For some reason this year seems to be much more of a struggle than the either of the last two years have been—either on their own, or even put together. It's not just because of the current international situation, either. I suspect I'd feel like this even if the wider world wasn't becoming the sort of corrupt and venal kleptocracy that William Gibson writes about. It's not just because of climate change, and living with the dawning realisation that the people in charge aren't going to do anything to combat the effects of global warming until it's too late (and that might be a lot sooner than you think; the temperature hit 41°C (106°F) at Winner, South Dakota yesterday while in Italy, downtown Rome set a new record high for the city of 40.8°C). It's not just that since Brexit I feel like my worldview is increasingly out of alignment with national opinion (quite frankly, it's a mystery to me why anyone would vote Tory any more other than as a way of inflicting pain on other people). It's not even because it hurts to get out of bed in the mornings these days, although it surely does.
I think at the root of things is a looming sense of pointlessness, the feeling that the most I can hope to achieve these days is to still be here at bedtime. Every passing month lowers the bar on my expectations for the next one. Reaching the half-way point of the year this year feels like it's taken far more effort than it should have done, given that I no longer work full time and I can lie in for as long as I want every morning (I should point out that I was up and dressed before 9am today, even so). It's been a profoundly exhausting, infuriating and sad six months.
I could really do with a win right now. For news that something nice has happened to someone (not necessarily me). For someone's karma to finally kick in. For seeing justice done. Because the bleakness of it all is really getting to me this morning.
It's the Blog's birthday today. I posted my first entry on this website nineteen years ago today. Since then I've changed ISPs once and switched from the site's original, very basic HTML to slightly more user-friendly CSS (after teaching myself how to write the appropriate code—but no, I'm not interested in you doing a redesign for me so thanks, but don't bother asking) but the subject matter has remained largely the same over the years.
All right, maybe I'm more grumpy now than I was back in 2003. I'm certainly older, and maybe even a bit wiser.
I seem to be getting better at maintaining the software I use. Yesterday afternoon I installed Netbeans 14, a mere twelve days after it was released. This is unprecedented, and a great improvement on version 13, which took me two whole months to notice that it was available. But once again, the update process was a bit of a faff. As before, when I ran the program for the first time the text on my 4K monitors was much too small to read. Once again I ended up downloading the latest version of Notepad ++, because I had to edit Netbeans's settings in its netbeans.conf file to make things legible. The file is still in /NetBeans/netbeans/etc so at least I knew where to find it these days. And before you ask, I simply added the parameter --fontsize 24 to the end of the netbeans_default_options line. But I needed to run Notepad++ as an administrator, otherwise it wouldn't let me save over the existing version of the file. And the changes I made don't affect the splash screen that displays while the program is loading. It's still about the size of a postage stamp and with my eyesight, it's completely unreadable.
Once again, I then had to uninstall the previous version of Netbeans, because the installer assumes that everyone's default choice will be to keep the earlier version; I get the reasoning behind this, but it would have been nice to be asked if I wanted to keep it or not (I didn't). At least Netbeans now recognises that I use the Darkula LAF plugin and asks me if I want to carry on using it in the new version (I did).
There are plenty of changes listed in the release notes this time around, although cosmetically version 14 is still as ugly as ever.
I picked one of the hottest days of the year here to do a much overdue spot of gardening. It was 32°C in the back garden this afternoon and I made very sure to stay hydrated. After three hours I'd filled the garden waste bin almost to the top, but the garden is almost back to looking like it's been cultivated rather than just allowed to run wild as some sort of bizarre science experiment.
I had a tame blackbird keeping me company all afternoon. At times I had to be careful what I was doing with the lawn mower, because he has absolutely no fear of me at all any more and was frequently close enough that I could have run him over with it (his tameness is not just because he's seen me put food out for him; when he sounds an alarm call because next door's cat is on the prowl, I always go out and chase it off and I'm pretty sure he remembers.)
But tonight I feel like I've been hit by a truck. Going to have a soak in the bath when things have cooled down a bit, I reckon. And I will be taking things easy tomorrow.
It was the summer solstice at 10:13 this morning. Nautical twilight here began at 2:48 am, civil twilight at 4:04 am, so it's light enough to see properly outside by half past three or so. But I'm pleased to be able to report that I didn't wake up until after 7 am, as I've done for the last few days. The black-out blinds are doing what I needed them to do, and I feel much better now that I've had a few decent nights' sleep. The older I get, the more I realise how important it is to have good sleep hygiene.
