| 190917 RX100m4 01071 | |
| Camera: SONY DSC-RX100M4 | Date: 17-09-2019 16:44 | Resolution: 4650 x 3100 | ISO: 500 | Exp. bias: -0.7 EV | Exp. Time: 1/80s | Aperture: 3.5 | Focal Length: 25.7mm (~70.0mm) | |
Something in the note on my desk this morning hinted that my day was not going to go quite as planned… 🙂
| The E Class gone green! | |
| Camera: Panasonic DMC-GX8 | Date: 03-05-2016 10:19 | Resolution: 5176 x 3235 | ISO: 200 | Exp. bias: -33/100 EV | Exp. Time: 1/100s | Aperture: 8.0 | Focal Length: 16.0mm | Lens: LUMIX G VARIO 12-35/F2.8 | |
Please bear with me – I’m trying to improve the process of generating new content for my blog, and then cross-posting it to Medium, FaceBook, LinkedIn and/or Twitter as appropriate. Over the next week or two you may see a few test posts like this one. Please feel free to ignore them!
One of the challenges is easily generating and uploading the photo blog, which holds all my images. I’ve created a tool which automates generating the blog, uploading it and then downloading links in a form ready to use in the blog. As a test, here’s a picture of my car before the paint job:
| My E-Class before its makeover | |
| Camera: Panasonic DMC-GH4 | Date: 14-05-2015 20:09 | Resolution: 4367 x 2457 | ISO: 500 | Exp. bias: -66/100 EV | Exp. Time: 1/60s | Aperture: 2.8 | Focal Length: 12.0mm | Lens: LUMIX G VARIO 12-35/F2.8 | |
And here’s another of the old VW Eos:
| A randomly selected image this morning – my old VW Eos | |
| Camera: Panasonic DMC-GX7 | Date: 30-01-2015 17:00 | Resolution: 4894 x 3059 | ISO: 320 | Exp. bias: -66/100 EV | Exp. Time: 1/60s | Aperture: 5.0 | Focal Length: 17.0mm | Lens: LUMIX G VARIO PZ 14-42/F3.5-5.6 | |
| That's more like it - hand-held at 1/3s | |
| Camera: Panasonic DMC-GX8 | Date: 13-11-2015 15:44 | Resolution: 3888 x 5184 | ISO: 100 | Exp. bias: -66/100 EV | Exp. Time: 0.3125s | Aperture: 10.0 | Focal Length: 21.0mm | Location: Views of Rinpung Dzong from Pa C | State/Province: Paro | See map | Lens: LUMIX G VARIO 12-35/F2.8 | |
There’s a common line in much of photography writing. Set aside those actively trying to sell you something, and most will at some point claim that "kit doesn’t matter”. The idea is that a good photographer can get excellent results with any equipment. He or she will understand and work within any limitations, potentially even making an artistic feature of them. Conversely the mediocre photographer chasing improvement through better equipment is wasting time and money better spent on training in technique.
However that has not been my experience…
I’ve been a photographer since my teens. Until the mid 00s I muddled along with low-end 35mm film cameras and a variety of relatively cheap lenses, typically “kit” lenses or items purchased randomly from shop displays of used items. Over the years I’d worked up from a “manual everything” camera with fixed 50mm lens to an entry-level Canon EOS SLR which provided automatic focus, exposure and film winding. The photos had their limitations, but I lived with them. They were better than most friends and relatives managed with “point and shoot” cameras, but they didn’t really relate to the sharp, colourful large-format images I saw in magazines or at exhibitions.
That changed with the advent of my first DSLR, a Canon 350D. Now I had a tool capable of producing high-quality digital images, in theory up there with the best of them. OK, auto-focus was slow for anything but well-lit static subjects, and the maximum usable ISO was 800, but by using the histogram I could reliably get correctly exposed and focused shots almost every time, banishing most of the technical issues which had limited my film photography. At the same time I realised that because of the constraints of my work, photography, and travel for it, was really my main hobby, and I wanted to become good at it.
Score 1 for a kit upgrade, but I suspect we all overlook this one.
I knew I needed to improve my compositional skills and my eye for images, but I read widely, attended courses, got some mentoring, and practiced. I do say so myself, but my ability to see, frame up and capture an image improved steadily. I learned to shoot RAW, and started to develop an efficient toolkit to work through and develop my pictures. I took the better ones proudly to my mentor…
…At which point he made a comment about sharpness, and I realised that it was true, many of my images seemed a lot softer than they should. Ignoring those with motion blur (due to subject movement), depth of field limitations or environmental constraints (haze or low light), quite a lot of straightforward static shots seemed to lack “bite”. I tried fiddling with the processing, but to little avail. I wondered if the problem was camera shake, but that seemed unlikely as I have a steady hand, and I proved that I could get sharp shots with my telephoto at full stretch and a moderate shutter speed, and from my non-stabilised wide-angle lens. That should have told me something, but it didn’t.
