|
| Greg's diary |
| February 2018 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Thursday, 1 February 2018 | Dereel → Ballarat → Dereel | Images for 1 February 2018 |
| Top of page | ||
| next day | ||
| last day |
|
Peugeot 307: no thanks
|
Topic: general, opinion | Link here |
Did a bit of research on the Peugeot 307 today. The results were surprising. It seems that a car like the one I saw yesterday has a street value of $2,900, not the $7,999 that Bailey had been asking, nor even the $4,999 to which he had marked it down. That's enough to give up already. Later I looked in at Ballarat Automotive and spoke to Paul Sperber, who told me that he didn't service them, and echoed my concerns about maintenance of old European cars. So that one's dead. A pity for one thing only: the large, low rear cargo area.
|
Now that's a bone!
|
Topic: animals | Link here |
Off walking with the dogs, in a bit of a hurry before leaving for town. I had given them until 13:42 to turn back. But we got half way and Nikolai jumped onto the side of the road and came out with this (taken later after we returned):
|
|||||||||||||||||||||||
That's all that remained of the kangaroo that had caught itself in a fence and died a month ago:
|
|||||||||||||||||||||||
|
Haircut and doctor
|
Topic: general, health | Link here |
Into town yet again, mainly for Yvonne to see the doctor, but while we were there we both had our hair cut.
Paul Smith (the doctor) had, to his obvious annoyance, not received the documentation of the case from the base hospital, but he had seen the X-rays taken yesterday, and based on them he was surprised to see Yvonne on her feet. Another confirmation that things are on the right track, though he took the trouble—probably necessary—to remind her that just feeling better doesn't mean that she has recovered, and there will be at least another 3 months before things are back to normal.
While I was there I asked for a renewal for my telmisartan (for high blood pressure). That requires an obligatory blood pressure measurement, which proved to be 105/60, a little low for Paul's liking. So there's the possibility that I will have to reduce the dose or stop taking the stuff. First I need to measure more often. In the evening I measured 122/66 and 126/68. That's definitely not too low for systolic, but what about diastolic? Or is it just my device?
|
Mexican food, and chilis
|
Topic: food and drink, opinion | Link here |
Seen at Woolworths while looking for chilis:
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
Now I really don't like jalapeño chilis, but how do these people come to consider them as anything else than chilis?
| Friday, 2 February 2018 | Dereel → Ballarat → Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More eBay pain
|
Topic: technology, general, opinion | Link here |
Various mail from eBay today. First, one that was completely unintelligible. The Subject: line was “Subject: eBay will begin intermediating payments on its Marketplace platform”.
What does that mean? Looking up in the OED didn't help. The verb “intermediate” (with long last vowel: /ɪntəˈmiːdɪeɪt/) has four meanings, only one of which is in common use: “To act between others; to mediate.” So we need a direct and an indirect object. But here I only get one:
eBay is happy to announce plans to further improve the customer experience by intermediating payments on our Marketplace platform. In doing so, eBay will manage the payments flow, simplifying the end-to-end experience for both buyers and sellers.
It takes a fair amount of reading and thought to discover that they use the word “intermediate” to mean “processing”. What a Marketplace platform is is left to the interpretation of the reader, who is (as they say) experienced.
Put that into the “too hard” category until I received a mail from Statista:
Going back to the eBayese message, it became clear: they're changing their preferred payment processor. And somewhere in there, that's what they say, but only after your eyes glaze over.
More interesting, though, is the graph. PayPal is growing by leaps and bounds, while eBay is stagnating. Not surprising, I suppose, based on my “customer experience”.
As if to emphasize that experience, I received another message:
Account security notice - Immediate action required
To help protect your account from unauthorised use, we’ve reset your eBay password and secret questions. We do this when we believe that your eBay password and username or email address have been obtained without your permission. Currently, eBay doesn’t have any knowledge about how this information may have happened. There are a number of ways to obtain this information. For example, someone may send emails claiming to be from a company that you do business with and then coerce you to reveal your personal information. Visiting harmful web pages that install malware on your computer can also put your personal information at risk.
They've changed their text, or maybe this is code for something else. Maybe it's just to punish me for having such bad thoughts about them. Previously this happened when I accessed their web site in a manner which, presumably, they didn't like. This time, based on the sender timestamp, it happened while we were walking the dogs, definitely not close to any action I took on the site. Could it be that somebody is really accessing my account? I have my suspicions about klearview_au, about whom I have heard of other fraud, but it seems unlikely that they could access my password information, which is only stored at home.
And the “secret questions”? I've ranted about their stupidity in the past, not just on eBay's part: “Who do you work for?”, “Who is your best friend?”, and “What is your favorite vacation spot?“. What good are they? I have never had the pleasure of using them. I can post them here because the answers are in no way related to the question, which I consider as a key. And now I have one more password that has nothing in common with the others except that it reflects my opinion of the company.
|
Ballarat yet again
|
Topic: health | Link here |
Off into Ballarat for the fourth time this week, Yvonne for physiotherapy, and I to the dentist. In each case, there wasn't much to report. It looks like this will be a weekly occurrence for Yvonne for a the next few weeks.
| Saturday, 3 February 2018 | Dereel → Sebastopol → Clarendon → Dereel | Images for 3 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Skill at arms
|
Topic: animals, photography | Link here |
Off to Unicorn Park in Clarendon, near Scotsburn, where there was a “Skill at arms” clinic today. Yvonne had been enrolled, but as a result of her accident she wasn't able to participate, and she gave her place to Amber, a recent friend, and went herself as a photographer, armed with her Olympus E-PM2, my E-PM1, the latter with Leica Summilux 25 mm f/1.4 and M.Zuiko Digital 45 mm f/1.8. It would seem that she's gradually making friends with those two lenses. Photos to come.
|
Down to one TV
|
Topic: multimedia, general | Link here |
A few days ago I put up our old, flaky TV for giveaway on Freecycle, and I had a couple of people interested in it. The first had planned to pick it up today, but I heard nothing from him, so when somebody in Sebastopol showed interest, I tried to contact him by email and phone, but failed on both counts. OK, we're going to town anyway, so why not take it with us? The place was in Albert St, the main road, so it was really no difficulty.
Finally only one TV!
|
Building priorities
|
Topic: politics, opinion | Link here |
Found by Yvonne:
|
Kangaroo: the other paw
|
Topic: animals | Link here |
On Thursday, Nikolai found a foreleg bone from the kangaroo that died last month. I had thought it was this one:
|
|||||||||||||||||||||||
But no, today I discovered that it was still there—some of it, anyway:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It was really jammed in tightly, and Niko wasn't able to extricate it:
|
|||||||||||||||||||||||
|
Four for dinner
|
Topic: food and drink, photography, opinion | Link here |
Saturday evening dinner tonight, with Chris Bahlo (late as usual), but this time also with Amber. So I took a panorama, which failed. I didn't close the space between Chris and myself, with disastrous results:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The discontinuity is at the top of the painting between Chris and me. I should be able to fix that by adding the missing part from another series, but it'll be challenging.
What did we eat? Oh. Smoked salmon, “Giros” and some of the cake that Yvonne got for her accident.
| Sunday, 4 February 2018 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Quiet day
|
Topic: general | Link here |
The last week has been particularly active. I've taken over all of Yvonne's “workload” in addition to my own, and we've been in Ballarat almost every day. Finally a day of rest.
|
Ports maintenance
|
Topic: technology, opinion | Link here |
For some weeks now I've been getting automated reports of build failures of enblend on the FreeBSD test boxes:
OK, it behaved as advertised, so I didn't pay much attention. But now I have a bug report about the same thing, with the additional information that this was with clang 6.0, something that I haven't been able to find in the automated report.
This has all the hallmarks of a C++ incompatibility. The source code hasn't changed for a year, and then it was only a patch. It must be the compiler.
Dammit, once upon a time it was possible to understand the syntax of a program. What we have here, as far as I can see, is all inside vigra—all we have done so far is read in the header files—and “it works for me” with an older version of clang. I'm beginning to develop a deep hatred of clang and C++: clang is so concerned with itself that it accounts for the majority of the buildworld target when building a new system, and it seems so sensitive that each new version causes some incompatibility.
And I learnt C++ nearly 30 years ago, and gradually came to the conclusion that it didn't offer what it seemed to, so I reverted to C. I discussed this in this diary nearly 14 years ago, but since then it has only got worse. So now I need to climb in to the maze of twisty little header files, all different, and work out what this version of clang doesn't like.
First thing was to try it on my -CURRENT box:
So 6.0 is clearly something very new. In the Ports Collection?
Sigh. This is a machine that I have kept as close to the original installation as possible, just upgrading from source every so often. What went wrong here? How do I fix it? I think the obvious thing is to grab a recent ISO image from the ftp site and install that. But why is this all so painful?
By coincidence, there's a new version of Hugin available, 2018.0.0. Or is that 2018.0? It proved to be the latter, a POLA violation, since the previous version was called 2017.0.0, and the pathname for the source tarballs included parts of both schemes. Built it anyway. Apart from that and the deliberate breakage in not allowing to build the binaries in the same directory (that's easily patched, not nearly as easy to convince T.Modes that it's breakage and that portable software doesn't need patches), it built with no problems.
And then in the evening I received confirmation: yes, the name was wrong, should have been 2018.0.0. Fixed. So tomorrow I can do it all over again.
| Monday, 5 February 2018 | Dereel | Images for 5 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More network investigations
|
Topic: technology | Link here |
It's been a few days since I've updated my network statistics page, and it's now showing a very clear pattern of congestion from 18:00 to 24:00. There are still a number of loose ends, in particular the question of whether ICMP packet loss has any particular relationship to TCP packet loss.
Well, netstat can show TCP packet loss:
=== grog@eureka (/dev/pts/5) /usr/src/usr.bin/netstat 161 -> netstat -p tcp -T
That's only a small part of the total, omitting sockets that function normally. OOORcv are out-of-order receives, in principle harmless, but Rexmit (retransmit) implies a packet drop. It's interesting to note that there are dropped packets on local NFS connections too.
Clearly netstat must get this information from the kernel somehow. Spent some time looking through the code, with the rather confusing result that I couldn't find where it got the information from. It must be done by some library call, and I haven't found out where yet. But that's clearly the way to go to find real-world statistics.
Also ran speedtest, for which Aussie Broadband has its own server, 180.150.1.126. Results so far:
| Time | Latency (ms) | Jitter (ms) | Download kb/s | Upload kb/s | ||||
| 13:37:33 | 20 | 16 | 23358 | 4514 | ||||
| 13:39:25 | 21 | 3 | 23573 | 4569 | ||||
| 13:41:54 | 24 | 20445 | 4530 | |||||
While not spectacular, that's acceptable for a 25/5 Mb/s line. Now to wait for the quality to deteriorate.
|
Planting creepers
|
Topic: gardening | Link here |
It's been a whole week since I picked up some cuttings in Ballarat. Kai Peters tells me that this one is Campsis radicans:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Clearly it's high time to take them out of the water and plant them in soil. We'll see how they far.
|
Leonid sick?
|
Topic: animals | Link here |
Leonid vomited copiously a couple of times today. The mass included much grass, bones (presumably from chicken frames), and sinew (presumably from beef). Apart from that, he looked fine, but he wasn't hungry in the evening. It won't do him any harm to go without food for a day, but clearly we need to keep an eye on him.
| Tuesday, 6 February 2018 | Dereel → Sebastopol → Dereel | Images for 6 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More network problems
|
Topic: technology | Link here |
Into the office this morning to find that my packet loss rates had gone through the roof round midnight. Looking at the raw data, I found packet drop rates of up to 75%, and several TCP sessions had timed out. This doesn't match the regular pattern I had been seeing so far. But why don't Aussie Broadband and the National Broadband Network know about this stuff? It seems that neither of them have monitoring in place.
Time to complain? No, I don't like those packet drop rates for a different reason: they're implausibly high. Even 2% or 3% causes severe performance issues, but I'm not seeing anything that severe. Time to look at netstat again and coax it to produce useful real-world (TCP) packet loss information.
|
Car repair, finally
|
Topic: general | Link here |
Chris Bahlo had time to bring me home from town today, so in to Ballarat Automative with the Hyundai Elantra, in the process starting it for the first time in nearly 2 months.
