWh. And Saturday's usage was more than Friday's. Time to look at the detailed reports, which show:
Consumption from 4/24/19 to 4/30/19 [sic]
| 24/04 | 25/04 | 26/04 | 27/04 | 28/04 | 29/04 | 30/04 | ||||||||
| Self-consumption ratio | 57.47 % | 48.66 % | 43.28 % | 27.48 % | 38.15 % | 44.39 % | 48.40 % | |||||||
| From PV | 10.77 | 10.72 | 12.54 | 5.97 | 7.52 | 9.31 | 8.38 | |||||||
| From storage | 9.47 | 12.26 | 13.23 | 11.23 | 11.04 | 11.60 | 9.63 | |||||||
| From public grid | 11.07 | 18.98 | 27.59 | 37.59 | 23.60 | 21.01 | 15.09 | |||||||
| Total | 35.20 | 47.21 | 59.53 | 62.57 | 48.63 | 47.10 | 37.21 |
Generation from 4/24/19 to 4/30/19 [sic]
| 24/04 | 25/04 | 26/04 | 27/04 | 28/04 | 29/04 | 30/04 | ||||||||
| Consumption | 10.77 | 10.72 | 12.54 | 5.97 | 7.52 | 9.31 | 8.38 | |||||||
| Battery Charge | 6.13 | 7.81 | 7.62 | 4.19 | 5.11 | 6.50 | 6.30 | |||||||
| Grid feed-in | 7.15 | 3.66 | 0.02 | 0.01 | 0.03 | 8.32 | 7.37 | |||||||
| Total | 24.05 | 22.18 | 20.18 | 10.17 | 12.66 | 24.13 | 22.05 | |||||||
This is full of interesting information:
These numbers bear little relation to the graphs. For example, on 27 April the first graph shows roughly 54 kWh consumption, compared to 62.57 kWh in the report on the same page, and on 26 April the graph shows about 53 kWh, compared to reported 59.53 kWh. The generation statistics are correspondingly vague:
| Date | 26 April | 27 April | ||
| Use, from inverter | 67 kWh | 69 kWh | ||
| Use, from report graph | 53 kWh | 54 kWh | ||
| Use, from report | 59.53 kWh | 62.57 kWh | ||
| Generation, from inverter graph | 35 kWh | 22 kWh | ||
| Generation, from report graph | 20 kWh | 10 kWh | ||
| Generation, from report | 20.18 kWh | 10.17 kWh |
What does all this mean? Clearly the usage graphs are Just Plain Broken. The generation graphs are less so, I think. The values in the report are as good as the same. And the inverter output seems to lump battery charge of any kind together, so if I take just the “Photov.” column, it lines up quite well. But clearly it's difficult to trust the usage values; I'll have to brave Powercor and compare what they report with what Ingecon reports.
The report format (in PDF) is also particularly emetic: normally you can cut and paste it into an editor, but here the entries have been added out of sequence, requiring considerable corrections.
In passing, it's interesting to note the comment at the bottom of the report:
(*) System could not connect to inverter. It will try later.
That doesn't apply to this report, and they're too polite to draw attention to that fact, but it's a confirmation of what I experienced yesterday.
As if that wasn't enough, there was no report information for today. Why? My guess is software problems at their end. As it happened, I was tracing communications, so it was clear that there was (surprisingly much) traffic.
|
What power in the middle of the night?
|
Topic: Stones Road house, opinion | Link here |
The other issue with power consumption was: how much power do I use in the night? The background was whether the RCD trip had something to do with power consumption. And how about that, there were some surprisingly high peaks. That's visible from the daily summary:
|
|
|||||||||
|
|||||||||||||||||||||||
The highest one was when I turned the air conditioner on round 7:00. But by then eureka was down. According to the graph, with its bizarre X axis markings, I must have used up to 3.8 kW round 4:00 and maybe 3.5 kW round 6:00. Why?
It's difficult to be sure from memory. Did I turn the air conditioner on at 7:00, or just turn the living room on, while the bedrooms had been heated through the night? I'll have to keep an eye on that for next time. What I did see was:
So there were three occasions on which I used more than 4 kW, corresponding roughly to the three spikes. But those three spikes on the graph show considerably lower consumption than the inverter. Smoothing? And what are those negative W1 values? W1 is the “feedback to grid” value, separate from PacGrid, the feedback to the grid (which was 0 in all these records, and thus not listed). In particular, though, the values don't add up. In principle we have consumption (Pac) and three sources: PacPV (from the PV array), PacBat (from battery) and PacGrid (from the grid). They should add up, and at 05:24:34 they most certainly did not.
Could this have been a power surge? It would have potentially explained the RCD trip. Wouldn't it be nice to see at least the inverter's view of grid voltage?
|
Autofocus fun
|
Topic: photography, animals, opinion | Link here |
More kangaroos in the garden today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And one that showed the autofocus of the Olympus OM-D E-M1 Mark II to perfection:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
But why? I decided a long time ago to set the autofocus to the middle of the field of view, and clearly that's where the kangaroo was. Panning, maybe? Still, it was the only autofocus failure.
| Friday, 3 May 2019 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Ingecon on line again
|
Topic: Stones Road house, technology, opinion | Link here |
So why didn't I have any output for my solar electricity system yesterday? I had guessed a problem at their end, and it seems that I was right. Today it was there, along with the data for today as well.
And how do I access that data directly? Spent some time sniffing the traffic, but nothing interesting went past. What about the source for the local display? An ngrep showed:
What's useful there? http://inverter.lemis.com/dashboard/ems_dashboard.js looks good, so I pulled that down:
One line, 28247 bytes (roughly the size of a Unix Sixth Edition kernel). And of course optimized for transmission, not human intelligibility. How can I re-indent that and make it readable? A quick web search brought me to https://beautifier.io/, which helped, once I worked my way round its insistence on using the clipboard for everything, giving me something marginally readable:
Of course, there are no comments (for once they can even claim that this is a feature, not a bug), and there's a lot to look through. What I was really looking for was other URLs, and unfortunately that didn't work.
|
Accessing my money
|
Topic: general, opinion | Link here |
So I have $17,000 to pay to Effective Electrical for the solar electricity installation. OK, call up CVI and find out how to transfer that.
Connected with Sarah, told her my name, and she told me my balance. Who needs an account number? I'm still a little surprised. But sorry, they don't do transfers to third parties any more—presumably some legal issue. So she had to transfer it to my own designated account.
OK, we can do that. Send an (unencrypted) email and it was done in 5 minutes.
What are the security implications? No account numbers went over the wire, and clearly it still belongs to me. Yes, the email was signed, but the Real World doesn't understand signed emails.
|
Everybody knows me
|
Topic: health, general, opinion | Link here |
Time for a doctor's appointment. Called up the clinic, gave them my name, and was given an appointment at 14:45 on Thursday. And again the receptionist (Rebecca?) recognized me and just mentioned my address for confirmation. How does everybody know who I am?
| Saturday, 4 May 2019 | Dereel | Images for 4 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Ploughing in autumn
|
Topic: general, animals, opinion | Link here |
Walking the dogs today, saw an interesting sight:
|
||||||||||||||||||||||||
Where's the plough? There isn't one. That's Fiona Drayton training a horse for harness. It wasn't always happy about the situation:
|
||||||||||||||||||||||||
|
Mushroom season?
|
Topic: gardening, opinion | Link here |
Also while walking the dogs, saw this on the side of the road:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Yvonne thought that it might be a Cèpe or Steinpilz. It looks like the same kind of mushroom that we found in Kuitpo Forest thirteen years ago, and which we ultimately identified as Suillus luteus (also Smörsopp, Butterpilz or масловка). But clearly that's a different mushroom:
|
|||||||||||||||||||||||
It took me more time to discover that I had seen the same kind of mushroom in the house forest 2½ years ago.
|
Firefox annoyances
|
Topic: technology, opinion | Link here |
One of my best reasons for not migrating to newer software is that they've done everything they can to make data entry more painful:
|
|||||||||||||||||||||||
Why did firefox come up with that message now? It has been running for days. And what was disabled? Nothing followed. Too polite to say? The links were, of course, generic, telling me what a great service they have done me by removing functionality upon which I depend. Dammit!
|
High resolution: too high for lenses?
|
Topic: photography, opinion | Link here |
I've wondered in the past how much use the 80 MP high resolution mode on my Olympus OM-D E-M1 Mark II is really worth. Can the lenses handle it? This article says no, in some detail, but a little too categorically for my liking:
Inexpensive lenses performed as well as expensive lenses with good reputations!
He followed it up with this graph:
OK, that matches the statement. But it doesn't match what DxOMark says. Its methodology is a little strange, but it comes up with resolutions of 7 “P Mpix”, which apparently means “perceptual MPix” (nothing?). But it gives the 75 mm lens double the P Mpix of the 14-42, which isn't surprising. So I'm left wondering if there's something in his test setup that limited things.
He did his testing on an Olympus OM-D E-M1X, which has effectively the same sensor as my E-M1 Mark II, 5240 x 3912 pixels. It's not immediately clear from the graph, but his scale is equivalent to vertical pixels, so he's saying that the best lens in his test (the M.ZUIKO DIGITAL ED 40-150 mm f/2.8 PRO) has a vertical resolution of 2000 pixels with a sensor capable of double that, 3912 pixels. That corresponds to an absolute resolution of about 5.3 MP. So why should the increase in sensor resolution to 10368 × 7776 make any difference? And what he measures there is a maximum of 3200 pixels, or about 13.6 MP.
He also states:
It becomes more intuitive when one recognizes that FF camera sensors top out at about 50 MP and the Olympus High Res Mode generates a FF image equivalent of 200 MP – that lens optics and performance might become the limiting factor to the image’s final resolution.
“Intuitive”. This reminds me of the saying:
For every complex problem there is a solution that is simple, elegant, intuitive... and false.
Sorry, the lenses are designed for that sensor size. They probably have a higher absolute resolution than any “full frame” lens. I don't seen any relevance here. If you follow that logic to a logical conclusion, a mobile phone lens for a 12 MP mobile phone with a “crop factor” relative to “full frame” of 7.6 (typical, according to this page) would correspond to a “full frame” camera resolution of 690 MP. Clearly that's absurd.
In principle I agree with the premise, that high resolution is limited by the lens. But there are too many dubious details in this article.
| Sunday, 5 May 2019 | Dereel | Images for 5 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Now that's a bone!
|
Topic: animals, general, opinion | Link here |
Took the dogs walking down Grassy Gully Road again today, a place they haven't investigated much. Nikolai went sniffing and found the biggest pair of bones that I have ever seen:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Niko took us back home in a hurry, where he kept it all to himself:
|
||||||||||||||||||||||||
Leonid was in his typical “OK, I'm sitting, now feed me” position. I wonder whether he expected Niko to honour it.
And what kind of animal was it? The most likely candidates are kangaroo and sheep. I don't think that sheep have any bones that long, and the fact that Niko found it near a fence suggests that it was a kangaroo that got caught in the fence, like so many others that we have seen.
|
Air fried spring rolls
|
Topic: food and drink, opinion | Link here |
Yesterday Yvonne brought some pre-fried spring rolls along with sushi. We've had mixed experiences with spring rolls: they should be deep fried, but it's very easy for them to fill up with fat and become decidedly unpleasant. So yesterday we tried heating them in the smaller air fryer, with good results.
Repeated the experiment today, with deep-frozen spring rolls. Maybe not quite as good—probably because of the rolls themselves—but definitely better than with a deep fryer.
There's not much where air fryers are clearly the best choice, but we're finding a few. They're also much less messy than deep fryers.
| Monday, 6 May 2019 | Dereel | Images for 6 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Saffron!
|
Topic: gardening | Link here |
This time last year I bought some Crocus sativus (saffron crocus) bulbs and discovered only too late that they were intended for planting in spring. Only one flowered:
|
|||||||||||||||||||||||
I got a refund and thought no more of it. But today I found:
|
|||||||||||||||||||||||
Is that all from the one surviving bulb? At any rate it seems to have multiplied.
|
Microwave oven shielding
|
Topic: technology, general, opinion | Link here |
Question on Quora today: Is it true that if you put your phone in a microwave, then call it on another phone and if it rings then your microwave leaks?.
Clearly he's talking about a microwave oven. Interesting. In principle I'd say “yes”: the shielding on the oven is designed to keep out electromagnetic radiation in the order of 2450 MHz (wavelength a little over 12 cm). That's easy enough with the mesh in the door, which has openings round 3 mm, or λ/40. And mobile phones have lower frequencies (typically 900 or 1800 MHz, wavelengths 33 or 17 cm). So clearly a mesh that keeps out cooking microwaves should be able to keep out phone microwaves.
