There wasn't a single rumble of thunder here yesterday and the rainfall was limited to a few ten-minute showers. Perhaps it's just as well, as things further north got a bit wild.

Twitter's much vaunted redesign finally landed on my desktop this week. While I didn't end up frothing at the mouth in disgust as some users did, I can't say I was impressed with the new interface. It's full of half-assed compromises; while we finally got a dark theme for the interface (which I've had on Tweeten for years), the choice of colours for objects like the "Tweet" button is limited to just six, three of 'em are really ugly, and grey is not one of them. On my 4K monitors, the padding between tweets and columns is all over the place and there is such a ludicrous amount of blank space on the screen that I was left wondering whether the UI team were only working with VGA monitors. Oh, and nobody seems to have bothered with Twitter's analytics pages, which are still served to my browser with the old look and feel.

Needless to say there is still no "edit tweet" button for Tweets that you've already posted.

To understand just why the company has proved so tone deaf to the requests of its users, we should look at one of its biggest rivals. Twitter seem to have taken their lead in the redesign from Facebook, who carry out tests with their users (perhaps imposes is a better word, because the users don't volunteer to take part in the tests—then again, Facebook has never concerned itself unduly with inconveniences like ethics). These tests are conducted to see how users respond to different versions of their product, both good and bad. One outcome of the tests was that Facebook acquired a lot of information about just how horrendous the user interface of a social network needs to get before its users stop using it entirely. The answer was, depressingly but not surprisingly, "really bad". This finding has proved significant to all the other social networks, and all the reasons why are bad ones because they indicate strongly that servicing the needs of the company and its advertisers can be prioritised. It's not necessary to meet the needs of its users, because they'll keep on using the site regardless. In other words, social networks know they can dick around with stuff, because their users are hooked. It doesn't matter how badly networks mess up the user experience, because usage is habitual and entrenched.

You may have noticed in recent months that what appears in your feed on Twitter has changed. Like Facebook, Twitter now curates what you see, favouring some tweets over others, rendering others effectively invisible. Twitter judges everything through tweet engagements, and the name of the game is building numbers for the top few per cent, not in helping the vast majority of users with just a few followers to grow an audience. Why? Because Twitter can inject advertisements into feeds from the most popular users and get lots of eyeballs looking at them for very little effort. Why bother doing that for Joe Bloggs if he only has half a dozen followers? It's not worth their effort, and neither is he. Like Facebook, Twitter now insist on showing you a selection of tweets selected by their algorithm rather than a chronological feed from everyone you follow. You've probably noticed that their algorithm also throws tweets from people you don't follow into the mix, and not just from advertisers who have paid for you to see their unwanted tweets. Twitter picks some items solely because somebody you follow may have "liked" the tweet. The good news is that for the moment, you can switch Twitter's behaviour back to how it used to be: change your feed to "Latest Tweets" by clicking the stars icon at the top of the interface, but the bad news is that Twitter clearly don't want you using the site in this way, because they'll change your feed back to the algorithmic version within 48 hours.

As an example of a service provider not giving a toss about its users' requirements, this is hard to beat, but the end goal for all the social networks is to get users to pay if they want their content to be read by a wider audience. This is, after all, what Facebook already does with its artist pages. Don't expect anything free to remain free for ever; this is the fundamental tenet of late-stage capitalism. If you continue to use it, Twitter will find a way to monetize you.

By the time I'd uploaded Thursday's blog update, the weather forecast had changed once again, showing an afternoon with no thunderstorms and the temperature rising to 32°C, which is what duly happened. Later that day in Cambridge, the UK record appears to have been broken (it's not been confirmed by the Met Office yet) after a temperature of 38.1°C was recorded at Cambridge University's Botanical Gardens. That's 100.6°F. Temperatures here in the house hit 32°C upstairs, which made for some very uncomfortable nights and I've been getting very little sleep. The house would have been even warmer, but I'm at home during the day so I was able to leave all the windows open. This is the UK, remember; we don't do air conditioning. Things have cooled off over the weekend and the temperature here at my desk right now is a much more comfortable 22°C. The temperature next week is expected to be closer to its "usual" levels, staying around 21°C or so, and I will be very relieved.

Weather forecasts are very fluid things at this time of year because the large amounts of energy in the atmosphere give plenty of potential for rapid and large changes in conditions. If you're putting together a forecast, getting your starting point wrong by even a small amount can quickly escalate into wild variations from reality. This "sensitivity to initial conditions" is a feature of chaos theory; the results of not taking into account something as seemingly insignificant as a butterfly flapping its wings can build and build until the effect becomes important at very large scales. That's why weather forecasters don't just run one model of the weather in their computer simulations—they run loads of them and see how much commonality there is between each one (and that's why meteorologists use serious amounts of computing power to bring you your daily forecast.) If many models agree, their forecast is likely to be reliable; if they rapidly diverge, as appears to be happening in the UK at the moment, it's far more difficult to predict the weather for the week ahead until a day or two before it actually happens.