At the time my main lens was the Canon EF-S 17-85mm f4-5.6 IS USM. This was a decent lens, neat, smooth in operation with good stabilisation, and received decent reviews. It did suffer quite bad chromatic aberration near its limits, but that was usually correctable. However I became convinced that simply because it was relatively cheap maybe it was the culprit.
So I set about trying to find a better mid-range Canon zoom. This proved easier said than done. Canon had multiple well-reviewed zooms which were 24mm at the wide end, but that’s only "wide" on full frame, and useless for my style of photography with APS-C bodies. I borrowed a 17-40mm lens, but that seemed heavy, lacked image stabilisation and didn’t seem to produce much better results than what I had. Ditto the 17-55mm, which is a good lens, but a big lump for what it is. At the same time these all equated to about 28mm at the wide end, and I was ideally hoping for something a bit wider. My options seemed limited.
At which point, Canon released the EF-S 15-85mm f3.5-5.6 IS USM lens. This looked like the answer to my prayers: similar range to the 17-85mm but a bit wider, marginally faster, and about twice the price, so it had to be better, didn’t it? I read reviews and tried one in a shop, all of which looked quite promising. I bit the bullet and purchased one.
Picasso had his blue period. This was the beginning of my blurry period.
I don’t mean that every shot I took was a fuzzy mess. In well-lit conditions straight on to a static subject with no vast challenge on depth of field the results were OK. Subject movement wasn’t a problem either – the lens played very well with the Canon 7D’s excellent autofocus on moving targets. However in terms of my images being a bit disappointing on the sharpness front, if anything the rate seemed to have gone up.
Nice colours, no shake, but still blurry
Some of this was down to my technique. I was arguably becoming too reliant on multi-purpose autofocus, and maybe not paying enough attention to depth of field complexities. Some was due to a straightforward mechanical weakness of the 15-85mm lens: pointed too far up or down the front element would move under its own weight and disturb zoom and focus, but I learned to recognise and manage that. However the fact remained that some images which should have been consistently sharp just weren’t.
Things came to a head on the first day of my Iceland trip, when I suddenly realised that only the autofocus lights in the bottom 2/3 of the viewfinder were active – the others never came on. The lens just wasn’t focusing properly on objects at the top of the image. I swapped to the 17-85mm lens and the problem went away, so that became my main lens for the remainder of the trip.
The 15-85mm lens was still under warranty, so it went back to Canon for repair. Actually it went back twice, as the first time it was returned "no fault found" and my carefully listed symptoms clearly ignored. The second time Canon reported that they had adjusted the front element of the lens. It was a bit better, but not right. I could point it straight at a wall, and either the top of the image would confirm focus or the bottom, but not both.
I went for a walk with friends, and took the shot below. It may not show up well at web resolutions, but it’s a very odd image. The roof tiles reflected in the water are sharp. The tiles photographed directly, without the challenges of reflection, but by definition at the same optical distance, are blurred.
The Canon 15-85mm lens’ failings uncovered
After that walk I did another review of the market, but was still unexcited by any other option. I sold the lens (for a fairly low price to a happy buyer, I checked), and bought another, brand new. It was a further step in the right direction, but I still couldn’t be sure that I was getting the images I should.
One problem is that it’s very difficult to understand the limitations of your kit if that’s all you have to compare. It’s a bit like trying to assess the benefits of Blu-Ray via an advert on a DVD – you only have a DVD quality image to judge. I call this the “can’t tell through current medium” problem. At the risk of channelling Donald Rumsfeld, you don’t know what you don’t know…
I was getting frustrated, and it shows in my portfolio. After the walk which generated the top image, the Canon 7D hardly contributed apart from sporting events, where coupled with the 70-300mm lens it continued to shine. Everything else was taken with other, supposedly "lesser" cameras.
When I bought the Panasonic GH2, I wasn’t intending to buy a "better" camera. I’d become attracted to the idea of mirrorless cameras, and I wanted a "full capability" camera kit which was genuinely small and light. In truth there was also a bit of gadget lust, partly bought on by my growing frustration with Canon, who were also very tardy in upgrading the 7D. Driven by the small/light mantra, I chose as my first micro four-thirds lens the 14-42mm "power zoom" (which folds itself down to a pancake when not in use), a lens which requires prodigious geometric correction in the camera or RAW convertor.
And the images it produced were so sharp, they just "popped" off the screen. Casual grab shots with the GH2 had a clarity of detail and colour I had rarely matched with the Canons, and only readily achieved with the excellent EF 70-300mm f/4-5.6 IS USM lens, rather than the mid-range zooms. I had found my reference!
My first real shot with the GH2 – sharp!
It’s surprising in hindsight, but even with this realisation, it was a long 18 months before I completely moved on. I was both personally and financially invested in the Canon system, and couldn’t change instantly. The GH2 was only 12MP (only 9MP at some aspect ratios). It struggled if the subject was actually moving, by comparison a great forte of the Canon 7D. In addition the early micro four-thirds cameras and lenses were tiny but felt fragile, and I was more disposed to expose the Canon kit to rain, dust or the sands of the Erg Chebbi. Underneath it all I suspect I was still somewhat in denial that a much cheaper, as well as smaller kit was capable of superior results.