That's easy enough to test. Put my phone in a microwave oven. Call it. Hear it “ring”.
What? Not what I expected. Try the other microwave oven. Call it. Hear it “ring”.
What's wrong here? Put it in a saucepan. Call it. Get voicemail.
So it seems that mobile phones do work in a microwave oven. How much attenuation is there? Surely there must be an Android app that measures this stuff.
OK, check that and find How To Measure Cell Signal Strength on Android Phones. As usual, with instructions that don't relate in the slightest to the menu structure on my phone. Thanks, Android developers, for your Tower of Babel.
Ah, I don't need that, says Peter Jeremy. Just look at the debug logs. Where? What, you don't have developer mode enabled? No, how do I do that?
Now isn't that intuitive? But it worked. And what I didn't see until later was that it doesn't give me any logs; for that I need the equally intuitive workaround of generating a bug report.
Gave up instead and went looking for apps. The one I found was Network Cell Info Lite, and on startup it showed:
|
|
|||||||||
OK, that looks straightforward enough, and it tells me that the -113 dBm signal level corresponds to two bars (as opposed to the one bar that the phone shows at the top of the screen). How do I measure inside an oven? Ah, a nice graphic display:
|
|
|||||||||
That was inside the Panasonic microwave oven. Outside it registered -121 dBm, inside it was -136 dBm. An attenuation of 15 dB, corresponding to a factor of 30. If that were energy from the magnetron, 1100 W would be attenuated to about 36 W, considerably more than I would expect.
Try the ALDI microwave oven. That did better:
|
|
|||||||||
140 dBm, an attenuation of 19 dB, or about 80 times. That would take the 720 W down to about 9 W, still more than I would expect.
OK, so that's at what frequency? Time to RTFM and find out what all these abbreviations mean:
|
|
|||||||||
That might also explain what the blue line in this graph means:
|
|
|||||||||
That's the result of putting the phone into the saucepan; it seems that it dropped back to UMTS mode, but more investigation is needed.
So how do I measure attenuation at 2450 MHz? You'd expect the same app to be able to do it, but if it can, I haven't found out how. Instead downloaded a couple of other apps of dubious quality, which made up for it by smothering me with advertisements. One, which is not called “Wifi Signal Meter”, gave me these results:
|
|
|||||||||
|
|
|||||||||
And of course it's too polite to show any axes on the graphs. But clearly the first one (Panasonic oven) shows significant attenuation, while the second one (ALDI oven) is barely recognizable.
Why? The Wi-Fi access point is directly behind the microwave oven:
|
|||||||||||||||||||||||
Can it be that the signal is so strong that even after attenuation it appears nearly full strength? Or are microwave ovens less well shielded at the back?
So how much does this tell me about stray radiation from the ovens? Clearly I can't repeat the test with the ovens running, but is there an app that measures the radiation? Yes, many, though I didn't find a good one. The two that I tried were useless: they were so smothered in advertisements that I wasn't able to establish with any certainty what I had. It might have been EMF Radiation Detector - Magnetic Field Detector (look at that URL! https://play.google.com/store/apps/details?id=com.tm.radiation.meetr), which calls itself “Radiation Meter” in the icon. But it was so difficult to avoid the advertisements that I gave up. When my stomach has recovered I can try again.
| Tuesday, 7 May 2019 | Dereel | Images for 7 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
MARION'S Laksa revisited
|
Topic: food and drink, opinion | Link here |
Curry laksa for breakfast today, this time with “MARION'S KITCHEN” paste again. I've been eating this on and off for a couple of years, but forgot to check my notes, and diluted it far beyond what it can handle. I need to make a table for the various varieties.
|
Bill shortens pensions
|
Topic: politics, opinion | Link here |
The Federal elections are coming up at the end of the next week, and every morning the news carries the latest promises from the major parties. Bill Shorten of the Australian Labor [sic] Party has promised to address the negative gearing that has priced city dwellings out of the reach of most people: good. And he has promised to end cash payouts of franking credits, the tax credit on dividends from many stocks.
BAD! That's what makes the superannuation investments of many people usable. Without the franking credits, my income would be about $1000 a month less.
After checking, the sum is closer to $500. Still a very large sum of money for a pensioner
Does he understand that? Took a look at his justification page and came to the conclusion that he doesn't know what he's doing.
And so, of course, did the Coalition, which prepared this page.
What now? Maybe things aren't as bad as they looked. If pensioners can't redeem the franking credits, then it will have an effect on the value of the stocks. For them the alternative is obvious: don't frank the dividends. But it makes me wonder if Labor has any better idea of what they're doing than the Coalition.
|
Epazote progress
|
Topic: gardening, opinion | Link here |
It's been six weeks since I planted my epazote seeds, and they took forever to germinate. Nearly 2 weeks later I discovered that my recently planted kniphofia seeds had overtaken the epazote.
But that was last month. Somehow the Kniphofia (on the floor) are no longer doing well, while the epazote (on the stand) are finally growing fast:
|
|||||||||||||||||||||||
Why are the Kniphofia not doing any better?
|
Still more “air fryer” success
|
Topic: food and drink, opinion | Link here |
The original idea for an “air fryer” (really an intense forced convection oven) it seems, is to cook chips. In my experience, that's a complete and utter failure. But in the course of time I've seen other things where they work well.
First, let's consider the advantages of an “air fryer” over a real deep fryer: it's much cleaner (no fat), uses fewer raw materials (no fat), and it uses less power. If you include the heat-up time for the deep fryer, it's also often faster.
After yesterday's success with spring rolls, today I tried with papadams. How hot? Try 200° for 7 minutes. After 4 minutes I opened and found:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Clearly too hot, but still edible. Interestingly, though, they had not increased in size. Tried again at 170° for 4 minutes. Before and after:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That looks better, and it doesn't taste too bad either. But of course it's an order of magnitude slower than a deep fryer for any reasonable number of papadams: 2 minutes per piece adds up quickly.
| Wednesday, 8 May 2019 | Dereel | Images for 8 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Solar electricity: more labeling
|
Topic: Stones Road house, general | Link here |
Call from Tony of Effective Electrical this morning: it seems that the surprising number of signs on the components were still not enough for Powercor, and he had to come and add more. Things have evolved. Here is what it looked like on 3 April 2016:
|
|||||||||||||||||||||||
After installation, it grew:
|
|||||||||||||||||||||||
And here it is today:
|
|||||||||||||||||||||||
Somehow I think that a circuit diagram somewhere handy would be safer.
In passing, discovered that the energy that I have fed back into the Powercor grid has not been credited to my account: that won't happen until their inspector (finally!) comes and takes a look; that should have happened at least 2 weeks ago. I think what this is really saying is that yes, Effective have received my voluntary payment, so they're (only) now doing the next step.
While the covers were off the switchboard, took a look at the wiring:
|
|||||||||||||||||||||||
Yes, as Tony confirmed, I can remove the RCD in the middle and replace it and the three circuit breakers to the right with three RCDs, thus limiting the effect of any future trip like the ones we had last week and three months ago:
|
Aussie Broadband: not what they used to be
|
Topic: technology, opinion | Link here |
In the mornings I download videos from German TV broadcasters. It's not always plain sailing, and today a couple failed on me.
Why? The reasons are endless. But for once I received a specific error message: the proxy server was down.
What? Checked. Yes, no access. We were off the net! Dammit, this happens far too often (the last time was less than 4 weeks ago). And the procedure was the same: check my saved mail. No, no planned outage. Call Aussie Broadband support. “You are caller 33. Estimated wait time is 15 minutes and ... 40 seconds”. Damn, and that with a mobile telephone. Why can't they at least have a “press splat if you want a call back”, like many others have?
OK, fire up their app, and though the error status was clear on the NTD (ODU error indicator), go through their test, and once again hang:
|
|
|||||||||
|
|
|||||||||
What's the difference between these two displays? Ten minutes (top right hand corner: 11:31 and 11:41). Clearly nothing is going to happen there. OK, give up and hope that the system will come back soon.
It finally came back an hour later. Try the app and see what it says this time:
|
|
|||||||||
No current outages. The last one started 5 hours ago and will end in 7 hours' time. The next one will start on 22 May 2019 at 7:00 and will end on 29 May 2019 at 20:00. A total outage of one week, 13 hours!
Yes, it's clear that we won't be off the net anything like that long. But this level of reporting and reliability is just not acceptable. Currently I have:
Yes, that's (marginally) better than my satellite availability:
But look at that average duration. The satellite outages were short and frequent. I've had far too many outages of an hour or longer with the NBN.
And Aussie Broadband, once exemplary, is gradually dropping significantly in my estimation.
|
The mildness of autumn
|
Topic: gardening, opinion | Link here |
The north side of the north garden bed has been less than satisfying. I planted a Salvia microphylla there ten months ago, but while it didn't die, it didn't grow much, and only ever had one or two flowers.
That has changed, though it's still not overly effusive:
|
|||||||||||||||||||||||
But it is autumn; with a bit more fertilizer and watering it might look better next year.
The Eremophila nivea next to it is also looking better than it was:
|
|||||||||||||||||||||||
And the Banksia integrifolia next to it is doing very well:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Maybe I should consider planting more Banksias. It's about the only tree that is growing normally.
The other salvia, though, appears dead:
|
|||||||||||||||||||||||
But it has self-seeded into the succulent bed, so all is not lost:
|
|||||||||||||||||||||||
And the “Gallipoli” rosemary is also growing:
|
|||||||||||||||||||||||
I think I really need much more fertilizer for this soil.
|
“Air fried” chicken skewers again
|
Topic: food and drink, opinion | Link here |
Chicken skewers for dinner tonight. As last month, I cooked them in the ”air fryer”. Last month I cooked for 12 minutes at 200°, and they weren't really crispy enough. Today I did them for 15 minutes at 220°. Yvonne's verdict: still not crispy enough. Next time 20 minutes at 230°?
I wonder how accurate the temperature settings are.
| Thursday, 9 May 2019 | Dereel → Ballarat → Dereel | Images for 9 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Mediathek reliability
|
Topic: multimedia, technology, opinion | Link here |
The German „Mediatheken“ (multimedia libraries, it seems) offer a large amount of TV material for free download, and one of my first actions every morning is to check the updates provided by MediathekView. Today I discovered that they're re-running (so to speak) the first year of „Um Himmels Willen“. OK, start the download. Episode 2 failed for some unknown reason.
That has happened before, and there's an alternative: go straight to the broadcaster's web site, in this case ARD, and download from there with youtube_dl. Search for „Um Himmels Willen“. Nope, no hits:
|
|||||||||||||||||||||||
Huh? They've been broadcasting more recent years. What does their “Series” page say? It comes with this horrible server-side slider that you have to press several times to view, but then found the page, tastefully half obscured:
|
|||||||||||||||||||||||
|
|
|||||||||
Yes, there are episodes, but only recent ones. Why don't they list the other ones? And why didn't their search function find any mention of the series?
It seems that the original issue was some kind of network glitch; after downloading other videos, MediathekView went back and retried, and it worked. But this leaves me with at least three problems with the ARD site:
To be fair to ARD, they're not the only people with this stupid scroll system. Just look at smart TVs and wonder what kind of brain damage led to the idea.
|
Anniversaries
|
Topic: history, opinion | Link here |
In my morning mail:
I remember the date, of course, but 44 years! How time flies!
But there was more:
I didn't remember that date. And just by coincidence, it was exactly half the time between my wedding with Doris and today.
|
To the doctor again
|
Topic: health | Link here |
Into town today to hear the results of my blood tests. Nothing unexpected: HbA1c 5.9, in the normal range, but fasting blood glucose 7.4 mmol/l. That's nothing new; I've had that for 12 years now. But Paul still can't make up his mind whether I'm diabetic or not.
Apart from that, MCV 104 fl again. Somehow that's the only anomaly, and since the Vitamin B12 parameters all seem normal (for once!), there seems to be no obvious reason for concern.
Had a influenza innoculation while I was there.
|
Internet signs of the times, 2019
|
Topic: general, technology, opinion | Link here |
I had nothing much to do after leaving the doctor's, so I took a somewhat roundabout way back to the car park, via the old Block Arcade, which has been looking down-at-heel for some years now. It looks quite different now from the photo they have on their home page:
In particular, the premises on the right are empty and to let, not the only one in the relatively short arcade; I'd guess that 40% of the premises are empty.
