Happy backup day, digital decluttering and my main apps
Happy World Backup Day.
I solemnly swear to back up my important documents and precious memories on March 31st. #WorldBackupDay
I will also tell my friends and family about World Backup Day – real friends don't let friends go without a backup. =)
In this context, a backup simply means a copy of digital data, usually stored in the cloud (as often phrased ‘someone else’s computer’) or on removable media, that can be used to recover from data loss.
Although for some things it’s fairly obvious that keeping a backup is good idea – like photos and videos – there’s other things that I often forget to think about. So as a reminder to me, and perhaps a warning to others, I’m making a blog post on it, and also having a bit of a chat on digital decluttering and my current apps.
Peekaboo bee. One of the best of hundreds of photos taken that day.
Where to put the backups?
The most obvious backup solution is a cloud storage offering that syncs files between your home computer (and phone) to servers located elsewhere – whether that be NextCloud or a big-tech solution like Google Drive, OneDrive, iCloud, Dropbox etc.
Although that will protect you in the sense of if you have a catastrophic computer failure, the cloud version will still be there for you to restore stuff from, these systems won’t necessarily protect you from accidental deletions or file overwrites, or in the event of some malware attack going on a spree of encrypting stuff.
The nature of folder syncing often means if a file is deleted, it’s deleted in both locations; if a file is encrypted, it’ll be encrypted in the copy too.
So, what you need is that plus at least one secondary copy. One answer is a cloud backup system with some sort of versioning (also called ‘file history’ or ‘snapshots’, basically keeping old copies when files change). If you do that though, it’s all too easy to end up with ever-expanding numbers of files, needing extra pricy storage space, and for these to get more and more disorganised. We’ll come back to digital clutter below.
Then you’ve got to think a little bit about the format of what you’ve stored too – do you have anything snuck away in a proprietary format (MS Office files are the most likely culprit) where you might not have access to software for it in future?
It all gets a bit complicated.
What I currently do is use a synced cloud storage solution, plus an external hard drive (one not standardly connected to the main machine or internet) with new content added every now and again.
If my computer or external hard drive were destroyed (say by a fire), then I’d have the cloud version; if my computer and cloud version were attacked, then I’d have the hard drive version. And I’d have to be very unlucky to have that fail at just the same moment.
Plus, I do have a small number of really important things on DVDs and in a second independent cloud storage location too (noting that DVDs degrade over time, so aren’t a forever solution).
My list of things to back up
My list of things to remember to back up, other than those photos and videos, is:
- Documents. From contractual stuff to appliance manuals.
Only this very weekend I discovered I did not have a copy of the manual on our toilet seat, which I suddenly needed after it broke; fortunately, I was able to hunt down its product number in an email from 2017. - Emails. Some email clients download emails locally from your host, effectively meaning you have a second copy, but if not, this is one to keep on top of. Particularly if you’re in a free ‘as in beer’ big-tech system (of course it’s not really free – you’re paying with your data) where support might not help much if you should get locked out of the account.
- Social media posts. For me, this means a backup of my fediverse (Mastodon etc.) follows and followers, as I only treat posts as ephemeral (occasionally you hear of an fediverse instance just disappearing overnight, not giving you time to export your stuff).
- Instant messages (if I might need to keep them).
- Data collected by my smart watch. We know there was a ransomware attack on Garmin back in 2020, which initially encrypted the data on their servers.
- Notes made on my phone, including content like recipes I’ve clipped from the web.
- Music and eBooks (non-DRMed stuff).
- Bookmarks.
- My (this) blog.
- Copies of historic energy use (to work out best tariffs).
- And last but not least is also making sure I have a note of all the backup 2FA codes to online accounts, in case I ever lose access to my authentication app (although beware storing these online if not themselves encrypted, and then beware losing that encryption key etc.!).
Organising the data
Of course, backing up is one thing; as I’ve alluded to, it’s organising your data that’s really the bigger task.
