4 ms·

I still have a disk image of the Mac Plus I used until I was 10, when I upgraded to a new Blueberry iBook. The Mac Plus experience can be recreated with Mini vM

I still have a disk image of the Mac Plus I used until I was 10, when I upgraded to a new Blueberry iBook. The Mac Plus experience can be recreated with Mini vMac, an emulator. Files since 1999 are all still in a folder on my local drive. I've made an effort not to lose the photos and chat logs. My iTunes database dates back to 2002, and I'm disappointed that I lost my old SoundJam MP database that I used for 2 years before that. Unfortunately, my archives continuously come under threat from forced upgrades. Upgrading iPhoto to the Photos app will lose a lot of metadata. The latest versions of iTunes removed USB sync for contacts & calendars, which really bothers me when I want to maintain my files without the cloud. Many old documents are now unreadable (e.g. Clarisworks). I've realised that simplicity is essential for keeping long-term records. Keeping a copy as plain text is important for preservation. Just because "there's an app for that" today doesn't mean that 10 years from now, the app will still work. This is (especially) true for companies that should know better, such as budget tracking tools. I recently went through my old records to compile a list of every rock concert I've been to, along with the location and price when possible. With heavy use of archive.org, my iCal, and my iPhoto library, I figured out all the dates, but it was a major effort. Most of them were in the last 10 years. My dad still has his PhD thesis on a magnetic tape. Anybody who knows how to read that onto a PDP-8 should get in touch!

chromakode 11y ago
I feel your pain. Especially with tricky formats like Clarisworks (or in my case, old Cakewalk and Hash Animation Studio) reviving the data might require emulating the entire OS and environment they existed in. At least for full disk images, the backups contain copies of the software used to create the files. It's really disappointing when software uses unnecessarily complex file formats. Chat logs are an important example: I was dismayed to discover that my Yahoo Messenger chat logs on Windows are stored XOR-obfuscated! [1] There's a lot of compromises to be made between the convenience of making backups and their long term viability. My hope is that by greedily saving all of the bits, there will be enough context to make sense of the data if I really needed to, even if it is an intensive process. A good alternate strategy would be to scoop up the most likely interesting files when processing the backups and re-encode them in more future-ready formats, as derefr suggested. [1] https://cryptome.org/isp-spy/yahoo-chat-spy.pdf https://cryptome.org/isp-spy/yahoo-chat-spy.pdf
- ckcheng 11y ago
  I've got quite a few old ClarisWorks files as well. At some point I realized all that data would become unreadable (at least not easily) when that file format becomes unsupported. So I went through every ClarisWorks file (with a script) and saved it out into multiple formats (e.g. RTF, PDF). The hope was at least one of those would open... Batch re-encoding those files at the cusp of losing easy access to them turned out to be a lot of possibly error prone work. So to future-proof important documents, now I do that re-encoding continuously up front, whenever those files are saved. But for long term viability, on top of backups, the files themselves have to be in a format usable by programs that will be around for a long time on platforms that won't disappear. Practically, that means using simple file formats (TXT). Or else using a program (LibreOffice) that creates files in an open format that can be easily re-encoded up front into multiple formats like DOC and PDF. The MultiFormatSave extension for LibreOffice makes it easy to save into multiple formats for that purpose. After all that work though, I just found out LibreOffice can open ClarisWorks files. lol...
hvidgaard 11y ago
> Many old documents are now unreadable This is exactly why some people are so vocal about open formats. Regarding your fathers PhD thesis, take a look at http://www.pdp8.net/ http://www.pdp8.net/ I'd be surprised if they cannot help or point you to some people that can.
- Jaruzel 11y ago
  This whole thread has just triggered me into finding out why current versions of MS Office are failing to open some 20+ year old documents that I have. Turns out, MS Office no longer supports Word for DOS/WordPerfect/AbiWord files - which is fine I guess, but to rub salt into the wound, there used to be converter plug-ins available, but MS not longer host them for download (I guess the 400kb archive was taking up too much space or something). I just had to resort to hunting down the archive and downloading it from some unknown/untrusted file directory. Anyway, it works and I can now open all my ancient documents in Office 2013 - which is not bad considering the converter plug-ins were written for Office 97. Regarding the article - I'm not sure I understand the concept behind storing a set of the recovered files in a Cloud service. Surely a second HDD with the same contents stored at a second location would be cheaper in the long run. The HDD would last 20+ years if stored correctly - you can't guarantee that your consumer level Cloud service will still be around even in 10 years time.
  - hvidgaard 11y ago
    "Container" is simply a place where you can read the content today. That can be a HDD, it can be AWS, it can be a USB drive - doesn't matter as long as you have a system that can read it. Once that "container" starts to get near it's EOL you transfer it to a new "container". I wouldn't trust a HDD for 20 years. Keep a backup mirror locally and a backup offsite (cloud or at a relative), what matter is that the transfer is automatically running in a periodic manner.
acqq 11y ago
The old file formats which get to be unreadable just because they are "old" and the newer programs don't support them anymore is much more serious issue than many people think. I hazily remember having the Outlook PST files as the mail archives and then later not being able to recover most of the metadata of the mails, not even programmatically, as the PST was designed even before the "internet standards" were something they'd worry. GPG also removed the support for the old encrypted files and the old keys since 2.1: https://www.gnupg.org/faq/whats-new-in-2.1.html#nopgp2 https://www.gnupg.org/faq/whats-new-in-2.1.html#nopgp2 which signifies the dangers of using it in archival contexts. The disk encryption tools can be even more problematic, being dependent on the particular OS versions.
WalterBright 11y ago
My files from the 1970's were on magtape that was written to by a long-gone drive that was way out of spec. I threw it away. My next oldest files were from 1982 or so, and were on 8" PDP-11 floppies. My 11 was discarded long ago. I contacted an old friend, and he said he still had his, though it hadn't been powered up in years. Amazingly, it powered up, and there wasn't a single bad byte on my 30 year old floppies (!). He was able to recover it all for me. I have most of the stuff since, though floppy backups were erratic. I copied the 5.25" floppies to CD, and then to hard drives, before discarding the old computers. I try to save most files as jpg, pdf, mp3, mp4, or plain text, figuring those are the most future proof.
- eru 11y ago
  Postscript is better than pdf, if you have the choice. (They basically introduced pdf, because postscript was too open and other companies started beating them at it.)
  - icebraining 11y ago
    PDF/A is an ISO standard, with an EU-backed preservation project writing conformance testing tools and such. https://en.wikipedia.org/wiki/PDF/A https://en.wikipedia.org/wiki/PDF/A http://www.preforma-project.eu/media-type-and-standards.html http://www.preforma-project.eu/media-type-and-standards.html
    - sp332 11y ago
      The trick is finding an editor that will save in strict PDF/A.
      - X86BSD 11y ago
        Exactly! I've been mildly involved as my families archiver for years. PDF/A is the standard that the should be used for archiving docs. In fact it's what the library of congress recommends themselves for long term storage of docs. But as you said I can't think of a single app that does strict pdf/a support. I may have come across one it seems but can't remember what it was. I store everything in pdf and images as tiff. The LOC has good guidelines for archiving all sorts of formats.
cpach 11y ago
Good points! ”Upgrading iPhoto to the Photos app will lose a lot of metadata.” I’m still on Mavericks but will upgrade at some point. This makes me nervous though. Would you care to elaborate on this issue?
- lucaspiller 11y ago
  Apple have a guide on it here: https://support.apple.com/en-us/HT204478 https://support.apple.com/en-us/HT204478 One thing to bear in mind is Photos will create a new photo library, leaving the old one untouched. The metadata will still be there, but good luck getting it out.
chinathrow 11y ago
> Many old documents are now unreadable There is/will be a niche market for folks writing software to convert legacy formats into open/current formats. Remember how all of a sudden shops offered to convert your VHS-tapes to DVD? I bet they were making a killing.
- ashwinl 11y ago
  Just went into a CVS to pay for one VHS conversion for $25.99. CVS is using a service called http://www.yesvideo.com http://www.yesvideo.com
123jay7 11y ago
You are not your memories. Stop saving them. White people put way to much though on their mind and their memories. You need to stop doing that and stop thinking. Get rid of all your hard drives. Rm -rf everything, don't keep a single thing. A rolling stone gathers no moss. Be a rolling stone. Your old self is stupid anyway.