The end of the transition came suddenly, via an accident which was happy for me, less so for my friend David. He was trying to shoot the swirling floodwaters of Winter 2013-14, knocked his tripod, and in went his 7D and lens. He wanted time to choose an upgrade replacement, so offered to buy my 7D as an interim solution. He also took a couple of lenses, but wasn’t interested in the 15-85mm. He’d obviously heard me swearing once too often!
The rest of my Canon kit went on eBay. Most sold quickly and for good prices. There was one exception: the execrable 15-85mm took months to sell and achieved a very low price. I was slightly chastened, but not really surprised.
Somewhat before the end I had mentally and practically moved fully into the Panasonic system. I loved much about it, especially the image quality, but also my kit finally included something which Canon had never been able to provide, a lightweight high quality mid-range zoom (the wonderful 12-35mm f/2.8, beloved even of lens snobs). My blurry period was over.
Now I’m sure there are plenty of people doing good work with APS-C Canon cameras, working carefully within the limits of the lenses, or living with their limitations. I could always have invested instead in a bunch of primes, or maybe I might have fared differently if I had got better results from my trial of the 17-55mm lens. However the reality is that I just couldn’t believe Canon would sell bad lenses for good money, and tried to "stick it out", rather than moving on sooner. For every photographer who is constantly chasing the next big thing, there’s probably one like me, constrained by the "I’ve bought it, so I must use it" mentality (or maybe just limited financial resources).
What moved me on wasn’t any clever analysis of reviews or lens performance charts. It was a few quid burning a hole in my pocket, frustration, a bit of gadget lust and a couple of inspiring Panasonic adverts. Effectively "Gear Acquisition Syndrome" saved my photography. I’m not sure to whom that should be a lesson, and I can imagine that this article may well not go down well with partners of dedicated gear nuts, but this is a true story, and you will never hear me say "kit doesn’t matter". I don’t agree.
While this is an enjoyable read, it prompts one big question. Why did the author feel that a heavily fictionalised re-telling of this utterly thrilling true story was needed? In the preface Gross says that he wants to tell “the story of how only a few brave men put an end to that threat”, but but then proceeds to invent a cast of central characters who are at best “drawn from” the real players and have their names changed. My decision to read the book might have been different if I’d realised up front the level of fictionalisation.
The central part of the book (between the commando raid and sinking the ferry) is almost entirely fictional, involving “Kurt Nordstrom” in not one but two love affairs. Now I get that “Kurt spent the summer of 1943 on the plateau eating reindeer and dodging the Germans” isn’t going to fill a lot of pages, but a shorter more focused tale would have been fine. Once you realise that this section is what it is, it calls into question how much of the remainder is historic.
The irony is that a lot of this is unnecessary. By Gross’ own admission, the dramatic chase which separated one of the escaping Gunnerside team from the others actually happened, just to another character not the invented American, and the true story of how the plant’s night watchman interrupted the commandos setting the explosives not once but twice in search of his glasses is both funny and more dramatic than the way it’s told here.
Beyond that, the story has been told well, with less fictionalisation, several times in recent years. The BBC documentary accompanying Ray Mears’ excellent 2003 book was superb, with interviews of many of the real players. I thoroughly enjoyed the tri-partisan 2015 TV series The Saboteurs which succeeded in portraying the perspectives of not only the Norwegian commandos and their supporters, but also the British and Norwegian commanders, and key participants on the German side. Even the still enjoyable 1965 film sticks to the truth at least as much as Gross’ book.
The book was originally published under the title The Saboteur, which makes perfect sense, but then got re-titled The Spy, which makes none, as there’s very little spying involved, and a lot of sabotage. Maybe this was to avoid an obvious clash with the international TV series, but it raises another “why?”.
If you want to read an enjoyable wartime romp with some real key events, then this book is fine. If you’d prefer to understand the background, achievement and the real players, track down one of the TV series.
| A randomly selected image this morning - my old VW Eos | |
| Camera: Panasonic DMC-GX7 | Date: 30-01-2015 17:00 | Resolution: 4894 x 3059 | ISO: 320 | Exp. bias: -66/100 EV | Exp. Time: 1/60s | Aperture: 5.0 | Focal Length: 17.0mm | Lens: LUMIX G VARIO PZ 14-42/F3.5-5.6 | |
My programming project of the last few weeks has been to build my own “rolling portfolio”, which shows random images from my photographic portfolio as either a screensaver or a rolling display on a second monitor. I’ve implemented a number of features I’ve always wanted but never had from freeware/shareware options, like precise control over timing, the ability to quickly add a note if I see a required correction, and the ability to locate and review recent images if someone says “what was that picture you were just showing?”.
