|
| Greg's diary |
| October 2024 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Tuesday, 1 October 2024 | Dereel | Images for 1 October 2024 |
| Top of page | ||
| next day | ||
| last day |
|
The NBN outage from hell
|
Topic: technology, general, opinion | Link here |
I was up relatively early this morning, in time to see our network link drop just before 9:00. Damn! That must be one of many National Broadband Network outages threatened for the end of last month and most of this month. From the Aussie Broadband site, trimmed and sorted:
| From | To | Duration (hours) | Total duration | |||
| 23-09-2024 06:00:00AM | 01-10-2024 06:00:00PM | 12:00 | 27:00 | |||
| 02-10-2024 06:00:00AM | 11-10-2024 06:00:00PM | 12:00 | 10:00 | |||
| 08-10-2024 12:00:00AM | 08-10-2024 06:00:00AM | 6:00 | 0:15 | |||
| 15-10-2024 06:00:00AM | 24-10-2024 06:00:00PM | 12:00 | 10:00 | |||
Admire the short outage on 8 October overlapping with a much longer one between 2 and 11 October. Potentially they have forgotten to mention that the latter is only in the daytime, where it can do the most harm. But the only real issue is how long it is going to last. What does the Aussie Broadband app say?
|
|||||||||||||||||||||||
No outage? Can that be right? The lights on the NTD showed a red ODU status, suggesting that something was wrong, along with an orange status. OK, we're not connected anyway. Power cycle. Comes back with the same status.
OK, the Aussie app offers me tests for my connection. Clearly I'm down, but it might say why.
|
|||||||||||||||||||||||
Now isn't that helpful? But then there's a loopback test, whatever that does; it doesn't sound as if it would work here, and it doesn't:
|
|||||||||||||||||||||||
OK, to be sure that it's really an NBN outage, off to the tower to see if anybody was working there, like 5 years ago. Off with a couple of long telephoto lenses, but there was nothing to be seen.
Report a fault? Sure. But first it needs to “analyse” the results:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Note the times (top left). After an hour I decided that it wasn't going to do anything.
While all this was happening, a chat person, Ryan, asked if he could help. On a mobile phone keyboard! I asked him to call me, but he can't do that. OK, are they aware of my outage? OK, full name, address and date of birth please! You'd think that they were trying to insult me. To access the app I already had to provide biometric authentication. And there's no reason to believe that he could help. But no, I was wasting his time. End of chat.
What about the NBN themselves? They now have a network status page, though I had to search for it with Google, and it took forever to load, over 10 minutes. My guess is that this was due to mobile phone system overload. And finally, yes,
|
|||||||||||||||||||||||
Admire the unsubstituted metavariable {{maintenanceEndTime}}. Can't any site provide useful information? But at least they provide the information to anybody who wants to know it, without requiring full name, date of birth or sexual orientation.
OK, what's in store? Lots of screen shots, but after the link came back up I was able to look at it marginally more sanely:
| Wed 2 Oct | An nbn network outage will take place after {{maintenanceStartTime}}. Your service is estimated to be down for a total of {{maintenanceDuration}}. | |
| Thu 3 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Fri 4 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Sat 5 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Sun 6 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Mon 7 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Tue 8 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| An nbn network outage will take place between {{maintenanceStartTime}} and {{maintenanceEndTime}}. Your service is estimated to be down for a total of {{maintenanceDuration}}. | ||
| Wed 9 Oct | An nbn network outage will take place today. Your service might be impacted. | |
| Thu 10 Oct | An nbn network outage will take place today. Your service might be impacted. | |
And yes, for 8 October they had two entries.
What does this mean? I think that it's very likely that we'll have more pain tomorrow and next Tuesday; the others might just be sabre-rattling.
The link finally started coming up round 17:50, when the signal strength LEDs on the NTD lit up. I didn't get connectivity until 18:02, after 32778 seconds (9 hours, 6 minutes, 18 seconds), the longest outage I have ever had. And they're threatening more in the next couple of days!
In passing, had more trouble with this horrible My Aussie app. I had to log in all over again, and it didn't let me log in with a thumbprint any more. Did Ryan disable it because I refused to be insulted? Looked once again for instructions on how to enable it, and failed once again. OK, Microsoft solution: remove the app and reinstall it, something that took forever on this overloaded mobile phone network. Still no way to enable thumbprints. But after about an hour it came back with it enabled, with no further action on my part.
|
Old telephotos again
|
Topic: photography, opinion | Link here |
When I went to the National Broadband Network tower, I took two long telephoto lenses: the Leica DG Vario-Elmar 100-400 mm f/4.0-6.3 and my old Hanimex 300 mm f/5.5 with a 3x teleconverter, giving a focal length of 900 mm. Was the latter really worth the trouble? When I got back home I tried some comparisons. The first issue with the Hanimex was focusing: it's almost impossible hand-held. Even on a tripod it was very difficult, especially as focus peaking didn't want to work. Finally I got comparison photos of the letter box about 70 m from the house entrance:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Somehow even the thumbnail of the Hanimex doesn't seem as sharp. But a crop of the lock makes things much more obvious (I can't say “clearer”):
|
|
|||||||||
|
|
|||||||||
That's at full aperture. The image quality at f/48 is marginally better for the Hanimex, though the chromatic aberration remains extreme, but at corresponding f/16 it degrades for the Vario-Elmar, presumably because of diffraction. Clearly the Hanimex isn't worth the trouble. And of course I came to this conclusion years ago, though this time it's more obvious.
|
Mobile hotspots
|
Topic: technology, language, opinion | Link here |
The National Broadband Network may be down, but I'm not completely off the net: I still have my mobile phone.
That's a pain, of course, but it could be better than nothing. And since it runs a castrated Linux, it should also be capable of routing IP, assuming that they haven't castrated that component too.
But there's this silly term “hotspot”. What does that mean? Traditionally a hot spot has been something, well, hot, generally used to mean a place where there's a lot of activity. And somehow that has developed into designating an area where there's a lot of radiation. From there, people with limited vision have applied it to a mobile networking access point.
OK, an access point is a kind of router: it connects other systems together. And my phone has a “hotspot” function. So I enabled it, I think. Setup seems to have been a choice of “Network name” (by which I think they mean SSID, a term that also shows up on that screen), and the kind of encryption.
And now? I know the address of the phone, because it gets it from DHCP on eureka. But setting a route to it does nothing; I don't even get any ICMP messages. But then there's a list of “connected devices”. What's that about? Have they managed to break the protocols that we have known and loved for decades? In any case, I wasn't able to get it to work.
But then there are other functions, such as “tethering”:
|
|||||||||||||||||||||||
This one is particularly confusing. I think that it means using the phone as a router, but then what's the “hotspot”? And how can it offer Ethernet tethering if it doesn't have an Ethernet interface? And why should it be dependent on the link level hardware?
| Wednesday, 2 October 2024 | Dereel | Images for 2 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Another day without a net
|
Topic: technology, general, opinion | Link here |
Up early this morning just in case the National Broadband Network decided to take the net link down again. A good thing, too: it went down earlier and stayed down longer:
| Duration | from | to | ||||
| (seconds) | ||||||
| 32778 | 1 October 2024 08:56:29 | 1 October 2024 18:02:47 | ||||
| 36709 | 2 October 2024 08:46:42 | 2 October 2024 18:58:31 | ||||
That's two entire working days! I'm putting up with it better now, but it still drives me crazy.
|
Still more NBN outages!
|
Topic: technology, opinion | Link here |
Mail from Aussie Broadband today, which of course didn't reach me until the evening:
Date: Wed, 02 Oct 2024 15:31:09 +1000
Subject: NBN scheduled maintenance
NBN are planning network maintenance between **Thu 31st October 2024 06:00 AEDT** and **Sat 9th November 2024 18:00 AEDT**.
As part of a national rollout to upgrade the Fixed Wireless network, nbn® will be replacing hardware in your area. The program will future-proof sufficient capacity for new and existing customers on the network. As tower climbing is required, this work must be performed during the day. You will experience two interruptions of up to 12 hours. Some towers may have up to three further interruptions to remediate any faults from the upgrade works.
You'd think that they enjoy torturing their customers. But there was also a bit of light relief: an unplanned outage reported at 18:52 and reported as completed at 19:26. Nothing to be seen of it here. It seems that you can't even rely on unplanned outages.
And tomorrow? The NBN can't make up its mind:
|
|||||||||||||||||||||||
At 21:38 they tell me that there will be an outage after 6:00 for about 10 hours. What, again? No, they're referring to the same day. Tomorrow there's just a “your service might be impacted”, which I interpret as meaning that nothing specific is planned. But in passing it seems that this page has been able to substitute values for {{maintenanceStartTime}} and {{maintenanceDuration}}. Why not on mobile phones?
|
Tethering and hotspots revisited
|
Topic: technology, general, opinion | Link here |
This almost total non-availability of the National Broadband Network makes it all the more urgent to find an alternative. At least in the short term the mobile telephone network seems the obvious solution. Before the net failed, tried a query with Google Gemini, which told me the obvious about configuring the phone, and then
Now, other devices like your laptop, tablet, or another phone can connect to your Xiaomi's Wi-Fi hotspot using the provided SSID and password.
Somewhere there it mentioned USB. Why USB? This thing has a network interface. But it seems There Can Only Be One. And how you connect to an IP network using USB is so obvious that they don't mention it.
But it's not clear that it makes much difference. Network performance on the phone was so terrible that I had to wait for minutes for any reaction. One of the apps on the phone includes a speed test, so I tried that.
|
|||||||||||||||||||||||
That doesn't look very fast, but I don't know what to expect. It's interesting that the upload speed is much faster than the download speed. That suggests overload. It's also interesting that this glacial speed is considerably faster than any dialup has ever been, so the idea of using dialup for fallback is clearly not worth pursuing.
|
Investigating makefs
|
Topic: technology | Link here |
The lack of a network made it difficult to ignore the bug report about makefs that I have been dragging behind me for the past 12 months. Spent some time looking at it, and came up with some more issues with the man page. One of the big differences between makefs on NetBSD and FreeBSD is the file system block size: 8,192 bytes on NetBSD and 32,768 bytes on FreeBSD. There's a way to set the block size, though the man page doesn't make it easy to find out. I think that this should do the trick:
kimchi# makefs -d -1 -o b=32k usrbin-32k /hydra/sbin
But the result is gibberish. None of the utilities recognize it as a file system. There's
a whole lot more to investigate, and clearly some of the defaults need changing: by default
it generates a UFS FFS type 1 file system, something that has been almost obsolete for decades. Now
wouldn't it be nice to have a network to do some investigations?
|
Telephoto check again
|
Topic: photography, opinion | Link here |
Yesterday's comparison of the Leica DG Vario-Elmar 100-400 mm f/4.0-6.3 and the Hanimex 300 mm f/5.5 wasn't completely fair to the Hanimex: it included a teleconverter, which could considerably degrade the image. So today I tried a couple of photos without the teleconverter, at f/5.5 (maximum aperture) and f/16. Here the results:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Again, the lock, this time comparing 3 ways: first, the Vario-Elmar, then the Hanimex without a teleconverter, then with the teleconverter. Full aperture:
|
|
|
|||||||||
|
|
|||||||||
|
|
|
|||||||||
And then stopped down to f/16 (f/48 for the teleconverter):
|
|
|||||||||
|
|
|||||||||
|
|
|||||||||
The most obvious thing is that the teleconverter doesn't add any value: the blown up image is worse than without the teleconverter. But it's still much worse than the Vario-Elmar, especially with regard to chromatic aberration. About the only thing of interest is that the shots at (real) f/16 are almost comparable.
|
TV fail
|
Topic: multimedia, technology | Link here |
No news from the net today, of course, so like yesterday I tried to watch TV news. But today the TV claimed that the antenna wasn't connected. Checked everything, and the only things that could have gone wrong were the antenna cable and the TV itself. I thought I heard something on the wrong channel before I changed it. Could it be a problem with the TV? I don't really want to know.
|
KardiaMobile 6L
|
Topic: health, technology, opinion | Link here |
The KardiaMobile 6L that I ordered on Monday promised 2 day delivery, but it wasn't sent until yesterday evening. What chance that it would get here today? But it did. Tried to use it, but I couldn't download the software.
|
|||||||||||||||||||||||
Admire the refusal to take a screen shot. I did have marginal network access. Is this a device for fair weather networking?
About the only thing of interest was the device itself:
Which way round? Clearly the V shows the direction. Oh, but it's not a V, it's a Λ, which they seem to think is an Α, so you should hold it the other way round in order not to get erroneous readings. What a clever idea!
| Thursday, 3 October 2024 | Dereel | Images for 3 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
One outage? Three!
|
Topic: technology, general, opinion | Link here |
After complete lack of service during the past two working days, I was hoping for something better today. And maybe I got it. The net link stayed up until 9:41. But then it went down again for nearly 4 hours. Still, better than the last couple of days.
