|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Saturday, 12 September 2026 | Dereel | Images for 12 September 2026 |
| Top of page | ||
| next day | ||
| last day |
|
Web browser migration: stalemate
|
Topic: technology, opinion | Link here |
It's been two months since I started transitioning from firefox to chromium. It still isn't done, and I'm beginning to think that it never will be. Yes, firefox has bugs that chromium doesn't, but the converse is also true. Earlier in the month I had a case where chromium refused to perform a function because of perceived security issues. Dammit, I want a browser, not a minder. chromium also refuses to log in to the State Library of Victoria: any attempt fails silently.
And then today I read, from my subscription to The New York Times:
You have been blocked from The New York Times because we suspect that you're a robot.
Why am I seeing this? There are a few possibilities:
You are browsing and clicking much faster than is typical of a human beingSomething is preventing Javascript from working on your computerThere is a bot with the same IP address as you
None of that applied, of course: it's clearly a bug in the Times' software. But it even occurred when I logged in as a different user. But it didn't with firefox! They later fixed it without telling me, but one way or another it's another reason to avoid chromium.
|
Instagram pain again
|
Topic: technology, opinion | Link here |
I don't use Instagram except to look at photos that people post. Today was such a day. Log in, please. Sorry, wrong password.
That's another site that keeps forgetting my password (which was, in fact, something like WhyD1Dyouforgetit?). OK, reset again. “We will send you a code to your cell phone”. They didn't. Once again I'm out of contact with Instagram. Do I care? Only because of the nuisance value.
|
Bliss: ignorance?
|
Topic: general | Link here |
As we get older, our daily walking the dogs is shorter and shorter. A few years ago we would go 1 or 2 km, sometimes even 3. Lately it has been as short as 600 m.
But today the weather was nice, and Yvonne had energy. Round the loop Progress Road → Rozenstein Road → Bliss Road and back, a distance of about 3.3 km. It was interesting: there seem to be more houses there than I recall, particularly in Progress Road, though none has been obviously newly constructed.
But it wasn't until we were on Bliss Road that it occurred to us how many properties had “Keep out” signs or similar:
|
|
|||||||||
|
|
|||||||||
|
|||||||||||||||||||||||
None of these properties are particularly affluent. Here a wider view of the first image above:
|
|||||||||||||||||||||||
Why do they do it? Some of the signs look like they were quite expensive to erect.
|
Missing days in this diary
|
Topic: technology, general, opinion | Link here |
Part of writing up this diary is to compare the contents of the monthly diary with the previous contents. And oh horror, the entry for 8 September 2026 was missing!
Quick check: yes, it was still in the previous checked-in version, so I was able to restore it. But that's one of my horror scenarios. 25 years ago I lost the entries for 11 and 12 June 2001, and it still keeps me up at night. I have PHP code to catch this kind of problem. Why didn't it? The code is:
if (($yesterday != ($day - 1))
&& (($day != 1) || ($isdiary < 2)) /* don't complain on diary wraparound */
&& ($diary_complete != 0)) )
$missing_days .= "Incorrect sequence: $yesterday followed by $day\n\
";
That looks OK. So why didn't it catch it? Spent much time searching through the code before I discovered this in the header of this month's diary:
$diary_complete = 0; /* No gaps */
Oh. That should be 1, not 0. And it seems that there were many such entries. Fix them all? Or kludge the test? For the time being, the kludge is the way:
if (($yesterday != ($day - 1))
&& (($day != 1) || ($isdiary < 2)) /* don't complain on diary wraparound */
&& (($diary_complete != 0) || ($diarydate > "20000901")) )
...
And that worked. And almost immediately I got an email message:
Invalid entries in diary /grog/diary-dec2025.php:
Incorrect sequence: 14 followed by 17
Incorrect sequence: 17 followed by 16
Aaargh! But it was benign: a typo in the date. I had written 17 December instead of 15 December, so all that was needed was to fix the date.