Having previous blogged about the poor quality of “random” algorithms in Android music player apps (see How Hard Can It Possibly Be?), I decided to put my money where my mouth is, and write my own preferred random algorithm. This does a recursive, random walk down the selected folder tree, until it either finds an image file, or a dead end (and then tries again). This was refreshingly easy to implement, and as expected runs quickly without needing any prior indexing of the content.
Also as expected, the simplest implementation returned a disproportionate number of hits (and therefore a lot of repeats) from folders with a very small number of images, but that was easily fixed by adding a “weighting” at the second stage of the walk, to reduce the number of hits on smaller portfolios.
Job done? Maybe. I started to notice that I still see the same image selected twice in quick succession, and sometimes more than twice over a day or two. At first I thought this might be an issue with seeding the random number generator, so that I was re-generating the same random sequences, but a quick check confirmed that wasn’t the problem. The next most obvious possibility (to me!) was an issue with the Microsoft .Net random() function, so I added some logging to the app, recording each random number, and then fed a day’s worth through some frequency analysis in Excel. That got Microsoft off the hook with a clean bill of health: there’s a slight preponderance of zeros, which I can explain, but otherwise the spread of results looks fine.
At the same time, I also added logging for the selected images themselves. In yesterday’s work hours operation the screen saver showed 335 images, of which no fewer than 21 were duplicates. Given that I have over 3500 images in the portfolio, this seems very high, but maybe not.
This is a known problem in mathematics, a generalisation of the “birthday problem”. It’s so known, because a common formulation is the question “given a room of people, what is the probability that at least two have the same birthday?”. While you need at 367 people to guarantee a duplicate, the counter-intuitive result is that with just 23 people in the room, it’s more likely than not. The generalised equation for the solution is the following:
E = k – n + n(1 – 1/n)k
In this n is the number of items, k is the number of random selections, and E is the expected number of duplicates. Feed in k = 335 and n = 3500, and you get the outcome E = 16. That’s close enough to my observed value of 21 (this is all random, so any one measurement might be either side of the expected value, but the order of magnitude is right). Couple this with the way my mind works, looking for patterns, and I must therefore expect to see some repetition. However it’s clear that the algorithm is working fine, it’s just the normal workings of probability.
Another implication of this is that as the sample grows, some images will naturally appear several times, and others may not appear at all. If we take 3500 samples, the expected number of duplicates rises to over 1200, so over 1/3 of the images will still be unselected.
Do I fix this? The relatively simple resolution is to keep a list of selected images, and use that to discard any selections which are repeats during a given period. However I would rather run this without a data store and maybe, now I can explain it, I’m comfortable. Time will tell.
| A purple unicorn |
A fable, sort of…
The jester who wanted to be king asked the crowd “Do you want a purple unicorn?”
Almost half the crowd said “We are happy as we are, and we don’t believe unicorns exist”, but slightly more than half said “Yes please, we’d love one.”
The king wanted his people to be happy, so the king’s men spent three years looking for a purple unicorn. They spent much of the kingdom’s treasure, and annoyed many of the kingdom’s friends, constantly asking for unicorns which they did not have, because unicorns don’t exist. Other realms laughed at the king and his kingdom, and important matters in the kingdom went without attention.
Eventually the king’s men came back and said “There are no unicorns. We’ve found a nice horse we can paint purple and glue a fake horn on its nose, will that do?”
All those who had never believed in unicorns said “told you so”. Many of the others said “OK, we agree”. But others were angry, blaming those who had never believed in unicorns. Some were so obsessed with the idea of a unicorn that they wanted to shoot all the horses, just so they would not get a horse painted purple with a fake horn glued to its nose.
A wise king could simply have stopped the search, saying “unicorns do not exist”. A wise king could have told the crowd “You may vote again. But just to be clear, unicorns do not exist, so you are voting for a horse painted purple with a fake horn glued to its nose.” But the foolish king thought that the only important thing was not upsetting those who wanted a unicorn, so the search continues…
I’m slowly working through, and very much enjoying the BBC series Icons. There’s been a lot of discussion about whether it makes any sense to have a "final" in which "iconic" sportsmen, politicians and scientists go head to head, but that’s not the real issue. The big problem is that the series has wholly the wrong name, and should be called not Icons, but Heroes.
To qualify as a 20th Century icon, a person should be:
The simplest and least controversial example is probably Albert Einstein. Everyone knows that smiling face and wild hair. Through a series of seminal papers in 1905 and the following years he established not only relativity theory but also key elements of quantum theory, the two major planks which ensured that 20th Century Physics diverted strongly from the Victorian version. I can use the single word "Einstein" or draw a very crude cartoon of a smiling face with spiky hair and a bow tie, and it immediately invokes a range of concepts in the beholder.
Einstein was also a hero. He completed his early work despite a number of personal and academic setbacks. A Jew, he escaped Nazi Germany and helped the allies to defeat the axis powers, but then became a strong proponent of denuclearisation and peace. He qualifies both ways.