But then it went down again! Admittedly only for 13 minutes, but it had just come back again when it went down for a third time! So far this week we have had
| Duration | from | to | ||
| (seconds) | ||||
| 32778 | 1 October 2024 08:56:29 | 1 October 2024 18:02:47 | ||
| 36709 | 2 October 2024 08:46:42 | 2 October 2024 18:58:31 | ||
| 13329 | 3 October 2024 09:41:06 | 3 October 2024 13:23:15 | ||
| 771 | 3 October 2024 15:28:22 | 3 October 2024 15:41:13 | ||
| 135 | 3 October 2024 16:01:17 | 3 October 2024 16:03:32 | ||
That's a total of 83722 seconds or 23¼ hours. Now they're talking about only 15 minutes on Tuesday. Will that make a total of over 24 hours?
|
Understanding Aussie Broadband
|
Topic: technology, opinion | Link here |
Yesterday I received no support from Aussie Broadband, but after the net came up again I received an email:
Date: Tue, 1 Oct 2024 13:51:12 +1000
From: Aussie Support <support@aussiebroadband.com.au>
Subject: Re: [Ticket#2024100110021591]
Thanks for raising a query with Aussie Broadband.
If the query is urgent, we suggest giving our friendly Australian team a call on 1300 880 905 (8am to midnight Australian Eastern Time).
OK, Aussie, I'll bite. What query? Sent off a reply and got the surprising response:
Date: Thu, 3 Oct 2024 16:04:35 +1000
From: Aussie Support <support@aussiebroadband.com.au>
Subject: Re: [Ticket#2024100110021591]
Hi there and thanks for reaching out. Unfortunately I cannot find an account
with the details you have provided. Please give us a call so we can check your
particular area and service for any issues.
Huh? That's your ticket. Why can't you find anything? Another response, but I don't know why I bother.
While trying to investigate the eternal outages, got this:
|
|||||||||||||||||||||||
That proved to be Aussie's way of saying “no network connection”.
|
ΛliveCor KardiaMobile software
|
Topic: health, technology, opinion | Link here |
So now we have a network connection, so I can try again to install the app for the KardiaMobile 6L. And yes, after offering a lot of unrelated stuff and they typical Android pain, it installed and asked me to sign up for some service that would end up costing more than the device itself. Ha ha, only joking. You don't have to sign up.
A bit of playing around, and how about that, I was able to generate an electrocardiogram that gave me the reassurance that it had detected nothing wrong with my heart. And yes, I can save it as a PDF—somewhere. Where is a question for another day.
|
Mobile phone speed test
|
Topic: technology | Link here |
Yesterday's mobile phone speed test showed particularly poor performance. About the only surprise was that the upload speed was much higher than the download speed (1.2 Mb/s and 182 kB/s). What's it like under normal conditions? Tried again today:
|
|
|||||||||
|
|
|||||||||
Not only comparable to the National Broadband Network link: the upload speed is greater than the 5 Mb/s limit on my NBN link. Yes, that's artificial, since I could easily go higher, but it's clear that the speed is sufficient—if it were available when the NBN link is down.
| Friday, 4 October 2024 | Dereel | Images for 4 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
A network!
|
Topic: technology, general, opinion | Link here |
Finally National Broadband Network is back to normal. Somehow it seems to have disrupted the entire week.
|
Mobile phone speed test revisited
|
Topic: technology | Link here |
So were yesterday's mobile phone speed tests typical? Tried again today:
|
|
|||||||||
Still comparable with the NBN, but it's puzzling that the upload speed is so much higher than the download speed.
|
Real-life heart monitoring
|
Topic: health, technology, opinion | Link here |
Finally got round to using the KardiaMobile 6L in earnest, on Yvonne. It works! And the good news is that it didn't detect any problems.
And now? Display the ECG? First you need to “download” it and choose not to encrypt it! What kind of nonsense is that? All you get is a PDF, but I knew that. At least we were able to display it. How can I back it up? I spent most of the afternoon trying to find out, but failed.
|
Android copy, yet again
|
Topic: technology, opinion | Link here |
Why couldn't I save Yvonne's electrocardiogram? All I need to do is copy it from her phone to a Real Computer. What are the options?
So: connect the phone to distress via USB. No files! Not even the ones that the other apps find! Somebody has wilfully destroyed the access method!
OK, Gemini, How do I access an Android file system via USB from Windows?. Five steps, including enabling USB debugging! And for that you need to be a “developer”. The instructions didn't work for me. how do I set developer options on Xiaomi HyperOS? (“HyperOS” is the way Xiaomi spell “Android”). Ah, for that there's a different solution, and Gemini still doesn't quite get it right.
So, enable USB debugging. Microsoft no longer even sees the phone! WHY do modern “devices” make it so impossibly difficult to perform probably the most basic of all manipulations?
More investigations. I've never seriously wanted to root a mobile phone, but if it's the only way to access data on the device, I may have to choose it. And how about that, there are instructions that include an official guide for rooting my Redmi 9T.
But why do makers do this? It looks like an attempt to lock people in to their services. And in this case, it's my (OK, Yvonne's) personal data. There must be laws against that.
|
Finally! CAPTCHAs decoded by AI
|
Topic: technology, general, opinion | Link here |
It's no secret that I hate CAPTCHAs with a passion. So of course I'm delighted to hear that Andreas Plesner of the ETH has cracked CAPTCHAs with an AI program. It's interesting that it's best at recognizing (presumably US) fire hydrants: in Switzerland they're almost certainly the same as in Germany.
In passing, it's interesting to note that this it's ReCAPTCHA version 2 that is so painful. Version 3 is invisible to the user, so I don't have an issue with it.
| Saturday, 5 October 2024 | Dereel | Images for 5 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Finding my personal data
|
Topic: health, technology, opinion | Link here |
How do I find Yvonne's ECG? KardiaMobile and Android don't help at all. But I found an unlikely ally: firefox. “Download file”. After fighting my way through the Kardia menus to “display PDF”, I chose firefox to display it, and then I could “Download” it (again). And it put it in the /Downloads directory at the top of the directory tree shown by WiFi File Transfer.
=== grog@eureka (/dev/pts/7) ~ 38 -> PHONE=fossil fh
Connected to fossil.lemis.com.
ftp> ls /Download
drwx------ 3 user group 0 Oct 1 18:40 MiuiFastConnect
drwx------ 3 user group 0 Jun 24 2022 downloaded_rom
-rw------- 1 user group 457209 Oct 3 15:08 ecg-20241003-150659.pdf
-rw------- 1 user group 460875 Oct 5 11:07 ecg-20241004-092324.pdf
-rw------- 1 user group 461878 Oct 6 12:38 ecg-20241006-113438.pdf
So I have worked around this particular problem, though I have this nagging feeling that they have personal data on their servers to which we don't have access.
|
Copying files: the pain
|
Topic: technology, opinion | Link here |
Spent a lot more time today trying to perform that most basic of operations, copying files, on Android. In addition to the apps that I mentioned yesterday, came across two more:
So no progress on that front. Let's try to root the Redmi 9T. Three simple steps:
Step 1. Download the Mi Unlock app to PC, and sign in with your Mi Account;
Step 2. Shut down your phone manually, and hold Volume down key and Power button to enter Fastboot mode;
Step 3. Connect your phone to PC using USB cable and click "Unlock".
Oh. “Sign in with your Mi account”. What Mi account? First I had to create one. But why? Because it needs an app! Why, again? Into the valley of death, and then “click “Unlock””. Where?
Fool! The real instructions are much longer, and include the requirement that...
You have already inserted a SIM Card into your device and can connect it to the internet using mobile data.
Well, no, this phone doesn't have a SIM card. So that's dead in the water. Only later did I read
Click on Unlock » Unlock anyway. On your first attempt, Mi Unlock will flash the message Couldn’t unlock. Please unlock 168 hours later. Follow the timer / waiting period to unlock successfully.
A whole week! Clearly they're trying to annoy people.
OK, what else can I do? Looking through the amazing number of developer options, found (three levels deep) a setting “Default USB Configuration”, which produced a display reminiscent of some cameras:
Can't open /home/grog/www.lemis.com/grog/Photos/20241006/photolist.php: fopen(/home/grog/www.lemis.com/grog/Photos/20241006/photolist.php): failed to open stream: No such file or directory
|
|
||||||||
OK, that's worth a try. Set to “File Transfer” and connect to hydra:
Oct 5 14:21:29 hydra kernel: ugen0.3: <Xiaomi Redmi 9T> at usbus0
Oh. No associated disk device? My camera produces (inter alia)
Oct 5 11:52:47 hydra kernel: ugen0.3: <OLYMPUS E-M1MarkII> at usbus0
Oct 5 11:52:47 hydra kernel: umass3: <OLYMPUS E-M1MarkII, class 0/0, rev 3.10/1.00, addr 21> on usbus0
Oct 5 11:52:47 hydra kernel: da3 at umass-sim3 bus 3 scbus11 target 0 lun 0
Oct 5 11:52:47 hydra kernel: da3: <OLYMPUS E-M1MarkII 1.00> Removable Direct Access SCSI-2 device
Oct 5 11:52:47 hydra kernel: da3: 29935MB (61306880 512 byte sectors)
O
Clearly there's not enough information to make it recognizable as a disk device. Still more searching brought me to the recognition that I need to “unlock” the phone before it can work, something that the config page tastefully hid. And there we have:
|
|||||||||||||||||||||||
That looks like it's taking me back to the previous pain. How I hate this stuff!
On a more positive note, Juha Kupiainen has been playing with using his phone as a “mobile hotspot. It looks like he might have a winner:
Oct 5 15:38:21 dv5 kernel: urndis0: <motorola moto g(10), class 0/0, rev 2.00/4.19, addr 2> on usbus2 [15:40]
Oct 5 15:38:21 dv5 kernel: ue0: <USB Ethernet> on urndis0
Oct 5 15:38:21 dv5 kernel: ue0: Ethernet address: 1a:82:45:10:8f:c5
I suppose I should try that.
|
Roast chicken again
|
Topic: food and drink | Link here |
Roast chicken again today, something that we don't eat very often. The cooking times stated to keep the breast covered in aluminium foil. But that's not the way to brown it, so I removed the foil when the meat temperature was 60°. Somehow that fell flat: the skin was still pale, but the meat (cooked to a breast temperature of 78°, which after cooking rose to about 81° before sinking again) was overcooked. Next time I should remove the foil at the end and grill it just before removing from the oven.
|
Carlotta: colic?
|
Topic: animals, health | Link here |
Finally the new season's grass is growing, and today Yvonne put the horses in the “sewage paddock”, the paddock where the septic tank drainage is located. The grass was good, and at the end Carlotta lay down, contented. My comment to Yvonne during dinner: “Hopefully she doesn't have colic”.
After dinner, out to take a look. She way lying down again! Off to look more carefully, and she got up and started eating. She had stomach sounds, and she also farted, none of which is usual when a horse has colic. But she was shivering, and Yvonne thought that she might have got a chill when it started to rain: her rug was wet through. So she set off to get a new rug, and I called up Chris Bahlo, who agreed with our assessment.
Another check before going to bed. All seems well.
| Sunday, 6 October 2024 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
syslog pain
|
Topic: technology, opinion | Link here |
I have a display of eureka's most common system log files on hydra:0.0: /var/log/messages and /var/log/maillog. And other systems are configured to log to eureka as well. But lately it occurred to me that I wasn't seeing any log messages from other systems any more.
More investigation showed that syslogd was running with the -s option (Do not log messages from remote machines). OK, fix that.
Oct 6 09:14:29 eureka Forwarded from homephone: HT802 [c0:74:ad:37:66:d8] [1.0.21.4]CallRecord::writeCDRFile, No space! current file size =51158bytes, need extra 86 bytes.
Oct 6 09:14:29 eureka Forwarded from homephone: HT802 [c0:74:ad:37:66:d8] [1.0.21.4]CallRecord::writeCDRFile, No space! current file size =51158bytes, need extra 86 bytes.
It seems that every log message from another system repeated endlessly. I've seen that before. It seems that I inadvertently “fixed” it by disabling network logging. But why is this happening now?
|
Subtitle display pains
|
Topic: multimedia, technology, opinion | Link here |
We watch videos with subtitles where available, but for some reason there are multiple problems with the representation. Yesterday we were watching „Der Bergdoktor“, and the subtitles stopped in mid-episode. A quick investigation showed that the subtitles (in WebVTT format) had empty lines. After removing this, all was well:
=== grog@teevee (/dev/pts/7) /spool/Series/Bergdoktor/17 48 -> diff -wu /spool/Already/Series/Bergdoktor/17/17-05-Lebenswege.deu.vtt 17-05-Lebenswege.deu.vtt
--- /spool/Already/Series/Bergdoktor/17/17-05-Lebenswege.deu.vtt 2024-07-19 00:19:13.000000000 +1000
+++ 17-05-Lebenswege.deu.vtt 2024-10-06 14:30:53.270641000 +1100
@@ -2619,9 +2619,7 @@
sub577
00:36:48.680 --> 00:36:52.520
Dürfen Sie mir als Noch-Ehefrau sagen
-
was mit ihm ist?
-
sub578
00:36:52.560 --> 00:36:54.400
Arguably this is an issue with mpv. But how do official subtitles end up so broken? SRF is particularly bad in this respect.
|
Sweet and sour fish again
|
Topic: food and drink | Link here |
Sweet and sour fish for dinner this evening. More experiments: use fresh oyster mushrooms, roll the fish in flour before deep-frying. And both proved to be an advantage.
I've been making this for over 50 years, and I still haven't stabilized the recipe.
|
Asparagus and prawns
|
Topic: food and drink, opinion | Link here |
Sweet and sour fish is a Chinese dish, of course. But while preparing for it I discovered some fresh asparagus in the fridge. That's normally Yvonne's department, and she had planned it for yesterday, but forgot.
OK, Gemini, what are some Chinese recipes for asparagus?. It came up with a number, including Shrimp and Asparagus Stir Fry, which it called Asparagus and Shrimp Stir-Fry.