Spring feels a good time to do a digital declutter in the same vein as a physical spring clean, and given their sheer size compared to anything else, it makes sense to think primarily in terms of photos.
I’m not saying my emails aren’t also pretty cluttered (although not my inbox, as I’m a fairly strict inbox zero-type person), but emails are often just text and therefore the size saving for deleting unnecessary messages is tiny.
It caused much hilarity (incredulity?) last year when the UK Environment Agency suggested to save water by deleting emails at the same time as promoting accessing a generative AI tool (LLM) from every lamp post instead of a simple website (and yes, it’s GenAI use that’s a major water-use concern).
In fact, it seems deleting emails does not save water, until you’re in the position you have to move the data anyway, as content at rest doesn’t use much water.
But decluttering is also about finding what you want, not having it all buried it a sea of miscellaneous junk, as tends to happen with my photos. I expect many people are keeping excessive numbers of photos that aren’t ever going to be looked at again, such as the many many blurry bumble bees recorded on a visit to Hitchin Lavender when trying to learn manual focus (but also, who knew, at least one of them came in useful for this post after all).
I personally don’t currently use an organisation database for my photos. They simply live in folders, named by date, year and/or occasion. I’ve looked at organisation systems more than once, but one of the complicated things is getting that to work across all the systems I use (Linux, Windows, iOS, and yes in my backup too) and play nicely with my current folders.
To declutter my photos, I’m really trying to, working from the oldest, go through little by little and plough on with sorting them – mostly deleting – in spare moments.
I do use some tools to help me organise content. I’d definitely suggest that before you start organising, go through and use something (we used fdupes) to remove outright duplicates (exact matches). [Aside: this also proved a good idea before moving cloud storage purveyors, given the gigabytes of stuff I had was already a multi-day upload endeavour.]
That said, for the rest, no automation can decide for me which of the many many similar photos of, say, the bees or the other half trying to learn to surf are actually the ‘ones’. It’s only me (or him) that can do so.
Beware the AI agent who says otherwise.
Don’t ever give elevated permissions to an Avian Intelligence… by David Revoy − CC-BY-SA 4.0
I actually didn’t believe the story of Meta’s director of AI safety who managed to ask their OpenClaw AI agent to ‘sort’ their emails only to then see it start to delete them as they looked on, all the while vigorously pleading with it to stop by text message, followed by a literal sprint to their computer to pull the plug when that didn’t work.
When I first heard about it on social media (not the original, as that was X, which I don’t use), I thought it was another attempt at fake virality, such as the toothbrush botnet story. However, with KrebsOnSecurity reporting on it, it’s actually probably true.
My apps and stuff
OK, just in case anyone’s interested, I thought I would also note what services and apps I use. Blogs tend to do these posts at the start of the year, but hey ho, late is better than never.
I like seeing these lists because it gives me ideas on software I might check out for my own use, and maybe just to be nosy!
💻 Desktop OS: Linux Mint (Cinnamon) (known as a simple distribution for the non-techy), dual boot with Windows 11 (although quite rare I ever use that).
📱 Mobile: iPhone (not saying I like Liquid Glass or sneaky updates applying identity verification without warning).
💬 Chat: Signal, iMessages (I do not use WhatsApp).
✉️ Email hosting: Runbox
📨 Email client: iOS Mail and Thunderbird
📁 Cloud storage: NextCloud on Hetzner
📆 Calendar: my own design of paper planner!
📝 Notes: Obsidian (a fair few of my notes are recipes, so I use the Obsidian Recipe View plugin to see those).
This is proprietary, but the actual notes are not stored by Obsidian and are in an open format (markdown), so if the app disappeared tomorrow, I’m covered. I’ve been burnt before when Evernote was taken over by venture capital and locked stuff away.