At the other end of the scale, consider Ernest Shackleton. Shackleton is undeniably, absolutely a hero. The story of the Endurance voyage, and how his leadership bought them all back safely despite horrendous tribulations bears endless retelling. Shackleton is certainly a personal hero to me: I have read books about him, watched programs, travelled to exhibitions. I managed to track down a copy of the wonderful dramatisation by Kenneth Branagh and we re-watched it just a few weeks ago.
But is Shackleton a 20th Century icon? How many people would recognise a picture of him out of context? I might struggle myself. Also in many ways he represents not 20th Century exploration, but the end of the Victorian era: plucky white men opening up the dark areas on the world’s map. There is a case for considering the Endurance story as a precursor to Apollo 13, that other great 20th Century tale of explorers rescued, but it’s not a strong one.
If you want an iconic 20th Century explorer, you really have to focus on aeronautics or the space race. There are many heroes, but the best chance for an icon is probably Neil Armstrong. We may not all instantly recognise his face, but that picture of a man in a spacesuit standing next to the American flag, or those words announcing "a giant leap for mankind" are certainly iconic, and representative of a type of exploration which didn’t exist before, and no longer really exists now.
In the political space, Winston Churchill is certainly an icon. His name and face, even in caricature, immediately invoke concepts such as strong leadership, freedom fighting tyranny, a blend of conservative and liberal ideals sadly lacking today. He is a clear exemplar of one side of 20th Century politics. His qualification as a hero is more nuanced: his amazing talent for being in the wrong place at the right time, his determination to do the right thing, his dominant skill as a leader, orator and writer all support it. However I acknowledge that his position on issues like Ireland and India, and his errors such as over Gallipoli and Singapore do at least slightly offset his great successes elsewhere. His icon is also capable of being misused, for example by those who view him as a symbol of British independence, who carefully ignore his post-war advocacy of unifying international institutions such as the UN and EU.
However, if you accept Churchill as a 20th Century political icon, you also have to consider another: Adolf Hitler. As an icon he qualifies without question. We instantly recognise his name and image, even if it’s just a simple cartoon of the hairstyle and moustache. He also stands as a clear exemplar of the other side of mid-20th Century politics, and a clear warning of the risks of allowing the rise of his like again. Icons do not have to be heroes. They can be villains.
A basic qualification for iconic status is that someone, or something must be famous, or infamous. However the BBC series has been so determined to not just parade a series of middle-aged white men that they have made some odd choices with the candidates. I enjoyed the story of Tu Youyou, the Chinese lady who discovered an important antimalarial drug, but can you honestly propose as a "20th Century Icon" someone whose individual identity was carefully suppressed until well into the 21st?
I have just sneaked a look at the results, and I see that neither of my prime examples of unquestioned icons (Einstein and Churchill) got through to the final. That doesn’t matter: I also consider both Turing and Mandela among my heroes, and I will still enjoy the rest of the episodes. However even if it risked being confused with that series about people with imaginary physical super-powers, rather than just real mental ones, I think the series should have been called simply Heroes.
I was really looking forward to The Favourite. It had a lot going for it. The period – the reign of Queen Anne, the end of the Stuart dynasty and the wars with Louis XIV – is an important piece of British history but has rarely been tackled in modern popular culture. The core plot of battling rival Queen’s favourites was a good one, with great scope for a hilarious period romp. The three strong leading actresses, great costumes and the wonderful Hatfield House locations were all strong positives. And yet I came away feeling very disappointed, that I had just spent two hours watching a very average film, with a great film trying to get out.
The main problem is that the film seems to be suffering a major identity crisis. It can’t quite decide whether it’s a bawdy romp, a serious drama about interpersonal dynamics or a sombre historical portrayal. As a result it misses all three targets. I appreciate that Anne was quite a tragic figure, and there are some unavoidable poignancies, but a slightly lighter touch would have let the "bawdy romp" out to play, rather than keeping it somewhat repressed.
The music (I’m very tempted to write "the alleged music") really doesn’t help. It actually took me a while to realise that the incessant sawing on a couple of string instruments was actually meant to be incidental music, rather than just some weird background sound effect representing troublesome Stuart plumbing or the like. This gets to the point where you are listening to it rather than following the action and dialog, and then you suddenly get blasted by a full "Night at the Opera" fanfare on the organ, for even relatively minor twists in the plot. It reminded me of the ghastly alleged music in Dunkirk, but at least Christopher Nolan had the excuse that he was trying to tell three overlapping stories with time playing out at completely different rates, which isn’t the case here. Now I know that occasional twinkles on the virginal or the odd burst of something recognisable by Vivaldi are very traditional, but they would have worked here, whereas what was provided definitely did not.