It ran well enough, and despite temperatures similar to last time I drove it, the air conditioner worked normally. Did it maybe cut out last time because of overheating? In any case, when I got to Sebastopol, there was coolant all over the place:
|
||||||||||||||||||||||||
Where did it come from? There's nothing obvious, and it was spread from one side of the car to the other. Paul thinks that it might be the radiator header tank, and that it might not be that expensive to repair, so I'll wait for his verdict.
|
Leonid still not well
|
Topic: animals | Link here |
Leonid behaved fairly normally today, but in the evening he didn't want to eat at all. Clearly there's something wrong there—constipation?—and so we made an appointment with the Golden Plains Vet Practice tomorrow morning.
| Wednesday, 7 February 2018 | Dereel → Bannockburn → Dereel → Ballarat → Dereel | Images for 7 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Leonid much worse
|
Topic: animals | Link here |
Up in the morning to find Leonid looking much worse than yesterday evening. He had bled from some orifice overnight, he had foam at the mouth—I think because of dehydration—and he didn't want to get up. We had an appointment in Bannockburn, Victoria at 10:00, but in view of his condition we called up and arranged to bring him immediately, still only arriving at 9:20. Bridie Bond, now a qualified vet, took a look at him. By that time he was looking a lot better, almost normal, and it was difficult to tell what was wrong, though she established that the bleeding was from the anus, and that it was not fresh, suggesting it had come down from further up. She called in Greg Coates, another vet who isn't on their web page, and he thought it might be haemorrhagic gastroenteritis, apparently related to Clostridium, a genus I knew from Clostridium difficile. Since he hadn't passed any faeces since Monday evening, we considered a blockage, so they arranged to keep him, hydrate him and take some X-rays.
We're a little concerned that the problem may be related to his diet, in particular the raw chicken frames. There were clear bits of bone in his vomit on Monday, and I was concerned that he might have a blockage or a perforation. Greg noted that there had been a recent study from some institution in Melbourne that concluded that chicken was bad for dogs.
OK, what to feed them? They don't really like the chicken frames anyway. Later on, picked up the week's beef from Whitey's in Sebastopol, and spoke to Tim Winter, who tells me that the study from Melbourne was politically motivated to discourage the creation of some chicken meat processing facility in central Victoria. He also told me that he sells about 250 tons of chicken frames a year (that's 5 tons a week!), which is also interesting because they cost $8 for 10 kg, while Yvonne has been going to Davis Wholesale to pay $10 for what appear to be identically the same things. Left with 5 kg of beef cubes and food for thought.
Back home I looked up haemorrhagic gastroenteritis, which proved to be (marginally) related to Clostridium perfringens, though the symptoms described there are only marginally similar to what Leo has.
Call back from Bridie in the late afternoon: he's better, but not good, and they wanted to do more X-rays with Barium sulphate to confirm the lack of blockage that they didn't find with the first X-ray. And for that he'll have to stay overnight.
|
Shopping again
|
Topic: health, food and drink, opinion | Link here |
Later in to Ballarat to take Yvonne to the physiotherapists and to do the shopping. Most of that went pretty quickly, except for one item: black bread. Yvonne told me on the phone that it's in the “European” section, but I couldn't even find that, only “International” (as opposed to “Mexican”, meaning “Tex-Mex”, and “Asian”, meaning “East Asian”). But I couldn't find any there:
|
|||||||||||||||||||||||
So I asked. No, she had never heard of black bread, but it would be with the breads:
|
|||||||||||||||||||||||
No, nothing there. Assistant went off and asked. Came back: no, no idea where it could be. Is it Dutch, maybe? No, not really. Something like Pumpernickel? Well, yes, and in fact some of the stuff we bought had been incorrectly labeled pumpernickel. Ah, that's another matter. That's Dutch, so it's with the Dutch food:
|
|
|||||||||
That's a crop from the first image, wedged between Tex-Mex, (presumably fake) Turkish Delight, Italian crackers, Easten European “pretzels” and some varied Dutch cakes.
What's wrong with this picture?
It took over 15 minutes to find the stuff. That was longer than the rest of the shopping put together.
Bread is bread. They have plenty of it. Why put just one kind somewhere else?
Pumpernickel isn't Dutch, as the Dutch Wikipedia confirms: it's German, specifically Westphalian
They have “German” bread in the main bread aisle. Admittedly, it's not very authentic, but the name is intended to look German. Here from the second image:
|
|
|||||||||
„Bürgen“ is a German word meaning “guarantors”. I suspect a misunderstanding lies behind that choice of name.
|
More network congestion
|
Topic: technology | Link here |
My network performance is still not good, though it's not as regular as I thought. But how do I report it? I'm not happy with the packet drop measurements I have been making, and so far I haven't got round to hacking netstat to produce the kind of information I want. But there's a standard test that every [IR]SP seems to use: Speedtest. So updating Monday's measurements, I now have
| Date | Time | Latency (ms) | Jitter (ms) | Download kb/s | Upload kb/s | |||||
| 5 February 2018 | 13:37:33 | 20 | 16 | 23358 | 4514 | |||||
| 5 February 2018 | 13:39:25 | 21 | 3 | 23573 | 4569 | |||||
| 5 February 2018 | 13:41:54 | 24 | 20445 | 4530 | ||||||
| 6 February 2018 | 9:31:30 | 32 | 40 | 6239 | 280 | |||||
| 7 February 2018 | 19:19:13 | 25 | 4222 | 4026 | ||||||
| 7 February 2018 | 22:16:08 | 30 | 74 | 7783 | 1298 |
It's difficult to correlate these values with my graphs, but it's clear that the last three results are unacceptable. Yesterday's measurement (6 February) corresponds with a particularly bad spike in packet loss and latency, while the other two were in the evening and may be typical for the time. But what's causing it? Should I care? Maybe: it's good to be armed against spurious explanations, most mainly (strangely) blaming me.
| Thursday, 8 February 2018 | Dereel → Bannckburn → Dereel | Images for 8 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Stuck sprinkler
|
Topic: gardening, general, technology, opinion | Link here |
Looking out the window this morning when I got up, the garden bed looked particularly wet. Time to reduce the sprinkler time? Not quite: the sprinkler was still going, and had been for at least 3 hours.
Dammit, has my sprinkler program hung or crashed? A known bug issue is that if
the program stops abnormally, it won't turn the sprinkler off. But no, the program wasn't
running, and the sprinkler relay was off. No crash. Power cycled the relay board.
Nothing. The sprinkler solenoid itself had stuck on.
That's not the first time I've had that problem. In fact, this particular solenoid was a replacement for one that failed, purchased conveniently just over 12 months ago, so that the warranty had expired.
Turned off the water and waited several hours for the area round the solenoid to dry out. Nothing obviously wrong with it. Turned the water on again. Sprinkler off, so the solenoid was off.
What will happen tomorrow morning?
|
ALDI Phat Thai
|
Topic: food and drink, opinion | Link here |
This week ALDI had Phat Thai mix on special, two portions for only $0.98:
|
||||||||||||||||||||||
|
||||||||||||||||||||||
So I grabbed ten of them: if they're no good, I can always take them back. Today I tried one. Not bad, but not really more than a single helping the way I eat it, and I'm not sure it's enough even then. Admittedly, I put in 110 g of noodles instead of the specified 75 g. I'll try another some time soon.
|
The power connector conundrum
|
Topic: technology, opinion | Link here |
External disk drives now almost invariably come with a 5 V power adapter that plugs directly into the mains and delivers power to the disk via a 2.1 mm IEC 60130-10 type A connector (and not the 2.5 mm variety). And clearly I get one with every disk I buy.
But where are they? Disks die much faster than power adapters, but I have many more disks than adapters, and it's becoming a nuisance.
OK, off to eBay to buy a few. What do you call them? Wall wart? None found. Power supply? No, that has too many false positives. Finally eBay told me: Power adaptor. But even then, just about every adapter had a different connector, mainly ones I have never seen. I still haven't found something that I thought you could find on every street corner.
|
Picking up Leonid
|
Topic: animals, opinion | Link here |
Call from Greg Coates this morning, saying that Leonid was doing well, and that I could probably pick him up in the afternoon. That proved to be the case, so off to Bannockburn in mid-afternoon, where Greg showed us the numerous X-rays that he had taken, not all of which looked as good to me as they did to him. Asked for him to send them to us, but that doesn't seem to be normal any more—he suggested taking a photo with a mobile phone, but promised how to investigate getting us the originals. Also discussed Pene Kirk's suggestion that it could be pancreatitis. No, counterindicated by the blood.
Back home, Leo was pretty much as normal, and showed (not surprisingly) quite a hunger. Looks like we're over the hill, but arranged to take him to Pene Kirk tomorrow morning.
|
Promotional recipes
|
Topic: food and drink, opinion | Link here |
At ALDI recently I found a sheet containing some promotional recipes. I showed them to Yvonne, and she chose this one:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Look at the ingredients: Lee Kum Kee Panda Oyster Sauce, Lee Kum Kee Premium Soy Sauce, Lee Kum Kee Sesame Oil (for which I can't find a URL on their broken site). Clearly intended to promote their products.
But otherwise the recipe didn't look bad, if a bit simple, like a simplification of my sweet and sour fish recipe. I modified it a little, and added more of the sauce ingredients. It didn't taste bad, but there wasn't enough sauce! That's really surprising; I'd expect them to want more, not less. But then, maybe I don't understand the terms tbsp and tsp, or maybe I interpreted them incorrectly.
|
|||||||||||||||||||||||
I haven't made a page for this recipe. I think that by the time (if ever) I decide to keep it, it will have changed beyond recognition.
| Friday, 9 February 2018 | Dereel → Cape Clear → Dereel | Images for 9 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Examining Leonid again
|
Topic: animals | Link here |
Leonid is looking a lot better. He was still hungry after eating last night, not surprising in view of how little he had eaten over the last few days, so I gave him some more of the “sensitivity control” food that we had got at the vets. He didn't eat much of it, possibly because he wasn't used to eating at that time of the morning. After that off to Cape Clear this morning to show him to Pene Kirk, who confirmed that she thought he might have had a bout of pancreatitis, and that bleeding was no counterindication.
In the evening, I gave him some boiled beef and rice, and he ate normally, making an incredible mess. Nikolai came along to tidy up. And then I realized why Leo has such difficulty eating crumbs: he doesn't use his tongue, possibly because his canine teeth are too close together.
|
Another laksa
|
Topic: food and drink, opinion | Link here |
I haven't eaten much laksa lately, though I bought quite a number of spice mixes in Geelong last month. One of them was this one:
|
||||||||||||||||||||||
|
||||||||||||||||||||||
In the past I've noticed that there was not much difference between the various pastes. This one is definitely different: it's much worse. It's almost uncomfortably pedas, and has little flavour. I have two more helpings, which I'll try to temper with more coconut milk. And then never again.
|
Polyridiculous chicken dangerous for dogs?
|
Topic: animals, opinion | Link here |
After some investigation, found the report about the dangers of raw chicken that Greg Coates mentioned yesterday, by way of a rebuttal that Yvonne has already come up with. It suggests that the report didn't follow due procedure. Amusingly, the rebuttal comes from Ireland, while the original report comes from Werribee, about as far from the vets as we are. It seems to look at the problem from a different aspect: what role does Campylobacter play in acute polyradiculoneuritis? What I see is:
“Twenty-seven client-owned dogs suffering from suspected APN and 47 healthy dogs, client-owned or owned by staff members” took part in the investigation. That doesn't look overly exhaustive.
“In addition, a significant association was detected between dogs affected by APN and the consumption of raw chicken (96% of APN cases; 26% of control dogs)”
This is a declaration of bankruptcy. If they want to investigate the effect of raw chicken, the control dogs should also have been chosen to eat raw chicken. They're not hard to find. As Tim Winter tells me, he sells chicken frames by the ton.
There's no discussion of the prevalence of APN. They write “we are under the impression that ... the incidence of APN is possibly higher in Australia than in other countries.”
“Under the impression”? That's a cop-out. Measurements, please. The ... relate to the consumption of raw chicken, so they're basing this assumption on the apparently higher consumption in Australia. That's upside down. If you want to prove that APN is related to consumption of raw chicken, you should measure things, not gain impressions.