Interesting, but certainly not directly useful. Lemon juice and zest. Ridiculous quantities of sugar (2 tablespoons, which Gemini tells me is about 24 g) for 450 g of asparagus). Onions (“white onion”, presumably not garlic). But no real garlic.
After some thought I came up with this recipe (adjusted for sane quantities)
| quantity | ingredient | step | ||
| 100 g | asparagus | 1 | ||
| oil to fry | 1 | |||
| 100 g | prawns | 2 | ||
| 15 ml | soya sauce | 3 | ||
| 10 ml | oyster sauce | 3 | ||
| 10 g | garlic | 3 | ||
I used the prawns I use for breakfast, which weigh about 15 g each. I think slightly smaller ones might be better.
And the result? Excellent! It's been a long time since I've had such a success. Yvonne liked it too, but she was more surprised by my reaction.
And now? How do I improve it? Change the sauce, I think: it was rather saltier than I would like.
And the photos? I was in a hurry, so I didn't take any. But Jane Ashhurst is coming on Thursday, and we'll do it again then.
But the real surprise was that I've been there before earlier in the year:
|
||||||||||||||||||||||||
The recipe looks similar but marginally more complicated. I think that the asparagus pieces were too long then. But I can't recall it, and there's no indication in my diary that I was so happy last time. Why?
| Monday, 7 October 2024 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
I hate Israel!
|
Topic: politics, history, opinion | Link here |
A year ago today Hamas staged an attack on Israeli citizens, killing 1193 of them and taking about another 250 hostage. An estimated 100 hostages remain under their control.
A barbaric act, one that demands a firm response. But the act pales by comparison with the Israeli response. They have reduced the Gaza Strip to rubble and killed round 42,000 mainly innocent people, round 2% of the total population. Israel's response sickens me. I have sleepless nights (well, a couple of hours) as a result, and I'm not even directly involved. With shock I discover that I feel a certain satisfaction when the Israel Defense Forces suffer a setback, though rational logic tells me that that is unjustified.
How do I describe my pain? I have started a couple of times, but there are so many facets, so I'll describe them one by one as time goes on. I'll collect the text, hopefully improved, here.
First: impunity. Israel as a country can do whatever it wants, in defiance of international law and agreements. Why? They've been doing it for nearly 100 years, since long before the foundation of the State of Israel. Even when they founded the state, they took over areas that had not been assigned to them, and the head of state, David ben Gurion, was probably an unconvicted murderer. And any criticism is labelled antisemitism, even when they murder other Semitic people such as the population of Gaza.
How did they get away with it? Clearly the massacres of the Hitler régime made people much more sympathetic towards the Jewish people. But that only went so far: the United Kingdom had had enough of their behaviour, and they only continued to get away with things because of the help of the USA.
This has to stop. If Israel has to bombard Gaza Strip, they should ensure the safety of the population, not boast of their atrocities online. The USA must stop supplying Israel with weapons for pursuit of genocide.
Will that happen? I have ever-decreasing hopes.
|
Begone foul cage
|
Topic: animals | Link here |
We've finally removed the cage that we put in the lounge room four months ago for Mona's protection:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Mona still doesn't get on with the dogs, but she's no longer overly scared. Today Larissa came within 30 cm of her, and though she hissed at her, she didn't try to escape.
|
Cascading problems
|
Topic: technology, multimedia, general, opinion | Link here |
Removing Mona's cage had a strange side effect. The weather station behind it, on top of the loudspeaker, started communicating unreliably with the outside station:
|
|
|||||||||
OK, try to move it slightly to see if it would fix the problem. Bloody USB cables! But I got it done.
Later I wanted to watch the news (I don't know why; all Israeli death and destruction). “No signal” from the TV. Damn, what's wrong there? teevee wasn't running! Somehow my messing around with the cables must have dislodged the power cable. Pulled the machine out of the cabinet, checked all the cables, but I couldn't see anything.
Put it back in, rebooted. Still no display! Bloody HDMI cables! Wobbled a bit, no improvement. Put the connector back in as straight as I could make it, off to hydra in the office, and restarted X from there. It worked! Comparing the log files shows, for the successful version:
[ 365.333] (--) NVIDIA(GPU-0): HISENSE (DFP-1): connected
[ 365.333] (--) NVIDIA(GPU-0): HISENSE (DFP-1): Internal TMDS
[ 365.333] (--) NVIDIA(GPU-0): HISENSE (DFP-1): DFP is not internal to notebook
[ 365.333] (--) NVIDIA(GPU-0): HISENSE (DFP-1) Name Aliases:
[ 365.333] (--) NVIDIA(GPU-0): DFP
[ 365.333] (--) NVIDIA(GPU-0): DFP-1
[ 365.333] (--) NVIDIA(GPU-0): DPY-1
[ 365.333] (--) NVIDIA(GPU-0): HDMI-0
[ 365.333] (--) NVIDIA(GPU-0): DPY-EDID-95ec323f-2338-78ae-30fb-f83daba49f6f
[ 365.333] (--) NVIDIA(GPU-0): HDMI-0
[ 365.333] (--) NVIDIA(GPU-0): Connector-1
[ 365.333] (--) NVIDIA(GPU-0): HISENSE (DFP-1): 600.0 MHz maximum pixel clock
[ 365.333] (--) NVIDIA(GPU-0): --- EDID for HISENSE (HDMI-0) ---
[ 365.333] (--) NVIDIA(GPU-0): EDID Version : 1.3
...
But the failed version, all 490,032 lines of it (about 8,200 pages if printed) consisted mainly of
[ 176.606] (--) NVIDIA(GPU-0): DFP-1: disconnected
[ 176.606] (--) NVIDIA(GPU-0): DFP-1: Internal TMDS
[ 176.606] (--) NVIDIA(GPU-0): DFP-1: 165.0 MHz maximum pixel clock
That repeated roughly 500 times a second. It did recognize the TV when I wobbled the connector correctly:
[ 330.954] (--) NVIDIA(GPU-0): HISENSE (DFP-1): connected
[ 330.954] (--) NVIDIA(GPU-0): HISENSE (DFP-1): Internal TMDS
[ 330.954] (--) NVIDIA(GPU-0): HISENSE (DFP-1): 600.0 MHz maximum pixel clock
[ 330.954] (--) NVIDIA(GPU-0):
[ 330.958] (--) NVIDIA(GPU-0): DFP-0: disconnected
[ 330.958] (--) NVIDIA(GPU-0): DFP-0: Internal TMDS
[ 330.958] (--) NVIDIA(GPU-0): DFP-0: 165.0 MHz maximum pixel clock
And that's all. It didn't seem to make any attempt to talk to the TV, maybe because it had set its hopes on DFP-0.
But I wasn't done yet. firefox couldn't access any sites, not even eureka. Why? Started a different profile, and it worked. But my standard profile, with open windows and saved passwords, wouldn't work at all.
And the other profile also didn't work properly. JavaScript was disabled. It's quite possible that I really disabled it during some experiment, but how do I reenable it? I have forgotten, and it seems that firefox settings never knew about it; at least the search function found nothing.
What does Chromium say?
=== grog@teevee (/dev/pts/5) ~ 2 -> chrome
[63211:10885812789248:1007/163437.444925:ERROR:process_singleton_posix.cc(353)] The profile appears to be in use by another Chromium process (53583) on another computer (tiwi.lemis.com). Chromium has locked the profile so that it doesn't get corrupted. If you are sure that no other processes are using this profile, you can unlock the profile and relaunch Chromium.
[63211:10885812789248:1007/163437.444948:ERROR:message_box_dialog.cc(144)] Unable to show a dialog outside the UI thread message loop: Chromium - The profile appears to be in use by another Chromium process (53583) on another computer (tiwi.lemis.com). Chromium has locked the profile so that it doesn't get corrupted. If you are sure that no other processes are using this profile, you can unlock the profile and relaunch Chromium.
Huh? Chromium doesn't work properly across the net. Checked on tiwi, where there was no process 53583, and it worked normally enough. Back on teevee it still produced the same message, and was too polite to tell me how to “unlock” the “profile”.
ktrace to the rescue, which showed:
63342 chrome CALL readlink(0x2102657af140,0x2710c046280,0x400)
63342 chrome NAMI "/home/grog/.config/chromium/SingletonSocket"
63342 chrome RET readlink 50/0x32
63342 chrome CALL readlink(0x2102657af1c0,0x2710c046110,0x400)
63342 chrome NAMI "/home/grog/.config/chromium/SingletonCookie"
63342 chrome RET readlink 20/0x14
63342 chrome CALL readlink(0x21026561ce90,0x2710c046110,0x400)
63342 chrome NAMI "/tmp/.org.chromium.Chromium.MeEmg7/SingletonCookie"
63342 chrome RET readlink -1 errno 2 No such file or directory
63342 chrome CALL readlink(0x2102657af180,0x2710c0460d0,0x400)
63342 chrome NAMI "/home/grog/.config/chromium/SingletonLock"
63342 chrome RET readlink 20/0x14
63342 chrome CALL __sysctl(0x2710c046550,0x2,0x2710c046580,0x2710c046548,0,0)
63342 chrome SCTL "kern.hostname"
63342 chrome RET __sysctl 0
63342 chrome CALL __sysctl(0x2710c0465f8,0x4,0,0x2710c046528,0,0)
63342 chrome SCTL "kern.proc.pathname.53583"
63342 chrome RET __sysctl -1 errno 3 No such process
What are all these “singletons”?
=== grog@teevee (/dev/pts/8) ~ 8 -> cd .config/chromium=== grog@teevee (/dev/pts/8) ~/.config/chromium 9 -> l
...
drwx------ 3 grog wheel 512 16 May 2023 ShaderCache
lrwxr-xr-x 1 grog wheel 20 3 Feb 2024 SingletonCookie -> 18034524446356579708
lrwxr-xr-x 1 grog wheel 20 3 Feb 2024 SingletonLock -> tiwi.lemis.com-53583
lrwxr-xr-x 1 grog wheel 50 3 Feb 2024 SingletonSocket -> /tmp/.org.chromium.Chromium.MeEmg7/SingletonSocket
drwx------ 3 grog wheel 512 16 May 2023 SSLErrorAssistant
So these symlinks (which must be symlinks, it seems) are pointing to long-dead processes, and Chromium is too polite to remove them and too secure to tell me their names. Bad Chromium! Remove them and all works as well a Chromium has ever worked for me. In particular, it had no difficulty with web site access.
And what's wrong with firefox? More investigation, but in the middle of it all it thwarted my investigations by starting to work normally again. “I didn't change anything”.
How I hate surprises when I'm trying to do something else!
| Tuesday, 8 October 2024 | Dereel | Images for 8 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More thoughts on Israel
|
Topic: politics, history, opinion | Link here |
Somehow I can't get past the issues with the current Israel conflict. Today I watched a recent documentary about Israel's war crimes:
It reinforces my views about their sense of impunity. Here are shots which they say (and I believe) were posted on social media by the mainly named Israel Defense Forces members:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
There's much more in the content (the video is 80 minutes long, and I haven't even finished watching it), including much wilful destruction of buildings and targeted killing of civilians. One of the most disturbing facets is how many Israeli citizens are in favour of these atrocities. The video seems to suggest that it's all Israelis, but we know that's not the case. How many? Still, in Israel the main concern seems to be getting the return of the remaining hostages, of whom probably fewer than 50 are still alive, many having been killed by Israeli attacks.
How can they get away with that? Clearly they're not being punished by Israeli authorities if they publish evidence of their crimes so openly. It's high time that Israeli leadership be prosecuted by international courts. Again, for that to happen something needs to change radically.
| Wednesday, 9 October 2024 | Dereel | Images for 9 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Tidying up the music room
|
Topic: technology, history | Link here |
I've been thoroughly demoralized these last few days, but I can't stand looking at more Israeli genocide, and I don't have the enthusiasm to do anything with my computers.
But there's always a thing waiting for me, tidying up the mess I have generated over the last nearly 10 years:
|
|||||||||||||||||||||||
There are three main categories: old cables, the photos that I inherited from my mother 5 years ago, and various books and data that I haven't been able to classify, even two Qume daisy wheels: I haven't used daisy wheels in over 30 years.
Spent a couple of hours tidying up cables, managing at least to separate still usable ones from the ancient and moving just about all out of the room. The usable ones will have to find space in my office, and the rest will come into the shed. At the end things didn't look that much better:
|
|||||||||||||||||||||||
And as the photo dates show, I spent less than 2 hours doing it all, mainly frustratingly untangling cables. But at least I can now move around. And while moving things around, found some documents about my father's activities in the Second World War, which my mother, for some reason, requested in late 1996. I still need to scan them in, but they show a relatively detailed description of his activities in the Royal Australian Navy (between 17 February 1943 and 3 July 1946, nearly a year after VJ Day). It confirms that he was on the HMAS Shropshire at the end of the war, including being present at the signing of the Instrument of Surrender on 2 September 1945. He had told me this, but it's good to have independent confirmation.
| Thursday, 10 October 2024 | Dereel | Images for 10 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Jane Ashhurst returns
|
Topic: general | Link here |
Jane Ashhurst back today for another few days' visit.
|
Asparagus and prawns again
|
Topic: food and drink, opinion | Link here |
For Jane's sake I made the second version of asparagus and prawns again, this time with photos:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Only a slight problem: I accidentally used the wrong kind of soya sauce, dark instead of light. It didn't spoil the dish, but somehow it wasn't as exciting as last time. Maybe I was just surprised then.
| Friday, 11 October 2024 | Dereel | Images for 11 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
A week of doing nothing
|
Topic: general, politics, history, opinion | Link here |
Somehow I seem to have done nothing all week. Why? Missing motivation, I suppose. Somehow the international situation is concerning me much more than it has any right to. I had intended to write down my thoughts on the current Israeli aggression, but I can't even manage that. The current situation reminds me of the saying “When in a hole, stop digging”. Currently Israel is in some form of conflict with Hamas, Hezbollah, the Houthis and Iran.