The dialogue was unnecessarily crude. I have no problem with swearing, but it needs to be in the right context and period. If I watch something about 21st Century New Jersey gangsters (for example), and every other sentence has a f*** or c*** that’s fair enough. Take two incompetent hit men, stick them in an enforced holiday in a charming Belgian town, and you can even make that a hilarious core part of the work, as in the incomparable In Bruges. But in The Favourite it just feels out of time, and out of place. It’s also wholly unnecessary. 18th Century English was rich with frequently hilarious euphemisms for sex, body parts and so on, and a bit of effort could have sprinkled these into the dialogue to much greater effect. As Upstart Crow demonstrates so well, you don’t even have to bother with the detailed historical research: I have no idea whether Elizabethans actually used the term "cod-dangle", but we all get what it means, it sounds right, and it’s genuinely funny.
I did enjoy some bits of The Favourite, and I did laugh in many of the right places. It’s not The Revenant. But just as the central character in that film spends a grim time hiding from his protagonists in a horse’s corpse, The Favourite feels like a better film has been partially hidden by a layer of pretentiousness and crudity which it really didn’t need.
I realised a couple of weeks ago, much to my horror and chagrin, that I had been walking around with a potential incendiary bomb. Not that I had done anything wrong – this is a more common practice than we’d like to admit, and it’s quite possible that some readers are doing the same, equally unaware.
The culprit was the battery in my 2015 MacBook. Now unlike the batteries on older laptops, this is sealed inside the alloy case, and not immediately visible to inspection. My laptop was still working fine, with battery life still around 2 hours even under quite hard use, which is not bad for a hard-used 3 year old machine. It was running a bit warm, especially in Namibia, but not so much that it indicated any real problem.
The only thing which was a bit suspicious was that it no longer sat flat on a table. The middle of the base-plate seemed slightly raised relative to the edges. At first I blamed myself, thinking that when I had taken the base off to check to see if you can upgrade the hard disk (you can’t, but that’s another story), I hadn’t screwed it down straight. However over time the problem seemed to be getting a bit worse, and I also started to note that the lid didn’t always close completely flush.
I would probably have let this go on a bit longer, but I happened to mention it to two others on the Namibia trip, who immediately suspected the possibility of a dying and swollen battery. Now this can be a serious issue, so as soon as I was back I went on eBay to order a new battery (fairly readily available at about ÂŁ70), and opened the laptop up for inspection. If I wasn’t already convinced of the problem, I reached that point when I had undone about three screws and the base literally “pinged” open. With all the screws removed I could see that not only one, but all six sections of the battery were badly swollen. Yikes!
So the battery definitely needed replacement, and a new one was on the way. I carefully discharged the old one by playing a movie until the battery was below 2%, and switched my work to my spare machine. I then started researching the process of replacement.
Now pretty much every laptop I have owned or used in the last 20 years has the following simple process for battery replacement:
In some Toshiba and Dell/AlienWare machines you can even do a “hot swap” without powering the machine down. The 2011 MacBook gets a bit more complicated, as the battery is inside the case, but it’s still pretty straightforward:
So surely, it wasn’t going to be that difficult to do the 2015 MacBook battery? Surely not?
I should coco. Like the 2011 Macbook has standard memory boards, and the 2015 device has soldered chips, or the 2015’s SSD has a unique connector, Apple have made battery replacement deliberately difficult. This is the one component which is highly likely to fail through age before the rest of the machine, but it is glued to the baseplate, with key components then mounted over it. The iFixIt Guide has no fewer than 72 steps (I’m not making this up), at which point you have stripped almost the entire laptop, used some quite powerful solvents to melt the glue, and have your new battery in place with the instruction “To reassemble your device, follow these instructions in reverse order”. The last time I followed 72 instructions and then “reassembled my device following the instructions in reverse order” it took me two days, and I ended up with a Renault 5 with working engine and clutch, but 5 large bolts left over. Not keen.
Should I get professional help? For a machine up to about 5 years old, Apple will do a battery replacement, for about ÂŁ300. Apparently they strip out all the components from your MacBook and mount them into a new chassis complete with new battery, keyboard and trackpad. Presumably the old chassis and related components go to the skip. However apart from the time this might take, I could see my MacBook coming back with all my keyboard customisations undone, and my hard disk which boots into Windows carefully wiped and OSX installed. Not keen.
[Aside: this is still a better position than if you go to Apple with a 5+ year old machine seeking service. Their official position is apparently “We are happy to recycle this for no cost. Here’s the price list for a brand new one”!]
I could look for a specialist, but again I was concerned about timing, and whether I’d get the machine back as I left it. So I decided on a self-fix, but trying to find a solution that didn’t mean stripping out the motherboard and all the peripherals. Now I could see that it might be possible to get a lever under the outer battery cells (the six cells are largely independent) without major disassembly, so I decided to try that route, hoping fervently that the iFixit guidance was overkill (as it appeared to be).
Obviously it’s a bit risky levering up already damaged lithium ion batteries, as you don’t want a fire, but hopefully the risk would be small since they were almost fully discharged. I took the precaution of having a heavy saucepan and lid sitting on a metal skillet at the end of the desk as a fire bucket, and used plastic tools as far as possible.