Just watching the skies brings home how volatile summer weather can be here in the UK. Several times in the last week I've watched large cumulus clouds build up out of a clear blue sky in a few minutes. In the summer months these "fair weather cumulus" (the sort of clouds that you see in the opening credits of The Simpsons) can either vanish just as suddenly or grow further, feeding on heat and humidity and evolving into the huge, anvil-shaped cumulonimbus clouds that produce storms. It's very difficult to predict which way things will go on a local scale and each time that I've checked the forecast for here over the last few days, I've seen something different. The forecast for Friday was for thunderstorms, then fine weather was expected instead; in the end it was overcast, with a brief shower of rain. The first forecast for next Tuesday afternoon that I saw predicted thunderstorms, but by Saturday morning Tuesday was expected to be dry and thunderstorms weren't expected until Thursday. When I checked again on Saturday evening, the forecast for Tuesday showed thunderstorms once again. This morning, the Met Office issued a yellow warning for thunderstorms here that starts from three o'clock in the morning on Tuesday and lasts for the rest of the day. I suspect that this means the computer models for Tuesday now agree sufficiently for a warning to be issued. It's been interesting (if somewhat frustrating) to follow the dynamics of all of this. There's clearly going to be an awful lot of energy in the atmosphere if it's expected to fuel storms for the whole day. But let's see what Tuesday eventually brings...

I was pleased to discover this morning that the Met Office had revised their forecast for the highest temperature here today down from 31°C to 29°C, but by the time I'd had breakfast this morning it was already 28°C (82°F) outside. The reason for the downward revision is that thunderstorms are now expected in the area from lunchtime onwards.

I love a good storm, but any recent ones have passed the village by. I was woken by an intense storm on Tuesday night and, of course, fired up LightningMaps to find out what exactly what was going on, and where. The main activity was over Chippenham, heading north to Cirencester. There were more than 150 lightning strikes being recorded every minute. I hadn't seen a storm of that intensity for at least twenty years and from the bedroom window it looked spectacular, but the storm was far enough away that it was largely silent. It only rained here for ten minutes or so: just about enough to fill up the bird bath. Perhaps this afternoon's storm will be closer; I will wait and see.

The Met Office is expecting to see a new record high temperature set in London and Kent today, and it could end up as high as 39°C (102.2°F). The West Country will escape the worst of the heat (which is being brought up off the continent by southerly winds), but even so the desk thermometer in front of me tells me it's 25°C indoors at the moment. Upstairs in the studio, it's a good few degrees warmer than that and I won't be recording any new music until things have cooled down. It's too hot to think, let alone come up with new songs.

I must admit that when I heard yesterday that Rutger Hauer had died after a short illness, I sat there in disbelief. He was one of those people that you assume will always be around; the idea of him passing away was, frankly, ludicrous. And yet, suddenly, he's gone. The announcement of his death was only made after his funeral had taken place. I don't do obituaries or tributes on the blog any more, because I was writing far too many of the things, but for Rutger I am making an exception. Because Rutger was exceptional.

Yesterday, my Facebook timeline was full of pictures of him. More than one of my friends observed that he'd died in the same year that the film in which he played his most iconic role was set. But there was much more to Hauer than Roy Batty; I've watched dozens of films where his was the performance that you remembered afterwards. They weren't all critically acclaimed, to be sure—but the cultural impact that he achieved was considerable. This wasn't always in a good way, it has to be said; as the Guardian's obituary observed, his eponymous role in the 1986 film The Hitcher was responsible for single-handedly killing off the practice of hitch-hiking in the United States. Whether it was horror, grindhouse, SF or highbrow, Rutger always brought his best game to the deal. I can't think of anyone else who could have taken on the central role in Hobo With A Shotgun with as much relish as Rutger clearly did.

Aside from Blade Runner, my favourite films of his include Blind Fury, Wanted: Dead or Alive, and Flesh+Blood. In recent years he appeared in memorable cameo roles in several big-budget productions, bringing gravitas and watchability to characters that in other hands would have been eminently forgettable. When he cropped up in Confessions of a Dangerous Mind, he elevated the film to a higher level and his appearances in Sin City and Batman Begins are the best parts of both those films. I actually cheered out loud when he appeared at the beginning of Luc Besson's recent science fiction film Valerian, where, fittingly, he played the President of the World.

But when things have cooled down tonight I will be watching one of his most critically acclaimed performances, as the antihero of Ermanno Olmi's The Legend of The Holy Drinker, and celebrating the life of one of cinema's most charismatic stars.