Whomelse can they choose as adversaries? The United Nations! They've declared António Guterres, the Secretary-General, persona non grata because he didn't repeat verbatim their account of the current situation. They've decided to close down UNRWA, the only help organization active in the Gaza Strip. And they're directly attacking the United Nations Interim Force in Lebanon (UNIFIL) posts in southern Lebanon. Their excuse? “We told them to go away”.
How can they get away with it? You'd think that they're deliberately trying to fight the whole world. About the only thing that's more amazing is that the USA are still supporting them. Most of their military aggression is performed with US weapons. When is it going to stop? 5 November, maybe? But there lurks another danger so sinister that I don't even dare watch or read the news any more.
|
Samba crush
|
Topic: animals, health | Link here |
Yvonne came in in the afternoon, showing this:
|
|||||||||||||||||||||||
Samba had stood on her foot, the first time that that has happened to her in decades. Hopefully it isn't too serious.
|
Council election again
|
Topic: politics, general, opinion | Link here |
The Golden Plains Shire has a council election this month. Postal votes, and I need to vote for 10 candidates in order of preference.
I don't know any of them! How can I choose? About the only name that rings a bell is Helena Kirby, and that's probably a false positive. But the name helps to discover that I've been here before. And my position hasn't changed: I'm not able to submit a useful vote. Should I just send back the ballot paper unmarked (“informal vote”), or should I choose numbers at random? OK, Gemini, give me a random 10 digit number with no repeated digits. At least I can claim that my vote was based on artificial intelligence
| Saturday, 12 October 2024 | Dereel | Images for 12 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Strange Hugin problem
|
Topic: photography, technology | Link here |
I've been using Hugin on a regular basis for over 16 years, but somehow it keeps coming up with new surprises. The most recent was with enblend, the component that actually performs the stitching. It hasn't been solved, so for affected panoramas I still use the old version on eureka—one of the many reasons why I can't upgrade it.
My current method is to perform the first part of the processing, the part that requires manual intervention (masking and rotation mainly) on hydra with the newest version, and then run enblend on eureka. Today I tried running enblend on hydra and got an error message I hadn't seen before: “Invalid command line options” or some such.
Huh? Oh. It seems that the -m option (specify cache size) is no longer there. OK, remove from the preferences. It still tried to use it.
I've had problems with enblend options before, to the point that I wrote a wrapper script to save the arguments (the log flies by on the screen and into /dev/null), but I couldn't catch it. In the end I had to add
ARGS=`echo $* | sed "s:-m 8192::"`
echo Args are $ARGS
espeak Args are $ARGS
enblend $ARGS
And finally it worked. But where did that -m 8192 come from? It's not in the configuration file.
|
Rösti pain
|
Topic: food and drink, opinion | Link here |
Rösti for dinner tonight. It's a pain to grate the potatoes:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I had started, but I managed to cut my thumb open, and Yvonne continued. The big issue is the grater, which has a rim at the bottom that makes things even more difficult. There must be an easier way.
| Sunday, 13 October 2024 | Dereel | Images for 13 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Where's Yvonne?
|
Topic: technology, general, opinion | Link here |
One of the things that I find very useful is keeping track of where people are. Google Maps helps here, up to a point: it can broadcast a person's location to people of his/her choice. We have that set up for each other, so for example when I picked up Yvonne from the hospital last month, she was able to observe my progress and come outside at the correct time.
But what if the phone doesn't broadcast its location, either because it doesn't know it (no GPS signal) or because it has mobile data disabled? Clearly we can't say anything, and under those circumstances Google Maps shows the last known location, along with the time. But normally it looks like this, where she's represented by the icon in front of the shed:
|
|
|||||||||
I've come to identify places where Google Maps gets into trouble: on the Ballarat-Colac road between Enfield and Dereel, where the mobile phone coverage is flaky, and for some reason after visiting Chris Bahlo, where it can claim that she's still there (with up-to-date time) almost until she returns home.
But lately I've been seeing completely incorrect locations, and this one repeats:
|
|||||||||||||||||||||||
|
|
|||||||||
Why there? It's over 700 m from our house, and in a paddock where I think we have never been. I'm not sure how we could even get in there. I'll keep an eye out to see if it happens again.
|
Chicken and Moroccan lemon again
|
Topic: food and drink | Link here |
Another day of cooking: the dish with the unwieldy name “prosciutto-wrapped chicken with lemons, spring onions and potatoes”, which we've decided to call by the most important ingredients, which don't include prosciutto: chicken and Moroccan lemon. It seems to be an annual thing: previous dates were 25 September 2021, 9 October 2022 and 7 October 2023. Today we had to make more, since Jane Ashcroft is here, and it wasn't helped by the enormous chicken thighs we had: instead of an average of 125 g each, they weighed more like 170 g.
|
|||||||||||||||||||||||
Yvonne wasn't to be outdone with her egg with fish roe entrée:
|
|||||||||||||||||||||||
Jane was so taken by the dishes that she used her phone to send photos to Scott to remind him what he was missing:
|
|||||||||||||||||||||||
| Monday, 14 October 2024 | Dereel | Images for 14 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Yet another quiet day
|
Topic: general | Link here |
There's no doubt about it, I'm running out of steam. What did I do of note today? Nothing that I can think of. Should I attempt to do something so that I have something to report here? It's my life, dammit, and I'll do what I want with it.
|
Yvonne back in paddock
|
Topic: technology, general, opinion | Link here |
Yvonne and Jane out riding today, and I tracked their progress. But before they left, I found Yvonne in a familiar if unlikely place:
|
|
|||||||||
That's exactly the same place as yesterday! The coordinates are precise to about 1.1 m north-south and 0.8 m east-west. Why?
| Tuesday, 15 October 2024 | Dereel → Ballarat → Dereel | Images for 15 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Extracting Yvonne from paddock
|
Topic: general, technology, opinion | Link here |
Why is Yvonne so frequently in that paddock 700 m east of the house? Clearly she isn't, of course, but Google Maps shows her to be there. But is it Google Maps' fault? Maybe her phone is making that absurd claim.
OK, she has the Mendhak GPS logger installed. What does it say?
Oh. It's logging to /Android/data/com.mendhak.gpslogger/files/. And Android, in its infinite wisdom, won't let me access that directory.
But I've been there before, and I worked around it by logging to a readable directory (in the /DCIM/ hierarchy) on hirse, and also by getting the app to send the logs via ftp. And yes, they're there:
=== grog@eureka (/dev/pts/11) /var/spool/ftp/GPSLogger/fossil 123 -> ls -l 20241014*
-rw-r--r-- 1 ftp lemis 263702 14 Oct 12:05 20241014.gpx
-rw-r--r-- 1 ftp lemis 1284651 14 Oct 14:58 20241014.gpx.1
-rw-r--r-- 1 ftp lemis 1386020 14 Oct 15:59 20241014.gpx.2
...
-rw-r--r-- 1 ftp lemis 1386020 15 Oct 11:18 20241014.gpx.21
-rw-r--r-- 1 ftp lemis 1386020 15 Oct 11:30 20241014.gpx.22
I had set the thing to send the log files hourly, not a good idea. But even then there's something seriously wrong: there are only 23 files, they run from midday to midday, and they're all the same! They cover the time range 024-10-14T00:05:23.000Z to 2024-10-14T04:54:29.000Z, not quite 5 hours. Why?
In any case, it's clear that I'm going to have to do better logging. Set the FTP interval to 1440 minutes (24 hours), and store the files in /DCIM/GPS.
Damn, on albo it's /DCIM/gps. Fix it? Sorry, no can do. Can't create a directory there. It was fine with GPS. OK, I will have to live with that. There's a very good chance I will never catch this paddock bug anyway.
|
More tooth loss
|
Topic: health, opinion | Link here |
I've been having mild discomfort with a tooth in my lower jaw for some time. Last time I had something like that, I grinned and bore it, with the result that the tooth died and had to be removed, my first ever. So this time I thought I should go early.
Things weren't quite the way I expected. It's related to the work that Bernd Doroschan did decades ago, and which caused Leela Movva such concern. And there's an infection in the root, and the tooth is dead. Unlike last time, Mario doesn't think that it absolutely has to be removed. He suspects that it might be related to a lateral canal, a concept so obscure that Wikipedia doesn't know about it, and for which I haven't found a good link. He went to a lot of trouble to show me sample X-rays of similar issues, but in the end it boils down to (expensive) root canal work to remove the infection, which may end up requiring the tooth to be extracted any. Or remove it when it starts to become too uncomfortable. Or maybe just remove it and be done with that. The best news is that it isn't urgent, so I have time to think about it.
|
Misery Creek Road again
|
Topic: gardening, opinion | Link here |
I haven't been down Misery Creek Road in Enfield State Park for over 3 years: in principle things don't change, and I have the photos that I want. But today I thought it was worth taking another look.
I was right: nothing new to see, and not even as many flowers as last time. But the road has been severely eroded, and now it's quite difficult to get through. But Google Maps isn't to be outdone. Turn left into this track:
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Where will that take me? As the sign says, it's a dead end. But:
|
|||||||||||||||||||||||
We're at the red pin at the top, and trying to get home, almost due south at the bottom. But the devil's in the detail: this is a pedestrian route, as the dots on the eastern route show. Why that? Last time I looked at Google Maps on a Real Computer, I must have selected a pedestrian route. And There Can Only Be One, so Google Maps gave me pedestrian routes on my mobile phone. It was not only too polite to say so, I don't know how to change it. This Android user interface is really amazingly primitive.
| Wednesday, 16 October 2024 | Dereel | Images for 16 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Goodbye, Jane
|
Topic: general | Link here |
Jane Ashhurst left early this morning, with the result that Yvonne started her shopping earlier than usual.
|
Where's Larissa?
|
Topic: animals, general | Link here |
Phone call from our next door neighbour Graeme Swift this morning: he had found Larissa out on the street again. How does she get out? The mountings of the front gate had been loose, but I had tightened them up, and she shouldn't be able to get out that way. But after recovering her I checked all around the garden, and I couldn't find anywhere else where she could have got out.
Picking her up from Graeme's was also not easy. On my way over, Linda called me, and she was looking for him too. After some searching and yet another phone call, we found him in the property on the other side of their house, conveniently locked and chained up. But the fence was in poor condition, and I got through there. Lara was lying on the ground, and Graeme was sure that she had been through some sort of trauma, particularly since she had her tail between her legs. But she got up willingly enough and followed me.
How is she getting out? Yvonne thinks that she can still push the gate open. Maybe I should anchor the gate motor even more firmly.
|
Accessing Android files, yet again
|
Topic: technology, opinion | Link here |
More investigation of how to get the existing GPS files off Yvonne's phone. Google Gemini had some useful information:
Note: The /Android/data directory is typically hidden and contains sensitive app data. Accessing it directly might not be straightforward for most users.
Methods to Access:
Using a File Explorer App:
Download a trusted file explorer app from the Google Play Store. Popular options include ES File Explorer, Solid Explorer, and File Manager.
Enable "Show hidden files" within the app's settings.
Navigate to the /Android/data directory and explore the app-specific folders.
Using ADB (Android Debug Bridge):
Connect your Android device to your computer via USB.
Enable USB debugging on your device.
Open a terminal or command prompt on your computer.
Type the following command: adb shell
Navigate to the /data directory: cd /data
List the contents: ls
Access a specific app's folder: cd com.example.app_name (Replace with the actual app package name)
Copy files: Use adb pull to copy files from the device to your computer.
“Accessing it directly might not be straightforward for most users”. Indeed. Clearly people cleverer than I have gone to a lot of trouble to make the system anything but straightforward. But that's the first time I've heard anything positive about “File Manager”. Previously it gave me nothing but trouble. But OK, try it out. And how about that, I can get a toy directory tree. Only:
|
|||||||||||||||||||||||
That helps, doesn't it? To protect MY privacy! How does that protect MY privacy? And how does the suggestion help me access MY files? WHY does Android have to reinvent file security when it's based on a system that has had better security for decades.
OK, second alternative: ADB. I always thought that that adb was a primitive Unix debugger. But List of debuggers tells me that it stands for “advanced debugger”. It doesn't know about Android ADB.
Try that?
=== grog@hydra (/dev/pts/25) ~ 450 -> adb shell
bash: adb: command not found
Not surprising, and probably I can find a version that I can use. But this is all so incredibly frustrating.
| Thursday, 17 October 2024 | Dereel | Images for 17 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
A year of hydra
|
Topic: history, technology, opinion | Link here |
It's been a year today since I picked up hydra, my “new” computer. And I still haven't finished configuring it!
Why? A number of reasons: laziness, of course. But what I didn't expect was how many programs no longer run properly. So it makes sense to have eureka running alongside it with its 9 year old system. And no, there's no way to exploit from outside any vulnerabilities that it might have.
|
Android file access: finally!
|
Topic: technology, opinion | Link here |
So how do I access my Android devices from FreeBSD with adb? Where's the port? Off searching. There isn't one!