On through Myer, which has been looking in need of renovation for some years now, and in to Central Square It looks a lot better than the other two, but the writing is on the wall, literally. Coming in from the east, the place looks like this:
|
|||||||||||||||||||||||
What's in the food court? Nothing much, but I discovered one shop, rather more tastefully covered up than in Block Arcade:
|
|||||||||||||||||||||||
That's the white wall in the background, which should be selling some kind of food. It's not the only one in the panorama above: on the right, in what should be one of the best positions in the shopping centre, is this:
|
|
|||||||||
That's roughly what I predicted in my paper The future of the Internet over 5 years ago. I've seen some of these shops change hands frequently, and I wonder if the current Flight Centre shop (barely visible on the left of the panorama) will last long. It's interesting that even food courts have difficulties, though the shops that are there don't look very exciting.
|
Strange bedfellows
|
Topic: technology, opinion | Link here |
I forget why, but today I discovered this page, a relatively recent photo of Julian Assange with... rms:
The original was at https://ergoemacs.org/emacs/i/Richard_Stallman_and_Julian_Assange_2013-07-12.jpg, but that's no longer available.
What's unusual about that? At the time Julian was in self-imposed exile, so rms must have gone to visit him. Clearly he never read the NetBSD fortune files:
I'm sure there are others, but I can't find them. I was left with a feeling of great antipathy from Julian towards all that is GNU and rms in particular.
|
Still more kangaroos
|
Topic: animals | Link here |
Seen from the lounge room:
|
|||||||||||||||||||||||
|
Real network problems
|
Topic: technology, opinion | Link here |
A rather surprising email this evening:
Date: Thu, 9 May 2019 17:52:30 +1000
From: Aussie Broadband <sales@aussiebroadband.com.au>
Subject: [Outage Ref: 49072] Our apologies - we're experiencing an outage
Apologies from the Aussie Broadband team - we're experiencing an outage in your region that may be affecting your service.
Our network team are investigating as fast as possible. We don't have an ETA for resolution yet.
Please keep in mind that our tech support call centre is likely to be pretty busy.
You can find updates on our website support page at https://www.aussiebroadband.com.au/support/, or on our Facebook page.
And yes, the support page did show an outage in the Ballarat area, and so did my network status page, here shown after the event (and thus also that it finished round 22:47):
Interestingly, despite the “outage”, I didn't completely lose connectivity. My status page didn't even register one.
|
More spring rolls
|
Topic: food and drink, opinion | Link here |
Had another go at cooking spring rolls in the “air fryer” today, smaller ones (about 1.3 cm in diameter) at 170° for 5 minutes. Not enough, I think; I'll try 200° and 8 minutes next time.
| Friday, 10 May 2019 | Dereel | Images for 10 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
The rains come!
|
Topic: general, opinion | Link here |
Somehow it seemed to have rained all night. That's probably because it did:
According to the weather station (which is very inaccurate and on the low side), we've had a total of 141.3 mm of rain this year, of which 81.6 mm were in this month. Today the weather station registered “only” 40.5 mm, but the rain gauges came up with closer to 60 mm. And again the effects were to be seen:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Spent a lot of time pumping water out of the pit in front of the house, being careful not to let it run dry, until I discovered that the float switch was working correctly. But I must have pumped hundreds of litres of water out from that area along. Time to talk to Warrick Pitcher about better drainage.
|
Ken Thompson interview
|
Topic: technology, opinion | Link here |
Saw an interesting interview recently:
Brian Kernighan interviewed Ken Thompson on a range of subjects, only barely touching Unix. I had some new insights, though I suspect that Ken might have been deliberately misleading in some of the stuff he said. For example, in (northern) summer 1969 he spent 3 weeks hacking while his wife visited relatives. Peter Salus writes that he spent this time writing a file system, which seems a tough enough challenge. But Ken said in the interview that he spent a week each writing an editor, an assembler and an operating system. No mention of a file system, though clearly that would have been a big part of it.
Other things of interest: he barely mentioned dmr. Those three weeks in summer 1969 he was on his own, and then he mentioned people who joined the effort, including Doug McIlroy, Bob Morris and Dennis Ritchie. But from his interview you wouldn't have thought that Dennis was any more important than the other two.
And the there was his T-shirt:
|
|||||||||||||||||||||||
What does that comment say? The closest I can come is “ΕΠΙΤΕΛΟΥΣ ΤΟ ΚΑΤΑΛΑΒΑ!”, which doesn't make much sense to me. Clearly you're not supposed to understand this: the real clue is the code underneath, capitalized for some reason, and that rang a bell: it's from the Sixth Edition Unix kernel, /usr/sys/ken/slp.c, lines 325 and on.
But to be on the safe side, I put it through Google Translate, which indeed was able to make some kind sense of it: it translates it as “the successful”. But pull the words apart and it changes its tune and says “At last the understood”. So somewhere there's a grain of truth in the statement. Now I need to parse it. And somehow I'm disappointed: it seemed cleverer before.
A few weeks later I received a message from Andreas Leitgeb, who translated it as “Finally I understand this”.
The funny thing is that the code is gradually starting to make sense. Together with dmr's comments it appears that they were hand-crafting a longjmp() with implicit reliance on the compiler implementation of stack frames.
| Saturday, 11 May 2019 | Dereel | Images for 11 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Facebook pain again
|
Topic: photography, technology, opinion | Link here |
Question on Quora: “Does Facebook remove EXIF data from uploaded images?”.
That sounded like an interesting question, and simple enough to confirm or deny. OK, enter a post. But how? Nothing echoes in the box provided.
Gave up and tried again with Chrome, which echoed. And Facebook sent me a message:
We noticed a recent login for your account @Groggyhimself.
Device: Chrome
Location: Traralgon, Victoria
*Location is approximate based on the login's IP address.
If this was you
Great! There's nothing else you need to do.
Now isn't that clever? Yet another report of me being in Traralgon, 300 km away. But what if somebody there really tried to break in? It says “Location is approximate based on the login's IP address”, which is doubly irritating: firstly, since when do “logins” (an action of logging in) have an IP address? And what is the IP address that they used? That would make things much easier. Why do people omit things like this?
Entering my “post” was easier, up to a point. First, of course, I had to upload a photo, a pain
in itself because all these stupid browsers don't have file name completion nor any concept
of directory folder hierarchy.
But finally I climbed the trees and pressed whatever the button is that means "upload":
|
|
|||||||||
|
|
|||||||||
OK, people, what was the error (again)? And what kind of work are you doing to fix it? The obvious work would be to say what the problem is, but that's too obvious for modern systems. Once upon a time you'd end up with messages like this:
|
|
|||||||||
Isn't that ugly? Clearly we can't have anything like that in a modern system.
Finally, of course, there was the issue: how do I submit the message? No “submit” button. Ah, that's an old, worn-out magic word. Now it's “Share”, the same word these people use for remote file system access.
Finally, the result: yes, Facebook is too polite to publish the Exif data. Who knows what harm it might cause.
| Sunday, 12 May 2019 | Dereel | Images for 12 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
New stables
|
Topic: general | Link here |
Seen at the gate this afternoon:
|
|||||||||||||||||||||||
That was CJ Ellis bringing a couple of posts for some new extension to Yvonne's stable area. At the end we had:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's going to be some kind of covered area to protect against the rain and wind, which come from that direction. I wonder if it will be more useful than this pen, which was built for Gabriela, the miniature horse, and which is apparently of no use for anything else:
|
|||||||||||||||||||||||
|
Sheep parade
|
Topic: animals, opinion | Link here |
While walking the dogs today, saw:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Why are the sheep lined up like that? Preparing for inspection? No, clearly they have been issued with hay, and they're eating it.
| Monday, 13 May 2019 | Dereel | Images for 13 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
International money transfer pain
|
Topic: technology, general, opinion | Link here |
Yvonne has booked a course in Queensland in early spring. Nothing unusual in that, but we have to pay in Canadian dollars, and they wanted something like a 10% surcharge for transfers from outside the country.
OK, I've been trying to arrange for cheaper international fund transfers for a couple of years now, and somehow I haven't got round to it. Most recently I found Transferwise and signed up with them, but never got round to doing any transfers. Finally started to do so a couple of days ago and ran into issues specific to the Canadian banking system—I think. So I sent off mail with questions and get replies today.
So: now do it for real. First, I needed to sign in again, and their web server was too polite to talk to my firefox, so I had to fire up a separate Chrome. Log in. “We have sent you an SMS”. Dammit, every time? It's a good thing that we now have mobile phone coverage.
Then filled out the form exactly as discussed with the other side. Nope, don't like your postal code (the one confirmed by the receiver of the transfer):
|
|
|||||||||
|
|
|||||||||
Tried various combinations, including Canada Post's postal code search. It took me a very long time to realize that I had misparsed the text:
ON is the province abbreviation and not part of the postal code. But how was I to know that?
Then more pain transferring money from my bank, but finally it was done. Only 2 hours! And the fee was $4.32. That looks like more than the 0.35% or 0.45% that I was quoted last year. What about the conversion rate? They gave me $0.938139, which I suppose compares favourably with the $0.9365 that OANDA gave for today's Interbank rate.
|
Red-flowering gum
|
Topic: gardening | Link here |
Seen while walking the dogs today:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
It's a pity they flower for so short a period. But maybe we should get one too.
|
Preparing for the Federal election
|
Topic: politics, opinion | Link here |
There will be a Federal election on coming Saturday. The satirists are preparing:
At least I hope that it's the satirists.
| Tuesday, 14 May 2019 | Dereel | Images for 14 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
What network speed?
|
Topic: technology, opinion | Link here |
Almost exactly 4 years ago I took part in a free trial of NBN's then-new 50/20 speed rating (50 Mb/s down, 20 Mb/s up). It seems that the trial lasted several months, because six months later I tested the speed pinging Google's name server 8.8.8.8 and came up with these results:
=== grog@eureka (/dev/pts/14) ~ 28 -> ping -c20 8.8.8.8
...
round-trip min/avg/max/stddev = 38.503/51.678/62.721/5.839 ms
Things could have changed, of course: in particular, we now have two links from the Radiation Tower to the POP. So for the fun of it I tried again (actually paying $10 for the privilege). Now it looks like this:
=== grog@eureka (/dev/pts/16) ~ 74 -> ping -c20 8.8.8.8
...
round-trip min/avg/max/stddev = 34.691/46.752/59.228/5.698 ms
Yes, that's marginally better. And the 1400 byte payload?
=== root@eureka (/dev/pts/9) ~ 21 -> ping -c20 -s 1400 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 1400 data bytes
72 bytes from 8.8.8.8: icmp_seq=0 ttl=55 time=59.788 ms
wrong total length 92 instead of 1428
...
round-trip min/avg/max/stddev = 42.306/52.864/59.993/4.588 ms
What happened there? It seems that 8.8.8.8 no longer responds with big payloads; the same ping to www.lemis.com returns the payload as requested. That makes the test pretty useless, except maybe to indicate that the performance problem that I mentioned last time was possibly not due to the link. With www.lemis.com I get with 1400 byte and 56 byte packets:
round-trip min/avg/max/stddev = 299.498/306.932/316.082/5.683 ms
round-trip min/avg/max/stddev = 289.057/298.828/314.335/5.624 ms
That's a difference of 8 to 10 ms, still more than you'd expect, but much better than the 100 to 130 ms I noted last time.
Of course, what I really need is a speed test site. But which one? Aussie Broadband have restyled their web site, and I can no longer find that sort of thing. And the page that I found via a Google search was of little help:
|
|
|||||||||
Third-party tests were of limited value, of course. But one of them complained about my ad blocker. OK, I have a spare browser. That worked—and so did the Aussie speed test! So it seems that this was their polite way of saying “please disable your ad blocker”. Why did they want to spam me with advertisements?
The results were nothing to write home about:
Download Speed: 30122 kbps (3765 KB/sec transfer rate)
Upload Speed: 17710 kbps (2213 KB/sec transfer rate)
Latency: 30 ms
Download Speed: 31682 kbps (3960 KB/sec transfer rate)
Upload Speed: 17286 kbps (2160 KB/sec transfer rate)
Latency: 29 ms
Surprisingly, though, other servers did better. An unspecified Telstra server (“Windsor”) gave me a download speed of 42.7 Mb/s, but latency of 70 ms and upload of 15.28 Mb/s.