I also sourced a Torx T5 screwdriver for the internal screws. While the case screws and the inner screw heads look similar, the former have 5 points, and the latter 6. Just to make it a bit more difficult. Actually I’m not surprised Apple have a five-pointed design – the pentagram fits well with their generally Diabolical attitude to service, maintainability and the risk of immolation from faulty batteries…
So here’s my rather shorter process for replacing a 15″ Retina MacBook battery:
It worked, I didn’t set fire to anything, and my laptop now sits absolutely flat on the table. We’ll know shortly how life of the new batteries compares with the old.
However, it really doesn’t have to be this way. If Apple cared remotely about their customers and the environment instead of screwing everyone for the maximum revenue then the battery replacement would be a simple clip or screw process similar to the 2011 version, optimised for repairability rather than designed to actively minimise and inhibit it. I’m not impressed.
| Sleepy Cheetah! | |
| Camera: Panasonic DC-G9 | Date: 16-11-2018 17:29 | Resolution: 3345 x 3345 | ISO: 800 | Exp. bias: 0 EV | Exp. Time: 1/500s | Aperture: 5.6 | Focal Length: 193.0mm | Location: Okonjima | State/Province: Okonjati, Otjozondjupa | See map | Lens: LUMIX G VARIO 100-300/F4.0-5.6II | |
The last day of any trip is always a bit sad, and hard work with the travel. However this year three separate organisations covered themselves in something which is not glory, and I have to get this out of my system before I write the traditional tail piece for my blog…
As a bit of compensation, here’s a nice picture of a cheetah, feeling about like I did at 1am on Friday.
Thanks.
The recurring dysfunctional ingenuity of hotel designers never ceases to amaze me, and provides an endless supply of material for this blog. On our way back through Windhoek, we stayed at Galton House, which while still quite smart overall and in the communal areas, was probably half a notch down from the Pension Thule, where we stayed outbound. My room was a bit poky and had a couple of major challenges, including very noisy air conditioning, and an Iceland-class duvet (in Windhoek, in the summer!). However the worst fail was that it had one accessible power socket, to the right of the bathroom door, while the desk, the only place I could rest laptop and things on charge, was to the left of the same doorway. I therefore had to spend my stay with a power cable stretched right across the bathroom doorway, limbo dancing under it when I needed to use the facilities.
One wonders what sort of hotel designer comes up with a room with a desk, and no power socket on the appropriate side of the room. That’s up to the standard of the Calais hotel I once stayed in where the lift worked but the stairs were out of order (due to a 10 ft gap half way down.) Admittedly about 20 years ago I did stay in a hotel in England where the only place you could get simultaneous power and modem connectivity was in the hot tub in the middle of the room, but that was an adapted medieval abbey, and plain weird. Galton House is a smart new purpose-built venue. Not a clue…
And to add actual injury to potential injury, most of us had got the whole way around Namibia without many bites, and several of us, including myself, woke up covered in nasty little red marks. Blast.
I’m fully in favour of Virgin holding onto the "full service airline" concept when BA and others have abandoned it. However, if you are running a night flight which leaves Johannesburg at 21.00 local time, and arrives in London at 06.00 local time, I would suggest that your highest priority is to try and enable your passengers to get as much of a decent night’s sleep as airline seating and turbulence allow. This is not promoted by serving, slowly, drinks, followed by a rubbish collection, followed by tepid towels (they were probably hot when they left the galley, but I was at the front of Economy), followed by a rubbish collection, followed by "supper", at about 00.30, followed by hot drinks, and finally followed by another rubbish collection at gone 1 in the morning!
…Followed by inedible breakfast, at about 05.00…
Would it really not be better to just give everyone some booze and turn the lights off?
I wasn’t impressed by the 787 on the flight out, but my assessment reduced a further notch on the way back. That plane revealed a number of areas where the new technology has aged very badly. One example: the window dimming switch on my window had obviously been jabbed so frequently and hard that the rubber cover had completely failed and peeled away. Worse, the toilet is supposed to retain the seat upright via some magnetic mechanism, with a nearby "non contact" switch operating the flush. In the loo nearest my seat the seat retainer had completely failed, meaning that I had to sit holding the seat upright with one hand, and every time I moved, the flush mechanism triggered randomly.
This was all on a "nearly new" plane which has by definition only been in service for a couple of years. How that plane will look after 10 or more years use I shudder to think.
I suspect that the 787-200 or whatever they call the "2.0" version will be a good plane, but I’d hate to be in charge of maintaining the oldest versions.
In fairness to Galton House, Virgin and Boeing, I arrived back at Heathrow at 06.00 safe, sound and slightly ahead of schedule. In the words of Old Blue Eyes:
It’s very nice to go trav’ling
To Paris, London and Rome
It’s oh so nice to go trav’ling
But it’s so much nicer, yes it’s so much nicer, to come home…
| Poly-tickle commentary | |
| Camera: Panasonic DC-G9 | Date: 24-11-2018 09:49 | Resolution: 2409 x 3213 | ISO: 200 | Exp. bias: 0 EV | Exp. Time: 1/500s | Aperture: 4.9 | Focal Length: 193.0mm | Lens: LUMIX G VARIO 100-300/F4.0-5.6II | |
Sorry it’s a bit fuzzy and not properly focused, but that’s nothing to do with my photography!