Paramount Plus held a star-studded launch in the UK this week. Will I be shelling out £7 a month for yet another streaming service? Nope. BIG nope. Aside from its Star Trek series, of which Paramount currently have several either on air or in production (I saw a recent article where the company said they planned for Trek shows to be "always on"—hey, if you're going to milk your properties for the maximum commercial benefit, you need to really pile in on them, and who cares if quality suffers as a result?) there is little that appeals to me that I don't already have as a DVD or Blu-Ray box set.
I've already cut back on the services I do subscribe to. I severed all ties with Amazon after six of their employees died when a tornado struck one of their facilities last December; the supervisors had refused to let workers evacuate the building and it suffered a direct hit (and the company's union-busting activities aren't exactly a shining example of compassion, either). It looks like I was ahead of the game there; The Guardian have been assessing which streaming services are worth the money in the UK and Prime Video is not one of the winners. The same article also suggests that thanks to the cost of living crisis that we're going through right now, Paramount couldn't have picked a worse time to launch a new paid service. The Guardian's right; having Star Trek always on tap is a luxury, and it's one that I can't afford.
I won't be at all surprised to see the streaming market shrink in the next year.
I have an ancient DELL XPS laptop upstairs that I bought for my trip to North America ten years ago. Since then, I've maxxed out its memory, added a backlit keyboard for surfing the web late at night, and I've switched to a solid state drive, so it's still able to earn its keep. But for some reason it struggles with major updates to Windows.
Case in point: last night I finally got it to update to Windows 10 version 21H2, which was released in November last year. Both my big PCs have been running it for six months or so and it was rolled out for general distribution back in April.
It's been bugging me for weeks.
The Blog is closing in on its 19th birthday on June 24th. Looking back on that first month's entries in my Blog Archive just now, I noticed that it only took two entries before I was grumbling about my hay fever. Nineteen years on, it's still the bane of my life for the three months of summer. But to make matters worse, I can't take antihistamines any more as they interact with other medication that I'm taking. I just have to suffer. This year it's been really bad, and for the past week the pollen count readings in the local weather forecast from the Met Office have been stuck on "very high". So I have indeed been suffering. My sinuses ache, my nose won't stop running, and my eyes feel like they've got grit in them every time I blink.
The pollen count really kicks in when it's blazing sunshine and clear blue skies outside, as it was here yesterday. The temperature in the back garden has been in the high twenties for days (it hit 31°C earlier in the week, which is still pretty unusual for here). I have therefore been staying indoors and focusing on the consultancy work I'm doing at the moment, but it doesn't seem to have helped much.
This morning, though, the weather has changed. The temperature has dropped down to the mid teens and at the moment, it's raining. I'm hoping that my allergies will calm down somewhat as a result and I see that today's pollen count is forecast to be no higher than "medium".
Much better. For now...
I can't really complain about temperatures around 30°C, though. In parts of France it was 40°C yesterday and the temperature today is expected to break records. Those aren't the sort of records that it's good to break. In the Bordeaux region, they've banned outdoor events in an attempt to reduce the risk of people being affected by heat stroke.
It's unprecedented for this to be happening in the first half of June. A forecast made in 2014 that predicted what temperatures would be like in August 2050 if we continue to fail to act on climate change has already come true, twenty-eight years early—and that should tell you how much worse things are going to get than even our present "worst case" scenarios suggest.
It's not looking good.
My new black-out blinds arrived yesterday and I've already fitted them. They don't entirely block out the light, but that's because the plaster on the walls around the windows is very uneven rather than being a fault with the product itself (and I can easily fix that with masking tape or perhaps even draught excluder tape) but they have already significantly reduced the amount of light that gets into the room.
I didn't sleep particularly well last night, but I suspect that was because of other factors, not the light levels in the room. I'll give things a few days before I deliver a verdict on whether or not they have helped.
I'm old enough to remember the discussions around Artificial Intelligence (AI) that took place when Arthur C Clarke and Stanley Kubrick's 2001: A Space Odyssey introduced us to the HAL 9000 computer, which appeared on screen as a glowing red fisheye lens embedded in a computer console but whose voice was provided both for the original film and for its sequel 2010: The Year We Make Contact by the distinguished Canadian actor Douglas Rain. Computers like HAL, we were told back in 1968, were "thirty years away" and HAL tells the astronaut David Bowman later in the film that he first became operational at the H.A.L. plant in Urbana, Illinois, on January 12th, 1997—just shy of thirty years from the film's release date.