chromakode 11y ago

I feel your pain. Especially with tricky formats like Clarisworks (or in my case, old Cakewalk and Hash Animation Studio) reviving the data might require emulating the entire OS and environment they existed in. At least for full disk images, the backups contain copies of the software used to create the files. It's really disappointing when software uses unnecessarily complex file formats. Chat logs are an important example: I was dismayed to discover that my Yahoo Messenger chat logs on Windows are stored XOR-obfuscated! [1] There's a lot of compromises to be made between the convenience of making backups and their long term viability. My hope is that by greedily saving all of the bits, there will be enough context to make sense of the data if I really needed to, even if it is an intensive process. A good alternate strategy would be to scoop up the most likely interesting files when processing the backups and re-encode them in more future-ready formats, as derefr suggested. [1] https://cryptome.org/isp-spy/yahoo-chat-spy.pdf https://cryptome.org/isp-spy/yahoo-chat-spy.pdf

ckcheng 11y ago
I've got quite a few old ClarisWorks files as well. At some point I realized all that data would become unreadable (at least not easily) when that file format becomes unsupported. So I went through every ClarisWorks file (with a script) and saved it out into multiple formats (e.g. RTF, PDF). The hope was at least one of those would open... Batch re-encoding those files at the cusp of losing easy access to them turned out to be a lot of possibly error prone work. So to future-proof important documents, now I do that re-encoding continuously up front, whenever those files are saved. But for long term viability, on top of backups, the files themselves have to be in a format usable by programs that will be around for a long time on platforms that won't disappear. Practically, that means using simple file formats (TXT). Or else using a program (LibreOffice) that creates files in an open format that can be easily re-encoded up front into multiple formats like DOC and PDF. The MultiFormatSave extension for LibreOffice makes it easy to save into multiple formats for that purpose. After all that work though, I just found out LibreOffice can open ClarisWorks files. lol...

hvidgaard 11y ago

> Many old documents are now unreadable This is exactly why some people are so vocal about open formats. Regarding your fathers PhD thesis, take a look at http://www.pdp8.net/ http://www.pdp8.net/ I'd be surprised if they cannot help or point you to some people that can.