Google helped, showing this page, which needed some reading before I discovered that the Ports Collection spells adb as devel/android-tools. Installed that and followed some now-lost instructions and got:
=== root@hydra (/dev/pts/4) ~ 145 -> adb devices
* daemon not running; starting now at tcp:5037
* daemon started successfully
List of devices attached
4lzlrwtwei55cacu device
=== root@hydra (/dev/pts/4) ~ 146 -> adb shell
iron:/ $
Well, almost. First I got a message on the phone, which I accidentally dismissed before I read it. I later saw it again on another phone:
|
|||||||||||||||||||||||
That makes sense, I suppose.
Still, now I have a shell! And there's almost a real computer hiding in there!
=== root@hydra (/dev/pts/4) ~ 120 -> adb shell
iron:/ $ uname -a
Linux localhost 5.10.198-android12-9-00085-g226a9632f13d-ab11136126 #1 SMP PREEMPT Wed Nov 22 14:16:37 UTC 2023 aarch64 Toybox
More to the point, though, I can access the file system, though of course it has yet another root. What shows up as a root file system on most utilities, including ftp, is now /sdcard. Is this stuff really on the SD card? I recall having big issues with that a few years back.
So: now I can finally access those GPS logs on fossil. Oh:
|
|||||||||||||||||||||||
And I couldn't connect. The “Troubleshooting” is useless even by Android standards:
|
|||||||||||||||||||||||
FOOL! You haven't enabled USB debugging! Isn't that obvious? OK, do that:
|
|||||||||||||||||||||||
Yes, OK, do that. And I still get the message! But at least now I can talk to the device. Yes, I can't transfer the files (or at least I didn't try), but I was able to firstly create the directory I really wanted, /sdcard/GPS, and move the files (all 500 MB of them!) there. And then I was able to use ftp to pull them down to hydra.
Somehow, though, it's a big relief. Finally I can access the device with normal tools. There seem to be almost all normal utilities, even ifconfig, something that Linux doesn't seem to like. And top!
Tasks: 834 total, 1 running, 833 sleeping, 0 stopped, 0 zombie
Mem: 7752932K total, 7576240K used, 176692K free, 2952K buffers
Swap: 6291452K total, 1846312K used, 4445140K free, 3665348K cached
800%cpu 47%user 3%nice 47%sys 692%idle 0%iow 8%irq 2%sirq 0%host
PID USER PR NI VIRT RES SHR S[%CPU] %MEM TIME+ ARGS
1889 system 18 -2 16G 407M 196M S 38.0 5.3 709:41.74 system_server
1544 gps 20 0 2.5G 6.5M 4.9M S 30.0 0.0 485:13.63 mnld
1281 system 20 0 2.1G 3.4M 3.1M S 10.6 0.0 703:45.89 android.hardware.sensors@2.0-service.multihal-mediatek
16426 u0_a302 20 0 9.1G 171M 128M S 5.0 2.2 0:08.88 com.wilysis.cellinfolite
666 root 20 0 0 0 0 D 2.6 0.0 414:48.37 [chre_kthread]
16985 shell 20 0 2.0G 6.2M 4.5M R 2.0 0.0 0:00.68 top
16815 u0_a230 20 0 8.1G 71M 47M S 2.0 0.9 49:52.02 com.lbe.security.miui
Can I believe that? 8 GB of memory seems plausible, but 6 GB of swap? I suppose it's possible. This is all for some other time. One thing is interesting: 834 “tasks”! And that seems reasonable. It also has an insane number of file systems:
iron:/ $ ps -ef | wc -l
839
iron:/ $ df | wc -l
73
By contrast, on hydra, the most powerful machine I have ever used, we have:
=== grog@hydra (/dev/pts/11) ~ 387 -> ps aux | wc -l
359=== grog@hydra (/dev/pts/11) ~ 388 -> df | wc -l
18
In passing, it's interesting to see that each phone has its own “host name”. I didn't exactly expect it to access DNS, but the names are interesting. Do they maybe relate to the phone model? So far I have:
| Internal | External | |
| name | name | |
| iron | albo | |
| lime | hirse | |
| cannong | fossil | |
For some reason establishing connections is flaky. Once I'm connected there are no problems, but trying to connect can hang, and my only recourse so far has been to shoot down the adb daemon process. There's an adb command kill-server, but it doesn't seem to work reliably. But after all, this is Android. Here an example that I saved:
=== root@hydra (/dev/pts/4) ~ 113 -> adb devices
List of devices attached
4lzlrwtwei55cacu unauthorized
=== root@hydra (/dev/pts/4) ~ 114 -> adb shell
adb: device unauthorized.
This adb server's $ADB_VENDOR_KEYS is not set
Try 'adb kill-server' if that seems wrong.
Otherwise check for a confirmation dialog on your device.
Still, I'm so happy that I have a sane connection to the device. It seems ridiculous that this is hidden as a “debugger”. I'll play around with it more some other time.
|
Tailor-made job descriptions
|
Topic: technology, general, opinion | Link here |
I'm still subscribed to LinkedIn, for reasons that I don't really understand. I retired decades ago, and there doesn't seem to be much else going on there. But an email today stood out:
33 N + 17-10-2024 To Greg Lehey ( 457) LinkedIn N + Greg, 19,102 individuals in Committer, core team member. roles like you are getting views
Now isn't that exactly what I would be looking for?
|
Finally a Grevillea rosmarinifolia
|
Topic: gardening | Link here |
Not much grows in the part of the garden east of the house. We've been trying for nearly 10 years to grow Westringias and Hebes. We have also planted a number of various Grevilleas, all of which have died. But now we have one that looks like it's going to make it:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
| Friday, 18 October 2024 | Dereel | Images for 18 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More Android investigations
|
Topic: technology, opinion | Link here |
As planned, more playing around with the Android debug tools today. Some of the first issues were the various hangs I get when connecting and disconnecting. What I have discovered so far:
adb is too polite to report anything except the most serious errors. My guess is that the hangs are due to some unreported error.
adb needs to be run as root. If not, it appears to run normally, but it doesn't find any devices, and the shell won't run.
=== grog@hydra (/dev/pts/11) /src/GPS/Tracklogs/Phones/fossil 393 -> adb devices
* daemon not running; starting now at tcp:5037
* daemon started successfully
List of devices attached
=== grog@hydra (/dev/pts/11) /src/GPS/Tracklogs/Phones/fossil 394 -> adb shell
adb: no devices/emulators found
The important thing to see here is that the “List of devices” is empty. This may be a permissions error on the USB port. I should try setting adb setuid root.
If I exit the adb shell before disconnecting the USB connection, I can restart with no trouble. If I just disconnect, I can't, and I need to stop and restart the adb process before I can access it again. This may be related to these console messages, which I only discovered later:
ugen0.3: <Xiaomi Redmi Note 13 5G> at usbus0 (disconnected)
usbd_setup_device_desc: getting device descriptor at addr 51 failed, USB_ERR_STALLED
adb does not report anything here.
I can't decide whether adb kill-server hangs or only works very slowly when the server hangs. I'll observe again.
Further investigation of the system is interesting. Here a few observations:
While programs like ifconfig and top are supplied, less isn't (more is).
It does understand DNS. I haven't established where it's getting its information from, but it seems to know that it belongs to lemis.com:
lime:/ $ ping www
PING www.lemis.com (45.32.70.18) 56(84) bytes of data.
64 bytes from www.lemis.com (45.32.70.18): icmp_seq=1 ttl=55 time=233 ms
...
^C
--- www.lemis.com ping statistics ---
7 packets transmitted, 6 received, 14% packet loss, time 6008ms
rtt min/avg/max/mdev = 199.814/232.871/300.492/33.398 ms
I was unable to set the correct host name. It seems that host also doesn't exist:
iron:/ $ host albo
/system/bin/sh: host: inaccessible or not found
127|iron:/ $ hostname albo
hostname: set 'albo': Operation not permitted
1|iron:/ $ sudo bash
/system/bin/sh: sudo: inaccessible or not found
127|iron:/ $ su
/system/bin/sh: su: inaccessible or not found
ls reports some interesting information. Here some of the listings that caught my eye:
lime:/ $ ls -l
total 84
drwxr-xr-x 52 root root 1100 2024-10-18 13:30 apex
lrw-r--r-- 1 root root 11 2009-01-01 11:00 bin -> /system/bin
lrw-r--r-- 1 root root 50 2009-01-01 11:00 bugreports -> /data/user_de/0/com.android.shell/files/bugreports
drwxrwx--- 8 system cache 4096 2024-10-17 13:08 cache
lrw-r--r-- 1 root root 17 2009-01-01 11:00 d -> /sys/kernel/debug
drwxrwx--x 55 system system 4096 2024-10-17 13:08 data
d????????? ? ? ? ? ? data_mirror
drwxr-xr-x 2 root root 4096 2009-01-01 11:00 debug_ramdisk
lrw-r--r-- 1 root root 11 2009-01-01 11:00 etc -> /system/etc
l????????? ? ? ? ? ? init -> ?
-????????? ? ? ? ? ? init.environ.rc
...
d????????? ? ? ? ? ? linkerconfig
drwx------ 2 root root 16384 2009-01-01 11:00 lost+found
d????????? ? ? ? ? ? metadata
drwxr-xr-x 16 root system 340 1973-06-26 18:27 mnt
drwxr-xr-x 2 root root 4096 2009-01-01 11:00 oem
d????????? ? ? ? ? ? postinstall
dr-xr-xr-x 809 root root 0 1970-01-01 10:00 proc
drwxr-xr-x 3 root root 4096 2009-01-01 11:00 res
lrw-r--r-- 1 root root 21 2009-01-01 11:00 sdcard -> /storage/self/primary
drwx--x--- 4 shell everybody 80 1973-06-26 18:27 storage
drwxr-xr-x 17 root shell 4096 2009-01-01 11:00 vendor
drwxr-xr-x 2 root root 4096 2009-01-01 11:00 vendor_dlkm
-????????? ? ? ? ? ? verity_key
In particular, the modification timestamps 1973-06-26 18:27 and 2009-01-01 11:00 both crop up multiple times. I can't see any significance in either, beyond the fact that the 11:00 in the second is the time zone offset. And of course, what are all these stupid ? entries? Complete inability to access the directory entry?
|
Revisiting Google Maps locations
|
Topic: technology, opinion | Link here |
So now I have a log of Yvonne's movements on Monday, where Google Maps had placed her here, to the east of our property:
|
|
|
|||||||||
But no, the track log showed a lot of noise round the house, along with their track along the “coathanger”, but nothing east of Stones Road:
|
|||||||||||||||||||||||
So the fantasy location looks like an idea in the mind of Google Maps. As if to confirm this, she suddenly appeared at almost exactly 38.8° S, 145.75° E:
|
|||||||||||||||||||||||
|
|
|||||||||
She stayed there for half an hour, to the point that I thought that she might have accidentally dropped her phone there: it is on our property, and in an area where she goes relatively frequently. But no, it was on her desk in her office. Time to generate another track log.
Oh. Somehow I don't have one on the phone. What did I do wrong this time? I'll have to wait for next time. But my guess is that the track log would not show her being there at that time.
|
More air conditioner pain?
|
Topic: general | Link here |
Despite 45 mm of rain (about 7.5% of our annual rainfall) today, it wasn't that cool, and I only turned the heating on briefly in the evening. After turning it off before going to bed, I checked the room temperature. 1. Huh? The zone controller showed E3.
I've seen that before. It means “Room Sensor Error (open or short circuit)”. Last time it happened after an E5, “Communication Error between Outdoor and Indoor units”, but I didn't see it this time.
I power cycled the air conditioner, and it came back sane. I wonder if it would have gone away by itself; the 1 in the room controller is the zone number, and it occurs normally when the unit is powered on. If it happens again (hopefully not), I'll leave it for a few minutes before doing anything.
| Saturday, 19 October 2024 | Dereel | Images for 19 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Keeping the gate shut
|
Topic: animals, general | Link here |
Earlier this week I tightened up the actuator for our front gate, which was loose enough to allow Larissa to escape. I thought that it would be enough, but she escaped again.
OK, one of the issues was the mounting of the actuator. It's screwed to the gatepost, and in retrospect it would have made sense to screw it parallel to the gate, but in fact it was at 90°, maximizing the leverage. An obvious solution is to put in a star dropper behind the arm:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It's not much work. Put a dropper hammer over it and ram:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
We'll see if that will work. Yvonne still has her concerns, and she wants to extend this already baroque construction on the other side:
|
|||||||||||||||||||||||
I really wonder why the opener kit doesn't come with a electrically controlled latch to hold the gate in place when it's shut.
|
How to live with Android
|
Topic: technology | Link here |
My recent experiments with Android have been relatively successful, and I won't do much more with it now: I want to maintain my sanity. But it's worth writing up a HOWTO for next time.
|
Chicken kebabs again
|
Topic: food and drink, opinion | Link here |
Chicken kebabs for dinner this evening, a dish so simple that I didn't even have a recipe for it. But Yvonne, who prepared them, wanted them cooked in the “hair dryer” “air fryer”, and we have times there, from 26 July 2019: 18 minutes at 230°.