“Fecal [sic] samples were positive for Campylobacter spp. in 13 of 27 of the APN cases compared with 11 of 47 cases in the control group”. Is that such a big difference?
“Sequencing was available in 10 of 13 Campylobacter-positive cases in the APN group, showing that 60% of the samples were classified as Campylobacter upsaliensis and 40% as C. jejuni. In the control group, DNA sequencing was available from 5 of 11 of the Campylobacter-positive cases showing that 80% were classified as C. upsaliensis and 20% as C. jejuni. No C. coli was isolated from any dog in our study.”
It's difficult to see how they can conclude that C. upsaliensis is involved here.
There appears to be nothing in the study that compared different parts of the chicken, nor whether the chicken in question had been declared fit for human consumption (like the chicken frames we can get in Ballarat). The statement “Chickens are a natural reservoir of Campylobacter, where the bacteria colonize the mucosal layer of the gastrointestinal tract and can be transferred between chickens by the fecal-oral route” suggests that they're not affected, but I don't want to just gain an impression.
Based on that, and despite the limited and apparently one-sided control group, the best anybody could say is that Campylobacter appears to be involved in APN. That doesn't make it a cause.
OK, now look at the rebuttal. They find:
Despite claims by the authors, they weren't the first to investigate the causes of APN. Holt et al. did 7 years ago:
... these authors investigated the link between 6 microbes (Ehrlichia canis, Borrelia burgdorferi, Toxoplasma gondii, Neospora caninum, Campylobacter and distemper) with APN and found that T. gondii was the only significant culprit.
OK, the authors of the current study don't need to agree. But to maintain credibility, they do need to refer to the study.
They also criticized the sample size and selection criteria of the animals in the investigation, for the same reasons I stated.
One thing that I didn't catch was that only 48% of dogs with APN had Campylobacter in their faeces. Why not all of them? And how can this information be used to put the blame on Campylobacter?
Interestingly, the discussion diverts to the causes of pancreatitis. Conventional wisdom has it that its due to too much fat. But the article claims that it can also be due to too many carbohydrates, more than the dog's digestive system can handle. Given that we don't give Leo any significant amount of fat (arguably less than he needs), that might be worth thinking about.
|
VicEmergency app
|
Topic: general, technology, opinion | Link here |
Petra Gietz told me that there was a fire in Napoleons yesterday. Did I hear about it? Of course not. I don't have the mobile phone app necessary to get the reports.
But why not? I have a mobile phone, and (presumably) the app is free. So why not install it? Now I know:
First, try to install:
|
|||||||||||||||||||||||
Why does it need access to my photos? That's insulting.
OK, accept and install. After all, I don't have anything of use on my phone.
|
|||||||||||||||||||||||
OK, that can happen, and I don't need those apps, so sure, go ahead:
|
|||||||||||||||||||||||
Why “1 others”? There's plenty of space to mention the name, and clearly from the size it's a more important one.
This thing is enormous! 33 MB! What does it need all that space for? It downloaded at about 70 kB/s, taking nearly 10 minutes. And then:
|
|||||||||||||||||||||||
How I hate software that can't calculate its requirements.
Delete some more stuff manually, try again:
|
|||||||||||||||||||||||
What's that? Not for mere mortals.
OK, how much memory do I have, anyway?
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
How I love fragmented memory! But here I have 238 MB “device memory”, 11.5 GB “USB storage” and 58 GB on the SD card. What's the problem there?
OK, try the Microsoft solution: reboot, retry. Nope. Still not enough memory. I wonder if it's main memory (“RAM”).
In any case, that's yet another indication of how appallingly bad this stuff is. Went to the Toyshop with a real browser and vented my spleen.
|
|
|||||||||
And then I read what others had written:
|
|
|||||||||
I suppose it's nice to know that—for once—others agree with me. I wonder who the anonymous person at top right was. It seems that he might have been favourable, possibly somebody on the team who wrote it.
|
Crickets up close
|
Topic: photography, animals | Link here |
The other day I found a cricket walking across the floor of the bedroom. OK, collect it, put it in the fridge to cool it down and slow it down, and take some macro photos.
Unfortunately, the cricket died. Why? The temperature in the fridge was about 5°, which a cricket should be able to survive with no problems. Am I wrong in that assumption? Or did it run out of air or moisture? Or would it have died anyway? One way or another, I didn't have to worry about it moving while taking my focus stacks. Here a couple:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
TV for Bruce: no go
|
Topic: multimedia, general | Link here |
Sad request on Freecycle today: Bruce, the bloke I gave the TV to on Saturday, is looking for a 26" analogue TV. Clearly he didn't get the other working. In the meantime I had heard from the original person interested in it, so I sent Bruce a message asking if he wanted to get rid of it. No answer yet.
|
X-rays for Leonid and Zhivago
|
Topic: technology, animals, opinion | Link here |
While I was in Bannockburn, I asked Greg Coates for copies of the X-ray files, something that they had done in the past. He seemed confused, and suggested taking photos of the screen of his laptop with a mobile phone! I explained that this had been done in the past and suggested that he send me the files as email attachments. He said that he would see what could be done.
Later I received a mail message from Charyn, surname unstated, giving links for the images, starting with http://itxviewer.com:8080/itxviewer/index.html. No response from the server. nmap showed:
So why did they give me port 8080? Does the site only respond to specific IP addresses? Then I saw the reverse domain: googleusercontent.com. No, they wouldn't do that, would they? I puzzled for quite some time and came to the conclusion that it could be one of these silly local fake domains, like I've seen elsewhere: they create an RFC 1918 private network and attach a host name to it that mirrors a real address on the Internet.
In the meantime, though, I did the more logical thing and asked again to have the images attached to an email address. And how about them, that's what I got, sort of:
|
||||||||||||||||||||||||
Look at the Exif data:
That was taken with a mobile phone, just what I didn't want. Why can't they just send me the images? I'm sure it's not because they're unwilling. They just don't know how—maybe. Or they think that I won't be able to process the original images, which—Greg says—are in some format that I had never heard of, and which I have forgotten. But I'm sure I'll find image conversion software for it.
The real problem, I suspect, is a general lack of understanding of basic operations like copying files. That's why we now need to move data to the other end of the world before bringing it back.
| Saturday, 10 February 2018 | Dereel | Images for 10 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Jägerschnitzel revisited
|
Topic: food and drink, opinion | Link here |
Chris Bahlo had expressed interest in Jägerschnitzel earlier in the week, so that's what we had for dinner tonight.
Problem: we don't have a really good recipe. Last time, coincidentally almost exactly a year ago, I used a recipe that I found on line, but this time I had the desire (probably encouraged by Chris' recollection of the butcher near the railway station in Gerolstein, where she used to go after school) to have more interesting mushrooms. To my surprise I found that Woolworths have fresh oyster mushrooms—not exactly what you'd find in a German forest—so I chose them.
Quantities? The old question. The recipe I used last time wanted 500 g mushrooms („Champignons“, i.e. normal cultivated mushrooms) and 4 pork schnitzel of unspecified size. We had 7 Schnitzel, weighing 650 g, and for that 500 g mushrooms seemed excessive. We had only 150 g oyster mushrooms:
|
|||||||||||||||||||||||
After frying, they diminished a little, so in the end I reconstituted 15 g of dried porcini and added them.
And the sauce? Again, the recipe looked dubious, so I winged it. 125 g old gravy from last week, 125 g sour cream. And how about that, it worked well. The only issue was that it really wasn't enough for our 7 schnitzel. But that was no problem either, because the schnitzel were too much for us, and the sauce was just right for the 5 that we did eat.
|
Feeding Leonid
|
Topic: animals, opinion | Link here |
I've been preparing special cooked food for Leonid until he gets over his bout of whatever it was. Today it was minced beef and rice. He doesn't handle it well:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's all he managed: he just bites, and because his snout is angled, he can't get it all in. Instead I had to call in Nikolai, who took over after Leo was finished:
Image title: Nikolai eating 6 Dimensions: 4608 x 3456, 2390 kB Make a single page with this image Hide this image Make this image a thumbnail Make thumbnails of all images on this page Make this image small again Display small version of all images on this page All images taken on Saturday, 10 February 2018, thumbnails All images taken on Saturday, 10 February 2018, small Diary entry for Saturday, 10 February 2018 Complete exposure details
Image title: Nikolai eating 7 Dimensions: 4608 x 3456, 2443 kB Make a single page with this image Hide this image Make this image a thumbnail Make thumbnails of all images on this page Make this image small again Display small version of all images on this page All images taken on Saturday, 10 February 2018, thumbnails All images taken on Saturday, 10 February 2018, small Diary entry for Saturday, 10 February 2018 Complete exposure details
Image title: Nikolai eating 9 Dimensions: 4608 x 3456, 2343 kB Make a single page with this image Hide this image Make this image a thumbnail Make thumbnails of all images on this page Make this image small again Display small version of all images on this page All images taken on Saturday, 10 February 2018, thumbnails All images taken on Saturday, 10 February 2018, small Diary entry for Saturday, 10 February 2018 Complete exposure details
He uses his tongue, something that Leo doesn't seem to do at all.
| Sunday, 11 February 2018 | Dereel | Images for 11 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Disk power supplies: the solution
|
Topic: technology, opinion | Link here |
I've spent several days looking for power adapters for my external disks. I've found a couple on eBay, but they all have connectors with 2.5 mm internal diameter (“guaranteed to fit both 2.1 mm and 2.5 mm sockets”). I know better.
But then it occurred to me: ask Chris Bahlo. And sure enough, she had exactly the power adapter I was looking for, on indefinite loan if not to keep. Finally I had enough power adapters for all my disks. And then I discovered:
Dammit, it was the disk that Rani destroyed two years ago. So once again I can't back up my TV shows. Still, that means that there's a companion 4 TB disk that's still OK: these were my photo backup disks in the days when I could get them all onto 4 TB. The other one must be at Chris' house. Mañana.
|
Pork schnitzel: boring
|
Topic: food and drink, opinion | Link here |
We had two pork schnitzel left over from yesterday, enough for one portion. Yvonne was happy enough to eat fish, so all I needed to decide on was how to prepare it.
I couldn't come up with anything interesting. In the end, I made a Jägerschnitzel again, this time with ordinary mushrooms. It didn't really taste much different.
| Monday, 12 February 2018 | Dereel | Images for 12 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Backups and image integrity
|
Topic: technology, photography, opinion | Link here |
Over to Chris Bahlo's house today to get my backup disk number 6, which I used for my photos from November 2013 until March last year, when it became too small (only 4 TB). Since then it's been sitting on a shelf at Chris' place.
That's just what I need for the video backup I've been working on for the last week or so. OK, newfs and start backing up.
But what about the content? It contained are all my photos up to a year ago. Yes, I have three other copies, but what if they're all corrupt? How can I even tell if they're corrupt at all? First attempt: give up and just compare size and modification timestamp:
=== root@teevee (/dev/pts/8) ~ 3 -> find . -type f|xargs ls -l|awk '{printf "%s %s %s %s %s\n", $5, $6, $7, $8, $9}' > /eureka/home/var/tmp/disk6-files
Do that on both disks and compare the output, something like:
Problem: the files weren't in the same sequence. OK, sort them. But somehow there were still far too many differences. And just because the EOF and modification timestamp are the same doesn't help: I've seen disk problems that have replaced whole blocks of data with binary zeroes. The obvious thing is to create checksums when you read them in and compare them from time to time. Too late for that now, of course, but at least I can compare checksums.
Started that, but it's slow. After a couple of hours, it became clear that I wouldn't get finished today.
Dammit, I didn't find any errors. Probably the stuff's OK. If not, I still have the remains of disk 7. Time for a newfs.
How big is the average file on this disk? It proved to be round 400 MB (after all, most of the data is videos). So:
=== root@teevee (/dev/pts/8) ~ 4 -> newfs -g 400000000 /dev/da0p1
=== root@teevee (/dev/pts/8) ~ 5 -> mount /backups/; df -i; umount /backups/
That seems like a lot of inodes. Was it right? Tried a couple of others, and discovered that newfs doesn't pay any attention to the average file size when calculating the number of inodes.
