Re:Unlimited is like a Ponzi Scheme
by gravitas on Sunday December 13, 2020 @09:35AM
(#25406807)
The organization and data which is in that takeout is a complete mess of years of design changes from picasa -> photos, empty folders, duplicate date folders, hidden old crap from your Google+ account (like banner pics) which you can't see anymore to delete, just stuff everywhere inside those zip files. $0.02 if you ever have to actually deal with them, add time to first normalize what you unzip.
[ Parent
| on HN ]
In all fairness, organizing old photos in general (especially scanned in pre-digital ones) is pretty much a horrid task in most cases. I'm dealing with it right now and stuff is linked to all over my hard drive, dupes in different formats at different levels of quality, limited metadata, etc. New stuff suffers from there being just so much of it--much of it repetitive--but at least it's in some semblance of order.
Afew years ago, I ran all my family photos through a script that renamed them based on the year, month, and date. I may have retained the old name in the metadata.
I'm looking for the script I used. I also have some tools that compare hashes and find duplicates.
You can google that stuff, or if you want to hire me to help, PM me.
I sorted stuff by ages and, very roughly, by content ages ago. So it's not horrible. By biggest issue is that when I first started using Lightroom I didn't really have a good orderly ingestion workflow so there's a duplication of raw scans, PSDs, JPGs, different versions of edits, etc. It's hard to automate and what to save and what to toss from the catalog is somewhat a matter of judgement.
>I ran all my family photos through a script that renamed them based on the year, month, and date.
I actually don't care about the filename, although maybe once things are cleaned up I'll do an export of everything. It's all in a Lightroom catalog now. It's "just" a matter of adding more primarily location-based metadata and making sure everything I especially care about is in appropriate Collections.
Years ago when Flickr made a big change and I needed to extract my original uploaded content (almost same exact scenario as this thread), I found this script as a starting point (it uses jhead): http://neverfear.org/blog/view/148/Rename_all_jpeg_files_by_...
For me, Flickr has always just been about copies of local files. At some point though, I do need to look at what all I've uploaded there. At one point, my Lightroom database got corrupted. I was able to recover it but I lost the local service mirror of what I had uploaded. (And a fair bit of stuff pre-dated using Flickr in any case.)
I discovered ratarmount the other day (pip3 install ratarmount), which fuse-mounts tgz. Fancier, you can glob all your Takeouts into a single merged directory (ratarmount *.tgz ~/takeout). Even fancier, PhotoStructure will then read and de-dupe that directory hierarchy. I just added person and album metadata parsers (those will be in the next release), which were the last bits that we're missing.
zip files suck for jpg, it's already compressed. If you're storing raw formats, it might help some.
I agree but I use zip as a handy container format. Assuming I have 50 image files I use the -0 option for no compression. It runs a lot quicker and it is easier to send/upload/download 1 file than 50 individual files.
zip -r -0 file.zip dir