Status-Q

Here's an unsolicited recommendation of a Mac utility... but first, a quick explanation of why I need it.

One of my bookshelves contains a large number of bare hard drives in protective cases. These store data that I am keen to keep, just in case I need it again, but which would consume too much space if stored on my normal machines or servers. Much of this comes from past video projects: the raw footage captured for a friend's wedding video, for example, or for one of my YouTube travel vlogs. Some of it is other data kept for a few years 'just in case': a clone of the hard disk of an old machine before it is wiped for sale on eBay, or an off-site backup of my mother's laptop.

In total, this adds up to several tens of terabytes. If it were more important, I would include it in my 'proper' data storage, most of which is on a Linux server using a ZFS filesystem (so it can be regularly and automatically checked for any errors) and snapshotted and backed up to another machine locally and at least one off-site location. But cloud storage starts to become really quite expensive once you're measuring it in reasonable numbers of terabytes, and this data isn't that important... it isn't even important enough that I want to keep all those disks spinning all the time and paying for the hardware and electricity to do so.

So a lot of it ends up on good old hard drives sitting on my shelf. Now, it's worth noting that these are actual hard disks, not SSDs. Much as I love solid-state drives for normal day-to-day use, one thing that's not emphasised often enough is that they are not good for data archiving, unless you power them up regularly and, ideally, run some scans on them. An SSD that's just sitting, unpowered, on a shelf will generally start to lose data much faster than an old-fashioned spinning-rust hard drive in the same state. And you should especially never rely on a cheap USB stick for archival storage.

However, hard drives do suffer from mechanical failure: they have moving parts, and if you power them up and those parts don't move as expected, you're stuck (though not quite as stuck as you would be with an SSD, because if the data were really important you could pay a lab a lot of money and probably get most of it back). But if the data were really important you wouldn't be storing it this way anyway.

I mitigate the risk somewhat by keeping two clones of most things, using my old favourite utility SuperDuper. (Carbon-Copy Cloner is a popular alternative.) And the drives are stored in archiving boxes, which keep the dust and moisture out, and provide some shock-absorbing.

I also don't want to use off-the-shelf external drives in enclosures, because they're likely to have cheap power supplies which fail, or an interface which quickly goes out of date. So I use bare drives, and when I need them, I currently put them in a toaster-style dock with a USB3 interface. But I know I could connect them up to Thunderbolt or eSATA or whatever in the future, or slot them into some hot-swap enclosure.

However, the real challenge with having all this offline storage is how you find data on the occasions when you do need it. If my video editor tells me that it's missing a file called VO1257.wav which was used in this project, how on earth would I know if I had it on any of the disks on the shelf?

So this is why I've become a fan of a Mac utility called NeoFinder, which has had a dedicated following over the last 30 years, and you can get an idea of its vintage from the fact it used to be called CDFinder! You mount your external storage, ask NeoFinder to index it, and it keeps a catalog of all the files on your local disk, which you can still search when the external storage is back on your shelf (or in your remote server, or your fire safe, or wherever). So now you can about VO1257.wav and it will tell you which of your disks have a file of that name, how big the file is, when it was last updated, and so on.

In addition it understands a lot more about many types of file. If I search for 'Fleetwood', for example, it will tell me all the drives that have copies of Fleetwood Mac tracks on them, the recoding bitrate, duration, etc., and show me cover art for the album. If I search for 'sailing', it will show me my photos that are tagged with that keyword (along with a thumbnail and the metadata). It will also show me PDFs that contain that word. And it finds a track by Rod Stewart. You get the idea.

It's very handy. But to make it work properly, you need to be disciplined about two things.

Firstly, whenever you update the data on a drive, you need to tell NeoFinder to update its index for that drive before you unmount it and put it away. It does this easily and efficiently, and I just click that button before putting any drive back in its protective case. Or you can select several catalogs, tell it to update them, and it will prompt you to plug in the disks one at a time if they're not already mounted. And similarly, if you dispose of a disk, remember to delete it from NeoFinder too!

Secondly, you need to label your disks properly – not just with a handwritten PostIt note! – and use the same name in the catalog. I typically have a title like 'Video Projects Archive 2' stuck on the disk and on the case, and the label also says 'Indexed in NeoFinder'.

More than a decade ago I bought a Dymo Labelwriter USB printer, which seemed like a silly extravagance at the time, but turned out to be one of my most useful computer accessories and has been in regular use for all sorts of things ever since. You may not have a need for NeoFinder. But I bet you'd find you have a need for one of these!

