Arlie's Journal

I got my first performance issues on the Ubuntu system this morning. Culprits appeared to be Steam and Firefox.

Steam took a bit of killing - after I closed all the windows I could find, it was still running, and using enough cpu cycles to have 2 processes highly visible on top. I applied "kill" (not "kill -9") to the one that seemed more likely to be the main one (name didn't include anything like "helper"). There was a flash or redraw on the right monitor, as if it was getting rid of a buried or invisible window, and things improved.

Note to self: do not leave Steam running overnight, and do double check with "ps" or "top" after I think I've shut it down.

Firefox is still using more cycles than I'd prefer, even while I'm not doing anything in any firefox window - i.e. it's windows calling home, updating themselves, and generally using my cpu resources for their own purposes. Kubuntu/Firefox doesn't appear to have the ability to semi-easily identify the offending web page possessed by MacOS/Safari, where their system monitor program will show something about the web page, not just that the offender is a web content sub-processs. So identifying which pages not to leave open will be a bit more difficult - I'll have to kill overly active web content processes by pid, and hope this leaves evidence like a dead window/tab, or an error message in a window or tab. (At least, I presume that the processes that show in top as "Isolated Web Co" and "Web Content" are the equivalent of safari/macos' web content processes, intended to prevent the main safari (or firefox) process from hanging when a single web page is a hyperactive pig hog.)

Other than that, I'm having essentially the same issue with every single game I've been able to run - the screen resolution I got by default is fine for every other type of task I've tried so far, but anything that displays text in a graphic window, presumably as graphics, comes out just a bit fuzzy, straining my elderly eyes. I don't really want to change global scaling to alleviate this, as I'm enjoying the large amount of text space I can fit on my two monitors,courtesy of what I think is improvements going from X11 to Wayland. Also, a naively chosen scaling factor might appear less crisp than the one the installation process chose for me.

Ah well, nothing is ever perfect. It's still way better than Pop!_OS, and often more congenial than modern MacOS, though I really didn't want to deal with performance issues this morning before coffee. (For the record, I rebooted 2 days ago, after installing some updates, so I don't think this was up-too-long bitrot.)

One of the traditional MacOS features is a single dock, that appears at the bottom of one of your monitors. From time to time it moves to another monitor; it took me a year or two to figure out that the moves were not random, but happened when you moved your mouse too close to the bottom of some screen that didn't currently contain the dock. This would be interpreted as a request to move the dock to that screen, on top of whatever you were trying to do there.

The KDE equivalent is called a panel, and can contain more than just the sort of things MacOS puts in their dock. You can have as many panels as you want, on as many of your monitors as you want, even 2 or 3 on the same monitor if you prefer. They don't have to have the same contents - you can put a panel with clock and various status widgets on the top of your screen, approximating what MacOS has there, and a second panel at the bottom, with task bar and related items, approximating what MacOS has there. Or you can do what I did, and put all of these in one panel, at the bottom of the screen, but put such a panel on all the screens you have.

There are some glitches, and not all of them may be learning curve. But most of them involve configuring the extra flexibility MacOS lacks.

This morning I started the day on my Mac, taking advantage of Safari's existing history to conveniently reorder a product I'd just run low on. The dock made one of its unintended moves. That's been an aggravation approximately forever. But this time instead of snarling at Apple, I smiled happily at the thought that soon I won't be dealing with this any more.

Apple's Sequoia release destabilized my Mac Mini, in ways that suggest I need a beefier, more recent machine to run everything they've shoveled into the OS. I get overnight Safari crashes, semi-randomly, at least once a week and sometimes daily. It also broke parts of my ordinary workflow. I'm reliably informed - by a friend who's a long time Apple developer - that the release after Sequoia, named Tahoe, is even worse. He's personally regretting "upgrading" to it.

So the time has come to find my parachute. Windows is not a reasonable option - it has all the MacOS flaws, plus a whole pile of its own. I fled from them to MacOS when they proudly released a version designed to look and feel like a cell phone, presumably to encourage purchasers to buy a tablet instead (sic).

I've known MacOS was going downhill for a while, and would eventually cease to meet my needs. A couple of years ago I bought a mini pre-installed with Pop!_OS, probably 22.04, the then current Long Term Support (LTS) release. It was a dog. Routine window manager hangs combined with a UI from Hades, that wasn't configureable in any way that mattered to me. Woof! Woof! Woof! I've been using it only for tasks I simply couldn't do on any other operating system - bookkeeping (I use gnucash) and retro games that require support for 32 bit apps (MacOS killed that).