=== root@teevee (/dev/pts/8) ~ 6 -> newfs -i 400000000 /dev/da0p1
=== root@teevee (/dev/pts/8) ~ 7 -> mount /backups/; df -i; umount /backups/
That looks better. And a little bit more space—only by comparison: it's a difference of 110 MB. I wonder if the -g flag does anything in addition.
Then off to back up the /spool file system. Problem: teevee doesn't have any USB 3 ports, and so it ran at a snail's pace, not quite 30 MB/s. How about that, for once the network is faster: plug the thing into eureka and back up across the network, where I should be able to get speeds approaching 100 MB/s.
No such luck. For whatever reason, eureka doesn't want to recognize this disk as a USB 3 disk, so there was no difference. OK, back to teevee. When will the backup be finished? Tomorrow, hopefully.
|
What bird?
|
Topic: animals, photography | Link here |
Walking past Fiona Drayton's (épouse Mitchell) property, saw a couple of birds, interesting mainly because they were the only two, and they appear to be of two difference species:
|
|||||||||||||||||||||||
|
|
|||||||||
What are they? One must be an Ibis, but which? And once again I was caught with the wrong lens. I really should pay more attention.
Margaret Swan tells me that the big bird is an Ardea pacifica, or white-necked heron, common to wetlands throughout Australia.
|
Camera and lens sizes
|
Topic: photography | Link here |
Olympus has brought out a new camera, the E-PL9. It seems only yesterday that they brought out the E-PL8, itself not much of an increment on the E-PL7. For the fun of it, I tried comparing the E-PM2 (Yvonne's camera), the E-PL7 and the E-PL9:
|
|||||||||||||||||||||||
The good news is that the E-PL9 isn't that much bigger than the (out of production, without a successor) E-PM2, so it would be a potential replacement for the E-PM2 when it dies.
Also, out of interest, a comparison of 25 mm lenses (f/1.8, f/1.7, f/1.4 and f/1.2). The size difference is amazing:
|
|||||||||||||||||||||||
And here three 17 mm lenses (f/2.8, f/1.8, f/1.2):
|
|||||||||||||||||||||||
How much difference is there between the E-PL7 and the E-PL9? Very little, it seems. They have the same sensor resolution. The E-PL9 does 4K video, and it has better image stabilization. It also has a built-in flash, probably the best thing from my point of view. Maybe there's something else that I haven't noticed.
But the E-PL9 is missing one thing that the others have: the ability to connect a viewfinder like my VF-2. Clearly the jack (under the flash shoe) is expensive, but the ability to connect the viewfinder is worth quite a bit, to me at any rate.
|
|
|||||||||
Maybe they expect modern people to use a mobile phone instead.
|
The demise of the file
|
Topic: technology, opinion | Link here |
More and more my objection to “modern” digital devices is pointing in one direction: inability to manipulate files. Thirteen years ago I grumbled that this new firefox browser had lost editing functionality. More recently I've been grumbling about inability to move data from one place to another without going via a server, more than likely at the other end of the world.
Today I had another example, also involving another rant favourite, Android. I received an SMS message from somebody, something that I don't see in a timely manner. And I wanted to answer it. With an SMS? Heaven forbid! That's what email is for.
So how do I get the message to my email? After two hours of trying, I still don't know. People on IRC helped a bit until they got fed up with my grumbling. “Cut and paste it!”. That's a declaration of bankruptcy! But I did learn how to cut and paste, at least entire texts.
The obvious thing to do would be to go to the messaging app and select “forward to:”, specifying an email address. But that's an old, worn-out magic word, and modern messaging apps (not in the Internet space, as Callum Gibson pointed out) want no interoperability with the Internet.
OK, cut and paste into my mail app. That worked, modulo the pain of toy keyboards, and it told me that it had sent it. But no message. The “outgoing mail” log (if that's the word) told me that nothing had been sent. This appears to be an issue with my mail app, though it's not clear what's wrong.
OK, there must be apps to do this. One installed and accepted my email address, but it wanted to forward only via mobile phone data. Maybe it can only forward to phone numbers, and didn't parse the email address enough to realize that something was wrong.
Two other apps seemed better, at least for future SMSs. But they wanted to send only to specific email addresses, and to use my Gmail address as the sender. Somehow this is confusing. I want to send it to me, not from me.
Finally I had something that seemed like success:
Just what I wanted! And the message content? HTML, of course:
And that's all! Well, no, there were two of them. Was this an SMS? Probably not, just the notification that I had configured it (twice, because it didn't seem to do anything). Probably it will only forward future messages.
So: how do people in the Real World do this? My guess is the same way that I have chosen: give up.
| Tuesday, 13 February 2018 | Dereel | Images for 13 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More sprinkler problems
|
Topic: gardening, general, opinion | Link here |
Our Salix babylonica is looking less than happy. Granted, it's planted (deliberately) in an inappropriate place in the entrance, but in the last few days the leaves have dried up. Sprinkler problems?
Yes! And they were my fault. After testing the sprinklers last week, I had forgotten to power on the sprinkler controller again. That's something that I should guard against in the sprinkler program. As a result, though, nothing had been sprinkled for 5 days.
Turned it on again, and discovered that the solenoid is still sticking. Damn! That certainly doesn't fulfil the “durability” requirement of the Trade Practices Act. Still more annoyances ahead.
|
DBDRV covers arse
|
Topic: Stones Road house, general, opinion | Link here |
The DBDRV has responded to the letter that I wrote them at the beginning of last month, after only 5 weeks. I had complained about a number of things, including:
There were many more: my letter was 188 lines long. The DBDRV response was shorter:
We confirm that careful consideration has been given to the Owner's submissions and it has been determined that:
And that's all. No mention of nearly all of my issues. What happens if the house burns down because of a fire caused by the inappropriate range hood? What mention of the fact that the assessor just plain didn't do his job measuring the floor levels? But clearly the response itself shows that they consider it normal not to address issues that are raised.
And what about the range hood? One of the issues I had was that he had claimed that the item was not defective because more than three months had elapsed since handover (a date that he stated incorrectly, and which still has not been corrected). That's patently absurd, since I complained 3 weeks after installation. They didn't address this issue, and they didn't remove it, but they made the change from “no defect” to “undetermined” anyway, though the statement about the maintenance period would mean that there was no grounds for any claim. I get the impression that they didn't read the text before changing things.
It was signed by Catherine Clocone, a Legal Adviser. Clearly the purpose of the letter is to ensure that I can't sue them, though I don't see that it has served its purpose.
I'm disgusted. While I don't believe that they are necessarily biased towards the builders, they're clearly incompetent to the point of being worse than useless. This is not the first time that I've seen the Victorian Government agencies demonstrate this level of incompetence. If this is the best they can do, they should close down the entire department.
|
Washing dogs
|
Topic: animals | Link here |
Finally got round to washing the dogs again. It was high time.
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
| Wednesday, 14 February 2018 | Dereel | Images for 14 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Even more extreme macro?
|
Topic: photography | Link here |
Even before I have come to terms with focus stacking, Kev Russell has come up with the next step: extreme macro. The site's difficult to navigate and read, so I took the easy way out and procrastinated. But it occurs to me that there's a big difference between focus stacking done the Olympus way and conventional focus stacking done by moving the camera: the first alters the distance between lens and sensor (v in the lens equation, which Wikpedia calls S₂), while the latter alters the distance between lens and object (u or S₁). That doesn't seem to make much difference, but the depth of field is roughly constant with changing v, while it changes dramatically with changing u.
|
Another dropped eucalypt branch
|
Topic: gardening, general | Link here |
Very windy today, with the result that various eucalypts dropped branches, as they are wont to do. This one clearly fell right across the road:
|
||||||||||||||||||||||||
Looking at where it broke, it's clear that there had been damage for some time:
|
||||||||||||||||||||||||
| Thursday, 15 February 2018 | Dereel → Ballarat → Dereel | Images for 15 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Saffron?
|
Topic: food and drink, opinion | Link here |
Mohamed Ifadir has come up with some strands and wanted to know if they were saffron or not:
|
|||||||||||||||||||||||
My first impression was “yes”, but more careful examination raised doubts. What's that branched strand on the left? Saffron is made from the stamens of Crocus sativus, and they don't have branches. And the end of the strand at bottom left looks like a dried-out flower. It's not the only one. Stamens don't have flowers.
|
|
|||||||||
|
|
|||||||||
OK, take my own photos—more fun with focus stacking. The out-of-camera photos looked like this:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
They both look different. The one on the left is the saffron, and the one on the right is the safflower that I received last July. Neither look very similar. As expected, the saffron had neither branches nor “flowers”. The safflower clearly has flowers, but they look different. Could it be a different kind of fake? Mohamed tells me that it smells strongly and gives him hay fever. Safflower (mine, anyway) smells of dried grass, while saffron smells strongly of... well, saffron.
|
Ballarat yet again
|
Topic: health, general | Link here |
Off to Ballarat yet again for shopping and to take Yvonne to the doctor and physiotherapist. Everything went smoothly, and I even had time to get to Bunnings, do some shopping, and get back only about 30 seconds late, something that I hadn't really expected.
Another sign of the times: the ANZ bank closed down its Sebastopol branch a few months ago, but left the ATM outside in operation, as this photo from Google maps shows:
|
|||||||||||||||||||||||
No more:
|
||||||||||||||||||||||||
I can understand shutting down branches; it's part of the Internet revolution that I described in my paper the future of the Internet, though I missed the issue of specifically banks closing down. But why remove the ATM? That doesn't match the expectations in my paper. Possibly it's a matter of property ownership, and they're planning a new ATM somewhere nearby. But why no sign to tell people what has happened, what is going to happen, and where the next ANZ ATM is? That smacks of stupidity.
| Friday, 16 February 2018 | Dereel → Ballarat → Dereel | Images for 16 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Weeding, not my way
|
Topic: gardening, opinion | Link here |
Mick the gardener along this morning to do some weeding. Well, what I wanted was for him to remove the big weeds so that I could spray the remains. What I had was pretty terrible:
|
||||||||||||||||||||||||
After 1½ hours he had done a reasonable job of tidying up one rose bed by hand:
|
||||||||||||||||||||||||
So I explained again to him that I wanted him to remove the big weeds so that I could spray the remains. Off he went, and when I came back he had cleared part of the lawn:
|
||||||||||||||||||||||||
So I explained again to him that I wanted him to remove the big weeds so that I could spray the remains. Came back later to find him trimming the edges of the lawn. So I explained again to him that I wanted him to remove the big weeds so that I could spray the remains. Came back later to find that he had been weeding the succulent bed, and had done several square meters:
|
||||||||||||||||||||||||
By then it was time for him to knock off. While I was talking with him, pulled out several bushes myself, apparently much faster than he had been doing. Ah, I was lucky and found soil that was a little moist. Five hours, $175, and he didn't do anything of what I asked him, and only got about a quarter of the way through the beds. Apart from the fact that I can't get him to do what I ask, I can't afford this. He's due back next Saturday, but I think it's time to find another gardener.
|
Too many doctors spoil the broth
|
Topic: health, general, opinion | Link here |
Checking with Yvonne about her medication, discovered that she didn't get a renewal for Celebrex, an anti-inflammatory, which ran out this morning. Why not? It proved to have been prescribed by Dr. M. Denton of the Base Hospital, so Paul Smith didn't have any record of it. Call to Health First to talk to Dr. Paul Smith. Sorry, he's not in today, won't be back until Monday. OK, can some other doctor prescribe it for her? No, sorry, not without an examination, and we're full today.
More discussion, in which the receptionist wasn't very helpful. She suggested an appointment on Monday. More discussion. “And so Yvonne spends the weekend in agony? This is an emergency. What do you do in an emergency?”. She suggested that we go to suggested that we go to Eureka Medical Centre! What, doesn't Health First deal with emergencies? Yes, but mumble (I really had difficulties understanding her speech). After a while, I persuaded her to ask a doctor, and they would get back to me.
In the meantime, Yvonne called the Base Hospital, who told her that they couldn't prescribe anything because the case had been handed over to the GP.
Five hours later, at 13:57, I called again and was told that they couldn't do anything because mumble wasn't back from lunch, and she had talked to Dr. Chewey, who was with a patient, and so we would have to wait. mumble was due back in 15 minutes, and their appointments were calculated for 20 minutes each, so it seemed reasonable to expect a call back in half an hour.