The blog's been taking a back seat recently as I've been up to other things, and not just gadding around London. I've been preoccupied with sorting out one or two grown-up things which are, frankly, much too boring to feature here.

But I've also been back making new music for the 50/90 challenge, and I now have eight tracks under my belt. I expected to be further forwards than this, but at least I'm on target.

Today I will be staying indoors; it's been raining here since the early hours of the morning. More significantly, today's also the day that the schools break up for the summer holidays, so the roads out there will be insanely busy this afternoon. Best stay in.

Everyone should have a favourite science experiment. It's a great way of understanding how science works and helps you to discover more about the world. If you're an engineer, you might pick the intricate and incredible ATLAS detector at CERN in Switzerland, built to investigate the Standard Model of particle physics. It was instrumental in the discovery of the Higgs Boson. If you're a chemist, the Elephant's Toothpaste reaction is always good for a laugh as it couples an energetic and violent expansion of compounds with extreme messiness; Camille Schrier used it this month to win the 2019 Miss Virginia beauty contest. But my favourite sort of experiment is the type popularised by Albert Einstein, which he referred to as a "gedankenexperiment" or thought experiment. It involves thinking through the implications of your theory and imagining what might happen if it were accurate, even if it would not be possible to perform the experiment in real life. In Einstein's most famous "what might happen if..." gedankenexperiment, which he conceived at the age of sixteen, he imagined what he would see if he could somehow chase a beam of light through space and see how it behaved. His sense that the conventional view of what should happen was wrong eventually led to his development of the theory of special relativity.

Today I was reading about another elegant gedankenexperiment originated by Jagjit Singh Sidhu and Glenn Starkman of Case Western Reserve University, and Robert J. Scherrer from Vanderbilt University. Their thought experiment involves the mysterious and unknown material called dark matter, which exerts gravity on the rest of the universe but which is otherwise invisible and undetectable. The "what would happen if..." question they asked was a doozy: "what would happen if dark matter moved around in lumps big enough to interact with a human being?" The answer they came up with was that there should be records of people dying from mysterious gunshot wounds. As this does not seem to have ever happened, they have been able to constrain the potential properties of dark matter without building any experimental equipment at all.

It's amazing what just sitting down and thinking about a problem can reveal.

I was in London last weekend meeting up with my friends from the William Gibson Message Board, a.k.a The WGB. It was the first time I'd seen some of them for several years, and I was reminded strongly just what a lovely bunch of people they are. Aside from drinking copious amounts of coffee and tea and enjoying several memorable meals, we'd gathered together in London to celebrate the work of another creative powerhouse: the Design Museum in Kensington is currently playing host to the incredible Stanley Kubrick Exhibition, which has been touring the planet for nearly half a decade now. The WGB folks (me included) tend to be deep cinema nerds, and Kubrick is a favourite of pretty much all of us. After all, in 2001: A Space Odyssey, Kubrick and Arthur C Clarke introduced us to the first really plausible Artifical Intelligence in science fiction. Coming face-to-face (or face-to-camera) with HAL himself is quite an experience.

The show was every bit as good as I'd hoped it would be; we all geeked out over the collection of material from the director's own archive, including his director's chair, a ridiculous collection of esoteric camera lenses and the Steenbeck editing machine that Stanley used on Full Metal Jacket. There was the letter Stanley wrote in 1964 to Clarke, suggesting that they collaborate together to make

"The proverbial "really good" science-fiction movie. "

There were pages of annotated scripts from The Shining and Dr. Strangelove. There was a sixteen-foot model of the spaceship Discovery from 2001, complete with a pod waiting in front of it that had a tiny model of astronaut Frank Poole in its arms. There was an intricate scale model of the Overlook Hotel's Maze that Adam Savage made specially for the exhibition back in 2015. There was even the Durango 95 sports car from A Clockwork Orange.

I will have to go back and see the show again before it closes in September, because there was just too much stuff to take in over the space of a couple of hours. The level of coverage of Kubrick's career is astonishing, but the immense detail, reflecting Stanley's own approach to making a movie, rapidly gets overwhelming. You need to see it, but set aside the whole day if you're going to do it justice.

After two nights when I got no more than four hours' very restless sleep and two days being much more active than usual (I hit record step counts on my fitness tracker on Saturday and Sunday), when I got home on Sunday I collapsed in a heap. On Sunday night I slept for nearly eleven hours. I did the same on Monday night, too. I've felt pretty much wiped out so far this week, but it was totally worth it.

However I have managed to drag myself into a very hot studio and make some more music. I've now got three tracks under my belt for 50/90 and the prog tendencies are beginning to assert themselves again, which is a good sign. As always, you can track my progress and listen to my efforts here.