Jaruzel 11y ago
This whole thread has just triggered me into finding out why current versions of MS Office are failing to open some 20+ year old documents that I have. Turns out, MS Office no longer supports Word for DOS/WordPerfect/AbiWord files - which is fine I guess, but to rub salt into the wound, there used to be converter plug-ins available, but MS not longer host them for download (I guess the 400kb archive was taking up too much space or something). I just had to resort to hunting down the archive and downloading it from some unknown/untrusted file directory. Anyway, it works and I can now open all my ancient documents in Office 2013 - which is not bad considering the converter plug-ins were written for Office 97. Regarding the article - I'm not sure I understand the concept behind storing a set of the recovered files in a Cloud service. Surely a second HDD with the same contents stored at a second location would be cheaper in the long run. The HDD would last 20+ years if stored correctly - you can't guarantee that your consumer level Cloud service will still be around even in 10 years time.
- hvidgaard 11y ago
  "Container" is simply a place where you can read the content today. That can be a HDD, it can be AWS, it can be a USB drive - doesn't matter as long as you have a system that can read it. Once that "container" starts to get near it's EOL you transfer it to a new "container". I wouldn't trust a HDD for 20 years. Keep a backup mirror locally and a backup offsite (cloud or at a relative), what matter is that the transfer is automatically running in a periodic manner.

acqq 11y ago

The old file formats which get to be unreadable just because they are "old" and the newer programs don't support them anymore is much more serious issue than many people think. I hazily remember having the Outlook PST files as the mail archives and then later not being able to recover most of the metadata of the mails, not even programmatically, as the PST was designed even before the "internet standards" were something they'd worry. GPG also removed the support for the old encrypted files and the old keys since 2.1: https://www.gnupg.org/faq/whats-new-in-2.1.html#nopgp2 https://www.gnupg.org/faq/whats-new-in-2.1.html#nopgp2 which signifies the dangers of using it in archival contexts. The disk encryption tools can be even more problematic, being dependent on the particular OS versions.

WalterBright 11y ago

My files from the 1970's were on magtape that was written to by a long-gone drive that was way out of spec. I threw it away. My next oldest files were from 1982 or so, and were on 8" PDP-11 floppies. My 11 was discarded long ago. I contacted an old friend, and he said he still had his, though it hadn't been powered up in years. Amazingly, it powered up, and there wasn't a single bad byte on my 30 year old floppies (!). He was able to recover it all for me. I have most of the stuff since, though floppy backups were erratic. I copied the 5.25" floppies to CD, and then to hard drives, before discarding the old computers. I try to save most files as jpg, pdf, mp3, mp4, or plain text, figuring those are the most future proof.

eru 11y ago
Postscript is better than pdf, if you have the choice. (They basically introduced pdf, because postscript was too open and other companies started beating them at it.)
- icebraining 11y ago
  PDF/A is an ISO standard, with an EU-backed preservation project writing conformance testing tools and such. https://en.wikipedia.org/wiki/PDF/A https://en.wikipedia.org/wiki/PDF/A http://www.preforma-project.eu/media-type-and-standards.html http://www.preforma-project.eu/media-type-and-standards.html
  - sp332 11y ago
    The trick is finding an editor that will save in strict PDF/A.
    - X86BSD 11y ago
      Exactly! I've been mildly involved as my families archiver for years. PDF/A is the standard that the should be used for archiving docs. In fact it's what the library of congress recommends themselves for long term storage of docs. But as you said I can't think of a single app that does strict pdf/a support. I may have come across one it seems but can't remember what it was. I store everything in pdf and images as tiff. The LOC has good guidelines for archiving all sorts of formats.

cpach 11y ago

Good points! ”Upgrading iPhoto to the Photos app will lose a lot of metadata.” I’m still on Mavericks but will upgrade at some point. This makes me nervous though. Would you care to elaborate on this issue?

lucaspiller 11y ago
Apple have a guide on it here: https://support.apple.com/en-us/HT204478 https://support.apple.com/en-us/HT204478 One thing to bear in mind is Photos will create a new photo library, leaving the old one untouched. The metadata will still be there, but good luck getting it out.

chinathrow 11y ago

> Many old documents are now unreadable There is/will be a niche market for folks writing software to convert legacy formats into open/current formats. Remember how all of a sudden shops offered to convert your VHS-tapes to DVD? I bet they were making a killing.

ashwinl 11y ago
Just went into a CVS to pay for one VHS conversion for $25.99. CVS is using a service called http://www.yesvideo.com http://www.yesvideo.com

123jay7 11y ago

You are not your memories. Stop saving them. White people put way to much though on their mind and their memories. You need to stop doing that and stop thinking. Get rid of all your hard drives. Rm -rf everything, don't keep a single thing. A rolling stone gathers no moss. Be a rolling stone. Your old self is stupid anyway.