A full and intense day archiving my Gmail, a backup procedure I've generally neglected. Until now, to save space rather than preserve data, I've sought out emails beyond seven years (chosen because it's the storage requirement for tax purposes), and deleted certain classes of email from that period, such as files over a certain size ("older:2015 larger:3Mb") or similar, but this is haphazard as I had no set rules or standards of what to keep and what not. In particular, small emails are often not worth keeping.
Most of the emails I receive are notifications, often automated of the 'no reply' sort. With any online order there is also a slew of emails; the order placed, an invoice, a delivery notification, a 'delivered' notification. I generally kept those as a record of what I bought and when, but in my ancient text-based backups (I have text emails going back to 2000), I only included conversations and not notifications. It seems over-cautious to keep records of purchases, akin to filing receipts from local shops; and online orders are often preserved anyway - Amazon's are at least. For unusual items, I often make a personal note or review of what I bought and how good it was anyway, so there's less and less of a case for keeping these sorts of email.
I also filed the occasional newsletter, or press release, or even (shudder) mp3 attachments sent to me for ArtsLab. Nowadays I'm much more aggressive when it comes to culling mail. I tend to download attachments and delete the email. It's a great shame that we can't delete attachments but keep the text.
So the day started by using Gmail searches "newer:2015 older:2016" for example to sort emails by year, then creating a label for each year. Then searching each big email and deleting any that don't need to be that big. Generally every email over 2Mb is deleted anyway, and those over 1Mb scrutinised for important content that I want to file. I had 2000 emails a year in the 2010s, but now it's more like 1000; partly because I delete more, partly because I'm less social and less e-social these days.
Then I started to delete, and have decided to delete most or all of the notification type emails; many of which are from the big tech companies like Amazon, eBay, Google. I'll keep last year's from them, but beyond that the information they convey is minimal. For anything I purchase that may be used in financial records, like invoices, I'll now download and file all of those. I often do that anyway, but can keep a second copy in email - not any more.
Newsletters, offers, posters, art or poetry competition forms are deleted, and perhaps filed under their event. If I'm sent images, those are also filed separately, so there's no need to keep an email too.
After this, Google Takeout can be used to download each year, one tedious year at a time. I prefer it that way rather the one giant block. It makes it easier to add a neat new file to the archive each year. Not sure how much space this has save me overall, perhaps half a Gb.
It's amazing how huge the Gmail email archives are. My text emails from 2014, the last year I stored them 'by hand', amount to 13.9Mb. The Gmail archive is 141Mb! Ten times bigger for the same thing.
Doing all of this filtering for the years 2015 to 2021 has taken all day, and the process is not complete. It's hard to be sure that this is a worthwhile job, but filing and organising is a fundamental purpose of life. I expect that nobody else does this, and I expect that in 100 years time (if you are reading this, reader of that future), my email record might be one of the few that still survives, but who can say? I might be wasting my time, but there's no harm in preserving such data, the most intersting and culturally important of which will, perhaps, be my every day conversations with friends.
Soon, to other jobs. Perhaps 2024 is the year I finally put my old games on Steam. Gunstorm II is an obvious choice, being my lasts and most techncally proficient game, although it's a plain-old shooter. There's probably more of a case of it than Future Pool or Snooker being on Steam, but what held me up is that I'd really need to release the first Gunstorm game first, and that's merely average. All of this costs a lot of money and the chance of getting it back slimmer than ever, but this is a cultural investment and in the long term is likely to break even. Other games that need releasing are: Bool, Breakout Velocity, it's sequel Fallout (I could probably combine both into one game), Firefly, Trax (which is very old and never been revamped), Arcangel (similarly ancient, and never released at its best), Outliner (one of my favourite to play, yet simplest games).