Half an hour later I called back again and was connected with mumble, who told me that her name was Gail, that she was a nurse, and that she was having difficulty because the Base Hospital hadn't sent them the records. Called the number specified on Yvonne's next appointment (orthopaedics, 5320-4221), and got the information that they could get the records department to fax the details again. Shortly later got a call back from Gail to say that she had been trying all afternoon to contact the records department, but was only getting voice mail.
There must be something that can be done in an emergency. Called up the main number and asked for the number of the records department, but that's secret. While I was talking to the receptionist, got a call on the other line from Jacinta Smith, a coordinator in some unspecified department, who told me that Health First was being unreasonable, but that she would fax them the letter of 31 January. I asked about the admission records, but no, they don't give that information to the GPs.
Called back to Health First. Yes, they had received the letter of 31 January, but they already had it. What they needed were the admission records, which are normally sent as a matter of course.
Somehow we're going round in circles here. People are so tied up with bureaucracy that they have lost sight of the fact that there's a potential emergency here. Decided to give up and take Yvonne to the emergency department of the Base Hospital. Called Health First again and asked Rebecca to tell Gail that she should call us on my mobile number if she needed us.
Half way to Ballarat, got a call from Gail, who had heard nothing from Rebecca, but had somehow found my mobile number anyway after two failed attempts at home. She had finally found the records, stored in a place she wasn't expecting. So finally we picked up our script round 15:45. A whole day chasing up the prescription!
|
Garden flowers in late summer
|
Topic: gardening | Link here |
Summer is drawing to an end, time for another monthly garden photo series.
The lack of watering last week has taken its toll. The Salix babylonica is looking particularly unhappy:
|
|||||||||||||||||||||||
Hopefully it'll survive. Others are also not looking very good. Here the Alyssum, normally a constant bloomer, last month and then this month:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And the smaller Carpobrotus, which normally blooms in November, is flowering here and there:
|
|||||||||||||||||||||||
Does that have something to do with less water?
The lack of water might also have something to do with the roses and the Grevillea, though that's a little certain. Some of the roses are flowering well, others less so, and others again have died:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The Grevillea bronwenae is looking a bit dry, but certainly surviving:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The Hedychium gardnerianum that we have had for years, and which we had to cut back several times at Kleins Road, is finally flowering:
|
|||||||||||||||||||||||
In the past I had thought it was a Hedychium coronarium, but clearly that's not the case. It's claimed to be an invasive weed; I'd be happy if it would just grow normally.
And we now have all of two Buddlejas flowering!
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The big one isn't flowering very strongly. Do I need more fertilizer? That might be a general issue.
The potted Mandevilla by the front door is flowering nicely, as are the self-seeded Mirabilis jalapa:
|
|||||||||||||||||||||||
I originally had yellow and red flowering Mirabilis, and what I have seen suggests that they will cross-pollinate and create other colours, including stripes. But not here, it seems. I still only have the same two colours:
|
|||||||||||||||||||||||
The other Mandevilla, the frost-hardy one that nearly died in the winter, is finally coming round to flowering again—at the end of summer!
|
|||||||||||||||||||||||
Another plant that only barely survived the winter were the two blue Salvias, now looking quite happy:
|
|||||||||||||||||||||||
The Clematis that I bought three months ago are gradually coming back to life, though they're still looking very much under the weather:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
The Tropaeolums are also not doing particularly well. For plants that should grow no matter what, this one has only maintained itself the way it was a couple of years ago:
|
|||||||||||||||||||||||
And this one, the black-flowering one (if it ever flowers) is clearly surviving, but not much more:
|
|||||||||||||||||||||||
On the other hands, some things are doing better than I expected. We didn't plant any petunias this year, because they didn't do well last year, but one seems to have survived:
|
|||||||||||||||||||||||
Indoors, the Phalaenopsis has almost finished flowering, after nearly 6 months. Just this one flower is hanging on, and it has been the last orchid of summer for at least 3 weeks:
|
||||||||||||||||||||||||
|
Saffron: real or fake?
|
Topic: food and drink, opinion | Link here |
Yesterday's discussion about saffron continued today. Mohamed Ifadir came up with an interesting page with far more detail about the saffron harvest than I had known before. In particular, various parts of the stamen are used for differing qualities:
Firstly, the ends of the stamens are really wider, and there's some reason to believe something might protrude from them, like in this detail of Mohamed's photo that showed yesterday:
|
|
|
|||||||||
And that corresponds to the thickening at the ends of my saffron. But the other ends are different: mine is still relatively thick, while Mohamed's are thinner and have branches (here a later photo):
|
|
|||||||||
The diagram explains this: my saffron is Sargol (or possibly Negin):
Mohamed's appears to be Pushal, including some of the branch between the stamens. For it to be saffron, it must be a three-way branch, which is the case here.
So what's the difference in quality? I'd expect Sargol to be the best, then Negin, then Pushal, and then increasingly dubious parts of the stamen, ending with Konge.
| Saturday, 17 February 2018 | Dereel | Images for 17 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
GPS pain
|
Topic: technology, opinion | Link here |
What time did we pick up the prescription at Health First yesterday? It was somewhere between 15:30 and 16:00, but I didn't note the exact time.
Never mind, I have a GPS navigator—two, in fact—so I can use something like GPS Visualizer to show where I was. But that shows two serious issues. First, the track is extremely inaccurate. In Ballarat, we parked between Eyre St and Dana St, a block south of Health First, and walked north along the west side of Doveton St. When we left (by car), we turned left into Dana St and left again into Dawson St. The track tells a different story:
|
|
|||||||||
That's not even the worst. Somehow I traveled 10 km or so west and east of the house (green pin) without even noticing it:
|
|||||||||||||||||||||||
Surely things don't need to be that inaccurate.
The other thing is that this view doesn't tell me when I was (supposed to be) there. Sure, I can look in the raw log, but that way madness lies. It's in UTC, so I should be looking at the time between 5:30 and 6:00:
And that would tell me, if I could visualize it. Maybe the <speed> specification can help. Those values suggest that, at the very least, we were not in a car. But wouldn't it be nice to find a good visualizer? I'm sure they're out there, but how do I find a good one?
|
Saffron qualities
|
Topic: food and drink, opinion | Link here |
I spent a fair amount of time looking at saffron yesterday, but I didn't get round to looking at all the details. The Flora saffron pages have measurements for their saffron, along with values according to ISO 3632. ISO politely refuses to divulge the values (cross their palms with silver), but Flora gives what are presumably correct details. According to that, we have (ISO specification in brackets):
| Picrocrocin | Crocin | Safranal | ||||
| Sargol | >= 85 (>= 70) | >= 230 (>= 190) | >= 30 (20 - 50) | |||
| Negin | >= 85 (>= 70) | >= 230 (>= 190) | >= 30 (20 - 50) | |||
| Poushal | >= 85 (>= 70) | >= 230 (>= 190) | >= 30 (20 - 50) |
Well, that's helpful, isn't it? All the values are the same. What differs is the spelling. On the page for Poushal, we also have the spellings “Pushali”, and the link itself uses the spelling “Pushal”.
There are others, like “bunched” saffron, of which they say:
The Style of the saffron plant, has less CULINARY VALUE, that means less AROMA, FLAVOR or COLOR. If is left attached to the red stigmas, it adds 30% to 50% DEAD weight to the saffron flower, and you pay for it.
But the analysis? The same as the others. My guess is that the >= is the most important part of the pages.
|
Fixing the sprinklers
|
Topic: gardening | Link here |
One of my garden sprinklers is still sticking, but the people at UPI had suggested that it might just be dirt in the valve. Why that? I have a filter in front of it, and I never had issues like that in Kleins Road. But out to take a look this morning, marveling at the way the valve had been installed so that it was difficult to access the solenoid:
|
|||||||||||||||||||||||
And sure enough, on removal it had a significant amount of rusty sludge on it. Cleaned that off, and it worked. At least one problem solved.
| Sunday, 18 February 2018 | Dereel | Images for 18 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
E-M1 USB problems
|
Topic: photography, technology | Link here |
I read photos from my cameras to the computer using a script called syp. It uses a feature of the Olympus file naming scheme. The names are of the form PMDDNNNN.JPG, where P is an (apparently) arbitrary letter, M is a one-character abbreviation for the month (1 to C), DD a two character representation of the day of the month, and NNNN a number that is incremented for each photo. The P can be changed to just about any character you want, so I have changed it on all my cameras to identify the camera (1 for the E-M1, 2 for the E-PM2, 3 for the E-PM1, and 4 for the E-M1 Mark II). syp analyses the file name and copies files to a directory corresponding to the date (20180218 for today), only if they're not there.
But recently things haven't gone as well:
=== grog@eureka (/dev/pts/26) ~/Photos/20180217 66 -> syp
=== grog@eureka (/dev/pts/26) ~/Photos/20180217 67 -> syp
I didn't change anything between the two attempts, but the USB subsystem had a field day, producing several hundred error messages, mainly repetitive. Here are the interesting ones:
This part is normal: camera connected to computer, configured.
What went wrong here? This all took place in the space of a second, though on other occasions it has happened later. The interesting thing is the apparent collateral damage, including (apparently) 3 keyboards and two mice. But it's not done yet:
That's the same thing all over again. And then, only 2 seconds later,
This time it stayed there, and I was able to read in the files and turn it off:
But what's causing that? It always used to work, and so far I have only observed the problem with this camera, so maybe it's some hardware issue. I'd put my money on cable problems, but the system response is strange.
|
Trying HDR Express
|
Topic: photography, technology, opinion | Link here |
My last attempts with commercial HDR software were with Franzis HDR Projects 4. Not an unqualified success: in fact, an unqualified waste of money. On the other hand, the method I'm using now, using enfuse, doesn't handle ghosting at all well, so it's probably time to look further.
The next one I found was HDR Express 3 from Pinnacle Imaging Systems. I downloaded a trial copy a few days ago, but have only just got round to trying it out.
The view on startup was unexpected:
|
|||||||||||||||||||||||
That's certainly minimal. But it was also completely unresponsive. Where's the manual? I couldn't find it on the web site, only video tutorials, something that I have always found to be a waste of time. So, try one of the undocumented Microsoft solutions: guess. Could it be because I'm using a remote desktop? The problem is that I don't have a display on this machine (dischord), something I should fix.
So: install on euroa, my laptop. Yes, indeed, it displayed something, and carried on doing so when I moved the session to a remote desktop. This may not be as silly as it sounds: I've seen it with other software, and it may be that Microsoft loads some necessary functionality only when it's started on the local display.
So, what do we have?
|
|||||||||||||||||||||||
|
|
|||||||||
As usual, of course, I had to select my directory (P:\HDR\6-HDR4-grog for hysterical raisins). But it didn't seem to like that:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Ha ha, only joking. The important part is the Select Folder at the bottom. And then I had to wait while it read in all the images, made me wait while it displayed them, and, in true HDR Projects fashion, misassigned them:
|
|||||||||||||||||||||||
|
|
|||||||||
That's the second row, as much of the 6 images (see top left) that it felt polite to display. The first three clearly belong together, but the fourth (and invisible fifth and sixth) are completely different. But for some reason (maybe time) it joined them together. OK, never mind, there's a slider “Auto-Stack images” at the bottom, set to 5. All my images are groups of three, so I changed that.
Click Merge. Off it went and came up with something else reminiscent of HDR Projects:
|
|||||||||||||||||||||||
|
|
|||||||||
What a mess! It's far worse than the original. And like HDR Projects, it offered me (far fewer, but still too many) silly ways of modifying it, not including the obvious higher dynamic range.
Round here I got the impression from part of the display that it wasn't trying to merge images at all, just extract the most from one image (though it didn't do a very good job there either). Time to RTFM. And there is a manual, and I could access it via F1:
|
|||||||||||||||||||||||
|
|
|||||||||
The second image is shown at the default size that it chose for a 2560×1440 display. Yes, I could expand the display, though it steadfastly insisted on two pages next to each other. But I wanted to display it on a different machine. Where is the manual? The file name was HDR_Expose_3_QSG.pdf, but in which directory? Adobe Reader politely refused to divulge:
|
|
|||||||||
Still, that's a start. Searched the entire Program Files hierarchy and found only one file: C:\Program Files/Pinnacle Imaging Systems\HDR Expose 3\license_en.pdf. It seems that they have deliberately hidden the documentation to restrict pirating and to significantly complicate legitimate use. Bad Pinnacle.