Hannah Fry is one of those annoyingly accomplished people that make you realise how little you've achieved in life; for more details, see her Wikipedia page. She's now a Cambridge academic, and I don't think we've ever met, though another thing that her page told me is that she grew up in my childhood home town of Ware, and attended the same school as many of my friends... but nearly 20 years later, which makes her scale of accomplishments since school even more annoying!

She's a great natural communicator – something not traditionally associated with Maths professors, but their numbers seem to be increasing! – and has made various short YouTube videos in the past.

But this month she released a longer one, which I'd highly recommend if you're interested in AI Agents but are (rightly) nervous about trying it yourself: watch Hannah do so instead! A great introduction to why they are both seductive and very scary.

(YouTube link)

"He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me."

― Thomas Jefferson, Selected Writings

If you're not a technically-inclined user of the Apple Photos app, you may want to skip this one!

The master copies of my photos live in two different worlds.

There are the quick snaps taken on my iPhone: handy to browse, to share, to sync across devices and, increasingly important for me, to search by using a map. I generally access these, like everyone else, using the Apple Photos app, and they're backed up to iCloud.

Then there are the 'proper' photos: taken on a 'real' camera, stored on a ZFS filesystem on my NAS, backed up automatically to two remote locations, and managed using Photo Mechanic and Capture One.

But increasingly, of course, the boundary between these is blurring, as phone cameras become capable of taking serious pictures, and as modern-day tasks such as sharing on social media and sending using Airdrop become something I want to do with my 'proper' photos too.

So I've started to experiment with ways to blend the two, and will talk more about them in a later post if they work well! But in the meantime, I just wanted to introduce Rhet Turnbull's excellent osxphotos software, which is really handy for anyone wanting to do something similar.

It's a command-line tool (and Python module) that can provide access to the Photos library and gives you a wide range of options. Here's just one example as an illustration:

osxphotos export ~/PhotoSync/FromPhotos \
--favorite --update --sidecar XMP \
--download-missing --exiftool \
--skip-edited

This will export any photos marked as 'Favourites' to the ~/PhotoSync/FromPhotos directory, only if they're new or updated since a previous export. It will create an XMP sidecar file -- a standard XML format for storing metadata alongside the images -- for each one. If the file is on iCloud but not on your local device, it will attempt to download it first, and it will skip versions edited in Photos because I just want the raw originals for my particular workflow. (There is a vast array of further options available, just on the export command! If, say, the Photos app has recognised the faces in your photos, you can turn the names into keywords in your exported files.)

And hey presto, I have a nice directory of favourite iPhone photos to incorporate in the library of my 'proper' photo-management software.

The import and export subcommands are the key ones for most people, but there are a couple of dozen other options...for example:

• osxphotos labels will print out all the automatically-assigned labels given to your photos; I was interested to find things like 'floodplain' and 'handwriting' in there; terms I wouldn't otherwise have thought of typing into the search box.
• osxphotos add-locations will let you find photos that have no location info and assign them a location if found on another photo taken within a specified time period: handy if, say, you took a photo with your good camera (which has no GPS) and would like to copy the geotags from the snaps you took at the same time with your iPhone.
• Some of the commands, though entered on the command line, will operate by default on any images you've currently got selected in the Photos app. I've just clicked a couple of pictures of Rose and Tilly, and set their captions using:
  osxphotos batch-edit --caption "Rose and Tilly {created.year}-{created.mm}-{created.dd}"

Anyway, I hope this gives you some ideas. I'll be playing with this a lot in future, I think!

I have friends who run marathons. Seems a bit daft to me, though I do quite like the idea of walking one. But I learned from friends who had recently done the London one that there's also a recent tradition of the Reverse London Marathon, which starts at midnight and follows the course in reverse, arriving at the start a little while before all the other competitors set off.

Some sites refer to this as the Nohtaram Nodnol . But surely a reverse Marathon should really be called an 'Athens'?

When I'm ruler of the world — something which is, by the way, on my list of retirement projects, but keeps getting usurped by urgent tasks like booking an MOT for my mum's car, so I haven't got around to it yet — one of my first actions will be to ban the phrase 'up to' from all marketing materials.

Last year, I complained here about our bathroom cleaner, which announced that it removes 'up to 100% of bathroom grime and limescale', presumably to avoid disappointment for customers who thought it might remove 120% of either. It might actually only remove 80%, but that's fine, because they said 'up to', and it's still clearly a superior product to any competitor which did the same but only claimed to remove up to 90%.