🌐 Browser: Firefox, Librewolf and Vivaldi, and Safari on iOS, with obligatory plugins UBlock Origin (at least UBlock Lite if that’s all you get) for tracker blocking and consent-o-matic for automatic cookie (non-)consent.
🔖 Bookmarks: Raindrop.io.
I’m not fully unworried about this one. Raindrop is a proprietary app run by a sole developer, but I haven’t as yet a better solution that doesn’t involve a quite pricy hosting package or self hosting, which would be needed for e.g. LinkWarden or Wallabag. I do export the content more often because of this.
🔎 Search: Qwant and DuckDuckGo.
These give me as good results as anything does these days, but such is subjective. Both do include AI summaries by default but you can turn them off, which is crucial to me. Google even changes article titles in the primary supposedly non-AI list to something the authors didn’t write, sometimes changing the meaning in the process, so it’s a no for that.
Aside: Google doesn’t currently index this blog – we haven’t prevented it doing so, but it’s known to not bother indexing personal blogs, or randomly stopping indexing just because. However, Google does index this blog’s Mastodon account and hence slops confidently and wrongly that that’s the blog itself.
🏞️ Images: Procreate (iOS, for doodles), Inkscape (vector), GIMP and Krita (raster).
We never use AI images in this blog, and I avoid blogs with them – if you use such images, how am I to believe you don’t slop the text (AI;dr)?
📜 Word Processing and markdown editing, 📈 Spreadsheets and 📊 Presentations: Libre Office.
🧑💻 Text and code editor: Visual Studio Code.
👨👩👧👦 Social media: fediverse (Mastodon, Bookwyrm).
🕸️ Fediverse app: IceCubes, Phanpy.
📖 RSS: NetNewsWire, fediverse RSS Parrot.
📝 Blogging: Hugo, hosted via Mythic Beasts.
I do note Hugo has been a bit much on the old breaking changes recently, but moving to another static site generator would also be a cost in time and effort. That I said I do agree with Brennan Day on 11ty→BlogAwesome issues meaning I’m not so sure that 11ty would be an improvement long term.
⌚️ Smart watch (running only): a basic Garmin Forerunner (I don’t need all the stats to tell me I’m slow).
I do check privacy settings to make sure the very sensitive health and location data collected here is not fed into AI training or route heatmaps (i.e. use for ‘product improvement’ is off).
🛒 Shopping lists: paper.
🎵 Music: Deezer and Bandcamp (and CDs from merch stalls when I go to something live).
📚 Books: Kindle.
🏡 Home automation: Home Assistant.
And finally:
☕️ Coffee: The Roasting House (although too good for everyday use, so it’s saved for guests and weekends 😉).
Extra shots
- The UK National Cyber Security Centre, which has advice on backups etc.
https://www.ncsc.gov.uk/ - A really in-depth take on all things backup by tuxwise
https://tuxwise.net/all-episodes/s01e03-secure-backups-and-a-time-machine-for-your-home-office/ - Simon Paul on if you can save carbon by deleting emails
https://newslttrs.com/does-deleting-email-really-help-with-climate-change-part-one/ - Blog by Shawn Hooper on using the open source photo manger Immich
https://shawnhooper.ca/2026/03/11/managing-photos-in-my-home-lab-with-immich/ - A simple guide to getting started with self-hosting NextCloud on YunoHost
https://blog.elenarossini.com/a-newbies-guide-to-self-hosting-with-yunohost/ - Another option for simpler self-hosting
https://www.pikapods.com/ - Google patent changing the actual content of your website, not just search result titles
https://www.forbes.com/sites/joetoscano1/2026/03/06/google-just-patented-the-end-of-your-website
Credits
The Avian Intelligence comic is copyright of David Revoy and reused under CC-BY-SA 4.0. Do check out the rest of the Gothic witch’s adventures too if you don’t follow that already!
App list section heavily inspired by a post on Forking Mad+. Also a great blog to have in your RSS reader.
Comments (0)
No comments yet.