But the ones she made today were much more consistent, and it seemed better to cook them at a lower temperature. In the end I gave them 19 minutes at 210° and then 11 minutes at 230°, which seemed to be correct. Here the results 5 year ago (cooked for 15 minutes at 230°) and then today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And yes, that seems a reasonable time. But it also brings home how important the quantities are.
| Sunday, 20 October 2024 | Dereel | Images for 20 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Pork and dofu
|
Topic: food and drink, opinion | Link here |
Somehow I'm getting bored with my breakfasts. Time for something new, and I found a recipe for “Spicy garlic tofu” from The Woks of Life.
Some of their recipes are not bad, but as I have commented in the past, the site is only bearable with an ad blocker. Today it had an additional trick up its sleeve: selecting “Metric units” just repositioned to the top of the page.
OK, put up with US units and link to it on my phone. No ad blocker! I really couldn't find the recipe hiding somewhere in the ads. Nothing for it, print out the recipe, showing various problems with my printing software—I really must set up printing on hydra.
Finally I ended up with:
|
|||||||||||||||||||||||
I'll maintain the recipe here.
| quantity | ingredient | step | ||
| 65 g | minced pork | 1 | ||
| oil for frying | 1 | |||
| 20 g | shallots | 2 | ||
| 25 g | garlic paste | 2 | ||
| 3 g (1) | Thai chili | 2 | ||
| 7 g | hoisin sauce | 3 | ||
| 10 g | bean sauce | 3 | ||
| 90 ml | chicken broth | 4 | ||
| 6 g | cornflour | 4 | ||
| 135 g | firm dofu, cut into cubes | 5 | ||
| 5 ml | sesame oil | 6 | ||
| 1 | spring onion | 7 |
And how did it taste? Not bad. I had thought that the quantities I used were excessive, but they seem to have been about right. Next time I should make the dofu cubes larger.
|
Another Ports Collection threat
|
Topic: technology, multimedia, opinion | Link here |
For some time now xv won't display PNG files. I found a workaround in feh. But it's not installed on teevee. OK, we can fix that:
=== root@teevee (/dev/pts/7) /home/grog 3 -> pkg install feh
The following 10 package(s) will be affected (of 0 checked):
Installed packages to be REMOVED:
ffmpeg: 6.1.1_6,1
firefox: 128.0.3,2
get_iplayer: 3.35
mediathekview: 13.8.1_1
mpv: 0.38.0_1,1
vlc: 3.0.21_2,4
I had thought that this package removal was a thing of the past. But this one is particularly aggressive: all the programs I use most on teevee! Begone, foul feh!
|
Where's Yvonne?, day 8
|
Topic: technology, general, opinion | Link here |
I still haven't found a reason why Google Maps displays Yvonne being far from her real location. It's not random: initially it was here, about 700 m east of the house:
|
|
|
|||||||||
But now it's roughly in the middle of our property:
|
|||||||||||||||||||||||
I've been trying to establish where these ideas are coming from. The best I can establish is that they happen when in fact her phone has no GPS signal (it's inside the house, and the receiver doesn't seem to be the best). But Google Maps shows “recent” locations that happen during these extended GPS outages.
So: Google Maps' fault? Or is the phone generating spurious locations? What's puzzling is that the locations are exactly the same.
|
Six months of Mona
|
Topic: animals, history, opinion | Link here |
Mona has been with us for six months. It has taken her quite a while to adjust, but things are good now, and she's almost everything we wanted her to be. She still doesn't get on with the dogs, but she no longer runs away. It's not clear if she'll ever make friends with her.
|
Onion soup
|
Topic: food and drink, opinion | Link here |
A while back I bought a big bag of onions, 2 kg: it was cheap. Not to be outdone, Yvonne bought another bag, this time only 1 kg.
But we don't eat that many onions, and they started to go bad. What do we do to use up the remainder? Onion soup, of course.
What recipe? Bocuse doesn't deal with such trivialities, but Mère Saint Ange goes into intricate detail. The onions must not brown, but the heavy pot makes that almost impossible to avoid:
|
|||||||||||||||||||||||
I think next time I'll use a lighter pot that doesn't hold the heat as much.
She also mentions that the onion slices should be as thin as a sheet of paper, and that the “roots” (the hard part at the base of the onion) should be discarded. I find that the mandolin does a good job. Exceptionally I disagree with her quantities: there's too much liquid. I've adjusted the recipe accordingly.
| quantity | ingredient | step | ||
| 300 g | onions | 1 | ||
| 100 g | butter | 2 | ||
| 30 g | flour | 3 | ||
| 1 l | water or broth | 4 | ||
| 15 g | salt | 4 | ||
| bread slices, about 1 cm thick | 5 | |||
| Parmesan cheese | 5 |
Slice the onions finely with a mandolin.
Melt the butter in a heavy pot, add the onions, and fry gently:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The onions should not brown.
When the onions are translucent, add the flour and continue to fry until well mixed:
|
|||||||||||||||||||||||
Add the water or broth and salt, and cook for 5 minutes.
Put a bread slice each into soup bowls, cover in the soup, sprinkle with Parmesan and grill in the hair dryer for 5 minutes at 230°:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
| Monday, 21 October 2024 | Dereel | Images for 21 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Nasi goreng: how?
|
Topic: food and drink, opinion | Link here |
I've been cooking nasi goreng for well over 50 years, and my recipe is pretty straightforward: cooked rice, leftovers, soya sauce and egg.
But lately I've been unhappy with the egg. I put it in raw at the end and heat up until it's cooked. That's the way we have always done it. Recently, though, I saw somebody cook up the egg first and add the rice later. So today off to see what recipes there are online.
None of the recipes match my method. So I tried frying and dividing the egg before adding rice. The result:
|
|||||||||||||||||||||||
I think it improves the dish. At least it's not soggy, as it tends to be otherwise.
|
Weeding
|
Topic: gardening | Link here |
Nick Macdonald along this morning with his mate Nathan to remove the weeds from the garden. They had planned to stay most of the day, but after 3½ hours they had had enough. They'll be back on Wednesday. And it seems that he wants to come back again in a month or so's time, something that he didn't tell me last time he was here.
|
Larissa still escaping
|
Topic: animals | Link here |
Despite all our efforts over the last couple of weeks, Larissa is still finding a way out of the garden. Is it really the gate? We'll have to try chaining it shut and see what happens.
|
Next case of software rot
|
Topic: technology, photography, opinion | Link here |
Yvonne called me into her office today. Chris Bahlo had told her that there was something wrong with her photos for 19 October. And indeed there was: no photos!
Check the sync operation. Volumes of vomit that she hadn't seen, indicating some problem with s3cmd. It looks as if it has lost libraries:
===== Mon 21 Oct 2024 16:30:19 AEDT on lagoon.lemis.com: make sync
s3cmd sync /home/yvonne/public_html/Photos/20241012/big/ s3://lemis/yvonne/Photos/20241012/big/ -P --delete-removed
Error loading some components of s3cmd (Import Error)
Invoked as: /home/local/bin/s3cmd sync /home/yvonne/public_html/Photos/20241012/big/ s3://lemis/yvonne/Photos/20241012/big/ -P --delete-removed
Problem: ImportError: No module named S3.ExitCodes
S3cmd: unknown version. Module import problem?
python: 2.7.17 (default, Mar 3 2020, 01:17:58)
[GCC 4.2.1 Compatible FreeBSD Clang 8.0.1 (tags/RELEASE_801/final 366581)]
environment LANG=None
ImportError: No module named S3.ExitCodes
For once a meaningful, if voluminous error report. There was really much more (here). But clearly it looks as if something has walked over the libraries. OK, try to replace it, running into the usual agony with the Ports Collection. But then, lagoon is ancient:
FreeBSD lagoon.lemis.com 12.1-STABLE FreeBSD 12.1-STABLE r355358 GENERIC amd64
That's the default version that's too polite to give build dates, but it dates from 12 December 2020. Before I go through more pain, time for an upgrade. That'll keep me busy for a while, but in the meantime I have the workaround of running the sync from eureka or hydra.
|
fvwm3 changes decorations
|
Topic: technology, opinion | Link here |
Strange problem on hydra today: the window manager (fvwm3) on hydra:0.2 changed its configuration. How? It had been running since 23 September. But I couldn't leave it the way it was, because the mouse key bindings had changed. So I restarted it, not a big problem in itself, but how did it happen?
| Tuesday, 22 October 2024 | Dereel | Images for 22 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Bubble foil?
|
Topic: food and drink, opinion | Link here |
Something seemed wrong with the mushrooms that Yvonne cooked for breakfast this morning:
|
|||||||||||||||||||||||
Is that a strip of bubble foil on the larger mushroom? No:
|
|||||||||||||||||||||||
I haven't seen anything that regular before.
|
Focus stacking again
|
Topic: photography, gardening, opinion | Link here |
Our curry tree is flowering again:
|
|||||||||||||||||||||||
Oh. Only the front is really in focus. There's a solution for that: my Olympus cameras have had focus stacking support for nearly 10 years, and it's been over 9 years since I first tried it. But gradually I used it less and less, the last time nearly 2 years ago.
Why? The main issue is to know how many images to take. I've been through it on many occasions, but the step calculations have always irritated me. Still, time to do it again, in the process also running into issues with the mounting hardware. And the camera firmware irritates me too. I have a saved configuration that takes 20 images, but for this I decided that 40 would make more sense. Oh. Focus highlighting doesn't seem to work in conjunction with focus stacking (why?). Switch to manual mode, focus. The number of images gets reset! It took me three attempts to get it right:
|
|||||||||||||||||||||||
Yes, the flowers are all in focus. The ones out of focus at the top belong to another bunch. But was it worth it?
|
Let me out!
|
Topic: animals, opinion | Link here |
We still don't know for sure how Larissa keeps getting out of the compound, but we're getting a kind of confirmation:
|
|||||||||||||||||||||||
That was after chaining the gate shut. So yes, even this small opening seems to be enough for her to squeeze her way through. Now to find a way to keep it shut automatically.
|
Samba: 20 years old
|
Topic: history, animals | Link here |
Yvonne tells me that today is Samba's 20th birthday. We were in Germany at the time, and I didn't note it in my diary.
| Wednesday, 23 October 2024 | Dereel → Ballarat → Dereel | Images for 23 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More garden work
|
Topic: gardening, opinion | Link here |
Nick Macdonald and mate along today to complete the weeding. I didn't get round to looking at the results beyond noting that they didn't do anything round the house entrance. But the bill (still $1,188) was less than I had feared.
|
More cardiology
|
Topic: health, opinion | Link here |
Off to Ballarat today for Yvonne's appointment with Rodney Reddy. They now have free parking in the parking house next to the Ballarat Base Hospital, just across the road. But it took me 10 minutes to find a parking space! It doesn't seem to be worth the trouble.
Rod's diagnosis: looks like further surgery will be needed in the foreseeable future, ablation with Professor Peter Kistler—but not until 4 February 2025! She can possibly have a telephone interview before that, but first we have to find Peter's contact details, which they forgot to put on the appointment confirmation (he's in Melbourne, but he'll be in Ballarat for the appointment).
I took along the KardiaMobile 6L to ask opinions. The ECG nurse had never seen it, and when I showed her a sample ECG, she said “Yes, that's an ECG, but I don't have my glasses on”. Rod was much more interested. Yes, excellent, do you have a sample showing atrial fibrillation? No, sorry. But he thinks it's a good idea. The subscription, however, is just money-grabbing. What the cardiologists get is the same as the PDF representation that we have anyway. Good to hear that my suspicions were unfounded.
|
A new bank account
|
Topic: general, opinion | Link here |
We've decided to try a new bank, Bank Australia. Yana has an account and is happy with them, so we had decided to try them to get rid of Bank of Melbourne and maybe ANZ bank. We were finished with Rod Reddy round 12:30, and Yvonne had previously been told to avoid the time between 12:00 and 14:00 if possible, but we decided to try anyway. We were received by a very effusive Di, who told us yes, Matt was out to lunch, but would be back in 10 minutes. OK, round the corner to look in vain for a new wallet, but did find some new shirts and thongs.
Back to Bank Australia, where it took nearly an hour to set up the accounts. And after that I need to look at the online details and transfer some money into it.
In passing, it seems that my suspicions a few months back were justified: though I had to identify myself with a driver license, it had to be a physical license. They won't accept digital driver licenses.
| Thursday, 24 October 2024 | Dereel | Images for 24 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Garden flowers in mid-spring
|
Topic: gardening, opinion | Link here |
It's gone a third of the time between equinox and solstice, high time for my monthly garden photos.
Last month there was almost nothing to be seen in the garden. It's not very different now. A couple of days ago we had Nick Macdonald of macGARDENS in with his mate to tidy up the garden. They certainly got rid of a lot of weeds, though much is in the spraying, so it will be a few weeks before we see a big difference. And in fact the photos don't show any improvement at all, just an indication that the bulbs have finished flowering in the centre bed:
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||
To be clear: that's last month first, and then today.
But one thing does stand out, the Clematis that I have been nursing back to health. Last month it was just forming buds:
|
||||||||||||||||||||||||
A few days ago the first of them flowered. But today I found
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
They had torn off not one, but two roots. Again! I can (only barely) forgive amateurs for making that kind of mistake, but these are highly paid professionals. I'm really very upset.
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
On the other hand, the roses are looking good:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That doesn't have much to do with the gardeners. It's too early for that.
A couple of months ago Jesse Walsh planted some Carpobrotus and some random plants that Yvonne got from the Nyaris. Now it looks like this:
|
|||||||||||||||||||||||
Somewhere in that mess of weeds (a part that Nick and mate didn't address) there's a still surviving Carpobrotus, flowering. The other ones seem almost all to have died. No wonder I despair.