Then a week or two ago, I was sent something that said, if I remember correctly, "Did you know your toilet could be costing you up to £581?" Well, I did, because it might be leaking through the floor, one drip at a time, into the dining room below and the water could seep from there into the secret trunk full of lost Gainsborough paintings that I have hidden below the floorboards. Oh, but, wait... then, it would be costing me much more than £581. So, no, I didn't know it could only be costing me up to £581. Perhaps I was the target market for this advertisement after all.

So I read a little bit further, where I discovered that what they meant (but hadn't said clearly) was £581 per year, but they made no mention of damaged masterpieces, and were really talking to people about the water bill incurred if their WC had a perpetual leak from the cistern into the bowl... people who presumably had previously assumed that this could only be costing them up to £580 per year!

So what I want to know is who came up with the precise £581 figure, and how they could be confident that the cost could never be, say, £600? I couldn't find the notice just now, and, doing a Google search for similar warnings, I discover pages stating that my loo could be costing me up to £200 per year, or up to as much as £1000. But, of course, it could be costing me 5p per annum, and all of their claims would still be valid.

So do watch out for these dubious phrases, but you only need to be on your guard for the next eight months or so, because by the end of this year I shall be ruler of up to 100% of the entire globe.

"Credit", I remember reading many decades ago, "is a nice new word for the nasty old situation we used to call Debt".

I've always liked that, and I can't remember its origin. I probably read it about the time that credit cards were first becoming popular, and it's stayed with me ever since!

In the world of IT, there's a thing called 'Technical Debt'. If you spend all of your software development budget creating quick shiny new features, and never maintain, update and improve your existing software architecture, then it will eventually come back to bite you, in much the same way as if you spend all your money on cleaning your car and neglect the basic maintenance of your engine, tyres and brakes. It often takes senior management to recognise this and overrule the demands of the sales and marketing department.

I was particularly impressed, back in 2009, when Apple released the latest version of their operating system and named it 'Snow Leopard'. It was two years after the previous release, 'Leopard', and they marketed it as having "zero new features"... focusing instead on performance and bug fixes: it was just a better version of Leopard. And this was back in the day when your OS upgrades came on a DVD and you had to pay for them (albeit only $29 for this one). If only more companies had the honesty and discipline to do this more often.

Mind you, we can't blame all technical debt on marketing departments. Generally it's more fun for developers, too, to create new code than to tidy up old code. They are more aware of the need for ongoing maintenance, but it can also be a tedious job. One of the most important benefits of AI coding assistants, I believe, is that they can make the tedious tasks less tedious. This gets less attention than the "we can create new stuff really fast" stories, but I've found real value in asking Claude to look at old code and say "I need to upgrade this software to be based on SomePlatform version 8.5 rather than version 5.6. What issues will I need to resolve?" It won't always find everything, but it will almost certainly do a better job than you will when it comes to looking at all the Release Notes of every version of SomePlatform between 5.6 and 8.5 and working out how each change might affect your software.

AI development comes with its own, new kind of debt, though, when you're creating new features: Cognitive Debt. This is where you describe what you want the AI system to build, and it creates it for you, and it all seems to work fine, but no human has actually taken the time to understand what it's doing 'under the hood'. And this requires a new kind of discipline, but now at the time of creation. At Telemarq I would try, though not always successfully, to ensure that the basic operation of any of our systems was understood by at least two people, so there was less disruption when somebody left or moved onto other projects. Now, it has become ridiculously easy for a single, junior developer to create vast amounts of code that not even they understand, and that can just as easily come back to bite you somewhere down the line. Are you really happy to ship software to customers when nobody in your organisation actually knows how it works? That's what today's senior software management people will need to appreciate when setting priorities.

Meanwhile, in my retirement, I have become painfully aware of the quantity of Office-Tidying Debt that I have built up over the years. I really will get down to fixing that soon. Really.

Well, at 1am on Saturday morning, we arrived home from our 'Tour de France' in the campervan. We had planned to travel for a bit longer, but having stayed in 15 different places for our first 17 nights, we were in danger of overload and so decided to turn for home and save some of the other places for a future trip! One of the joys of campervan travel, and especially in France, and even more so in France outside the peak season, is that you don't need to book anywhere in advance, and so can make the trip up as you go along, and change it on a whim. But we had a lovely time and clocked up nearly 2000 miles on the lovely, smooth, quiet French roads.

Some of our destinations were chosen for literary reasons. We love Neville Shute, and wanted to see Brest, Douarnenez, and the submarine pens at Lorient, mostly because they feature in his books. A fondness for Alexendre Dumas, and the references in The Three Musketeers to the Isle de Ré, and the siege of La Rochelle, added those locations to the list. As a fan of Shakespeare's Henry V, I am still looking forward to a future visit to Agincourt, but we did go once more unto the breach, dear friends, in Harfleur: it's now a roundabout, called La Brèque, at the place where the breach in the old city wall used to be!