I wonder how many more are hiding there.
|
The price of progress
|
Topic: technology, opinion | Link here |
It has been nearly 3 years since I bought hydra.lemis.com, my new main machine with ridiculous quantities of memory. And I still haven't completely migrated to it:
I had hoped to migrate my X configuration from eureka, which also had 4 displays. But the window manager (fvwm 2) no longer works reliably. And the “drop-in” replacement fvwm3 has lots of compatibility issues.
The 3840×2160 monitors require a surprising amount of software configuration to display text in a legible size.
The intention to run all my other “machines” as virtual machines (and thus the 192 GB of main memory) hasn't come to fruition. There are too many bugs, in particular the networking code in conjunction with Microsoft. So I'm still running ThinkCentres like distress.
Countless newer versions of existing software no longer work the way I want them to, notably Emacs and Hugin. I've fixed some of them, but there are so many changes needed. It's a pain, and there's a quick workaround: use the version installed on eureka, now nearly 11 years old.
The last point raised its ugly head again today. When checking my daily diary, I display a diff in an Emacs buffer. Then I commit the version, still in Emacs. Only things have changed: previously, the commit would work even if the diff was selected. Now it silently fails. I really need to position in the diary buffer. This was the background for my lost day's entry. Nothing for it: use Emacs version 24.5.1 of 1 October 2016 rather than the halfway modern version on hydra.
What a let-down!
|
How much bread for fondue?
|
Topic: food and drink | Link here |
Fondue de fromage for dinner this evening, as frequently before. One of the issues is quantities. We need about 225 g of cheese and about ⅔ of that quantity of bread. This evening I weighed the ingredients: 222 g cheese, 159 g of bread. We had 27 g of bread left over, giving us a ratio of quite close to 0.6:1. It can never be that accurate, of course: the 159 g of bread represented 3½ slices of bread. But at least it's something to aim for.
A good thing that horses like pieces of bread.
| Sunday, 13 September 2026 | Dereel | |
| Top of page |
|
More noodle cooking time changes
|
Topic: food and drink | Link here |
How long do you cook noodles? I have a page about that, 40 pages long. And I'm constantly changing it: I'm currently at revision 1.118. Today I ate the Jack Hua rice noodles, which I have been cooking for 5 minutes regardless of width. But today's were the thin 1 mm cut. Do they need as long? My guess was no. Cooked them for 4 minutes, and they were fine.
|
More ACDSee videos
|
Topic: technology, photography, opinion | Link here |
I've now finished watching the acdsee Photo Studio videos. A total of 11 of them, including a 10 part introduction for beginners, of which only 3 were of much use. How to use their database facilities. How to attach an image to an email (excruciatingly complicated, even before their bizarre requirement of “Microsoft Outlook”). How to back up your photos. I didn't even bother looking at that one, though maybe I should.
The other three were interesting for reasons that I hadn't expected. I get the impression that the product derives from two separate products: “Develop” (“non-destructive”) and “Edit” (“destructive”). In fact, they do very similar things, and neither is “destructive”. The big difference seems to be that “Edit” uses layers and “Artificial intelligence", while “Develop” doesn't.
Sadly for me, though, they haven't shown anything that DxO PhotoLab can't do better, with the possible exception of lens distortion correction, which is based on lensfun. And that offers corrections for the Voigtländer Nokton 25 mm f/0.95 and the Sigma 150-600 mm lens. That's not the same as my OM System M.Zuiko Digital 150-600 mm, of course, but it's closer.
So: at least I have an idea of how to navigate the application now. What I need to find is more about noise reduction.
| This page contains (roughly) yesterday's and today's entries. I have a horror of reverse chronological documents, so all my diary entries are chronological. This page normally contains the last two days, but if I fall behind it may contain more. You can find older entries in the archive. Note that I often update a diary entry a day or two after I write it. | 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. |
| Greg's home page | This month | Greg's photos | Greg's links |
| RSS 2.0 |
|