THE ONLY CONSTANT IS FAILURE

Well I have been perhaps a little depressed lately but nowhere near 
as bad as that heading makes out. I was going to use "the only 
constant is change" as a vague pun about PHP v8 no longer willing 
to accept unquoted variables as array indices, insisting that such 
unadorned labels must refer to a defined constant. It'll probably 
be a long while until I switch over to PHP v8, but I fixed a 
handful of such instances last week in a diversion while fixing 
bugs in my online store, probably a result of me either getting 
confused by, or copy/pasting from, the unquoted array variable
format inside strings (which I was relieved to confirm isn't
changing). But change and failure are the same, everything that 
works will break after enough change, and that's a better theme for 
my past week. This doesn't just refer to carefully-argued changes 
in programming languages, or even to human decisions in general. 
Change is rife as much in the things we ignore as in the few 
intellectual topics we feel passionate about. Wear and decay are 
change, eating relentlessly away at the physical realisations of 
designs once outlined in a coneptual state of pure perfection. 
Mistakes are change, the little differences in process that dart 
reality outside the bounds of our own intended outcomes, building 
and cancelling themselves in an unseen rhythum that has a tiny 
potential to one day peak in  absolute catastrophy.

In the real world these changes sometimes create success: Changes 
to designs are of course made to improve them, decay of designs can 
create its own beauty, and serendipitous mistakes frequently 
shortcut technological advancement. But none of this is guaranteed. 
The only constant with change is that beyond some unknown threshold 
it will lead to failure, of human concepts and intentions in all 
their physical manifestations. This is the weight of change that we 
all have to lift, raising it above that threshold of failure.


So I've been busy with that heavy lifting lately. The website thing 
killed me - fix one bug, cause another, and I only seem good at 
discovering bugs when I go back to test it one last time late at 
night. Then I stay up late while my tired brain insists on finding 
fixes to them before bed, which then have to be fixed again the 
next day because I was tired and doing a rubbish job. Though 
somehow tired-me is much better at least at getting to the root of 
a problem than awake-me who tries too hard to understand everything 
that's going on and ends up wandering down all sorts of dead ends. 
The result is that I only seem to make progress at this in 
day/night cycles, which frustrates the hell out of me, especially 
because before a night's testing I always think I've finished.

That got me into my late night / late start cycle again, so I made 
an effort to catch up over the weekend and hence I have time to 
write this phlog post in the morning for once. Though I'm running 
out of that time because I'm go extremely slow when I start on the 
philosophical stuff, so this is about to get rushed...

The weekend was stuffed up by my washing machine dying. It's the 
newest appliance that I use besides perhaps the A/C, so in my 
pessimism about modern technology I suspected it would be the most 
likely to break, and I was right. It was the control electronics 
that failed, so I set to fixing them. I'm pretty rubbish at 
debugging electronics as well - I inderstood it fine and quickly 
tracked the problem down to the switch-mode power supply (which in 
a typical show of excessive cost-cutting, isn't isolated from the 
mains so testing it safely is very inconvenient), but then got 
bogged down in checking every part of the circuit and was 
bamboozeled by an unusually-labeled Chinese-brand PTC thermistor in 
series with the AC input, substiuted for the fusible resistor in 
the datasheet's reference circuit. I also lost my place and started 
referring to the wrong schematic in the datasheet at one point, 
which at least caused me to _absolutely_ rule out every component 
except the one that has failed. Unfortunately that's the one 
component of which I haven't a suitable spare - the switch-mode 
controller chip. It's widely available at least, but to buy one 
that's probably not a Chinese fake I've got to pay $15 postage on 
top of the $1.70 cost for the chip, and it'll probably take a week 
for the "next day delivery" to my location outside of the 
internet's known universe. What really pisses me off with the 
design of this machine though is that they have a rotarty switch on 
it with an off position. Does the switch actually cut the mains to 
the power supply and power-off the microcontroller in the off 
position? No, the microcontroller turns all the lights off, but 
that circuitry is still on 24/7 waiting for that one day of the 
week where you actually use the thing for an hour or two. Why? So 
the manufacturer can save 50c by not needing to use a mains-rated 
rotary switch. It's bloody ridiculous. I might hope that 
environmental responsibility ideas would have had them fix that 
since my model was made in 2011, but odds are the current model 
probably uses even more power in its pretend switched-off state so 
that it can talk to smartphones and phone home via the internet. 
Anyway I never trusted it and always turned it off at the wall 
anyway. That chip probably would have burnt out many years earlier 
if I hadn't.

I also saw on the TV news, after they'd finally stopped rambling on 
about stuff in America, that Michaels camera store in Melbourne 
closed down after 106 years and gave all of their unsold stock away 
for free to people on the street. Further research reveals that 
they actually closed last year after getting knocked about by all 
the pandemic lockdowns, and this stock seems to have been 
everything that they've failed to sell online since. Anyway it does 
mean that I count myself lucky to have visited their camera museum 
in 2019 during my last visit to Melbourne. At 10,000 cameras they 
claim that it's the world's largest private collection, though they 
only had room to display 3,000. It was actually quite tucked away, 
up a staircase and past some offices, but it was very well 
presented in fancy custom-made display cabinets. It was the last 
attraction on that year's day trip and I spent so long there that 
they'd started closing the shop and had to let me out their secret 
door onto the street - it's funny all the places you can pop out of 
in these big cities. So as it turns out I discovered it just in 
time. There's some talk in old articles from last year about the 
collection being donated, but there doesn't seem to be any new news 
about that.

The webpage for the collection is here:
https/michaels.com.au/pages/the-worlds-largest-camera-museum
Actually it shows me that, after three years, I've forgotten most
of what I saw there anyway. Such is life. But hey, I did take
photos! I just have to get around to developing them.

Since then I haven't been back to Melbourne due to lockdowns, and 
although the date for my annual visit is approaching rapidly for 
this year, the requirement to wear a mask for the entire train trip 
there/back, and equally the presently rising COVID case numbers 
because most other restrictions have been lifted, rule it out for 
me again. I was always a bit suspicious that I'd catch something 
even before, given all the crowds (not in the camera museum though, 
I had that mainly to myself), one of many reasons I kept it to one 
visit a year. I'm guessing there's been a lot more change there 
since than just a 'failed' camera store. But it's probably time 
that I stumped up the cash to pay for accomodation and found some 
new places to visit besides those within day-trip range, and 
perhaps without quite the same rush to get around everything before 
closing-time (though that does add a sense of excitment).

Finished at lunch time because I did take too long in the end.

 - The Free Thinker.