More to the point, though: what do I get for this speed? Marginally faster downloads from SRF, but at typically 15 Mb/s they were never the issue; some of the German sites are only about a tenth of that speed, and not surprisingly there was no difference there. The upload speeds are of no particular interest, especially since I need to tune my remote systems to get even 5 Mb/s upload. So, somewhat to my surprise, it'll be back to 20/5 at the end of the month.
| Wednesday, 15 May 2019 | Dereel | Images for 15 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Solar electricity installation accepted
|
Topic: Stones Road house, general | Link here |
Round 9:30 this morning a friendly inspector from Powercor came along to inspect last month's solar electricity installation. He showed me what he was looking for as he went (in particular, good anchors for the panels, which sounds like an excellent idea), expressed surprise that it had taken so long for him to be called (I still think they waited until I paid), and noted that Effective Electrical were usually pretty reliable, unlike some others whom he didn't mention by name. He took a look (apparently his first) at the inverter, and we discussed the display: it shows round 5 V more than in my office. So he went off and measured the voltage at a suitably placed power point near by; no significant difference. But it seems that our line voltage, once round 237 V, has gradually crept up, and now we're round 245 V. That's irritating, since it will climb further when I'm feeding power in.
And next? We'll hear from somebody. His job is done, system accepted without any issues.
|
Where's my pump?
|
Topic: general | Link here |
It's been five months since I brought my then 10½ month old pump in to the Ballarat Pump Shop for repair. I called them in January and was asked to wait. But that's enough waiting. Called up and spoke to Bruce who was—understandably—surprised. They appear to have lost my contact details, but he found a pump, still in bits, that matched the description. Will talk to Jeff and call you back.
Hopefully it won't take another 5 months.
|
MediathekView display issues
|
Topic: technology, opinion, multimedia | Link here |
I use MediathekView on a daily basis to choose which videos to download. The download screen typically looks like this:
|
|||||||||||||||||||||||
|
|
|||||||||
The first image shows what it looks like on a full screen (1920x1080, and the second is reduced in size to make it legible. The bottom right is in two panes: the pane marked green in the second image (meaning “download finished”) is a list of selected videos. The pane below is a description of the selected line (dark green).
All well and good. But today, somehow, the description pane was gone. Why? Where? How do I get it back? Tried configuration and all sorts of potentially hidden sensitive areas, to no avail. Have they changed the software? There was no download, so it seemed unlikely.
OK, I have another instance running on teevee. Is it gone there too? No, it's there. And there's the x at the top left of the pane. OK, click on that. Gone? No, that would be too simple. If it is, I have two instances that don't have the pane any more. First save the configuration. Then click on it. Gone!
Stop program, compare configuration:
OK, that's clear: the display is controlled by the variable system-download-beschreibung-anzeigen. I know how to turn it off. I don't know how to turn it on, and I'm not sure that there is a way from the program. But stop the program, frob the config file, and it works again.
In passing, yes, this is what the config file really looks like. It seems that they really have a config variable with the typo system-anz-tage-filmilste.
|
The smart of Android
|
Topic: technology, photography, opinion | Link here |
Somebody mentioned Google camera recently. It's for phones, of course, so off to the toyshop, where I found the description. OK, try it on the phone:
|
|
|||||||||
|
|
|||||||||
|
|
|||||||||
Where is it? It comes up with all sorts of things, just not anything that's obvious Google Camera. What does Google Street View have to do with the search terms, for example?
I couldn't find it. I gave up. But when I went back to a real computer and checked, I found the fine print:
|
|
|||||||||
“This app is incompatible with all your devices. Want to install?”.
O brave new world, that has such technology in 't!
| Thursday, 16 May 2019 | Dereel | Images for 16 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More solar power documentation
|
Topic: Stones Road house, general, opinion | Link here |
An unexpected email this morning:
Who the hell are they? Somebody who read my diary and wants to cash in on something? No, I also got an email from Effective Electrical telling me that they would contact me. But why? I thought I would have an agreement with Powercor.
OK, follow the link and get a pretty normal web form, except that every address started with “MS: “MS, 21B Villiers DR, Wendouree VIC 3355”, “MS, 29 Stones RD, DEREEL VIC 3352” and so on. No way to fix it; just press “SUBMIT” to continue. “Thank you, your acceptance has been noted”. I didn't accept anything! Anybody could have “signed” this. And later I got a confirmation:
Digital signature? I did nothing of the kind. I wonder what legal validity the form has.
|
Flowers in late autumn
|
Topic: gardening | Link here |
Middle of the month, now late autumn. Time for the monthly garden flower photos.
The garden didn't fare well over the summer, and of course things still show. But now that it's cooler, some plants are recovering. I have a surprising number of hibiscus and similar:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The first two photos are the “Uncle Max” Hibiscus rosa-sinensis, which is still not very happy outside; possibly next spring will be better if it survives. The next is the Hibiscus with the erectile dysfunction and the prolific but small flowers.
Then there are the relatives:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The first two are the Alyogyne huegelii that I planted in the spring, and the last is a Hibiscus syriacus flower. I hadn't noticed until now how similar the flowers are. I planted two Hibiscus syriacus trees a couple of years ago. One suffered at the hands of Mick's whipper-snipper, but proved equal to the challenge; after it recovered a little, we transplanted it to the north side of the house, where it has grown a lot but barely flowered (the last image above and the first below). The other one has flowered a lot but not grown as well, presumably because of the wind (second image).
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The trees in the front garden have been a source of concern since we moved in. Two of three birches died, but this year the third has what looks like normal autumn foliage:
|
|||||||||||||||||||||||
Nothing special there. But that's not the way it was before. Last year it looked like this:
|
|||||||||||||||||||||||
But that was in February!. After that it lost all its leaves.
The situation is similar with the Paulownia kawakamii:
|
|||||||||||||||||||||||
I don't seem to have taken any photos last autumn, probably an indication that there wasn't much to see, but in the middle of summer 2018 it looked like this:
|
|||||||||||||||||||||||
Clearly it has grown a lot, but still far short of the way it should have (by now it should be 5 m tall).
The Clematis “General Sikorski” that was nearly strangled by the Tropaeolum is recovering well. I thought it might be forming a flower, but not quite. The first photo is from last month:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Preparations for spring are well on their way. All of these bulbs are outside my bedroom window:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And the first forget-me-nots are flowering in front of the library:
|
|||||||||||||||||||||||
On the down side, the blue-flowering salvias have almost completely died. Here last year; I didn't even bother to take a photo of the dead sticks today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
Strange flower?
|
Topic: gardening, animals, opinion | Link here |
Walking the dogs today, saw a strange red flower showing through the undergrowth:
|
||||||||||||||||||||||||
What was it? Not a flower at all: the rib cage of a dead wallaby:
|
||||||||||||||||||||||||
|
Alpacas
|
Topic: animals, opinion | Link here |
Walked the dogs down Westons Road for the first time in a long time. The people on the corner now have Alpacas:
|
||||||||||||||||||||||||
Not Llamas? No, the ears are different, amongst other things.
They paid a lot of attention to us, possibly wondering what kind of animal Nikolai and Leonid were. Then they made a noise. Warning? It sounded like I could imagine a Banshee, something between a gurgle and a scream. Quite eerie.
|
Microsoft update pain
|
Topic: technology, opinion | Link here |
A couple of days ago I tried to update the system on dischord, my Microsoft box. I failed: it hung about ⅔ of the way through the download. OK, yesterday I tried again. And how about that,
|
|
|||||||||
|
|
|||||||||
It stayed like that all afternoon. OK, Microsoft takes its time. Wait.
But then I saw another detail:
|
|
|||||||||
Last update 11/01/2019 at 7:22 PM? In other words, 11 January 2019 at 19:22? Surely I've updated since then. Follow the update history link:
|
|
|||||||||
|
|||||||||||||||||||||||
What are these people smoking? In any case, it carried on with no change for the rest of the day.
| Friday, 17 May 2019 | Dereel → Ballarat → Dereel | Images for 17 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Still more solar electricity documentation
|
Topic: Stones Road house, general | Link here |
Still more mail today regarding the solar electricity installation: a Certificate of Electrical Safety, a copy of “fully paid invoice for your solar system” (I had only done a part payment, and it's a photovoltaic installation, not a “solar system”) and an STC Assignment form, 4 pages of it, with serial numbers for each and every component, including each solar panel (with 24 digit serial numbers), but nowhere explaining what an STC is. I had assumed something like “Solar Trade Credit”, but it proves to stand for Standard Test Condition, which means barely any more than STC.
But the best was at the end: the confirmation of the “digital signature”:
Now isn't that a convincing proof that I signed it? Admire the RFC 1918 address. But where did it come from? Not from me. I don't use RFC 1918. My best guess is that this is an address in their internal network.
So what happens next? Will I finally receive credit for the power I put back into the grid?
|
Simplifying bank transfers
|
Topic: technology, general, opinion | Link here |
Yvonne is collecting money for the next Anke Hawke clinic. She received an interesting question: “Do you have PayID?”. Huh? What's that? Off looking and discovered https://payid.com.au/, which tells me that it's a way to replace two numbers (BSB and account number) with one number, apparently useful for people with limited memory. It wasn't until later that I discovered a real advantage, one that they don't seem to have bothered to mention: it's portable, so if you change banks, you can at least theoretically take it with you.
But why wasn't I informed? Logged in to my ANZ “Internet Banking” and saw nothing. OK, “What's new”? Yes, PayID is here:
Forget your BSB and account number. Create a PayID and get paid quickly by sharing your mobile number or email address.¹
Look at the small print. ¹ refers to a text:
“Technical interruptions may occur”. Doesn't that fill you with confidence? I translate it as “the application is buggy”. OK, let's set up, anyway. Select “Create PayID”. A too-small popup window appeared with the login screen. OK, log in again (why?). And there I was back at my “Your accounts” screen. No mention of PayID anywhere.
Clearly a broken application. In case of doubt, try Microsoft; they might have tested it there. Nope. Nothing. Help? No. Contact? Either call us on the phone or come into a branch (I can imagine how much help the latter would be). So basically, from this perspective, it Just Plain Doesn't Work.
This is supposed to be easier? Of course, maybe it's a technical interruption. What a wunch of bankers!
|
More physiotherapy
|
Topic: health | Link here |
Into Ballarat today for another round of physiotherapy with Heather Dalman. Apart from a misunderstanding on her part about the time of the appointment, nothing of great interest. She seems happier with my progress than last time, and it'll be a month before I see her again.
|
Microsoft update, day 3
|
Topic: technology, opinion | Link here |
Left the Microsoft update running overnight for a second night. It didn't help:
|
|
|||||||||
Like Schrödinger's cat, it has both failed and is still working. OK, do what I had planned and install only some of the updates.
|
|||||||||||||||||||||||
But it came back with the same error message as before. Or did it? Ha, ha, only joking. I left it like that, but next time I looked, it had reconsidered:
|
|
|||||||||
As far as I can tell, it didn't even download anything else. The installation succeeded, but being Microsoft, it would required a reboot and endless reconfiguration. Another 6 times? It's clear that something big has failed. So leave out the biggest one, the security “rollup” KB4499164, and try the remaining 5:
|
|||||||||||||||||||||||
That went off and did the Microsoft thing:
|
|
|||||||||
And that was it for the day.
| Saturday, 18 May 2019 | Dereel | Images for 18 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Microsoft update, day 3
|
Topic: technology | Link here |
So now I have installed 6 out of 7 Microsoft updates; it has only taken 2 days. On to the last one:
|
|
|
|||||||||
It stayed that for long enough that I lost interest, and the next time I saw the same picture, but claiming that updates had been installed today at 14:22 (sorry, 2.22 PM). Nothing that I saw, and definitely no reboot.
|
ON1 again
|
Topic: photography, technology, opinion | Link here |
ON1 Photo RAW 2019.5 (isn't the shouting nice?) has just been released. It offers a surprising number of features that I've been looking for, especially since it gives the impression of being an incremental release:
Terms like “A New Workflow for Layers” and “User Interface: The new user interface will have a fresh and modern feel” look ominous, but I can always try it.
Or can I? I noted that I tried it 18 months ago and discovered that it required OpenGL 3.3 and later, but didn't tell me how to install it. Today I had the same issue after finally installing it (and watching the install process just die without any message, leaving me to search the maze of twisty little icons, all different, for the one that started ON1.
