I have been cleaning out files and scanning meaningful letters. It occurs to me that I would like to preserve some meaningful emails to be future proof. Other than printing them, what options are there? I know I can archive, but I don’t trust that to be ‘future proof.’
Printing to PDF would be my first thought.
DEVONthink Pro Office is my choice. I keep the occasionsl email in Evernote. Very easy to search and you can save in pdf if you want.
To preserve meta information of emails I backup the mbox files (Unix mailbox format used by mail client, here Thunderbird). Really important emails get exported to single *.eml files. The DEVONThink approach I found unusable when emails have attachments as it stores the emails as *.rtf documents. So I save the attachments in the mail application to DEVONthink the rip the attachments off and Import the emails with DEVONthink.
If you have a windows pc (or a bootcamp installation or a virtual machine) at hands try mailstore (https://www.mailstore.com) which has a free Edition for private/personal use.
Thanks. All good thoughts.
I was wondering about saving to .txt and storing in a folder. . .? I have heard David talk a lot about NVALT. I’m not sure I trust either DTP or Evernote to be ‘future proof’. PDFs seem more secure.
What about someway to ‘batch process’? E.g. I would like to save all the emails I have from our daughter so that her daughters will be able to read at some point in the future.
I create a PDF of the email and any related attachments and send them to DEVONthink. Also, if I think I may want to edit the text of an email, I send it as plain text to DT.
I store emails by printing to PDF and storing in either Notes or in my regular file structure.
I currently backup emails via MailStore Home on Windows, but I am curious about Mac alternatives.
I thought this was a really interesting question and I really wanted to know if there was a way to automate this. So, I spent the last 4 hours figuring it out. This is what I was able to figure out. I think the readme does a decent enough job of explaining how to use it. I’m thinking of using this for myself now!
Forwarding to Evernote works quite well.
Are there no privacy implications with this approach?
Only if someone is preventing you from doing so, eg your boss or a regulator. Evernote is an online service and their security is as good, or as bad, as any other online service.
If you are required or determined to keep the files on devices you own and control then probably an mbox archive is your best option - at the expense of data protection, search and instant availability from every device you have.
You then need to have a robust regime of backups, ensuring your data is protected.
I rather like @dustinknopoff ‘s solution.
My initial thought also was to export the message in mbox or eml format and then parse it in some fashion for storage.
When I was actively using DEVONthink I would send emails there, but had similar issues with the content format as already commented.
Many times I just print to PDF, but that is not a complete solution for attachments.
I did not check out mail-parser, but now that the idea is in my head, I’ll probably throw together a Python scriotmto parse eml files and save the attachments with names related to the base email subject for clarity. A useful option might be to provide an option for what I think is called the “textbundle” format that (I think) Brett Terpstra had proposed with some others (maybe the Ulysses guys, I can’t remember offhand).
Python has really good libraries for handling email mime format, so this is a pretty easy script to throw together. Maybe out it together with Automater as a service for Mail.app…
If anyone is interested I’ll be happy to share the script, although I probably won’t play with this until later this week when I have a bit of free time.
As a side note, I did at one time play with using AppleScript to export messages from Mail.app, and I found that AppleScript was severely broken and cannot properly handle attachments. This was a while back, and as far as I know Apple has never bothered to fix this, so at the time I did mess with parsing up the raw email source, but I didn’t need it enough to write the script back then. Since, I’ve tried to get away from using Mail folders to store reference data, so now is the time to revisit this.
I’d definitely like to see your implementation especially since I didn’t even try to account for attachments.
Mail-parser was the simplest library I found. It’s the only one I played with that actually worked the way I thought it should.
I have used EmailArchiverPDF in the past, and worked quite well.
Haven’t done a backup in ages, since Devonthink does it for me now - but might be worth a look, if you’re not a DTPO user.
If you are using Gmail then I find CloudPull from Golden Hill software works great. It’s the same developer that is now supported the Unread RSS reader.
I am going to explore email archiving programs but am also looking to be a bit more low tech than that. It will take a long time, but I’m going to create a Day One journal called emails (at least, for now). Once I have the ones I want, I’m going to create a book from within the Day One app.
I use email archiver pro for the Mac - keeps the email and all attachments as PDF in a folder for each email. It works with 5 email accounts - and can be set to update as needed - I have all my emails since 2005 46,000 of them and its rock solid
I used to use CloudPull, but I got fed up with it pegging my CPU until I killed and restarted it. I took it off my Mac a couple of years ago. Has it gotten better since then?
I’ve been using CloudPull for several years and I have never had that problem. I just checked Activity Monitor and it is currently using 0.0% CPU. I have never had to restart it ever. If my computer has been asleep for a few hours when it wakes up CloudPull does use some CPU time for a few minutes, as it backs up the messages that came in when the computer was sleeping, but it certainly doesn’t interfere with operation of the computer (I only notice if I have Activity Monitor open).
Looking at CPU time since my last boot, CloudPull is using similar time to Little Snitch and Dropbox. Way less than WindowServer and kernel_task.
Bottom line, I haven’t had a problem with CloudPull dragging down my system. I do have a fairly fast computer with 16Gb of memory.