We visited harbours and grottos, lighthouses and cathedrals, beaches and Roman villas. We stayed mostly on campsites, but sometimes at an Aire de Camping-Car, which, for the uninitiated, is a parking area where a local authority allows you to stay overnight, often for free. There are huge numbers of these in France, often in small villages, and they encourage visitors to visit local shops and restaurants. We now have warm fuzzy feelings about a little village named Locmaria-Plouzané, in the Finisterre region, of which we would otherwise have been completely ignorant, simply because they kindly let us pass a very peaceful night here, a short walk from the village centre:

Very few British authorities are this enlightened, though CAMpRA, the Campaign for Real Aires, is working to get more aire-like facilities in the UK, and we've stayed on a couple of delightful ones.

Some French aires provide a few more services and require modest payment. Some are privately owned: we stayed at a lovely one in the Dordogne that was surrounded by a wire mesh fence with a sliding gate, and the owners asked us to make sure it was closed at night. I was surprised... I looked around at the rolling farmland and thought it didn't look like a high-crime area. No, no, they explained, the gate was to keep any local wild boar out so we weren't disturbed in the night...

We spent one night, for free, in a vineyard, courtesy of the France Passion scheme.

Complete, of course, with a small chateau...

And in Honfleur we stayed in what was basically a large car park... but which was absolutely peaceful at night, provided an electric hook-up, and was 5 minutes' walk from the wonderful old harbour.

For those contemplating the relative comforts of a car park and a hotel room, it must be admitted that our ensuite facilities in the van are somewhat compact. But we also have the benefit of sleeping each night in our own bed, with our own pillows and duvet, and our clothes conveniently in the cupboards without a suitcase in sight.

Our range of activities was a little constrained on this trip by the presence of Betsy, our five-month old puppy. At one point we came back to the van to find she'd realised the view was better from Rose's seat than from the floor. Bother. I fear this means we'll now have to purchase some seat covers...

Anyway, we're now back, and have tamed the jungle that just three weeks ago was our nicely-mown garden, so I can soon get round to editing my many hours of video footage into a YouTube video, which I do, primarily, to help me relive the trip and remember it for longer!

Well, yesterday's post got a lot of responses! Thank you, everybody!

Some of them are visible in the comments and some were by email. But most of them were from drivers rather older than me who said that my formula would have them driving at 40 on motorways (or something like that).

Now, notice that I did talk about the maximum speed you like driving rather than the maximum speed you would consider driving, but clearly I have more high-velocity octogenarians amongst my readership than I had previously realised, for which I feel duly honoured.

Ian Clark suggested an amendment which was essentially

V_max = 103 - (age * 2 / 3)

which would still suit me and give older readers a few more mph to play with, but I think I'll need an exponential component to get anything reasonable, perhaps something more like:

V_max = 100 - 0.6 *age^0.95

Anyway, there was agreement from several readers that the desire to break the speed limit certainly decreases with age, and Andy Davy observed that toddling along at 50mph is often fine as long as you have an interesting podcast playing; an assertion I can definitely support. In my case, though, it's often an audiobook: we're currently touring France to the accompaniment of The Count of Monte Cristo. (I'm now contemplating my next theorem about the duration of any journey being inversely proportional to the quality of your listening material...)

All this reminds me of, in my younger days, getting stuck in completely stationary traffic on the M25 while heading for an airport to catch a flight. The minutes ticked by, my anxiety rose, and then I realised that the pumping drumbeat of my music probably wasn't helping. I replaced it with some nice Brahms cello sonatas, and, all of a sudden, my equanimity was restored, my blood pressure declined, and I gazed calmly at the surrounding vehicles (about which I could, in any case, do nothing). And yes, I arrived in plenty of time.

When we started towing a boat behind our car, and were consequently limited to 60mph rather than 70mph on major roads in the UK, I found I rather liked travelling that way. Yes, it took a bit longer, but it was more relaxing.

And I've noticed that, whether due to a growth in wisdom or to a decline in testosterone -- I prefer to think it's the former -- I now tend to drive rather more slowly than I did a decade or two ago. Rose suggests I may just be subconsciously aware of slower reaction times...

But this has led me to propose Quentin's Law of Optimal Velocity, which is the maximum speed in miles per hour at which you like to drive, and is given by:

Vmax = 120 - age

but I freely admit that this is based on a rather small sample size (errm... one, to be precise) so would be grateful for more data.

Does it correspond to your own experience?