So why don't I have OpenGL 3.3? And how do I install it? It seems that it should be installed on just about all systems, but I couldn't find it in the program list, and it failed both on dischord (desktop) and euroa (laptop). Is this something to do with the fact that I'm running Microsoft “Windows” 7 on both? Or maybe because I'm running via a remote desktop? Too much pain for today, but maybe tomorrow...
|
The people have spoken... the bastards!
|
Topic: politics, opinion | Link here |
Federal election today. When should I vote? After a bit of consideration, decided that mid to late afternoon would be the best choice. Finally off round 15:00 and found the place closed, though a couple of people were standing round a stand in the car park, presumably selling sausages.
Oh. This was at the Dereel Hall, and we had always had elections in the side hall. Have they moved to the main hall? Yes, it proved that they had, and they were too polite to assume that anybody would make the mistake to go to the place where it was last time, so there was no sign on the door.
The place was pretty quiet, and I didn't have to wait at all to get my ballot. I had hoped to find supporters of the main parties there, but it seems that they had given up and gone home. Pity: I had plenty of opinions for both Labor (“why do you want to cut my pension by $1000 a month?”) and the Liberals (many, mainly human rights and climate). Clearly I didn't agree with either party, but given the small choice, Labor seemed better.
In the evening, Chris Bahlo and Margaret Swan came to dinner, and after the polling booths closed (20:00, 18:00 in WA), we followed the results, rejoicing as Tony Abbott lost his seat. But that was the only good thing. Not only did the The Coalition stay in government, it seems to have received more votes than before. How could that happen? As we discovered (and I confirmed on IRC), while we're not rabid Labor supporters, we're all agreed that the Coalition as it stands today is one of the worst things that has happened to Australia in recent years. Who are these silent majority? And what are they thinking?
|
Schnitzel: deep fry or pan fry?
|
Topic: food and drink, opinion | Link here |
(Crumbed) veal schnitzel for dinner tonight, along with gratin dauphinois. Normally we eat this as Wiener schnitzel, in which case it should be deep fried. But somehow that seemed wrong to me in combination with the gratin, so I made half pan fried in butter (left) and half in the deep fryer (right):
|
|||||||||||||||||||||||
The result? The deep fried ones were marginally overdone, but the consensus was that the deep fried ones were better.
| Sunday, 19 May 2019 | Dereel | Images for 19 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Power fail!
|
Topic: Stones Road house, technology | Link here |
Into the office this sunny morning, and almost as a matter of course looked at the “dashboard” display of the solar electricity inverter. No power! The PV generation was off.
What happened there? Out to take a look at the inverter. No, no problems there. Oh. PV array shows 478 V, which looks right, but 0 W input. Why? Played around with the menu system on the inverter and came up with a way to “Start” and “Stop” the inverter.
|
|||||||||||||||||||||||
OK, cautiously take the inverter out of circuit with the unmarked oscillator
isolator to the right of the inverter:
|
|||||||||||||||||||||||
Then stopped and started the inverter. No change.
OK, what happens if I disconnect the PV arrays from the inverter using the convenient isolators?
Nothing! It continues to display the same voltage! Something has clearly hung.
OK, let's isolate everything, including the battery. Surprise, surprise: the inverter shuts down completely. Well, what would you expect when you take away all its power? Turned it on again, and it took about 10 s to fire up again. But then it worked normally, and the PV voltage was 458 V, presumably because it was drawing a couple of kilowatts. About the only other thing worth noting was that I needed to start the “dashboard” at http://inverter.lemis.com/dashboard/main.html.
OK, what does the information in my database say?
So for no obvious reason the inverter status changed from “On-grid” to “On-grid (calibrating batteries - 1)”. Why? And why did it stay that way, even after I had restarted the inverter?
As if that wasn't enough, the inverter started charging the batteries both from the PV array and from the grid:
|
|||||||||||||||||||||||
Why? Yes, it's (currently) set up to start charging from the grid when the battery charge hits 20%, and not to stop until it hits 30%, but here we have a charge of 75%. Turned off the main switch, and sure enough, it charged normally. But clearly something's wrong, and I'll need to investigate further.
|
More RCD trips
|
Topic: Stones Road house, general | Link here |
As if the fun with the solar power inverter wasn't enough, round the same time I had not one, but 3 trips on the RCD. What caused that? It's easy to guess that the problems are related. Overvoltage, maybe? I should get some kind of mains voltage monitor. Or maybe I can find a way to get the inverter to supply the remaining information that it displays on its screen but not elsewhere.
|
Sprinkler problems
|
Topic: Stones Road house, technology, gardening | Link here |
I've been dragging my heels again on cleaning the bore water filter, and it shows: the garden beds are no longer moist. OK, clean the filter, then run the sprinkler:
=== grog@eureka (/dev/pts/21) ~ 47 -> netsprinkle 30 30 30 30
Nothing! What went wrong there? Something to do with the other two issues, maybe? Discovered that it was off the net, or at least it didn't respond to pings. My best bet was one of the network switches in the pantry, so went and power cycled them. No change. Finally got euroa (my laptop) and took it out to the shed and connected to the network cable. Works.
Damn! The power supply voltage looked OK, so it must be the relay itself. Is this related to the other issues? Power surge, maybe? The indicated voltage has been very high lately, just shy of the 253 V maximum allowed by the standard.
More to the point, what do I do now? Order another one? How long will that take? As it happened, I found one in Sydney for about the same price as I paid for this one, but with a case and mounting to hang on a wall. That sounded like a good idea, so I ordered one. It's a different board; hopefully I'll get enough information to control it.
|
ON1 tried again
|
Topic: photography, technology, opinion | Link here |
So why does ON1 Photo RAW 2019.5 want OpenGL 3.3? Is it installed on my systems or not? Or is it an issue with remote desktop?
Fired it up on euroa. Crashed. Do you want to report? Yes, of course. But it seemed that that referred to the last attempt. After that, sure enough, it started with no trouble and happily migrated to my remote desktop session. So maybe it is worth looking at after all. I just have to work my way through yet another maze of twisty little menus, all different.
|
Steak and kidney pie, again
|
Topic: food and drink, opinion | Link here |
Steak and kidney pie for dinner tonight. We had that pretty well worked out last time, but it turned out to be different today: I had assumed that cooking for 1 hour at 150° would be enough, but although the beef (topside) looked particularly good, so good that I felt I should have made roast beef out of it, it was still a little firm. Longer cooking next time. And I should make a note to make the pieces smaller, not more than 1 cm thick.
And finally, what weight? In the past I made individual pies in these forms:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That proves to take about 230 g of filling. But it was too much for Yvonne: she ended up eating only half. We'll have to find a suitable smaller form.
| Monday, 20 May 2019 | Dereel | Images for 20 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More power insights
|
Topic: Stones Road house, technology | Link here |
After yesterday's fun with the solar electricity installation, spent a lot of time observing and analysing the results. About the only definite conclusions were that it happened again, this time during normal charging:
It's interesting to note the changes in the fields Pac and W1. At the very least it means that the inverter has started charging the battery from the grid, although there was plenty of power to spare from the PV array. Took a screen shot shortly later:
|
|||||||||||||||||||||||
This corresponds to this status from the inverter:
OK, it's clear that W1, PacPV and PacBat are power to the inverter from grid, PV array and battery respectively. PacBat is negative because the battery is charging. But what's Pac? I thought that it was the power output of the inverter (positive meaning output in this case). Why is it showing a negative value? Does this always correspond with the “calibrating (batteries - 1)” status?
My first thought, though, was that I didn't want to charge a nearly full battery from the grid. So I disconnected from the grid, and a couple of minutes later I had:
|
|||||||||||||||||||||||
Now Pac does resemble the power output to the house. The minor difference of 4 W is probably because the requests to the inverter are asynchronous and the status is thus not completely identical.
Still later the status changed:
Here once again the values are strange. Pac is back into normal positive territory, so my hypothesis that negative Pac corresponds with the “calibrating (batteries - 1)” status is clearly incorrect. But now W1 is negative. That doesn't mean that it's feeding power back into the grid: that's the PacGrid value, currently 0. And the change in W1 after the status change isn't reflected in any other value. On the other hand, the “dashboard” display continues to display marginally plausible values.
So what do these values mean? Still more thinking to do. About the only other insight is that the “calibrating...” status and the lack of input from the PV array are not completely related.
Also sent a mail message to Fyodor (since it looks like a software bug to me) and Tomas (since he's the supplier and may need to send somebody out). But in fact Fyodor said that he would be along tomorrow to take a look. Now that's service.
Also calls from Red Energy, my grid power supplier, asking me to allow them to pay me for energy I forward to the grid. Sometimes I wonder what the world is coming to: after all, I did apply for it. But at any rate they took my word for it. Now Powercor has 20 business days to reconfigure the meter, for which they will charge me some apparently random sum a little over $60, or about the equivalent of 400 kWh feedin.
|
Game of Thrones!
|
Topic: multimedia, opinion | Link here |
Everybody's talking about Game of Thrones lately. It would be easy to assume relieft that it's because it has just broadcast its final episode, but the world-wide admiration it evoked surprises me. IMDB gives it a score of 9.5, very high.
I took a look at it a while back and decided that it wasn't for us. But I had forgotten the details, so I thought we should have another look. Grabbed the first season and started watching this evening.
We managed about 8 minutes, by which time the second head had been sliced off in defiance of the laws of physics. What is it about US American films that are so horror-oriented and unrealistic? It doesn't say anything good about their society.
|
Sprinkler revival
|
Topic: Stones Road house, general | Link here |
Writing up yesterday's diary, went to check the messages from the netsprinkle program when it tried to contact the dead sprinkler relay. The simple answer: there weren't any, and it's not clear why not. It fires up a curl process, which is normally unnecessarily verbose, but here it said nothing at all.
The more complicated answer, though: it did say something. The relay is working again. What happened yesterday? It was clear beyond any reasonable doubt that the thing was non-responsive. Only sleeping? Pining for the fjords?
|
Random power issues
|
Topic: Stones Road house, general | Link here |
Apart from the issues with the solar inverter and the sprinkler, for some reason my office UPS was screaming on random occasions all morning. That's related to the Eaton UPS behind it, and the voltages alternately showed 224 V (apparently what it shows for 230 V) and round 245 V (which would thus be 252 V). Are we having problems with overvoltage? First disconnected the grid, which left the voltage at “224 V”, but the screaming continued. Why? What is that UPS trying to tell me? Disconnected the office UPS and connected it to a non-UPS power point, after which I had peace, if not enlightenment.
| Tuesday, 21 May 2019 | Dereel | Images for 21 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Finally the kitchen stoves
|
Topic: general, food and drink | Link here |
Mike Simpson (“Mike for odd jobs”, phone 0419 935 147) along today to finally install the new stoves that I bought 6 months ago. That all went much faster than I had expected, only just over an hour. The change is obvious:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
In some ways, discovered that “it's just what I asked for and not what I want”: the contrast between the gas cooker on the left and the induction cooker on the right is rather more than I had expected:
|
|||||||||||||||||||||||
And of course the minimum setting on the gas flames is too high, just like it was with the previous cooktop:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
In case that's not clear, these are the maximum and minimum settings.
Took a quick look at the “operation manual”, which described installation and adjustment of 5 different models, one of which looked somewhat like ours. They referred to “Fig. 8” for adjustments. Where's “Fig. 8”? Not in this “manual”. Mañana.
The interesting thing is that we now have space round the cooktops, for the first time in three years, and maybe since we moved in to Stones Road; after getting my first induction cooker, I've had one to the right of the cooktop area. And before that, at least for much of the year that had elapsed between moving in and getting the induction cooker, I had had a toaster oven there. The toaster oven is now in a cupboard; for the number of times I use it it doesn't seem worth having it take up worktop space.
|
Gas and plumbing: talk to the experts
|
Topic: general, opinion | Link here |
One of the issues in installing the new cooktops was that the gas hose was no longer long enough. Looking at the prospect of having to call in three different specialists (Mike, the joiner, for the carpentry, a plumber for the gas and an electrician for the electrics), I asked Yvonne (coincidentally in town) to drop in at The Haymarket in Ballarat and pick up a hose that could connect the stove to a gas bottle until the plumber came. I gave her the threaded protective cover of the connector as an indication of what we're looking for.
She called me from there and put me on to Cheryl, who told me that she couldn't do anything without knowing the thread. A plastic cover wasn't enough. I pointed out to her that the thing had a thread, but it seems that she neither checked nor believed me. In any case it would have to be a special order.