It seems that they have “Easy HDR Creation”, basically what I did. They don't say how to stop it from attempting to merge unrelated images. So why did it create this real messy image from only one image?
But it seems that there's also a batch processing feature, so tried that. In principle it looks just the same, including perpetual insistence on the directory C:\Users\grog\Pictures: you'd think they had shares in the directory name. (In fact, you can set the directory name in the Preferences, not by entering the name, but by climbing a directory tree to get there, in true Microsoft fashion). Surprise, surprise. It won't output to JPEG (which they call JPG, presumably an opinion of said experts), only to various HDR formats. OK, do that first, ask questions later. After a while it had written a file called (in the real world) /Photos/6-HDR4-grog/e-from-house-0-1EV-0001.bef, and then I got:
|
|
|||||||||
|
|
|||||||||
|
|
|||||||||
It continued and created /Photos/6-HDR4-grog/e-from-house-1-1EV-0001.bef, in the process drawing my attention to the fact that it had ordered the files neither by name nor by modification time, but in the sequence 0, 2, 3, 1. Presumably this is its way of reporting that fact that it couldn't make sense of its choice of lumping the two source sets /Photos/6-HDR4-grog/e-from-house-2* and /Photos/6-HDR4-grog/e-from-house-3* together.
Round about here I gave up for the day. It's possible that it can produce good results, but so far it's just pain.
And the output files?
=== grog@eureka (/dev/pts/1) /Photos/6-HDR4-grog 51 -> l house-from-entrance-0-1EV-0009.bef
=== grog@eureka (/dev/pts/1) /Photos/6-HDR4-grog 52 -> identify house-from-entrance-0-1EV-0009.bef
This output was followed by 10 error messages complaining about the file. Clearly a 35 MB file, even TIFF, can hold a higher resolution than 256×192. Probably that's just a thumbnail. Once again, RTFM—but how, when I can only do it by running a program that is even fussy about where it will display at all?
Is that really the current state of the art? I almost feel like trying HDR Projects again.
| Monday, 19 February 2018 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Oil low, change it
|
Topic: general, opinion | Link here |
Last Thursday I received a voice mail from Lisa Sperber of Ballarat Automotive, telling me that my car was ready, but that <mumble> about the oil. I dropped in on Friday, but at 15:30 they were shut.
Today Lisa called again, during breakfast. The oil in the engine was dirty, and the level was low, so it needed changing. OK, more details please. No, that's enough. But I had the oil changed only a few 1000 km ago. In any case, I was in the middle of breakfast, can I call back? No, it'll only take a moment to check. I was in the middle of breakfast, can I call back? Line went quiet, so I hung up.
Called back later. It seems that the last service was in May 2016. How many kilometres have I done since then? Don't know, but that's enough. Car must be serviced. I pointed out that low oil level isn't a reason for a service: I can just top it up. OK, pick up your car and top it up yourself. <click>. And she hung up on me.
So is a service due? I still don't know. It seems that the last one was done at 165,000 km. What's the current mileage? Lisa didn't check. How often are the service intervals? Lisa says “less than 15,000”, but she didn't give any further details. Since she's so uncooperative, I'll have to find out for myself. Looking online showed a surprisingly complicated web page which, after many questions, told me that I needed an oil change every 7,500 miles.
That's because it's US American, and they still use archaic units. It corresponds to 12,000 km, inconveniently half way between the more plausible 10,000 or 15,000 km. But despite much searching I couldn't find a corresponding Australian page. I'll have to ask CJ Ellis, who has the same model with an instruction manual. And I'll have to check the current kilometres on Wednesday when I go in.
This isn't the first time I've had issues with Lisa. I've already commented on how she can be obstructive, though this is by far the worst I have experienced. They're a strange couple: Paul is competent, friendly and helpful, but just about every time I interact with Lisa, it's negative. It's also clear that the business is not doing well. When he bought it from Vic England, over 8 years ago, it was doing well, and there were two other mechanics working there. Now they're down to Paul and Lisa (who does the accounts and customer interface, fiercely insulating Paul from them), working only 4 days a week.
Why? It can't be the quality or the cost of the work; that's as good as I have had, and it's not like me not to complain about workmanship. My guess is that Lisa has driven them away. Under the present situation, it would be better for her to go home and stay home, and let Paul take over that side of things as well until business picks up. Then he could hire somebody of a more pleasant disposition.
|
More playing with HDR Express
|
Topic: photography, technology | Link here |
My previous experience with HDR Express 3 was not the most positive, but then, it is a Microsoft space program. Maybe I'm trying to do too much at once. Let's take one step at a time.
Fired it up again, this time remotely from euroa, and confirmed once again that I got a black screen. Started it locally and migrated to rdesktop, again with no problems. This time I took the normal interface. And once again it insisted on looking in C:\Users\grog\Pictures. But this time the Preferences screen looked completely different, and in particular didn't allow me to set a default folder.
Tried things anyway. It seems that when you have the contents of the directory displayed on the screen, you're not done yet: you have to select the ones you want (just the opposite of what the batch window does, it seems). And then it will merge things. I tried it with the sequence e-from-house-4-* and got what looked like a quite acceptable result:
|
|||||||||||||||||||||||
The horizontal bar towards the bottom seems to be due to incorrect screen refresh after I finally got rid of an irritating “Preset” bar. But the image looked acceptable, though possibly still a little too contrasty. In particular it was far ahead of my enfuse solution (on the left) regarding ghosting:
|
|
||||||||||
|
|
|||||||||
In particular, Amber's left foot is completely missing in the enfuse version.
OK, save it, and got an output file e-from-house-4-1EV-0003-Tone Map Linear.tif, which lost much of the Exif data. It also blows the mind of my standard viewer xv. OK, use ImageMagick to convert to JPEG:
=== grog@eureka (/dev/pts/1) /Photos/6-HDR4-grog 64 -> convert e-from-house-4-1EV-0003-Tone-Map-Linear.tif e-from-house-4-1EV-0003-Tone-Map-Linear.jpeg
The result? Not what I was expecting. Here again the enfuse output on the left, the HDR Express output on the right:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Why is this such a mess? The output image looks nothing like what I saw on the screen. And, like HDR Projects 4, it's nothing like what you could consider reasonable. Once again I'm left with the feeling that I have to outwit the software to get any useful results.
| Tuesday, 20 February 2018 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Panic!
|
Topic: technology, opinion | Link here |
Updated the system on teevee today, as I do about once a month. The last time was:
Nothing very interesting there, which is why I almost never mention it. Today it finished, I rebooted, went away briefly, and came back in time to see the system displaying:
Huh? I can't recall when I last had a panic with a FreeBSD -STABLE kernel. OK, watch what it does next. Finish rebooting normally, (automatically) start X and...
I couldn't read all that, of course, so in to look at the dump. After all, when it comes to FreeBSD dumps, I wrote the book. OK:
=== root@teevee (/dev/pts/5) /var/crash 2 -> gdb -k /boot/kernel/kernel vmcore.1
=== root@teevee (/dev/pts/5) /var/crash 3 -> kgdb /boot/kernel/kernel vmcore.1
What went wrong there? Really corrupt kernel stack? What appeared on the screen seems plausible enough. And then I found a new file that never used to be there:
=== root@teevee (/dev/pts/5) /var/crash 4 -> l
That seemed to be related to the previous panic, but presumably the panic was the same. Looking inside showed all kinds of goodies, far beyond what I needed, but including a proper gdb trace:
So it seems that since I last debugged a kernel, people have changed the method Yet Again. I had half expected that—that's why I tried two different incantations, but they're now both useless. Time for another RTFM.
That doesn't help much. It looks more like a case of WTFM, or possibly the issue is that this was the version of kgdb relating to the old kernel.
In any case, core.txt.0 (why isn't there a core.txt.1?) told me everything I needed to know: it's related to the nvidia X driver. Whose fault? Difficult to say, but my outstanding bug report means that I'm running an old version of the driver. Does it occur with the newest driver? That'll take a bit of testing. In the meantime the workaround was to revert to the old kernel.
| Wednesday, 21 February 2018 | Dereel → Ballarat → Dereel | Images for 21 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Shopping again
|
Topic: general | Link here |
Off to Ballarat again this morning for shopping, also taking Yvonne with me to go to the physiotherapist's. While she was there, I went on to some furniture shops to see what was available. Now that I have given up on wanting an ottoman, which is called tchéise in Australia, the selection seems much more straightforward. Either a conventional lounge room suite like we have always had:
|
|||||||||||||||||||||||
or an all-in-one round-the-corner job:
|
|||||||||||||||||||||||
On the whole I'm more in favour of the former, but first we need more endless discussions. The other thing that was surprising was how big a difference there was in prices of what appeared to be similar suites, up to a factor of 4.
More fun at Woolworths. Last week I discovered that dried white beans are international food, but it seems that Thai red curry is not. It's fish, at least if it contains any:
|
|||||||||||||||||||||||
|
|
|||||||||
Once again we needed the help of multiple assistants to find where to look for what we were looking for (sprats), which, unfortunately, we didn't find.
|
Car service intervals
|
Topic: general | Link here |
To Ballarat Automotive to pay for the repairs to my car. I can't pick it up until I have another driver with me, and currently Yvonne doesn't fit that bill. But I had already established from CJ Ellis that the oil change interval for the car is indeed 15,000 km. How many had I done since the last service nearly 2 years ago? 5,000!
OK, there's the second half: “or 12 months”. Why 12 months? When I was a lad, the answer was simple: people who do low mileages do short trips, and the engine never warms up properly, so it accumulates moisture. This is exactly what my ancient “Book of the Car” (1970) states.
But what were the time intervals in those days? Round 1 year, I suspect, though at the time oil change intervals were typically 5,000 km. The distance has increased, but the time hasn't. And in any case, I don't do short journeys: I do few trips, but they're typically 60 or 70 km or more. So I decided against, not hindered by the fact that the car is on its last legs anyway.
|
nVidia pain, next step
|
Topic: technology, opinion | Link here |
I've had a lot of pain with the nvidia driver for FreeBSD lately: first the performance bug I experienced last month, and then yesterday's panic. The two are not completely unrelated: as the result of the performance issue, I'm using an old version of the driver. Could it be that only this driver causes problems?
In any case, I had a rather strange request from the person handling the driver bug:
Can we get video showing performance drop when only single display is connected to the gpu?
I have tested with Ubuntu 16.04.2 + 384.111/390.32 drivers + GeForce GT 710 + Unity desktop . I observed these messages are generating in /var/log/Xorg.0.log file when display goes to sleep mode. I put display in sleep mode with command "xset dpms forece off" and checked Xorg logs. I observed below message generating after every second. But I didn't observed perf drop while interacting with desktop. Also I see below message when opened ubuntu applications like Power Management Settings, Abut this Computer, System Settings , Displays etc.
[ 8618.484] (--) NVIDIA(GPU-0): HP 22uh (DFP-0): connected
[ 8618.484] (--) NVIDIA(GPU-0): HP 22uh (DFP-0): Internal TMDS
[ 8618.484] (--) NVIDIA(GPU-0): HP 22uh (DFP-0): 330.0 MHz maximum pixel clock
...
OK, this is clearly a different monitor. In any case the presence of the messages should be enough to investigate. But I suppose I should humour them, though it's a fair amount of work. In this case, though, it would also make sense to see whether the panic still occurs.
So I saved the old driver and installed the latest. Ran as before. No problem! No messages.
Why? It's a new driver, 390.25, released on 29 January 2018, as the archive shows. When I entered the bug report on 21 January, the driver release was 384.111, though the archive now shows that version 390.12 was released on 4 January. Was it really? Potentially the ports maintainer didn't include it in the port, though the change log reports:
The reference to panics in revision 459638 is interesting. Following up the bug report shows that one of the versions in question was 340.104, the version I was running. But what was the panic? No mention, only that it happened about once a day, not quite the same situation.