The Alyogyne huegelii seems to be recovering from the mistreatment it has had over the years:
|
|||||||||||||||||||||||
I wish that I could say the same for the Hibiscus rosa-sinensis and the Strelitzia nicolai:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Again, they're not dead, but despite the mild winter they seem to have suffered unduly.
Our two real oak trees seem to be doing very differently. The one planted in the east garden is not dead:
|
|||||||||||||||||||||||
But the one still in a pot on the verandah looks much better:
|
|||||||||||||||||||||||
The Grevillea rosmarinifolia near the house entrance hasn't been there long, but it seems to like it, something that I can't say of previous attempts:
|
|||||||||||||||||||||||
We should consider putting more in.
The transplanted wildflower that seemed to have been recovering last month still seems to be recovering, but it's not in a hurry. Again, last month and this month:
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||
It looked much worse a few months back, so there's still (a little) hope.
Some (few) things seem to be doing well, though. The Abutilon in front of my bedroom window has grown far larger than I had expected:
|
|||||||||||||||||||||||
And the Camellia japonica that we transplanted last December still doesn't look good, but the new leaf growth is green rather than yellow, so that seems to be an improvement:
|
|||||||||||||||||||||||
We have a number of plants indoors in the lounge room at the moment. The curry tree has improved almost beyond recognition in the last month:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I will have to decide soon when to put it outside.
And the Hibiscus rosa-sinensis “Uncle Max” that looks so unhappy outside has a clone which is doing fine:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I have also put the Spathiphyllum in the lounge room, and that seems to suit it too. It has been flowering for some time, and a second round has just started:
|
|||||||||||||||||||||||
|
The proof of the gardener
|
Topic: gardening, opinion | Link here |
Taking the monthly photos of the garden also gave me an opportunity to look at what Nick Macdonald and Nathan did. I've already mentioned the shock of the destroyed Clematis, but there's more. Lots of Valerian, which I had specifically asked them to remove, and an obvious weed in with the daun laksa:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And then they knocked down a fence section and didn't reposition it:
|
|||||||||||||||||||||||
Is that acceptable? I've lowered my standards over the years, but it's disappointing that a gardener who charges so much should let that slip by.
|
Noodle experiment
|
Topic: food and drink, opinion | Link here |
Lately I've been doing various experiments with East Asian breakfasts. There are a whole lot of dishes like Beijing “Fried Sauce” Noodles, Hokkien Mee, KL Hokkien Mee, Mi udang and Pad thai. And then there are Japanese dishes that I just don't understand.
So: how about another experiment? This time I used Guan Miao noodles and some enoki mushrooms that I bought yesterday:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
| quantity | ingredient | step | ||
| oil | ||||
| 20 g | spring onion | |||
| 20 g | garlic paste | |||
| 10 g | ginger paste | |||
| 10 g | prawn paste | |||
| 3 g | chili | |||
| 70 g | mustard greens | |||
| 30 g | enoki | |||
| 150 g | Guan Miao noodles | |||
| 20 g | soya sauce | |||
The enoki mushrooms are interesting. I've never used them before. It's clear that the bottom part is to be discarded, but how much?
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
Mona and Bruno progress
|
Topic: animals | Link here |
Seen this afternoon:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's Yvonne's bed, where Mona has previously not dared because the dogs are so frequently there. She seems to be making progress, seen also earlier in the afternoon where Larissa approached her to within 20 cm, without Mona trying to escape or attack Lara.
| Friday, 25 October 2024 | Dereel | Images for 25 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Have I been pwned?
|
Topic: technology, opinion | Link here |
Message from haveibeenpwned, a service that checks for leaked credentials, mainly email addresses. Two lemis.com addresses have been leaked, from a breach at digiDirect, despite the name a camera sales company from whom I last bought something 5 years ago. At the time I grumbled about their IT infrastructure, but it seems that they have had a breach of 304,337 accounts. As haveibeenpwned puts it, frustrated-user@lemis.com
Compromised data: Dates of birth, Email addresses, Names, Phone numbers, Physical addresses
OK, what email address did I give them? Not my own, of course, something like iratecustomer@lemis.com, reflecting my satisfaction with their computer interaction at the time. Name? Physical address? Almost certainly. You don't need a data breach for that. Phone number? I don't recall giving them that, but again, not a problem. The all-powerful dates [sic] of birth? No, I have no record of giving them a date of birth. Clearly iratecustomer@ didn't have the same date of birth as I do, so I would have registered it.
The only issue I have is that they claim that I have had 2 email addresses exposed. And they're now too polite to say which; I have to ask for each of the 600-odd email addresses. Fortunately I didn't erase the name of the old email address that I had initially used (digidirect@lemis.com, only the address itself. I had to change 7 years ago as a result of their broken software. But clearly my strategy works. Addresses exposed, no harm done. And I'll find out if anybody sends email to iratecustomer@.
|
Teleconference with Professor Peter Kistler
|
Topic: health, technology, multimedia, opinion | Link here |
We had planned a teleconference with Professor Peter Kistler this afternoon. Bloody Zoom again! I didn't even try to set it up with FreeBSD. Back to ancient eucla, a Microsoft laptop. And I was able to set that up with relatively little pain. Only I had to type in meeting ID and password, and a couple of times it decided to go to sleep, and when it woke it had forgotten everything. By the time we were done, I could remember the password QRLw24, though not the 10 digit meeting ID with the shifting spaces.
The meeting had been scheduled for 15:15, though Yvonne had asked for 15:45: Martin, the farrier, had been due at 14:45, and we were concerned that he might come late and not be finished by 15:15. As it was, he arrived on the dot of 14:45, and he was done by 15:10. Set up the Zoom connection and waited. And waited.
That was too much for eucla. I accidentally jogged the power cable, and it proved that the battery is dead. Shut down. Reboot, and it really did manage to set up Zoom again before Peter called. But to be on the safe side, also set it up on albo, my phone, and mounted it on a tripod.
Finally Peter came online round 16:10, and we ended up with a 3 way conference. Yvonne tells me that albo obscured Peter's face with some text, and she only saw his hair, here somewhat tidier from his home page:
The discussion wasn't long, but focused. He asked for timings, suggested that yes, ablation is the way to go, and demonstrated a surprisingly large device that gets inserted into the heart. There's an 80% success rate, but that doesn't mean that the other 20% die: they just need a second iteration. Slight risk of bleeding after the operation. OK, how about next Friday?
That was a surprise. We hadn't expected anything in the next 4 or 5 months. But sure, I'll have to cart her to the Melbourne Private Hospital, next to the Royal Melbourne Hospital and apparently the place where I spent some time in an oxygen tent while recovering from mumps over 70 years ago.
She'll have to stay overnight as a precaution against potential bleeding heart syndrome, but in principle it's a daytime operation.
|
Melbourne hospitals: care for all
|
Topic: health, language, opinion | Link here |
Read in the Ballarat Courier:
An infant has been taken to hospital in Melbourne following an incident, believed to have involved a horse, in Learmonth on Wednesday, October 23.
They were taken to the Royal Children's Hospital.
I didn't know that they had a veterinary department.
| Saturday, 26 October 2024 | Dereel | Images for 26 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Steamboat again
|
Topic: food and drink, opinion | Link here |
On Wednesday I bought a couple of ingredients that I had been looking at for some time: enoki mushrooms and finely sliced beef for hotpot, which I know as “steamboat”; when I was a lad, hotpot came from Lancashire. The mushrooms have many uses, and the beef was frozen, good and cheap, so I hadn't really intended to use it right away. But it seems that the enoki mushrooms also go well with a steamboat, so today was the day. A search for recipes on the web brought Dad's Hot Pot at Home, Easy Chinese Hot Pot At Home and How to Make Chinese Hot Pot at Home. They all seemed reasonable, and I gained a few ideas, but nothing seemed very different from what I've done in the past. In the end I selected these ingredients:
|
|||||||||||||||||||||||
How was it? Alright. Somehow I still can't bring up much enthusiasm for the dish. It's too close to what I eat for breakfast. Maybe I should try sukiyaki next time, but the recipes I have seen suggest that it's too sweet for my liking. shabu-shabu maybe?
| Sunday, 27 October 2024 | Dereel | Images for 27 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Strange spam
|
Topic: technology, opinion | Link here |
A triplet of spam today. From my inbox listing:
7 N + 27-10-2024 To groggyhimself@ (1320) Shein Surprise Limited Time Only: Get Your Shein Mystery Box Reward Now!
8 N + 27-10-2024 To groggyhimself@ (1320) Woolworths Shipment ├=>Final Notice Coming for a Tupperware 36-piece set Reward
9 N + 27-10-2024 To groggyhimself@ (1320) Nespresso Surprise └=>Expiring Soon: Claim Your Nespresso Pixie Reward
That's with mutt, which groups messages together. But what do these three have to do with each other, apart from the fact that they're clearly spam? The => gives a clue: they all have the same Message-ID! For confirmation,
=== grog@hydra (/dev/pts/21) /src/FreeBSD/git/main/usr.sbin/makefs 15 -> grep ^Message-ID /var/mail/grog
...
Message-ID: <[10_n]uU6YbwTCQkizQoqaC1-QCw@geopod-ismtpd-4>
Message-ID: <[10_n]uU6YbwTCQkizQoqaC1-QCw@geopod-ismtpd-4>
Message-ID: <[10_n]uU6YbwTCQkizQoqaC1-QCw@geopod-ismtpd-4>
Somehow spammers are getting sloppier. On the other hand, looking at the other headers, it seems that it fulfils all SPF and DKIM requirements. So much for the latest lot of hoops to jump through.
|
The Great Controversy?
|
Topic: general, history, opinion | Link here |
Found in our letter box today (Sunday): a dozen duck eggs and a book.
The eggs are understandable. They were left there for Yvonne by Donna Watt. But the book? It's clearly brand new, and nothing we knew about. It's titled “The Great Controversy” and claims to be “The untold story behind the Vatican's rising influence in America”. Indeed, that seems to be particularly untold, but is it a controversy? It's clearly religiously motivated, and it took me a while to find the Wikipedia page The Great Controversy (book), written in 1888 by Ellen G. White. But how much does she have to do with this book? The typesetting is a little unusual (amateurs?), and it doesn't mention an author. There is a reference on the back of the title page, however: “Original text by E. G. White”. But in the original book the controversy is between Jesus Christ and Satan, not the USA and the Vatican. The clearest statement is that it comes from the environment of the Seventh-day Adventist Church
So many questions: who put it in our letter box? There's a Seventh-day Adventist church in Ballarat, and I picked up some plant cuttings from a person who lives nearby a few years ago. But is Lynn (the person who gave me the plants) a Seventh-day Adventist, just because she lives near the church? And that still asks the question: why?
Should I read it? Certainly the transformation of the controversy from Christ and Satan to the USA and the Vatican (which is which?) is puzzling enough. But it's 450 pages of dense text. Do I really care?
|
Mona: next progress
|
Topic: animals, general | Link here |
Couldn't find Mona today. She proved to be in Yvonne's bedroom again. The difference: Larissa was there too:
|
|||||||||||||||||||||||
Presumably Mona was there when Lara came in, but at least she didn't run away. Earlier she stood her ground when Lara wanted to pass her. We're not there yet, but there's progress.
| Monday, 28 October 2024 | Dereel | Images for 28 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Microsoft crash
|
Topic: technology, general | Link here |
Processing some photos today, Yvonne was using distress, the Microsoft machine that we normally use. OK, we have a backup: dischord, an older, slower machine. Start a remote desktop. Bang! It seemed to explode, and though it had been powered on (hibernated), it lost power. And there was a vague smell of burning electronics. I've never seen that before.
|
Patient registration: bad for the heart
|
Topic: health, technology, general, opinion | Link here |
Yvonne is due for her catheter ablation on Friday, and of course she needs to fill out patient registration. I've been through this before earlier this year. On that occasion it took me 3 days. Given Yvonne's cardiac issues, it was clear that I had to do it.
The good news: it was much faster this time, though of course they annoyed me in the same manner. All dates in their format, all numbers, with slashes and leading zeroes for single digits, so my birth date 19 August 1947 had to be written 19/08/1947.
|
|||||||||||||||||||||||
To help, it inserted the slashes automatically, even after I had already done so, so I ended up with a first attempt 19/81/947, which it then of course rejected.
And then enter the GP name. There's only one Paul Smith, though other sites manage to find his Doppelgänger at different locations and thus assume that he's somebody else. But this one wants me to enter just his surname: Smith. Sorry, SMITH. Then a drop-down menu that runs out of steam towards the end of the A Christian names.
Finally managed to enter things, and at the end was presented with the confirmation: “GP SMTIH, Paul”. Who cares? But they wanted all details, including the practice, with street address, phone and fax numbers. Surely they could have that on record. And the last blood test? Which lab? Last week Rod Reddy found it instantly.
Then they wanted a list of the medications that Yvonne is taking. Managed to enter that, but I also had to enter the purpose of each medication! How do I know? The cardiologist does. I had to enter something, so how about “unknown”. Yes, fine.
When did Yvonne stop smoking? Over 40 years ago. But no, they wanted the exact date. I see that I had that issue last time as well.
Finally it was done, only a little over an hour and 15 mm Hg blood pressure increase.
|
BUPA: What will you pay?
|
Topic: health, technology, opinion | Link here |
The other half of the issue with Yvonne's procedure on Friday is, of course: who will pay for it? In particular, I had asked for a single room for the night. Will BUPA pay that? And how much of the other costs will they cover? Tried to call them up, but I didn't get a dial tone. After a lot of messing around, discovered that this horrible Grandstream HT802 Analogue Telephone Adapter had hung itself up. It still responded to nmap, but it didn't respond to my web requests. Power cycle, and then it worked.