Told her to forget it, and off to tell Mike of my amazement. He was surprised, too: it's a perfectly standard ½" thread, just the same as the one for the old cooktop. He removed the old hose and confirmed that the other end, too, had a male ½" thread. Told me (indirectly Yvonne) to go to Ballarat Gas and Plumbing (for once a sensible name), where she found the item with no difficulty. Brought it back home, and Mike fitted it for us. No plumber needed. But I'm still amazed at the incompetence at The Haymarket. I won't forget that one in a hurry.
|
Firmware updates for solar electricity system
|
Topic: Stones Road house, technology, opinion | Link here |
Fyodor Torgovnikov along this morning and took a look at the solar electricity installation. My main concern had been that the inverter had stopped charging from the PV array, but his main concern was the battery. He had some reason: like all modern equipment that uses networks, it had to be its own Wi-Fi access point rather than join in to an existing network. I found an Android app that displayed network cells, but I lost my way through the maze of twisty little passages past the flood of advertisements, and couldn't find it again. What I did see was my own network, the inverter's private network (though it does exactly the same thing as http://inverter.lemis.com/) and the battery network—but only after Fyodor had power cycled the battery.
He installed new firmware both on the battery and in the inverter, and showed me ways through another maze of little twisty menus, the one that brought me to some limited configurability of the inverter.
In particular, I'm now able to set limits for at what SOC (state of charge) level to turn battery charge from the grid on and off. It had been on at 20% and off at 30%, but it seems that Fyodor changed it remotely some time in the last 24 hours. In the evening I discovered that it turned on and off at 20%, which effectively mean that it started charging when the charge dropped to 20% and turned off again when it hit 21%. That seemed even less sensible. Adjusted it to stop at 25%, but I think I need to start it at 19%, so it only ever comes on if there has been a grid outage that takes it below 20%.
One thing that did come out of of the discussion was that our issues in the last couple of days don't appear to be related to overvoltage: we barely hit the maximum of 253 V; Fyodor has seen others as high as 270 V.
|
Chicken filo “air fried”
|
Topic: food and drink, opinion | Link here |
Yvonne bought some chicken filo parcels recently. They're intended to be baked in a conventional oven (and nothing else, according to the label): 25-30 minutes at 180° in a normal oven and 160° in a fan-forced oven.
What settings do I use in an “air fryer”? Decided to set it as a conventional fan-forced oven and keep an eye on it. That proved to be a good idea:
|
|||||||||||||||||||||||
Apart from the fact that it leaked filling, the only issue was the food itself: about an appetizer portion, though Yvonne thought it was plenty. I sometimes wonder how she survives on the little that she eats.
| Wednesday, 22 May 2019 | Dereel | Images for 22 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Cooktop woes
|
Topic: food and drink, general, Stones Road house, opinion | Link here |
So how do I adjust the minimum flame on my new gas cooker? Back to RTFM, where I read, obviously added as an afterthought:
|
|
||||||||
OK, screw the screw tight. No change.
Damn these second-rate gas cookers! I should have tested it when it arrived 6 months ago, not after cutting holes into the work surface. But what choice do I have? I spent a lot of time in October comparing what's available, and this seemed to be about the only usable one. I can always use the old trick of setting the controls between “full” and “off”, so for the time being at any rate it stays, and I grumble.
And the induction cooker? That works, of course. The basic disadvantages remain: the controls are slow and barely visible—grey markings on black background, in the middle of the front, an area that's typically in shadow:
|
|||||||||||||||||||||||
But now it's under the range hood, so I can turn the light on:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I need to find a way to highlight the positions of the controls that won't wipe off.
|
How far to walk the dogs?
|
Topic: general, technology, opinion | Link here |
I've been walking the dogs further than previously; instead of the 600 odd metres to Bliss Road and back, we're going further afield. Today I was on my own, as Yvonne was in town, and I went to the end of Grassy Gully Road, a distance I estimated at 2.4 km. OK, try to process the GPS data:
|
|
|||||||||
Two problems: first, the obviously wrong track. I went up and down a road, so the two lines should overlap or at any rate be very close together. But this level of accuracy appears to be state of the art.
And then there are all those squiggles at the west end (in front of our house). They represent places I was that had nothing to do with walking the dogs. How do I get rid of them? I can't find a good site to help me, but I knew what time I left, and I know the data format:
The times are in UTC, 10 hours behind the local time. In this particular case I knew that I left shortly after 14:00, so I just needed to remove the trackpoints before this time, and also the ones after my return. Then I got:
|
|
|||||||||
Upload that to Wikiloc and I get:
|
|
|||||||||
That doesn't seem to be much of an improvement, but at least I now get details and even more details, which tell me that I had only just done 2 km. That must be because I stopped before the end (there are dogs in the property on the north side, and they overly excite our dogs). Now wouldn't it be nice for it to display things like speed and average speed? The whole thing took (only) 33 minutes, or a little under 4 km/h (mainly because the dogs were busy investigating the sides of the road).
| Thursday, 23 May 2019 | Dereel | Images for 23 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Understanding solar electricity charge transitions
|
Topic: Stones Road house, general | Link here |
I'm still trying to come to terms with the solar electricity inverter. As I've noted a number of times over the last month, it doesn't make sense to start charging from the grid when the battery charge level drops to 20%. So how about getting it to start at 19% and stop at 20%? That way it would only charge when it had been off the grid, which is what I would want.
But that doesn't work. There's no way to explicitly tell it when to start. There seem to be the following parameters:
So would it to make sense to set SOCmin or SOCrecx to 19%? Tried the former, without much success. It still switched in at 20%. And I'm concerned that if I do things wrong, I may end up with no power. In the end set the charge power to 50 W and the charge limit to 30%. The battery has a capacity of 6.4 kWh, so this means 640 Wh at 50 W, or 12.8 hours to charge.
That seemed to work, but when I came into the office this morning I had:
|
|
|||||||||
Oh. Charge was below 30%, so it looks like it was charging from both. In the database I had:
But why? Once it becomes feasible to charge from the PV array, it should stop charging from the grid, even if the limit hasn't been reached.
In passing, it's worth comparing these numbers. PacPV is 1276, matching “PV Generation” in the display. PacBat is -1331, matching “Battery charge from Grid” in the display. W1 is 2562, matching “Grid Consumption” in the display. But what's the relationship between the number in the middle and Pac? The number in the middle is the sum of PacPV and W1. Pac is PacPV + PacBat. The power supplied to the house is clearly PacPV + PacBat + W1, or 2507, a number that shows up nowhere.
While I was scratching my head, I got this:
|
|
|||||||||
Now PacPV, PacBat and W1 are as before, but Pac is the number in the middle + W1, and it matches the power actually being provided to the house. Again the “dashboard” doesn't show this value.
Still below 30%, but now it's discharging the battery! It took me a while to realize that we had had a state transition in between:
So that probably makes sense after all.
And then there's this:
|
|
|||||||||
Now PacPV, PacBat and W1 are as before. The number in the dashboard is the power being supplied to the house (for once!), and Pac is PacPV + PacBat.
The battery isn't fully charged, but it is only charging at a little over 500 W (from the PV array). It has enough power over to export the maximum permissible power to the grid. Why isn't it charging any faster? Some slow-down because it's nearly full? I have the feeling that I've seen it charging faster at this level.
Returning to the analysis of the various variables and displays: the values Pac and the number in the middle of the dashboard display are meaningless. There is no variable that returns the amount of power being consumed in the household, but that can be calculated: it's the sum of the values PacPV, PacBat and W1.
As I had suspected, the battery's Wi-Fi network has gone away again. Finally I have found the network display I was looking for yesterday:
|
|
|||||||||
From left to right we have the Marriott's network from next door, my own and the inverter. The inverter is so much stronger because I'm right next to it. But the battery is nowhere to be seen.
Apart from all that, there's the issue of how much charge. Basically I have two conflicting uses of the battery:
|
Fixing the gas cooker
|
Topic: food and drink, general, opinion | Link here |
Why is the “low” setting on the gas cooker so high? It was exactly the same problem with the last cooker, but I've never seen the issue before we came here.
Potentially there are three causes, by no means mutually exclusive:
The fact that the pressure regulator is the only thing in common makes it suspect. Out to take some photos:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Is the text embossed on the body intended to be read? It's very difficult, which is probably why it states “DO NOT ATTEMPT PREPAIR” [sic]. And “INSIST YOUR GAS CO. REGULARLY SERVICE REGULATOR” sounds like a non-starter, especially since there is none.
Even while I was taking the photos it occurred to me that they had already tested the gas pressure when I first complained about the issue. Is that enough reason to discount the cause? Somewhere I have an adjustable regulator. Maybe I should play around with that.
|
Where's my wine?
|
Topic: food and drink, general, opinion | Link here |
On Monday this week I finally managed to get my way through the Mclaren Vale Cellars web site and order and pay for some wine. At the time I had received a confirmation along with tracking numbers. And Australia Post told me that the numbers were invalid.
The next morning, ha ha, only joking, we have now received shipping information.
And today? So nice, so nice, we do it twice:
Shipping information approved by Australia Post, Tue 21 May • 9:21amShipping information received by Australia Post, Tue 21 May • 8:51am
No time zone, of course. We don't need no steenking time zones. But it looks as if the second entry (on top, of course could be Victorian time (UTC+10), while the first was South Australian time (UTC+9:30). Approved? Received? Believe that who may.
And that was all. Decided against calling them up and asking what was going on. Probably Australia Post is just confused again.
|
Revisiting the Big Oak
|
Topic: general, animals | Link here |
For a change, took the dogs for a walk to the „Dicke Eiche“ (“Big Oak”), really a clump of rather thin eucalypts to the south of our old house in Kleins Road:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Went quite some distance from the eagle's nest via Kleins Paddock:
|
|
|||||||||
Or at least it seemed quite a way. On return and processing, discovered that it was only 2.01 km, coincidentally 10 m shorter than yesterday's walk.
|
Air-fried ćevapčići
|
Topic: food and drink, opinion | Link here |
Ćevapčići for dinner tonight (really left-overs from making Bratwurst). In the past I had done this sort of thing in the toaster oven, but it sounded like an ideal match for the new “Air fryer”. After 10 minutes at 210° I had:
|
|||||||||||||||||||||||
That's actually quite good, and for once we didn't have lots of juice underneath the meat.
| Friday, 24 May 2019 | Dereel → Napoleons → Dereel | Images for 24 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Inverter configuration error?
|
Topic: Stones Road house, technology | Link here |
I've been playing a little with inverter settings. Today I set the “charge from grid” rate to 1000 W and a maximum charge of 40%, pretty much the opposite of what I did last time. When trying to save the configuration, I got a message: “Error writing config”, an indication of how modern this inverter is.
What went wrong? Hard to say. A retry worked.
|
Wine!
|
Topic: technology, food and drink, opinion | Link here |
Message from Australia Post today: my wine is waiting for me in Napoleons.
But yesterday it still hadn't been picked up. How did it come so fast? Clearly by net transfer. It still hasn't been picked up. The tracking information reads (upside down, of course):
Awaiting collection at NAPOLEON CPA
NAPOLEONS VIC Fri 24 May • 10:14am
In transit
WENDOUREE VIC
Fri 24 May • 5:53am
Unable to deliver - Item carded and transferred to post office for collection
WENDOUREE VIC
Fri 24 May • 5:52am
Shipping information approved by Australia Post
Tue 21 May • 9:21am
Shipping information received by Australia Post
Tue 21 May • 8:51amIs it really worth the trouble to maintain such superficial “tracking information”?
|
More clever error messages
|
Topic: technology, opinion | Link here |
While trying to create maps of yesterday's dog walk, got an unexpected error message:
|
|
|||||||||
Yes, that was a link from the same site. But on closer examination it proved to be a new way of saying “we can't find your input file”.
| Saturday, 25 May 2019 | Dereel | Images for 25 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Various botanical issues
|
Topic: gardening | Link here |
While taking the house photos today, discovered that the Alyogyne huegelii that I planted in the spring is finally flowering relatively profusely:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Why only now? Was it too hot before? Not enough watering?