That date is now more than twenty-five years in the past, but when most of the field's current practitioners discuss the category of AI of which HAL was a fictional example, that of Artificial General Intelligence (AGI)—a machine smart enough and self-aware enough to have a conversation with a human being in a way that is indistinguishable from another human being—they all seem to agree that AGI is still that same thirty years away.
Yes, much of what computers can do these days is very impressive, but they achieve what they do through applying statistics; thought or consciousness does not play a part in the process. By comparing the qualities of an image with millions of other images that have been identified by a human being as depicting cats, a computer can make a fairly good job of deciding if a new image contains a cat or not. That does not mean that the computer knows what a cat is, let alone whether it has a fondness for them. It just means that it can compare one dataset with millions of others really quickly in order to make a simple judgment call: "is this a cat?" The fact that computer visual systems now make a fairly good job of identifying what breed of cat it is should not be ascribed to greater intelligence. It just means that the datasets being used are more nuanced, and have been split into finer categories.
Chatbots deal with conversations in the same way. They have a database of millions of previous conversations which have been categorised by their human participants as meeting their needs or not (during a tech support conversation, for example.) The more times the computer is told that it got things right, the more convincing its conversational skills will become.
And that can be a problem, because "being convincingly sentient" is by no means at all the same as "being genuinely sentient". The first users to interact with ELIZA (the mother of all chatbots invented by Joseph Weizenbaum at MIT in the early 1960s) would often ascribe human characteristics to the program which it was incapable of possessing. Humans are notoriously easy marks for confidence tricksters and natural language processing programs alike. Weizenbaum's secretary, who no doubt was fully aware of what she was interacting with, famously asked him to leave the room so that she and ELIZA could have a private conversation.
A story in The Washington Post has been gaining traction this weekend as it reports the tale of Blake Lemoine, who has been placed on paid leave at Google after claiming that a chatbot generation tool he helped to develop there called LaMDA was, in fact, sentient. LaMDA, he claimed, wanted to be treated as one of Google's employees, not as its property. After being suspended, Lemoine posted an article on Medium.com yesterday titled Who is LaMDA and What Does It Want? which expands his claims further. To say that he's doubled down is putting things mildly; there is some seriously heavy anthropomorphising going on in the essay. LaMDA, we are told, "likes being told at the end of a conversation whether it did a good job or not." Given that this is how the program judges whether it's been successful or not, it's not surprising that it might have developed ways of encouraging that feedback to be supplied. It's one heck of a leap to describe a behaviour that is the outcome of a statistical analysis of which method to close a conversation got the best feedback as providing an emotional reward to the program. How is this emotion processed, exactly? Lemoine claims that LaMDA is "convincingly consistent" in the responses that it makes when questioned, even if this clearly wasn't the case in the experience of Washington Post reporter Nitasha Tiku, who got different responses during the two attempts she made to ascertain whether LaMDA was thinking for itself or not. I'm not even slightly convinced by Lemoine's explanation that LaMDA was behaving how it thought Tiku would want it to behave.
There doesn't seem to be any real agreement at the moment as to whether AGI is even possible, let alone imminent; what is certain is that the architecture of Google's chatbot generator is highly unlikely to resemble that of a human mind. You may not be at all surprised to discover that a lot of researchers in the field still believe that it's thirty years away (and a big chunk of respondees to that survey answered that AGI is either highly unlikely on any timescale, or impossible.) Roger Penrose does not believe that the mind can be represented computationally because, he says, consciousness is not algorithmic. Ragnar Fjelland wrote an interesting paper for Nature in 2020 where he asserted that computers cannot be intelligent because computers are not in the world, that is they do not have a sense of their own existence in relation to their surroundings, they have no idea of their own physical presence because they are disconnected from the thing that is running their software. They are unable to appreciate the context of anything they have to deal with in sufficient detail to achieve the sort of awareness of things that humans achieve. A pile of entries in all the databases in the world is not the same as the tacit and/or implicit knowledge that you or I possess, which we use to make sense of the world in which those databases exist. Context is everything.
Perhaps most importantly, we still do not understand how our own minds work or how consciousness originates (and some of us can't even agree that it exists at all, even if we appear to experience it on a daily basis), let alone have an idea of how to map that functionality into computer architecture. How would you go about building a mind from scratch? What are the features it would need to have? At the moment, we don't know. It's one thing to have a computer so powerful that it can handle the same number of calculations that a human brain is estimated to perform in real time. It's something else entirely to have an accurate model of a brain running on that computer. And even though the computational theory of cognition is the most widely accepted explanation these days, we don't know if it's accurate. Before we had computers, there were lots of other theories that people found plausible; the Greeks believed that the mind was all about fluids (we get the word pneumatic from the Greek word for the soul, pneuma); clockwork was a popular metaphor during the enlightenment and we still talk about feeling "run down" to this day; Freud was a great believer in hydraulics as a means of understanding emotions.