Yet another race condition, but hopefully I won't need to worry about it any more.
| Thursday, 22 February 2018 | Dereel → Sebastopol → Dereel | Images for 22 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Picking up the car
|
Topic: general | Link here |
Into Sebastopol with Chris Bahlo this morning to pick up the Hyundai Elantra. All was well until I got home, when I smelt coolant. Was it still leaking? A look under the bonnet confirmed that there was little wetness, if any, and after all, it had been raining. And clearly the radiator was new. But there were patches of dried coolant on the radiator cap. Called up and spoke to Paul Sperber, who confirmed that the radiator cap hadn't been changed. So so far all I have to go on is the smell of coolant. Wiped the cap clean; we'll see if it recurs.
|
HDR Express: still no luck
|
Topic: technology, photography, opinion | Link here |
My previous attempts with HDR Express 3 were less than successful. But OK, since I can't comfortably access their manual (it can only be displayed from the running program!), I was going by what seemed to make sense, not necessarily what you should do with Microsoft-based programs.
And then it occurred to me: I had come to this program because of a book, “The HDRI Handbook 2.0”, by Christian Bloch. The obvious thing (apart from the discrepancy in version numbers) would be to follow the instructions there. Went to some trouble (connecting a DVI cable from dischord to my right-hand monitor) to be able to start the program on dischord, something that I hadn't been able to do before because of the rdesktop bug. That way I had a pristine installation.
The book starts off by telling you to take the images (but not to read them into the computer), and then
“Drag and drop all images onto the HDR Express program icon at once.”
Aaargh! What a horrible way to process files! In my case, it was also very difficult, because it meant starting the program, once again using this horrible mouse. But OK, why not?
One reason was that I still haven't worked out how to select all files in a directory from Microsoft. Something to do with Ctrl and Shift along with button 1, some of the time, but it doesn't always work. And then there's Ctrl-A, which also works some of the time. Fortunately I had linked the files I wanted into a separate directory by themselves, so I didn't have to do too much mousework. Finally I got it to work, and it gave me pretty much exactly what I could have done more easily the way I had done before:
|
|||||||||||||||||||||||
Next: “Follow the on-screen prompts and click OK.”
Which ones? Does it mean these buttons? That's all I can see:
|
|
|||||||||
|
|
|||||||||
No OK. Still, Merge seems right, so I pressed that.
“Now you have your first HDR image. That’s the mythical 32-bit file that contains all information from all source images. HDR Express plays a little animation now, ramping through the entire range of exposures”.
Never mind what “mythical file” means. The rest wasn't what I saw. (Mercifully) no animation. All I got was what I had seen on previous occasions, just this time without the information panels on the right:
|
|||||||||||||||||||||||
That's far too dark.
Clearly there was no point in carrying on here with the instructions: they don't work for me. Why? I'm not a novice when it comes to HDR, but somehow this program baffles me, and the instructions don't help.
Further investigation showed what I had discovered before: it seems that it had only processed the first image, to judge from the tab at top left:
|
|
|||||||||
Last time I selected all three images and got a better result, at least on screen. Why didn't the the instructions tell me to do so? OK, try again, first selecting all the images (which causes their icons to highlight):
|
|
|
|||||||||
After that, I got a different result:
|
|||||||||||||||||||||||
Far too bright! But now I have two (and not three!) images, and it seems that it was only showing the highlighted one:
|
|
|||||||||
Why only two? And why is it not showing a merged image? I tried again with Manual instead of Auto, with no obvious change. And I can't see anything else I can do.
This is really amazingly frustrating. What I have so far:
HDR Express can't start properly with a remote desktop, though it runs as well as locally when started locally and then migrated to the remote desktop.
The manual isn't available as a PDF document.
It insists on the directory C;\Users\grog\Pictures even after reading in other directories, and usually I can't find a way to convince it to use any other default. Instead, I have to go tree climbing Every Time to get to my preferred directory.
Despite both the PDF documentation and the book, I have not been able to merge an HDR image after 5 days of trying.
As if to emphasize the point, my next attempt failed with something completely different: the program froze, leaving a broken display:
|
|||||||||||||||||||||||
I had to force stop it to get any reaction.
What do I next? The sensible thing would be to forget it. But I don't want to give up. Next time I'll try the videos, which may be more up to date than the book.
|
FreeBSD: how stable?
|
Topic: technology, opinion | Link here |
Discussion of my recent panic on IRC this morning, and John Marshall expressed surprise that I was running FreeBSD-STABLE on my machines.
Why? That's the most stable version. But not according to John:
Huh? What has he been smoking? But he came up with backing evidence:
Followed up that last quote. It's (currently) in Tracking a Development Branch in the FreeBSD handbook. But that's not the way I recall it. In The Complete FreeBSD I wrote:
FreeBSD-STABLE is an updated version of FreeBSD-RELEASE to which all possible bug fixes have been applied, to make it as stable as possible. Fixes are made on a daily basis. It is based on the same source branch as FreeBSD-RELEASE, so it has all the features and fewer bugs. It may contain additional features, but new features are tested in the -CURRENT branch first.
That's no proof of course, except that my opinion hasn't changed in the nearly 15 years since those lines were last updated. When did the text come into the handbook? Surprisingly, very recently:
But clearly that was first entered some time (months? years? decades?) before, so that doesn't help. Digging in old files I found, also in the handbook:
In those (far) days, the handbook was structured differently, and this file was:
So why the change of heart? Also, as some of the comments show, the change is not completely accurate. Bug fixes get applied to -STABLE, but not to -RELEASE.
| Friday, 23 February 2018 | Dereel | Images for 23 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Offensive code of conduct
|
Topic: technology, opinion | Link here |
It's no secret that I'm opposed to “codes of conduct”, specific interpretations of what the codifiers see as good behaviour. Apart from the fact that there are too many of them, all with loopholes and inconsistencies, they shouldn't be necessary in the first place. Six years ago I decided against attending linux.conf.au (seasonal URL) because of their insistence on a code. Two years ago I noted a perceived issue with the FreeBSD Code of Conduct, requiring (in their view) an overhaul of the code.
Now it's there, and we had a long discussion (as it happened, started by myself, though I didn't stay in the thread for long). It seems that there's a distinct difference in attitude between the USA-centric members of the project and the rest, though it's worth noting that the head of the code of conduct committee was Benno Rice, who comes from Adelaide, and who was one of the few participants of my first Hacker's barbecue, 17 years ago. Maybe he has been in the USA too long.
Today two committers took such offence at the code that they resigned. This was private mail, so I won't mention the names:
I stop participating in FreeBSD project because of such stupid things as
https://www.freebsd.org/internal/code-of-conduct.html
Until this is removed or changed. For example, the current CoC can be shown
in case of a negative question "Are you 12 and/or your upbringing and
intellect is OK?".
and
FreeBSD is more care about SJW than resolving conflicts. I stopped doing
anything and visiting events after conflicts with bsdgirl.
Thank you members who was helping me to solve issues. Happy to see you in
real life without strange restriction by CoC. Please keep my account, i
might get back on board in future.
In passing, I have learnt a new TLA: SJW.
In some ways this is similar to my decision not to participate in linux.conf.au 6 years ago. I wonder how many others feel the same way (but not enough to read the endless mail threads). As it is, 121 ports in the Ports Collection are now looking for a new maintainer.
In passing, it's interesting to note some of the language used in the code of conduct: “misgendering”, ““dead” or rejected names”, ““outing” of any private aspect of a person's identity”. I don't really know what any of those mean. “Don't cause pain to people” should be clear to everybody, and it seems to encompass the entire code of conduct.
|
Understanding lens caps
|
Topic: photography, opinion | Link here |
When I bought my first Leica lens in nearly 50 years, the Summilux 25 mm f/1.4, I felt a certain feeling of satisfaction at having a Leica lens:
|
|||||||||||||||||||||||
The only issue was the lens cap:
|
|||||||||||||||||||||||
Why not a Leica lens cap?
|
|||||||||||||||||||||||
So I went out looking for one. The prices are typical Leica: I could pay up to USD 40 for one. Finally I found one, but it didn't quite fulfil the quality expectations: it arrived in pieces that I couldn't glue back together.
Since then I've been looking for another, and finally I found one. It was even cheaper, only USD 10 (plus $11 postage!). And it arrived in one piece.
But it didn't fit! Unlike most modern lens caps, it was designed to fit round the tube on the outside of the lens. And this part was too small. The tube at the front of the lens has a certain thickness, matching the filters. I measured an inside diameter of 45.18 mm and an outside diameter of 50.52 mm:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And the lens cap has an inside diameter of 47.84 mm and an outside diameter of 50.56 mm:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Somehow it seems that something is telling me that I should just give up. The inscription on the new caps is very low key anyway.
In passing, it's interesting to consider how hard to use lens caps are. For my four thirds lenses I have 5 or 6 different kinds of lens cap, all with marginally different ways to be removed. Most require to be pressed in a certain place, like the Lumix cap above, and the lens cap for the Olympus M.Zuiko Digital ED 8mm f/1.8 Fisheye PRO also requires to be put on the lens in a particular orientation—just what you need when you're in a hurry. Rear lens caps are (not surprisingly) all the same: they have a short rotation bayonet fitting. Why can't people do that with the front lens caps?
|
Server down!
|
Topic: technology, opinion | Link here |
Mail from Callum Gibson this evening, telling me that I (my proxy) had disappeared off the IRC channel. Strange. The proxy bip, runs on the external web server www.lemis.com. It occasionally hangs (I think there's a file descriptor leak somewhere), but I had just restarted it. Off to take a look:
=== grog@www (/dev/pts/1) ~ 1 -> uptime
The server had gone down again! That's the second time this year.
Somehow RootBSD is not what it used to be. I've had nearly 5 years uptime in the past, but when the server went down last month, it had been up for less than two years. But this time it was up for less than 6 weeks! And no information as to why. I could enter a ticket or send mail, of course, but why bother? Somehow things have changed, and silly things like uptime are only of interest to old fogies like me.
| Saturday, 24 February 2018 | Dereel | Images for 24 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Nvidia
|
Topic: technology, multimedia, opinion | Link here |
Mail from Nvidia today asking me how I liked my support experience. I've described this in painful detail, notably the absence of a real online bug report submission and the fact that the support person wasn't aware that the problem was solved in a newer version. But what got me was this:
|
|||||||||||||||||||||||
“Submitted Online”? How? If they really have that option, they're hiding it well. Hopefully my comments will get read, and they'll fix that.
|
Dinner preparations
|
Topic: food and drink, general | Link here |
Margaret Swan is in town again. She's a vegetarian, and thus provides a good excuse to cook Indian (and Malay) food. But this time it was different: Yvonne is still out of action, and it showed. Somehow it took me all afternoon, and I didn't have time for photos.
We had alu masala, gobi pakora, ikan goreng, mixed dal (of which I have found not one, but two enormous bags of deep frozen portions), spiced green beans with coconut milk and (since I was out of time, deep frozen) ayam lemak. Not surprisingly, we have enough left over for at least two more meals.
|
Black tropaeolums!
|
Topic: gardening, opinion | Link here |
The black tropaeolums that I planted months ago are finally flowering:
|
|||||||||||||||||||||||
Black? They're never completely black, but these ones are not even close. Still, the real question is why they have taken so long to flower. I think I need to water more.
| Sunday, 25 February 2018 | Dereel | Images for 25 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Cleaning filters
|
Topic: photography, opinion | Link here |
How do you clean a filter? I do it with detergent and much running water to wash away all traces of the detergent. The result looks something like this:
|
|||||||||||||||||||||||
I'm using rain water, which has very few dissolved solids, but it's not as clean as distilled water. If I let it dry, there are still marks on the glass:
|
|||||||||||||||||||||||
So I need to dry it somewhere. What better than with lens cleaning paper?
As it turns out, just about anything is better. One of the photos that I took yesterday to describe my lens cap issues showed the result:
|
|||||||||||||||||||||||
It seems that the mess on the surface is fibres from the paper. How do I do it right? In the past I used a linen serviette, and that might be the way to do it again.
|
HDR Express, yet again
|
Topic: photography, technology, opinion | Link here |
I've been trying to use HDR Express 3 for a week now, and I have yet to create a single image. Yes, it's Microsoft space, but surely I can master even that.
After the last attempt I decided that I should watch the video tutorials after all. So I fired up the program, once again cursing the mouse, and selected “tutorials” from the main menu.