On the other hand, the Hardenbergia planted just 2 metres away is looking really unhappy:
|
||||||||||||||||||||||||
Is it going to survive? And why the problem? It's native round here, and there's nothing special about what has happened to it recently.
|
More solar energy inverter information
|
Topic: Stones Road house, technology, general, opinion | Link here |
While Fyodor was here on Tuesday, he showed me some other functions of the solar electricity inverter web server: http://inverter.lemis.com/#localinverters/datalogger gives a tailored display of parameters, going back apparently to the installation:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
This one includes all sorts of parameters that I haven't been able to find elsewhere, including grid voltage and frequency, and even “Inverter. Phi Cosine” (what I call cos φ, but is generally referred to as Power factor). And at top right there's a little green box
|
|
|||||||||
OK, I'll bite. What's the format? It proves to be:
How about that, all the data together, one entry per minute. There must even be some way to import this directly into a MySQL table. Did a bit of snooping and discovered that it came from http://inverter.lemis.com/inverter/log/1/2019-05-25. And sure enough, I can pull it down like that:
Oh. That's JSON, not CSV. But again, I can convert it. With this information it's not clear that my current data (one entry per second) is of any use.
|
GPS tracking: worth the trouble?
|
Topic: technology, general, opinion | Link here |
While writing up my diary for yesterday, processed this map of walking the dogs:
|
|
|||||||||
What is that nonsense? The pointed “path” at bottom right would have taken us a good 100 m through Fiona Drayton's paddock. I've seen paths like this when the app is allowed to guess location based on the network, but I had explicitly removed any such points. And the real track is ridiculously inaccurate. Is it even worth it?
Today I got something only marginally better:
|
|
|||||||||
Would a better GPS receiver do better? I had suspected that trees might play a part in the problems, but the last one shows that the accuracy is no better when there's no tree cover.
| Sunday, 26 May 2019 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Which lens do I need?
|
Topic: photography, opinion | Link here |
The better modern (contemporary is a better word) cameras have interchangeable lenses. Why? Because of limitations in the lenses. Mobile phones are showing that it's not possible to get by with one lens: the top-of-the line models have up to 3 lenses, each with its own camera.
So what lenses do you need? After collecting 21 lenses for my current Micro Four Thirds system cameras, it's worth wondering what is really needed.
The interesting thing is that you can really get by with just the kit lens. But there are several others that make sense:
Fisheye lens. You can't replace a fisheye lens (typical horizontal angle of view 125°) with a rectilinear lens, where the widest have an angle of view of about 103°: rectilinear projection breaks down if you go much beyond 100°.
Ultra wide angle rectilinear. My Zuiko Digital 9-18 mm f/4-5.6 has a maximum horizontal angle of 88°, and that seems enough.
Since writing this article, I have bought a M.Zuiko DIGITAL ED 7-14 mm f/2.8 PRO
Normal focal length: kit lenses do this well, but I've found a lot of use for my M.Zuiko Digital IS PRO 12-100 mm f/4.
Telephoto (long focus) lens. The Leica Vario-Elmar 100-400 mm f/4-6.3 takes over where the 12-100 leaves off.
Macro lenses. I use the 60 mm f/2.8 and the 30 mm f/3.5.
Wide aperture. With the exception of the fisheye (only partially coincidentally) none of these lenses has a particularly wide aperture. To get one you have to forget zoom lenses, maybe (the Zuiko Digital 14-35 mm f/2 and the Zuiko Digital 35-100 mm f/2 are obvious exceptions). And suddenly the number of lenses needed increases. How many? I have 5: 15 mm f/1.7, 20 mm f/1.7, 25 mm f/1.4, 45 mm f/1.8 and 75 mm f/1.8. Do I need that many? I don't think so. I don't use any much.
Of course, there are many wider aperture lenses for Micro Four Thirds system, with apertures as wide as f/0.85. There are at least 6 with an aperture of f/0.95.
So in summary I use about 6 of my lenses on a regular basis (probably 7 including the 20 mm f/1.7 on the Olympus E-PM1). Is that enough? Add the five primes and it's 12, still a long way from the 21 that I have. And if I were starting again, what would I buy?
That's interesting because of the number of new systems coming out, notably the Canon EOS R and the Nikon Z7. Both have come out with new lens mounts, which puts them in the same position as Olympus and Panasonic ten years ago. For obvious reasons they chose the same approach: introduce a couple of good lenses and supply backward compatibility to the DSLR lenses.
But now they're rounding things out. It's interesting to see what they have on offer, and how it compares to the Micro Four Thirds offerings.
When the cameras were announced, they came with:
The Canon RF 35 mm f/1.8 IS Macro STM looks like a fairly standard wide-angle lens. It corresponds roughly to the M.Zuiko 17 mm f/1.8. Nothing much to write home about there.
The Canon RF 50 mm f/1.2 USM is clearly a different beast. It's been a while since I've seen a lens that wide on a Canon camera. Of course, it compares directly with the M.Zuiko 25 mm f/1.2 PRO.
Then there's the Canon RF 28-70 mm f/2L USM, again a surprisingly wide aperture. And it corresponds directly to the Zuiko Digital ED 14-35 mm f/2.0 SWD, at least optically. My 14-35 has the disadvantage of relatively short zoom range, and the weight: 900 g. So the Canon would have the same disadvantages, and then one: it weighs over 50% more, at 1.43 kg.
Finally (for Canon) there's the Canon RF 24-105 mm f/4 IS USM. That corresponds to 12-52.5 mm on μFT. For that, the f/4 aperture is quite boring: I already have the Leica DG Vario-Elmarit 12-60 mm f/2.8-4 (wider aperture, marginally longer zoom range) and the M.Zuiko Digital ED 12-100 mm f/4.0 IS PRO (considerably longer zoom range, same aperture). Nothing worth writing home about there, especially as Olympus also has the M.Zuiko 12-40 mm f/2.8 Pro in its range.
How many of my list above does this address? No fisheye lens. No ultra wide angle lens. No macro lens. No telephoto lens. Just items 3 and 6 of my list above (and no, I didn't check the Canon and Nikon offerings before writing that).
But there's more in the pipeline, though I can't find good links for them:
The RF 85 mm f/1.2L USM compares well with the Olympus 45 mm f/1.2. It's not available yet, though “Real Soon Now”. It will weigh 1.2 kg, compared to 410 g for the Olympus.
The RF 24-240 mm f/4-6.3 IS USM compares better to the M.Zuiko Digital ED 12-100 mm f/4.0 IS PRO, except that it has a smaller aperture.
The RF 15-35 mm f/2.8L IS USM looks like a good match for the M.ZUIKO DIGITAL ED 7-14 mm f/2.8 PRO.
And that's it. Nine months after the announcement, there is still no fisheye, still no macro lens, still no telephoto, still no lenses that would challenge the μFT lineup. Sure, there are lenses that can be used with an adapter, but I've seen what a difference that makes with the comparison between the Zuiko Digital ED 8 mm f/3.5 fisheye and the M.Zuiko Digital ED 8mm f/1.8 Fisheye PRO. Despite the 2 stop wider aperture, the M.Zuiko lens is significantly sharper than the older (DSLR) lens.
And Nikon? They started out more conservatively:with a 24-70 mm f/4 (only a quarter the aperture of the Canon!), a non-spectacular 50 mm f/1.8 and another 35 mm f/1.8. All look relatively boring, but they have released a roadmap for the next three years:
A 14-30 mm f/4 ultra wide angle, close enough to the Olympus 7-14 mm f/4 PRO. Another “Real Soon Now”.
A 20 mm f/1.8 lens. There's really nothing in the μFT lineup to quite match this one. There's the Olympus M.Zuiko Digital ED 8mm f/1.8 Fisheye PRO and the M.Zuiko 12 mm f/2.0, but they're not directly comparable. Still, close enough? Maybe not. But then there's the Voigtländer 10.5 mm f/0.95, which should give it a run for its money.
An 85 mm f/1.8, comparable to my M.Zuiko Digital 45 mm f/1.8. A good choice, but no fireworks.
A 24-70 mm f/2.8. Still not as spectacular as the Canon or Olympus lenses.
A 70-200 mm f/2.8, a stop slower than my corresponding Zuiko Digital ED 35-100 mm f/2.0.
A 58 mm f/0.95 “Noct”. This one is interesting. Olympus and Panasonic don't have anything comparable in their lineup. You can get the Voigtländer Nokton 25mm f/0.95, or the Meike 25 mm f/0.95, but they don't have autofocus.
Oh. Neither does the Nikon. I wonder why. Do such wide apertures present significant problems for autofocus? In any case, that's only f/0.95, while there are f/0.85 lenses for Micro Four Thirds.
I later discovered the reason: autofocus restricts the optical choices available to a lens designer. Normally they're not overly restrictive, but for really wide aperture lenses it's too restrictive.
So, once again, Nikon currently only covers points 3 and 6 of my list, with a promise of 2. Neither Nikon nor Canon seem to be interested in fisheye lenses, macros or telephoto lenses. Potentially their DSLR range of telephotos would be good enough, though it's not clear how well autofocus would work with them. But then, who wants to break their back? Here a photo that Kev Russell put up on the Facebook M43 Tech Talk group:
Those are both 300 mm equivalent lenses, and thus considerably shorter than my 100-400 mm tele. There can't be much of a market for the Canon.
| Monday, 27 May 2019 | Dereel | Images for 27 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Miserable weather
|
Topic: general, opinion | Link here |
Gradually autumn is coming to a close, and the weather is correspondingly unpleasant. Over the last couple of days we have had 24 mm of rain, and though the temperature wasn't overly low (minimum 3.7°), the wind and humidity have made it feel really unpleasant. And of course the solar electricity system suffered correspondingly: only 8.57 kWh of power produced, but a total of 76.67 kWh consumed (because of the heating), nearly 4 times as much as 4 days ago.
|
“Air fried” Bratwurst again
|
Topic: food and drink, opinion | Link here |
Bratwurst for dinner tonight, mainly to finish a Weißkrautsalat. Again in the hair dryer (the larger of the two “air fryers”). Last time I had decided that 12 minutes at 210° would be enough, but I have reconsidered: this time 14 minutes seemed right.
|
Canon R macro lens
|
Topic: photography, opinion | Link here |
Following on from yesterday's article on lens choices, I discover that I'm not completely right about saying that Canon doesn't have a macro lens: it seems that the Canon RF 35 mm f/1.8 IS Macro STM is a macro lens, as I should have noticed from the description. So Canon does have a macro lens for the R mount, though it's hard to think that that was its purpose: nowadays macro lenses have a relatively long focal length. My lenses have “full frame equivalent” focal lengths of between 60 mm and 120 mm. I don't agree with this trend, but 35 mm does seem to be very short. And it's only borderline macro: the smallest area it can focus on is 48x72 mm, compared to 10.4x13.8 for my M.Zuiko Digital ED 30 mm f/3.5 Macro . That's 25 times the area.
| Tuesday, 28 May 2019 | Dereel → Ballarat → Dereel | Images for 28 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Foo yong hai?
|
Topic: food and drink, opinion | Link here |
Yvonne has had some bad luck with some eggs she bought recently: of 12 eggs, 5 yolks broke, making them useless for fried eggs.
OK, omelette. But what kind? As a child I remembered a crabmeat omelette, foo yong hai. But is that the way you write it? “Foo” should clearly be “fu”. Yong? No idea. “Hai” sounds right, and I discovered that it means “crabmeat”, which makes sense. But the Wikipedia page is Egg foo young, clearly a reinterpretation. And where does that u come from in “young”? It includes many spellings, but always with “egg”, and none with “hai”. I've decided on “fu yong hai” until proof of the contrary.
Finding recipes is just as difficult. The dish is so popular outside East Asia that it has changed almost beyond recognition. Even recipes from East Asia look suspect, like this one from Rasa Malaysia. The name (“Malaysian taste”) looks authentic, but the recipe itself, along with the title “Egg Foo Young”, looks like it was collected somewhere in the USA. Even the description makes that clear. Bukan rasa malaysia saya. And then there's Olivia's kitchen with a recipe, also for “egg foo young”, that at least explains that it's fu yung hai. But the result (with peas and gravy!) doesn't look anything like what I recall:
Once again I came to the conclusion that it's not easy to find recipes on the web. In the end I found one from the volume I of Pei Mei's Chinese Cookbook that I bought in Mountain View on 13 September 1987. It, too, panders to US measures with unit abbreviations like C. (I assume cups). And the other quantities were correspondingly vague. What we ended up with was:
| quantity | ingredient | step | ||
| 280 g (4) | eggs | 1 | ||
| 50 g | fresh mushrooms | 1 | ||
| 170 g | canned crab meat | 1 | ||
| 50 g | cooked ham | 1 | ||
| 50 g | spring onions | 1 | ||
| 20 ml | water | 1 | ||
| 0.4 g | stock powder | 1 | ||
| 4 g | salt | 1 | ||
| 20 ml | soya sauce | 1 |
Preparation was simple—too simple: I forgot to cook the ingredients first. Just whipped up the eggs and added the other ingredients. Cooked considerably more than a French omelette, until the surface was mid-brown.
And the taste? The lack of pre-cooking wasn't an issue, but it was boring! The crab meat was completely lost. I might just as well have left out most of the ingredients. If I ever do make it again, I should use more eggs, maybe 6.