Called them up on 134 135 and had to enter my date of birth and my policy number, all over my phone keypad. Why date of birth? Because it's secret! Finally got past a voice non-recognition system and was told I would have a 10 to 15 minute wait. Call back? Yes, please.
An hour later I gave up. To their web site, where I tried to sign up for myBupa, the customer access site. Give all the information they asked for, including of course the date of birth in their chosen format. Finally entered it all and got:
|
|
|||||||||
No, all the data was correct. The only thing that wasn't was their link to “contact us”, which just gave some not-very-helpful tips.
OK, elsewhere on the site there's a possibility to request a call back. Reason, please? None relevant, and I had to choose from one of them:
|
|
|||||||||
“We'll call back within 10 minutes”. After another half hour I braved their wait times and called again. This time I made the mistake of saying something unrecognized to the voice non-recognition system. “Is that something to do with <mumble>. Please answer “yes” or ”no”. Well, no, not quite. But I can't say that. After a couple of blood-pressure raising iterations, asked to be connected to a consultant. Ah, that's a word it understood, and I was connected relatively quickly to Mitch, who wanted to hear again the information I had painstakingly typed into the phone keypad.
He was helpful, though. Yes, Yvonne is covered for the procedures, she's entitled to a single room, and we don't even have to pay the $250 excess: that's only once a year, and I paid it in March for my cataract operation. Unless (as is likely) Peter Kistler and the anaesthetist come up with fees beyond the normal, we have nothing to pay.
He also explained the inability to register online: they don't know my mobile phone number. No matter that it had already sent at least one PIN to it; they do that anyway. He entered the number, and I was able to sign up.
What a contrast. Mitch was really helpful, but their computer setup is a disaster.
|
Another monitor dropout
|
Topic: technology | Link here |
Just to keep me on my toes while fighting health service web sites, my number 2 monitor dropped out on me. That's the one I bought three months ago to replace the old Matrix NEO 270WQ, which was dying of the same symptoms.
Cursed location? No, I think that it was Bruno wandering around behind the monitors, though I thought that DisplayPort cables were less susceptible to such problems. To be observed.
| Tuesday, 29 October 2024 | Dereel | Images for 29 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Israel: eliminate obstructions to genocide
|
Topic: politics, history, opinion | Link here |
The Knesset has overwhelmingly voted to declare UNWRA, the United Nations Relief and Works Agency for Palestine Refugees in the Near East, a terrorist organization, and ban their operations on Israeli soil.
Three weeks ago I started trying to put my disgust into words, but I failed. It seems that things just get worse all the time. Now Israel is pitting itself against the United Nations, essentially the rest of the world. Without knowing the background, who is more likely to be in the wrong? Once a beacon of light in the murky Middle Eastern nations, it is now a pariah—maybe it always was, but now they're showing their true face. They condemn the very institution that was founded to ensure that the horrors that happened to the Jews would never again be possible. They perpetrate those horrors against their neighbours! They break all laws of humanity.
Countries round the world have condemned the action, but the USA says nothing. Of course Bibi has timed things well: it's a week to the US Presidential election, and neither candidate wants to say anything that would rock the boat. God help us if Donald Trump should win.
How long will Israel remain a member of the United Nations? They have attacked UNIFIL, placed there inter alia to keep them out of Lebanon, they have declared António Guterres, the Secretary-General, persona non grata. Somehow once again I'm reminded of Hitler Germany, which left the League of Nations 91 years ago. When will Israel leave? And they (still) have the US iron-claddedly on their side, supplying most of the weapons they use to perpetrate their genocide. They should be clapped in irons.
How will this continue? I don't see a quick end. Will Bibi live long enough to be subjected to (and doubtless convicted by) a latter-day Nürnberg Trials? One thing's clear: Israel has greatly lost respect over this matter, and that will take decades to recover.
|
Bank Australia web site
|
Topic: general, technology, opinion | Link here |
Interesting (paper) mail today: two debit cards from Bank Australia, the first communication I've had from them since we signed up last week. OK, needs activation. The sheet of paper accompanying it told me what to do: Login to the app or internet [sic] banking; Go to Cards.
OK, I knew that I had a temporary password to change, so logged in (or is that loggedin?), did that and ran into the most extreme password strength checker I have ever seen, helped by the fact that, like Bank of Melbourne, their web site is too secure to show me what I have typed in.
Finally I chose a password that it found favourable. To the home page. “Go to Cards”. How? No mention of cards. Checking my account details, found why I hadn't received the promised documentation: they had misspelt my email address! That's what comes of filling out paper forms. OK, go to help, where I read:
Log in to the app or internet banking and select ‘Services’ > ‘Activate card’
OK, go to Services. No “Activate card”:
|
|||||||||||||||||||||||
Clearly that's their way of spelling “Card management”. OK, go there, activate card. Everything is blocked except “Wallet”, whatever that means. I can't use the card to pay for anything in the Real World. And there was no obvious way to enable it. Ah, yes, the card is “locked” (punctuation is original):
By activating your card you will make it available to add to your digital wallet (Apple Pay, Google Pay , Samsung Pay) . If you already have your physical card you can unlock your card after activating. If you have not received your physical card DO NOT unlock your card until it arrives.
OK, select “unlock”:
|
|
|||||||||
The important information is in the middle of the text: “This option requires Secure SMS validation”. Send me an SMS to enable SMSs! Yet Another site that wants to send me thousands of SMSs. Finally did that and unlocked the card. And then I was given the opportunity to set a PIN:
|
|
|||||||||
Clearly the web programmer didn't expect the customers to be able to type digits.
So how does this compare to other bank web sites? It's certainly no better. Is it worse?
Maybe. And maybe it's just that I'm behind the times. Who pays with a card any more? Use
your pain smart phone! For me, having sensitive information on a phone is a
security disaster waiting to happen. Am I wrong? Before I can decide that, I need to
understand the entire process, and I don't. My guess is that very few people do, and many
of them might be criminals.
|
Larissa catches a kangaroo
|
Topic: animals, general | Link here |
Walking the dogs today, Larissa caught a kangaroo:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Admittedly, it didn't try to run away.
| Wednesday, 30 October 2024 | Dereel | Images for 30 October 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
An easy spam indicator
|
Topic: technology, opinion | Link here |
A couple of days ago I received multiple spam messages with different content but the same message ID. It happened again today:
59 N + 29-10-2024 To groggyhimself@le (1163) Package delivery N + The delivery of your package is still pending - Confirm your shipping details
60 N + 29-10-2024 To groggyhimself@le (1155) BCF Surprise N + └=>Expiring Soon : Get Your Yeti Crossroads Backpack Reward Now!
That => in the second line is mutt's way of indicating that the messages are related, in this case because they have the same Message-ID. But that also helps identify spam without checking. In this case, it's clear that the second message is spam. But the first? I'd need to check, especially as I was expecting a package. But this => saves me the bother.
|
A slew of broken web sites
|
Topic: technology, general, opinion | Link here |
Over the past couple of days I have had pain with not one, not two, but three different web sites. They're not unusual, but somehow they seem to be ganging up on me. I was expecting delivery of a replacement battery for my ThinkPad. Not urgent—I hardly use it—but today was Yvonne's shopping day, and initially Australia Post had claimed that they would deliver it yesterday or today. But in the course of the morning I received lots of contradictory statements. This one at 11:30:
|
|||||||||||||||||||||||
That's a lot to read. The important parts are:
|
|
|||||||||
|
|
|||||||||
The first one says that the item is at the post office for collection. The second says that it won't arrive until tomorrow. And eBay is even clearer:
|
|
||||||||
They tried to deliver it! How? It's at the post office in “NAPOLEON“, as the incompetent people in Australia Post call it. And that was followed up by this message:
Date: Tue, 29 Oct 2024 20:04:47 -0600
From: Australia Post <noreply@notifications.auspost.com.au>
To: groggyhimself@lemis.com
Subject: Your parcel is ready for collection from the Post Office
And of course it was there, despite their prior claims to the contrary. But admire the time zone, which must be North American Mountain Time, 17 hours behind AEDT, our time zone. It corresponds to 13:04 AEDT today. Once again I'm amazed at the incompetence of Australia's national postage service. For the fun of it I ran a traceroute to the address of the mail server, and it ran into the ground after this:
10 be11--bbr2-phx1.net.sfdc.net (13.108.1.247) 204.994 ms 210.226 ms 200.658 ms
11 hu-0-0-0-8-0--bbr2-dfw1.net.sfdc.net (13.108.7.134) 218.300 ms 224.492 ms 229.923 ms
Those node names suggest Phoenix, AZ and Dallas, Texas. Phoenix is in the Mountain Time zone, but Dallas is in Central. Who knows where it went after that? In any case, it seems that Australia Post has outsourced its mail to the USA. What an advertisement!
But they're not the only one. The Bank of Melbourne (“we speak your language”, “a bank for Melbourners” or some such) sent me a confirmation of a funds transfer:
Date: 30/10/2024
Time: 10:59 AM Sydney Time
It's not “Sydney Time”, of course, it's the New South Wales time zone, which is not in effect in Melbourne. Why do they do that? Because they're a subsidiary of Westpac, presumably, previously the Bank of New South Wales. But why don't they use the normal term AEDT, which covers both states?
|
More Android strangenesses
|
Topic: technology, multimedia, opinion | Link here |
Recently I've discovered the Android apps of the German TV stations. Interestingly, though they clearly run on a mobile phone, they offer a download service. For the fun I tried it out. And how about that, they download round 10 times as fast as the same service on the web. On the web I typically get round 200 kB/s, while on the phone it's closer to 2 MB/s. Why? It's not a tuning issue on eureka: videos from SRF come down round 2 MB/s. It must be the broadcaster web sites.
But there's another strangeness: yes, once I have downloaded them, I can play them if I can put up with the tinny phone loudspeaker. But how about moving them to a Real Computer? Oh, can't do that. The “File Manager” knows the concept “Videos”, but it can't find any. And my FTP access shows “top level” directories Movies/ and Recordings/, but they're both empty. It looks as if I will have to search in the bowels of the apps to find what they have done with them.
That's not just irritating, it's just plain stupid. Each of the two videos that I downloaded (one from ARD, the other from ZDF) are about 1.5 GB in size. That's a significant part of the total storage of the device, and “normal” (for Android) tools can't even see them. It seems that I must remember what app downloaded them and delete them manually. And no, this is not a protection or copyright issue, since I can legally download them from the web site.
But in this age of 80" TVs, why should anybody want to watch a video on a display that was considered tiny even when TV first became available? In 1957 the Australian magazine “Radio, TV and Hobbies” had a build-it-yourself TV based on a 5" oscilloscope tube, and they apologized for the tiny display (because it was cheaper). albo's display has a diagonal of 17 cm, about 7", but the aspect ratio means that it's pretty much in the same range. At the time, the smallest TVs had a 14" display.
| Thursday, 31 October 2024 | Dereel | Images for 31 October 2024 |
| Top of page | ||
| previous day |
|
Bloody NBN!
|
Topic: technology, general, opinion | Link here |
Another all-day National Broadband Network outage today, from 8:32 to 16:25. Or at least, that's what we saw. Aussie Broadband's useless app told me that there was a danger, but that there was currently no outage. On the other hand, they threaten a total of 10 days of outages, starting this morning:
|
|
|||||||||
What use is this app?
|
More Bank Australia fun
|
Topic: technology, general, opinion | Link here |
Paper mail today, one each for Yvonne and me. Not our Bank Australia debit cards—they arrived on Tuesday—but the corresponding PINs. I had already set mine, so I gave the other to Yvonne:
Yvonne didn't like the PIN that Bank Australia gave her. Not a problem: she can change it. Just log in and follow the undocumented menu tree.
But she couldn't log in! Wrong password! But we only set it yesterday. OK, ask for a new password. Sorry, can't identify your account.
Huh? Off to my computer, where I was able to make the request with no trouble. Set the correct password. “Weak password”, which means what they're too polite to say outright: we won't use this password. That was exactly the same password that they considered strong on Tuesday! Finally worked round it by using a foreign language insult, which so far hasn't been rejected.
Why do people do these things? So far Bank Australia has managed to make us consider cancelling the accounts again.
|
Preparing for hospitalization
|
Topic: technology | Link here |
Yvonne is off to the Melbourne Private Hospital tomorrow for a catheter ablation, which will involve her being in overnight. Boredom? Time to set up the Facebook app on her phone. And to make things bearable, she should use voice input.
That works fine on my phone, but not on hers: “No permission to enable Voice typing”. Huh? Off to look at the keyboard settings, but without any success. And for once Google Gemini had nothing useful to say. Finally I found it:
|
|||||||||||||||||||||||
That's only approximate, of course, like all Android documentation. But it shows that it's an issue with the app, not the functionality. How I hate this stuff!
|
Curry tree flowers
|
Topic: photography, gardening, opinion | Link here |
I've been meaning to take photos of my curry tree flowers for a long time, but it's not easy. They're about 8 mm across, and the depth of field at that size is minimal:
|
|||||||||||||||||||||||
So another effort with focus stacking, not helped by the lack of clarity on the number of images required. In the end I ended up taking 60 images of each of these photos, only 35 of which were needed:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
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 |