It wasn't a video tutorial after all. It was the document hidden inside the program about which I complained about earlier. But this time I found a way to save it! So now at least I can look at the manual on another display. In the process I discovered that, though I downloaded HDR Express 3, it now calls itself HDR Expose 3.
OK, step by step. Damn, I can't copy the text from the manual! I have to type it in again, or show screen shots. This time I tried “Easy HDR Creation” on page 2. In the following, I show the manual text in italics. It states:
Selecting an image folder and filter using the Choose and Source filter functions.
What does that mean? Where's the verb? Still, surprisingly, my images in /Photos/6-HDR/foo were still selected, saving me the pain of tree climbing.
Next,
|
|
|||||||||
OK, I know that it misinterprets “similar images”, and I've already established that the “Auto Stack” doesn't work. That's why I have moved only these three images to foo. The important message here is “add images manually by clicking each image”. It doesn't quite work like that, as it says only further down: you need Ctrl and Shift in the usual Microsoft meaning.
OK, follow the instructions:
Check the Auto button.
Not needed: it was already selected.
If the images are static, check the Merge Static Photos checkbox.
Where's that? What I get is:
|
|||||||||||||||||||||||
But that's not what the manual shows:
|
|
|||||||||
Why is there a difference? This manual didn't just come with the program, it appears to have been bundled inside the executable, so there's no excuse for it not matching.
Select the key frame (image to base the HDR merge around) by clicking that images in the Select Key Frame area.
Again, where?
Select Merge.
Basically, this just plain doesn't work. The result appears to be identical to what I got last week: it seems to have only processed the first image, as (I think) the name at the top left shows:
|
|||||||||||||||||||||||
|
|
|||||||||
That's what I've had several times before. I'm not 100% sure that this means that only one image was used, but the appearance suggests it.
In fact, I now think that this doesn't mean that only one image was selected. But I still don't know how to tell, and if I'm incorrect in my assumption, I don't know what else this indication could mean.
OK, the next page of the manual tells me how to do it manually. Once again the image in the manual differs markedly from what I see. Here the manual, then my screen:
Check the Manual button.
There's no mention of the fact that the program goes off and does a lot of work at this point, including (on one occasion) crashing. It just goes straight on with:
Click Preview to open the Manual Alignment and Deghosting window.
Well, no, I got that automatically, this time with many more buttons than the manual shows:
|
|
|||||||||
|
|
|||||||||
The good news is that it seems to be using all three images, as the images at bottom right and the “Source Frames” display show:
|
|
|||||||||
The Manual Alignment tool (crosshair) creates an alignment point and opens a zoomed-in view of the selected area of each source image. You may click and drag one or more of the source images to align it. HDR Expose 3.0 allows you to select up to 8 points. Increasing the number of points allows more precise alignment.
I find this very hard to understand. The image shows a crosshair image at top right, and my image on the screen does too. But that's not what they mean. There are are two Adjustment Tools just below it, and you need to click on the left-hand one:
|
|
|||||||||
This seems to be optional, so I left it the first time round.
The De-ghost tool (ghost with a - sign) allows you to elect the area(s) from which to remove ghosting. To use this tool, select it and then click the desired location in the image preview. You may adjust the removal radius by dragging the handles.
The “ghost” is the jellyfish on the right. But what are handles? Still, after getting this far, I no longer have too much difficulty understanding. Like the manual alignment too, this appears to be optional.
Click Merge.
Off it goes, of course, and comes up with a result that looks like all the previous ones. Here first what it produced, then with the results from enfuse, and finally an unretouched image from the camera. Run the cursor over an image to compare it with its neighbour:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Despite the name, the out-of-camera image was overexposed by about 1 EV, which is visible in the driveway. But it's so much better than what I have managed with HDR Express 3 after a week of work. OK, I can tweak it, but that's just too much work. This is only one of about 35 images I merge every week. And you'd really expect the defaults to be at least usable, if not perfect. Based on these results, it's quite possible that the Auto results are also the (very poor) result of merging.
In summary,
So what do I do now? Give up. There's other software out there. It can't all be as bad as this or HDR Projects 4. But I suppose I should report this to the maker.
| Monday, 26 February 2018 | Dereel | Images for 26 February 2018 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
HDR Express: final sting
|
Topic: photography, technology, opinion | Link here |
So I've given up on HDR Express 3. As planned, sent them an email pointing to my diary articles. And they had one last surprise for me:
|
|
|||||||||
|
|
|||||||||
How do I enter the email address? So I filled out the (required) fields and hit Send. It flagged all the fields as “required”. So I tried again, and got:
|
|
|||||||||
So somehow they had moved the email address to the bottom. Somehow this is typical of what I have experienced.
OK, filled it out and sent it off. I wonder if I'll get a reply.
|
Next HDR software
|
Topic: photography, technology, opinion | Link here |
I've spentwasted over a week on HDR Express 3, but I'm
still no closer to a solution for my ghosting. What alternatives are there? Off to search
for best hdr
software, and of course got lots of responses. Interestingly, I didn't see HDR
Express in the list.
For some reason, picked on Aurora HDR, number two from this list. Download and installation went smoothly, and it seemed straightforward (“intuitive”) enough—until I tried to save a file. Ah, you don't save, you export. OK, I can live with that, though I'll probably continue to bitch and moan for a while. And then it offered me:
Save as "JPG files" (*.jpg)
Where did the experts go? And why “save”? That's (apparently) an old, worn-out magic word.
Still, on, and it seemed to do the job out of the box—my main complaint about HDR Express was that it didn't even come close. Here the default output same example that I agonized over with HDR Express. The sequence is HDR Express, Aurora, enfuse and out-of-camera. Run the cursor over an image to compare it with its neighbour:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
There's clearly a big difference, but which is best? Played around with the various presets, coming up with these:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Again, run the cursor over an image to compare it with its neighbour.
Clearly these are the typical “HDR look” that some people seem to expect. But the last one is surprising. It's the same preset as the first! It seems that there's some settings leaking going on here.
So which is best? Out of the ones I tried, clearly the first. But it's far more contrasty than the enfuse version, and possibly more contrasty than I want. This is one of the things I complained about with HDR Projects 4, so it warrants some attention. But overall the results are acceptable. Here another comparison with what I got out of enfuse, first Aurora, then enfuse:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
There's plenty of documentation—100 pages of it, somehow taking up 66 MB of PDF file. But the “batch processing” section is stuck at the end, almost as an afterthought. And what batch processing!
Load the Brackets. You can drag files or folders (which contain files) into the new window to load them.
Sigh. But then, this is Microsoft.
|
Installing drippers
|
Topic: gardening | Link here |
I've been meaning to add some drippers for the new plants I've planted this summer, and it has taken me at least since mid-November, but it took me until today to finally do something about it—and then only half of them. I really am getting lazy.
| Tuesday, 27 February 2018 | Dereel → Ballarat → Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
And now for something completely different
|
Topic: health | Link here |
Another week with three doctor's appointments for Yvonne this week. Today was with Mr. Kon Shimokawa, which puzzled me. In fact, I wasn't sure why he was involved with Yvonne's accident in the first place: he's a gastrointestinal specialist, and his main connection with her was for the colonoscopy last year. So why visit him now?
That became clear: it was indeed nothing to do with the accident, except for one of the MRI scans, which—by chance—showed an enlarged pancreatic duct and some cysts in the pancreas. Something worth looking at in more detail, in particular to rule out the possibility of IPMN, so there will be still more tests to be done—another MRI, specifically a magnetic resonance cholangiopancreatography (MRCP), an endoscopic ultrasound (EUS), a fine-needle aspiration (FNA) and of course blood tests. Yvonne not at all happy, mainly because of the possibility that it could lead to cancer.
| Wednesday, 28 February 2018 | Dereel → Ballarat → Dereel | Images for 28 February 2018 |
| Top of page | ||
| previous day |
|
More health issues
|
Topic: health, general | Link here |
Into town with Yvonne today for yet another doctor's appointment, this time with the orthopaedics unit of the base hospital. Left Yvonne there and went on to do the grocery shopping, and when I had finished that, called her to find that she was just finished, so back to pick her up. No problems, neither with the pickup time nor with the results of the examination.
|
Better than money
|
Topic: general, opinion | Link here |
ALDI and Woolworths now both have shopping trolleys that require a coin to use:
|
|||||||||||||||||||||||
When the trolleys are returned to a collection point and connected by a chain to the next one, the coin is released. The idea is to ensure that the trolleys are returned rather than just left wherever the customer finds convenient. I think the idea came from Germany via ALDI, who had them first here.
For reasons that I don't understand, Yvonne uses little tokens instead of coins. ALDI sells them, for $1 (ostensibly $0.99, but since the smallest coin is 5¢, that gets rounded up to $1). What's the advantage? They have a small hole in them, which she can use to attach them to her overly-large key ring. I find that yet another disadvantage.
But when I got to ALDI today I discovered that she had taken the $1 coin that I leave in the car for that purpose. Nothing for it, uncouple the token from the keyring. And the first trolley jammed! If they have these things, they should at least ensure that they work. Took the second trolley, did my shopping and thought no more of it.
Then, at Woolworths, the same thing happened! But this time not only did the second trolley not work, it swallowed my token. Person from the help desk (conveniently directly inside) saw the fun, came out to help and told me that ALDI tokens don't work in their trolleys—that's “common knowledge”. After quite some time trying to extricate the token, I did my shopping, and then we tried again after attaching the chain. She also gave me a Woolworths token, which apparently are free.
These tokens don't make much sense at the best of times. But if they don't work, they're downright stupid.
|
More stupid naming
|
Topic: language, general, opinion | Link here |
Over 10 years ago I discovered the stupid name wombok, which is used to mean Chinese cabbage. It hasn't gone away, but it doesn't seem to have stuck too well, either, as these items on display at Woolworths show:
|
|||||||||||||||||||||||
|
|
|||||||||
Now they always carry the text “Chinese cabbage”. When will it occur to them that the name “Wombok” is superfluous?
But wait, there's more. On the right of that image the cabbages are labelled “Savoy cabbage:
|
|
|||||||||
|
||||||||||||||||||||||||
My knowledge of cabbages is mainly German, and in Germany that's a Weißkohl (white cabbage). What's Savoy cabbage? Back home I found out: it's what the Germans call Wirsing, a kind of cabbage that I don't know in English-speaking contexts. It looks completely different:
In particular, the leaves are softer (you can wrap things in them), they're darker and there a somewhat bitter taste about them.
So why are Woolworths calling normal cabbage “Savoy cabbage”? This is an order of magnitude more stupid than inventing fantasy names. How can I make Kohlroulade with this stuff? It seems that you can, in fact, but the result would be completely different.
|
EUS: much more than colonoscopy
|
Topic: health | Link here |
Contact with the clinic of Dr (and not Mr) Peter Tagkalidis today to arrange the endoscopic ultrasound (EUS) and fine-needle aspiration (FNA) for Yvonne. Effectively it's the same procedure as a colonoscopy, but the details are much more complicated. It seems that nobody in Ballarat can do it, so we have to go to Melbourne, and it'll effectively take the whole day. They even said that Yvonne would have to stay the night in Melbourne, but after reading the copious documentation they sent me, this really translates to “don't drive after the procedure”.
|
Reading mail attachments
|
Topic: technology, opinion | Link here |
My MUA of choice is mutt, a descendent of the original elm. It can handle MIME attachments, but not well.
The mail from Dr Tagkalidis had about 8 PDF attachments, which were not the easiest to view, so I bounced the message to gmail and looked at it there, confirming that other MUAs don't do very well with this kind of document either. Each document was presented as an icon with three different sensitive areas, one to display directly, one to download, and another that took me a while to understand.
After reading one document, I got the message “safe to drive”, which I suppose made sense given that it didn't relate to a procedure. But they all said “safe to drive”. So why spend the night in Melbourne? Ah, I misread: “Save to drive”. How do you parse that? It seems that “drive” is some Google concept that I don't think I want to know, probably related to the concept of “disk”, “disk drive” or (shudder) “hard drive”.
Callum Gibson reminded me that Google Drive is a cloud storage service, one that I looked at in January. Why didn't I remember it? Clearly it's not a very memorable name.
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 |