Last week I had a rush of energy, possibly brought on by anger over current events. I finally started serious steps towards looking for a better distro. I acquired a number of cheap memory sticks, expecting to have to try out dozens before I found one tolerable. I canvassed my most knowledgeable friend for advice, and did so on a Discord server where other friends could weigh in. And then I made a post on reddit asking for suggestions, with a long description of what I do and do not like in an OS.

( Read more... )

From Jenn Dowd at Data for Health a reasonably decent substack blog in the medical space:

Does adolescence really last until age 32?

Capsule summary: comparing brain scans of individuals at different ages doesn't tell you very much about aging. For that, you need brain scans of the same individuals at different ages. also, "convenience samples" of people who just happen to have been scanned are unlikely to tell you as much as you think about the general population.

The study cited was published in something call Nature Communications, which is presumably unrelated to the well-regarded scientific journal Nature.

See also the same blogger's Are We Really Aging in Bursts?.

The United States elects people for a lot of roles I'd consider to require a technically competent, non-partisan person, whose skill at doing the job might be unrelated to their skill at winning elections - or worse, the need to run for (re)election might incentivize bad behaviour in the job at hand.

One of these roles locally is the county assessor - the person in charge of deciding property values for purposes of taxation. The previous incumbent has retired suddenly, or perhaps died, so we had a vote for a short term replacement. There were 4 candidates, only one of whom I'd ever heard of - he's been sending out small amounts of election ads. I don't like what he said about his background, and I like his proposed policies even less. His latest election email contradicted itself, presumably trying to be all things to all people. I also strongly suspect he's proposing to implement changes that are far beyond the scope of his office - i.e. he can't do what he says is the biggest reason to vote for him.

This individual was one of the top two in the recent vote. There will be a run off soon, with ballots arriving in early December. He'll be one of the two on the ballot. I don't know who his opponent is, but I'd vote for any of the other three ahead of him, and will make a point of doing so.

( Read more... )

Some weeks ago, we set out from my driveway to drive to Costco. I basically know the route(s), but the turns to get there from whichever major artery you take are a bit tricky, so I asked Google Maps to direct me.

Google first told me to turn left, then, before I was out of the driveway, to turn right. By the time I'd made it to the end of the block, it had alternated telling me to keep going with telling me to make a U-turn several times. Once I got onto one of the two obvious routes to a relevant highway, it suggested a change of route, then switched back to approving the one I'd picked, a couple more times.

It settled down once I got to the highway, and I did not find a slowdown on the route I'd kind of randomly picked. (And if there had been e.g. a major accident, it would normally have given me a suggested change of route, once, and explained why.)

And once I got near my destination, it handled the exit and its confusing turns perfectly well. So it did what I needed, along with adding extra amusement as we set out.

I've been meaning to post this topic ever since, but something else has always seemed more topical. So I've forgotten the exact number of times it changed its mind. (Yes, I was counting.) But with another Costco trip planned for tomorrow, I've finally remembered and posted.

There've been no recurrences since then - but I don't use Google maps all that often, so the sample size is small. I am somewhat curious whether it will lose its mind again tomorrow.

I subscribed to this blogger based on some recommendation or the other, and their claims of what they blogged about. This article has me giving them one strike, as in "three strikes and you are out" except that I'll probably only require one more strike to unsubscribe.

The topic is how health communicators need to act like online influencers, making themselves trusted based on their charisma and personality, eschewing evidence or anything that carries the negative taint of looking like normal health communications. Yep, they should communicate just like RFK Jr. That way they'll be believed more than they currently are.

( Read more... )

On 9/29/25 I called my cellular provider (T-Mobile), and after spending half an hour on hold, I asked about charges for calls from my home in the US to Canada. A person who called herself Aniya informed me that my plan (which I've had for 11 years) now gives me unlimited free calling to Canada (+ Mexico). I recalled it not having that feature, when I first set up the plan, which is why I made a point of asking. (I'd expected I might have to change plans.)