And the worst of all: after writing this article, I discovered that I already had a recipe, in which I noted pretty much the same things 6½ years ago, though on that occasion I used a different base recipe from “Makanan lazim Malaysia”, a Malay-language book which nevertheless wants peas and gravy.
|
A case for fertilizer
|
Topic: gardening, opinion | Link here |
Two months ago I planted some Epazote, and watched anxiously for it to germinate. It took its time, and it wasn't until six weeks later that I took these photos:
|
|||||||||||||||||||||||
Clearly there's not much difference between the two on the left (planted with a bit of soil cover) and the two on the right (planted on the surface). OK, what difference does fertilizer make? I put some fertilizer on the second and fourth (the smallest on these photos). Three weeks later I have:
|
|||||||||||||||||||||||
These are the two on the right. The one at the top is the second from the right. Clearly the fertilizer has made a big difference, and I've put some on the Kniphofia seedlings in the hope that they will recover (they're the ones on the floor in the first photo above).
|
Finances update
|
Topic: general, opinion | Link here |
Although I'm sad to see that the the Coalition is still in power, it does have one advantage: Bill Shorten's short-sighted tax reform plans, which would have cost me about $500 a month, are now off the table. But while discussing the matter with Peter O'Connell a couple of weeks ago, we decided that it was about time to get together again, so into town to talk to him, for the first time in nearly 3 years. It doesn't seem to have made much difference: at the time I noted:
Brexit has confused the markets
Plus ça change, plus c'est la même chose. But in general our finances are looking good, so we only did a minor tweak.
|
How to answer a mobile phone?
|
Topic: technology, opinion | Link here |
While I was talking to Peter O'Connell, my phone rang. Pulled it out of my pocket and somehow lost the “answer phone” display. How to answer the phone? Went to the phone app, but that just wanted to show me my contacts and a keypad option. I failed.
What is wrong with this system? Clearly there's a secret way to do this, but it evades me. No wonder they call them smart phones. From the Oxford English Dictionary (my reformatting):
smart: Mental pain or suffering; grief, sorrow, affliction; (sometimes) suffering of the nature of punishment or retribution. Also: an instance of this, a pang.
|
Pump finally ready
|
Topic: Stones Road house, general, opinion | Link here |
The phone call was from Yvonne to tell me that the Ballarat Pump Shop had called. After over 5 months, my pump has been repaired. Up there to find Bruce, who is apparently new and possibly something like a manager. Pump had been rewound, controller (which Jeff had said was OK when I brought it in) had been replaced. Bill: $591 odd.
But the pump was still under warranty when I brought it in, only 10½ months after I bought it. Oh. He did some searching for evidence of sale in his system, with which he didn't seem very familiar, and I told him I'd bring some evidence when I returned the loaner pump.
Back home and checked: yes, was booked with the information:
But look at the price! The repair bill was 10% more than I paid for the pump! Now to find the docket.
| Wednesday, 29 May 2019 | Dereel | Images for 29 May 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Answering Android calls: unsolved mystery
|
Topic: technology, opinion | Link here |
Yesterday's inability to answer my phone concerned me, of course, so today I tried to find out what I should have done. Spent some time, without success. What I have established is that, using Android 8.1:
If you pick your phone out of your shirt pocket, very carefully avoiding touching the screen, you get this display:
|
|
|||||||||
As it says, “swipe” up to answer. That works.
If the phone is active, you get a smaller display at the top:
|
|
|||||||||
|
|
|||||||||
This also seems to work.
If, however, you manage to press the wrong area on the surface, the displays go away. Now you have trouble: how do you answer it? There's an obvious solution that is both simple and elegant: select the phone app. Like so many simple and elegant solutions, it's unfortunately wrong. What I get is whatever was there before, like this:
|
|
|||||||||
Why doesn't it have an indication of an incoming call?
And so far I haven't found a way round this problem. Android has about 250 million users. Surely I'm not the only person who has this problem. But people on IRC weren't able to help me either.
What I did find was a workaround that might also have some advantages: this page describes an Android app called Easy Answer, carefully avoiding a link. It could be this one (for which they want money) or this one, or possibly there's another one of that name. Both of these suffer from surprisingly poor ratings, and the fine print “This app is compatible with some of your devices” doesn't sound encouraging either. And then there's the comment in the original article:
Easy Answer hasn’t been tested on Android 4.1 or higher versions. It may not work properly on these devices.
One of the many issues with Android is compatibility, and here it rears its ugly head again. Still, something to look at if I get bored.
|
Unhappy Hibiscus
|
Topic: gardening, opinion | Link here |
For whatever reason, I now have 4 different types of “Hibiscus” in the garden: two different Hibiscus rosa-sinensis, two of the same cultivar of Hibiscus syriacus, and an Alyogyne huegelii (“Native hibiscus”).
So far, one of the rosa-sinensis has been doing best. Here it was in the middle of last month:
|
||||||||||||||||||||||||
Not overly bushy, but plenty of flowers. And the others were hardly flowering at all.
But that has changed. Now the Alyogyne, the other rosa-sinensis and the syriacus are flowering:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And the first rosa-sinensis seems just to have died:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Why? I thought it might have been the recent rains, but it wasn't particularly wet there, and it should have dried out relatively quickly. Looking back, it seems that it was trying to tell me something two weeks ago:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
At the time I hadn't considered the yellow leaves to be a problem.
I wonder what it is? Something poisonous down below?
|
Greetings from Bendigo
|
Topic: general, opinion | Link here |
Call from the insurers of the Robinsons today, the people whose car I damaged in Bendigo on 5 March. Why that? We had agreed to keep the insurers out of it. He had called me in April to tell me that the repairs would cost $990, so I had reported it to my insurers and sent them not only a reply, but also a reminder.
OK, deal with the insurers. They sent me documentation, including a quote—dated 12 March—for $625! Somehow it's difficult to believe in their good will.
| Thursday, 30 May 2019 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More “air fryer” times
|
Topic: food and drink, opinion | Link here |
Ayam lemak for breakfast today, and with it the opportunity to cook no less than three items in the smaller “air fryer” and add the times to my times and temperatures list:
Both the peanuts and the ikan bilis came out of the device not completely crunchy: clearly there was residual moisture. And that's not surprising: the “air fryer” has relatively little air inside, and it doesn't get replaced, so in the course of cooking it must get much moister than air in a convection oven. After leaving the items to cool down, the ikan bilis became completely crunchy, but the peanuts didn't: thus my intention to cook them hotter next time. The chicken wing, on the other hand, was relatively dry.
In passing, it's interesting to note how slow this all is. The total cooking time was 52 minutes, not including preparation!
|
Finally I understand!
|
Topic: language, technology, opinion | Link here |
Mail from Andreas Leitgeb today, referring to Ken Thompson's T-shirt that I discussed a few weeks ago:
|
|||||||||||||||||||||||
Andreas writes:
I know a little Greek - enough to understand the caption as "Finally I understood this".
Fwiw, Googles translation is quite a bit off. "ΤΟ" is the Greek article "the", but without a noun following it, "this" would be a better translation. The next word is in 1st person singular, past tense for "understand". Google apparently utterly failed to recognize 1st person.
No surprise about the criticism of Google Translate, which still seems to think that syntax is irrelevant.
| Friday, 31 May 2019 | Dereel | Images for 31 May 2019 |
| Top of page | ||
| previous day |
|
Phone smarts
|
Topic: technology, general, opinion | Link here |
Mail from Tim Bishop and Oliver Musch today about my problems answering an Android phone. Tim noted from the screen shot that a call was pending. All I saw was:
|
|
|
|||||||||
But what I should have seen was in the top left-hand corner:
|
|
|||||||||
The lightning deflecting off the old-fashioned handset clearly means that a call is incoming. As Tim says:
Pretty sure if you swipe down from the top of the screen there'll be
something in the notification area that'll let you answer the call.
There's a phone symbol along the top of the screen that suggests there's
a call-related notification.
Oliver is using Android 9, which for some reason appears to work the same way as my Android 8.1. He writes:
wenn das Gespräch noch ansteht (Hörer-Symbol oben in der Leiste), kannst du
bei mir mit Android9 auf ZTE einfach die Übersicht von oben aufswypen und dort
auf "Annehmen" drücken.
After savouring the New High German word „aufswypen“, the question is: which direction? I would have thought that what Tim describes would translate as „abswypen“. But it didn't help: I couldn't reproduce the screen. It's frustrating enough that I didn't try very long, but next time I'll know what to try.
As if that wasn't enough, Petra Gietz was here and asked if we had mobile phone coverage. Yes, of course, but why? Ah, it doesn't work where she is in Enfield.
That didn't sound right, but sure enough, even here her phone (an iPhone 6) was displaying “SOS calls only”. Why? Out of credit? That shouldn't stop incoming calls, but when I tried from my phone (both using the same network), I got voice mail.
Dammit, where are the settings menus for iPhones? Finally found them, but there was nothing obvious. OK, remove SIM card, put in my spare phone. Yes, she's out of credit, but I can call the number. So why does Apple not allow incoming calls under these circumstances?
Why, does Apple not allow incoming calls under these circumstances? Maybe a problem with the SIM card contacts? Put it back in again, and it worked: incoming calls OK, no outgoing calls until recharge. And then Petra told me that she had dropped the phone earlier in the week. Presumably (but surprisingly) that had caused some mechanical issue with the SIM card.
So why did this take so long? Nothing on the phone display or in the settings menu reached out and said “Can't find a SIM card, you idiot”. It's far too polite for that.
Wouldn't good error messages be such a good idea? I wonder why nobody has thought of it.
|
Power fail?
|
Topic: Stones Road house, general, opinion | Link here |
It's been 6 weeks since we installed the solar electricity system, and in that time—I'd like to say “of course”—we have had no power failures. But what about Powercor? I've already established that the system doesn't tell me when the grid goes away. I'll have to ask the database for that:
What does that mean? Were they power outages? There's no “Off-grid” status there, but this 60 second (configured) delay in “Waiting to connect” suggests that there might have been no fewer than three short (1 second odd) outages on Wednesday.
|
ON1: OFF
|
Topic: photography, technology, opinion | Link here |
It's been a while since I installed ON1 Photo RAW 2019.5, which gave me headaches with claimed missing OpenGL 3.3, but I had established that I could start it if the display was the native monitor rather than my standard rdesktop. OK, the free trial is running away, so time to see how it works.
Surprise, surprise. It wouldn't start on dischord even on the native display. So I had to start it on euroa, the laptop.
One of the nice things about ON1 is that it comes with lots of videos, and I had looked at the basic ones. But even so, I couldn't work my way round the menu system. Click on an image in the preview pane. Nothing happens. Click again. Oh yes, it does, just taking its time: it crashed. Various other attempts had the same effect: as soon as I wanted to look at an image in detail, it crashed.
Why? Presumably because it really wants to use the graphics card processor, and can't work out how to do that in conjunction with rdesktop. OK, ON1, you're now OFF2 (OFF1 was 18 months ago). I suppose I should suggest to them that they get the thing to work over rdesktop.
|
Powercor consumption information
|
Topic: general, opinion | Link here |
Petra Gietz is now also keeping a careful eye on her power consumption. And she discovered this:
|
|
|||||||||
At 16:30 (“PM”, according to the Powercor web programmer), she used 1.9 kWh. That's not 1.9 kW; it's an average of 3.8 kW in the half hour between 16:30 and 17:00. And nobody was home! What happened? My best bet at this stage is that it was plumbers doing some work on the sewage pumps that they need in Enfield. But that's still an amazing amount of power.
|
Wotan unwell
|
Topic: animals, opinion | Link here |
Five years ago Yvonne had to give up her favourite horse, Wotan (and not Óðinn) because of back problems. Since then he has had a less than stellar time. First he went to a horse sanctuary, where they changed his name to Humphrey. That didn't work out either, and three years ago we took him back. Then he went to Julie Donaghy, who renamed him Bob.
And then a week or so ago we heard that Julie couldn't keep him. So we're planning to bring him back again just as soon as Chris Bahlo can fit it into her schedule. But then this afternoon we heard from Julie: Wotan lying down, looking sick, won't get up.
Damn! Yvonne over to take a look.
|
|||||||||||||||||||||
|
|||||||||||||||||||||
|
|||||||||||||||||||||
After some discussion, a vet from the Ballarat Veterinary Practice came along and diagnosed colic, though by that time he was clearly better. It looks as if he will survive.
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 |