It's always unpleasant when your company gets bought by a larger organisation that couldn't care less about the things that made your enterprise special. It's even worse when they subsequently decide to shut you down. I've been there; it's horrible. So my heart goes out to the folks at Vertigo Comics, where it was announced last month that the imprint is being shuttered. Vertigo, in case you didn't know, were the folks responsible for publishing Neil Gaiman's Sandman, Warren Ellis's Transmetropolitan, Garth Ennis's Preacher, and a host of other classic works including Lucifer, Hellblazer, and The Saga of the Swamp Thing (which was written by Northampton's very own Alan Moore). The closure of Vertigo is a very big deal for the comics industry, and not in a good way.

DC is now part of the conglomerate behemoth that currently calls itself Warner Media. Warner Media is itself owned by the telecommunications giant AT&T, because one of the world's largest international telcos is the obvious pick when you're looking for someone to run one of the comics industry's most innovative imprints, right? No, I didn't think so either. Weird. The company alienated a whole new tranche of consumers this week when it announced that the long-running comedy publication Mad Magazine will cease to publish any new material. Mad has been in business since 1952. Instead, the magazine is switching to recycling old material (with new covers) and will close completely in August, once its existing reader subscriptions run out.

When you're younger, you assume that the institutions you know and love are permanent fixtures. You never imagine that, one day, they might just go away. Unfortunately, as you get older you discover that the things you cherish don't last forever. Even so, I was surprised how upset I was by both these bits of news.

I've already got one song under my belt for 50/90 this year. I started out nice and easy with a bit of blues, because it's a style of music I love listening to (and playing) and after all, It All Comes Back To The Blues.

I've also started work on track number two, which is developing along proggier lines, but I can't quite figure out what it wants to be quite yet, so I'm leaving it to marinade for a while. I suspect that means it's going to fall into the "long and complicated" category...

Today, though, I'm feeling a a bit out of sorts. I don't know why, but I can't focus. So I have spent the day doing household stuff like laundry and I'll come back to the music when I'm a little more with it. At least I'm getting some excellent guitar tones out of my gear at the moment. The fiddling about with things over the last couple of weeks is really paying off.

The studio has been tidied up and organised. The guitars are ready and waiting. The software on the studio PC is as up-to-date as it's going to get. My fingers are itching. It can mean only one thing: I'm about to start on another of my annual songwriting challenges.

For Fifty/Ninety the goal is to write fifty songs in the ninety days between the fourth of July and the first of October. I've taken part each year since 2013, and each year I've managed to reach the target and write at least fifty songs (in 2014 I got more than a little bit carried away and managed to write well over sixty). It's a much longer haul than February Album Writing Month and the work rate required is faster; successfully completing FAWM only entails writing one song every two days for 28 days. Summertime also tends to have more demands on your time than the depths of winter, and as a result of all this the number of participants is much smaller, but the 50/90 community is enthusiastic and highly creative, and good fun to be around.

As regular readers of the blog will know, I spent some time last month getting the wrinkles ironed out of my current guitar effects rig, as I'd been having power supply problems. Last night I spent an hour or so playing guitar, and everything worked as it should, so I think I'm all set. I've also changed one or two pedal settings, resulting in my best-ever lead tone; thanks to the judicious addition of a noise gate in the G3's patch, there's no background noise in the signal at all when I stop playing. This is all the result of playing at louder volume during my jam session with Paul last month. As a way of showing how important playing with other musicians is in helping you to develop your sound, this sort of thing is hard to beat.

All my preparation time involves the technology I use to create music. I don't spend any time thinking about creative themes, or pondering the approaches I'm going to adopt until the challenge actually starts. It's not that I think it would be cheating to do so beforehand—it's that learning to "switch on" your creativity at a moment's notice is one of the most important lessons that taking part can teach you. When you've got a deadline to meet, you can't be precious about waiting for your muse. Right now, I have no idea what sort of music I will come up with on Thursday. But I will come up with something on the 4th; that much I do know.

I can't wait to find out what it sounds like.

I was very excited to learn this morning that Netflix is planning to develop a big-budget television series based around Neil Gaiman's Sandman series of comic books. The comics, which chronicle the adventures of Dream, a.k.a. Morpheus, the God of Sleep and his interactions with his siblings (Destiny, Delirium, Desire, Despair, Death, and a lot of other names all beginning with the letter D, collectively known as the Endless), are rightly regarded as classics of the genre.

We are living in a golden age of television at the moment. Thanks to competition from streaming services like Netflix and Amazon, networks have realised that they have to up their game with regards to quality. Studios have also realised that a television series provides a better format for really getting to grips with a story than a ninety-minute or two-hour feature film can manage. The extra space gives plots the luxury of slower pacing and resultingly larger payoffs. The recent adaptation of Neil Gaiman and Terry Pratchett's Good Omens is a prime example (you see what I did there?)

I'm still waiting for the adaptation of Larry Niven's Ringworld, though.