8 hours ago I made a long call to Canada, using the cell phone rather than the land line. About half an hour after the call ended, T-Mobile sent me a text: "As of 10/04/2025, your account reflects $135 in international calls, and/or call to premium-rated numbers in this cycle." That is a lot more than what I would have been charged for making the same call from the land line.

While I was on hold with T-mobile on Monday, I had an online chat with Verizon's customer service. They told me that their Unlimited Welcome plan would give me free calls to Canada - for essentially the same price T-Mobile is charging me. And if I were to authorize a switch then and there, they'd waive the $40 activation fee. Moreover, they'd give me a $5 monthly discount for 3 years if I brought my own phone.

I'll be contacting T-mobile next Monday to complain, and maybe convincing them to reverse the $135 charge. (I wonder how a claim against them in Small Claims Court would play out in practice...) I won't consider staying with them unless they both provide free calls to Canada *and* reverse this surprise charge.

I moved to T-mobile 11 years ago because at that time they were the company that wouldn't charge me an increased monthly free to cover price-reduced phone upgrades, since I wasn't going to buy a phone from a telco in any case. (Phones bought from a carrier generally arrive locked to that carrier, which isn't a feature I want.)

All this because ATT is killing POTS, so I'm inclined to get rid of my landline entirely. I valued the ability to use my phone in a power outage, but non-POTS landlines don't offer that feature, short of more complexity and expense than I want to deal with - whereas a cell phone *might* work, if its battery has charge and the cellular towers have backup power, or are outside my outage area.

At least T-mobile warned me, rather than leaving me to find out when my next bill arrived, after many many more expensive calls.

susandennis recently posted about several technical things she's trying. She's much more bullish on digital tech than I am, and even enjoys shopping (!), but she's been a great source of pointers for me of things I might like.

Today's post included a for-pay ad-free, surveillance-free search engine called kagi. She says she's been hearing about it a lot; I on the other hand have been hiding under my rock, and only just noticed it. (I really should read Ars Technica more frequently.)

After thanking Susan for bringing it to my attention, I just spent some time looking at it, finding myself on the fence about trying it. This Ars Technica article from Aug 5, 2025 says much of what I might have said myself, only better.

( Read more... )

I'd appreciate other people's comments.

Yesterday I had a rather inconclusive appointment with a physiatrist at Sutter Health. The appointment was inconclusive because he was unable to read the CD with the MRI results, from another outfit called SimonMed Imaging. I'd been referred to one of the physiatrist's colleagues by the doctor who'd sent me to SimonMed. Yet somehow her report had never reached Sutter Health. The text-based MRI report had reached neither Sutter Health, nor my primary care physician, nor me - but the referring orthopedic surgeon had explained the MRI to me verbally, so I knew basically what it said - she'd also had no problem accessing it, but she's in-network with SimonMed.

I spent a chunk of yesterday afternoon on 3 separate "my health" type web sites, for Sutter Health, Simon Med, and my primary care doctor. I cut and pasted the textual MRI report to both the physiatrist and the primary care doc, via each of their messaging systems - which had a character limit, requiring the report to be split into multiple parts, and AFAICT no "upload important medical info" interface.

My primary care doctor responded later that day via the messaging system, which sent me an email telling me to check there for a message. As of noon today, I had received no similar message from Sutter Health's system.

I had, however, received a request to review the physiatrist on Google. The message came from feedback@sutterhealth.org (I have since received an email telling me the physiatrist has responded on their system.)

I reacted extremely badly to the email, to the point of writing this name and shame about the medical organization. I'm rather disgusted about getting requests for feedback rather before I get anything resembling results. I'm not entirely comfortable with the idea of writing reviews for doctors, particularly ones one has only seen once. (I feel similarly about medical advertising, or maybe rather worse.)

I was already getting warning signals about Sutter Health, suspecting an excessive focus on profit along with low grade sloppiness about minor patient facing matters, compared to two other health organizations I've used in the past year or two. (I have no useful information about medical matters.) This request for review made those signals rather louder and harder to discount.

Meanwhile, I don't know whether Simon Med produced a bad CD - or whether that problem originates at Sutter Health. I also don't recall for sure whether SimonMed was definitely told to CC everything to my primary care doctor (the MRI was last April; I didn't schedule the physiatrist until after we knew that physical therapy wasn't helping). I'm suspicious of them too, but less so - in part because CCing the primary care doctor seems to be something the patient must explicitly request; otherwise results may just go to the requesting physician.