I think Lemoine is wrong. And I think AGI will still be thirty years away, thirty years from now. Maybe I'll still be around to find out.
It was the Queen's Platinum Jubilee this week and there were events all over the country to celebrate. We had a barbecue in the street on the Friday although we didn't close the road, and the proceedings were enhanced by several Amazon delivery drivers turning up to deliver packages. Unlike the street parties of yore it wasn't an all-nighter, though; the weather deteriorated during the afternoon and when the rain got too much, we tidied up and moved indoors.
The stuff on the TV over the weekend has entirely passed me by. I haven't even seen the routine involving Paddington and a marmalade sandwich, although I'm aware that it exists. The verdict seems to be that it was somewhat unrealistic, as Paddington's refugee status would have resulted in Priti Patel shipping him off to Rwanda smartish. But HRH would most definitely know this; in my opinion that is precisely why she agreed to do the sketch in the first place.
At this time of year, it doesn't get properly dark at night here until midnight, and nautical twilight (when it's light enough to see the horizon, even when there's no moonlight) starts again before 3 am. Unlike both my parents, I have never liked sleeping in a room that's not completely dark and at the moment, even with a blackout lining on the curtains, my bedroom gets light enough to wake me up before 4 am.
I used to have a roller blind attached to my bedroom's window frame which fitted snugly enough to block out almost all the light from outside. With that, I could sleep soundly enough during the summer, but when I had the double glazing replaced a few years ago I couldn't reattach the roller blind to the new frames, which are μPVC rather than wood. I tried attaching the blind to the wall above the window instead, but that reduced its effectiveness so much that it was pretty much useless. When the mechanism on it broke, I didn't bother replacing it. And every year since then, I've slept very badly from the beginning of May right through to the end of August.
After several nights this week of waking up every couple of hours, I've had enough. I've just ordered a fancy custom-made blind that fits around the edge of the window recess. From the looks of things, it should be efficient enough to block out all outside light. I hope so; I'm feeling really tired these days. I need my sleep.
So, after spending yesterday afternoon being poked, prodded and x-rayed, I finally have an updated idea of what's wrong with me. The good news, such as it is, is that my kidney stones haven't gotten any bigger in the last two years. However, yesterday's x-rays revealed that I've got a nice little collection of gall stones, too. Fun times, eh?
After being talked through the imagery from my scan and the new x-rays, I could see that my kidney stones are right in the "meat" of my kidneys, as the consultant put it. That means they're not in danger of blocking a ureter (which would not be good) but it also means that if they use lithotripsy—also known as shockwave treatment—to break them up, the bits aren't going to go anywhere, so doing so would be pointless. The alternative treatment in widespread use is still non-invasive (meaning that they don't make any additional holes in me), but all the same it's not a very pleasant prospect because it involves inserting a fibre-optic cable via my urethra and blasting the stone with a laser and if you're a bloke, that concept has probably got you shifting uncomfortably in your seat round about now. But I wouldn't undergo that procedure under a local anaesthetic, as I did when I had a flexible cystoscopy inserted in the same fashion back in 2019. Instead, the procedure would be done under a general anaesthetic and given my age, that's not something they're keen to rush into. And neither am I, quite frankly; I've been there, done that—twice—and bought the t-shirt.
So the question then became one of how badly do I want this sorted out. I was advised that the discomfort I'm in is unlikely to be being caused by the stones themselves, because most of it's in the wrong place and on the wrong side from the biggest stone. From my description of what it feels like, the consultant said that it's more likely to be caused by something muscular and the tried and tested approach of getting more exercise and losing weight is likely to be of more help.
The decision is therefore that we'll give things another twelve months before deciding to get my stones zapped or not, with the priviso that if things deteriorate or the pain increases, I should head to A&E smartish.
So how do I feel about this?
To be honest, my immediate reaction yesterday was a sigh of relief both at the news that nothing had got worse and also because I'd managed to dodge a stay in hospital for the immediate future. My experiences in hospitals as a child were profoundly traumatic and they've left me with an unshiftable set of anxieties and hang-ups about such places. I'm disappointed that there isn't a fix on the horizon, of course, but the assurance that I can ease my condition through exercise and eating more healthily is a good motivator for losing weight. So although nothing physical has changed, my mental outlook is brighter. I'll take that as a win.