Banks constantly tell us to do more to protect our financial details against online fraud, but we live in a world where there is often no alternative to exposing important financial information to potential misuse. The frustration is that there are some relatively simple services the banks could provide to avoid this, but for some reason, probably just their inertia, these are currently unavailable to a lot of users.
Paying for stuff online frequently involves a big act of trust – when you type in your credit card details you are effectively handing the receiving party the keys to thousands of pounds of your money. You want to hold the merchant to a very high standard of behaviour with those details, which is probably justified for a big household name, but what about other cases? A smaller organisation may be perfectly honest, but may hold your card details in a form which could be vulnerable to an unrelated attack.
Worse, the payee might not have honourable intentions for your card details. You don’t have to be doing anything very nefarious to come across potential examples: the other day I was trying to track down a manual for a second-hand watch, and the only download sites wanted me to "register a credit card" before proceeding. Possibly innocent, quite possibly not.
I really shouldn’t have to expose powerful payment credentials in such a situation. My strong preference is to use a trusted intermediary like PayPal, but that’s not always an option. The best alternative solution is the concept of a "single use credit card" – a set of virtual card details used for one specific purpose, with a short lifetime and very low "credit limit".
However while this is a well-established concept, actually getting hold of such details turns out to be very difficult. As far as I can see, no mainstream UK bank offers this service. Several of the big American banks do, but not to UK customers. Capital One have such a service built into their online support tools, and I have one of their cards, but I couldn’t access those tools with my credentials.
There are a couple of third parties offering the service in the UK, but often only with an expensive subscription. The honourable exception appears to be EntroPay. It’s a bit fiddly getting set up so that you can load their cards from your regular credit card provider, and cost me a 20 minute call to my bank, but I now have a virtual credit card with a ÂŁ5 credit limit and no other uses. Ideal, but harder than it should be.
This is not rocket science. The fact that several major US banks readily offer such services confirms that this is feasible. We pay substantial fees for access to banking, so why can’t UK banks follow suit?
In the move from cash and cheque to direct bank transfers even for small personal payments we have also adopted another behaviour which is perilously close to leaving your keys on your front doorstep. This is the practice of sharing your bank account details with anyone who offers to send you some money. This is another practice which leaves me deeply uncomfortable.
Again there is a relatively simple solution. Your account should have a second "shadow" number which can only be used for paying in money, not for withdrawals or other actions (although it might be the visible account number on payments you make). This becomes a "public key" which you are comfortable sharing, while the real account number remains a private secret shared only by yourself and your bank. That then becomes a useful piece of two-way authentication, whereas at the moment someone who knows your account details could have got them from a discarded email or similar. If someone only has the "public" number, then neither your nor your bank should take any instruction from them.
The idea of public and private keys is well established in the electronic world, and ironically the banking system has used physical versions for years – think, for example, of the "hole in the wall" deposit machines for which many people have a key allowing deposit, but only the bank has a master key for collection. However, I’m not aware of any UK banks offering this simple service.
The next is as much about error as fraud prevention, and may be specific to certain banks, but certainly in the Lloyd’s system if you are setting up a personal payment there is zero feedback on whether you have the right account number . The system doesn’t even require you to type in the number twice for confirmation.
Any party in the chain might have made an innocent error, and if the result is a valid account and sort code combination then the funds will be misdirected. If you received payment details via some insecure mechanism such as email, it is also not impossible that a fraudster could substitute their own details, and you would be none the wiser until the real recipient complains about the missing payment.
I suppose banks might argue that showing the account payee name could allow a certain level of account number "guessing", but that sounds specious to me. The simple solution is to combine this change with the payment-only shadow number concept above.
Finally a simple prophylactic against the "your money is in danger, please put it in this account (of mine)" scam. Banks could insist on either two days’ notice or a personal phone call before any transaction which either largely empties an account, or exceeds a certain threshold. Notice could be provided via the banking application to cut down on administration. For most users, most of the time, this would be no problem, and it would require that any more significant transaction is either planned, or has a "cooling off" period in which fraud checks could be carried out. "Instant access" would still be possible, but only after a phone call or bank visit in which you could be asked "has someone told you to do this?".
Credit card companies do this all the time – mine insisted on an exchange of texts and a call to OK a payment of ÂŁ5 to Entropay. Yet I know someone who emptied three accounts under a scammer’s instructions before a bank manager asked the key questions. There’s a bit of a mismatch there.
We all need to play our part in fraud prevention, but that goes double for the banks, and a few simple service enhancements along the lines above would make financial life much more secure for all.
List
Abstract
One+Abstract
Email me
Others
Thoughts on the World (Main Feed)
Main feed (direct XML)