The Best Photo Organizing App? I’m Still Looking
nytimes.com
nytimes.com
The old flickr was perfect for me, I could group things by continent/country/city give a title, drop it on a map to geolocate it, etc. But now they've flattened out the structure a bit and it's not as good for my purposes.
I completely agree. I also liked flickr's system, but I left a year or two ago, after deciding I would rather not put my photos on someone else's server. I would love a good desktop metadata system that felt like it would be long-lived and widely useable. gthumb and darktable both have tagging capabilities, for example, but I have not gotten the sense that either of their systems are worth going back and tagging the past 10+ years of photos I have taken. A great metadata system would have me doing that.
I know next to nothing about photography, so maybe there tons of tools out there to manage your photo library EXIF tags that I have never noticed.
So what I do is store a ".meta" file in the directory of each set of photos. That contains "location=" and "tags=foo,bar". Then I have a wee shell script to search.
If I need to I can have ".meta.IMG_FOO.CR3" to store per-picture data. But generally I order by date + location anyway, so having one file in a directory works out well.
On the "where it was shot" front, I'd love an app that could interpolate positions from iPhone photo EXIF data and use those positions to tag photos from my proper camera based on timestamp in a way that's easy to use and dead-simple. I've found methods of doing it, but none that are worth the effort...
This problem cannot be reasonably solved with a web app. Or even a native app running off a mechanical hard disk. With 20MB photos common these days, you need the data coming off a blazing fast local SSD. Otherwise, you can't flip through your photos in a reasonable way. It takes too long to move from one to the next, and every web app does that thing where they load this horrid blurry version first while waiting for the full-resolution version to load. Unacceptable (to anybody who's ever experienced instantly flipping through a hundred pics off a fast SSD, anyway).
So a web app can't do it, but you do need a web app component, and cloud storage strategy, to solve this problem. People's local SSDs are typically too small to store all their photos. Plus, you want your photos everywhere, on all your devices.
So not only do you do need to architect a web/cloud component, but you either need to create a native app for all devices, or at least create an integration with the local photo management facilities on each platform, so that the subset of photos people want are in fast local storage.
It's a hard problem. Meanwhile, my own 500GB of photos (I used to have only maybe 20GB, but then my kids were born) are just sitting in a big folder, with no organization at all, waiting for somebody so solve it.
Lightroom lets me do all the adjustments and metadata correction locally (something that only has to be done once anyway) and Picturelife gives me the cloud storage and browsing/sharing capabilities I need. It's not one application but it is a workable solution.
As for webapps being 'doomed', there's a pretty large spectrum between 'horrid blurry version' and 'full-resolution version'. Considering that my photos are generally ~4K in resolution and I don't own a monitor that can display them in full resolution, a 1080p scaled version is fine and loads entirely fast enough, even on my terrible ~8Mbps ADSL connection.
Sure it's not as great as flipping through an SSD locally but I think it's definitely acceptable and 'a reasonable way'. I think you're just setting your standards too high. Most of the world doesn't even own an SSD, let alone demand that speed in every application they use. Most users are entirely happy just flipping through albums on Facebook.
Once you get used to that (which doesn't take long at all) the old, painfully slow ways no longer seem 'reasonable'.
Right now, my equipment may be fairly high-end, but we are fast approaching the day when virtually all computers have fast storage and 4K+ displays.
In 1992, I thought it was reasonable to dial a modem, log onto the Interzone BBS, and have conversations like this one at 9600 baud in Zterm on a 512x342 pixel screen. Few people would accept that today, though.
Having said that, I do agree that Dropbox + Lightroom + Picturelife is about as good as managing photos gets today. (I canceled my own Picturelife subscription, though, because using it was too slow.)
To address the speed issue, Mylio converts 20+ MB raw images into 1 MB, 2048 px previews that are faster to load and can be automatically loaded on mobile devices or laptops with smaller hdds. Caveat, we use a proprietary compression technology to convert raw images into previews, so not all cameras are supported, but we are increasing camera support all the time.
A lot of the users of Mylio have terrabytes of photos, and Mylio will automatically provide you with different views of your photos (time, location, people, and customized albums) as well as metadata and rating based searches across your entire library of photos, on desktops or mobile devices.
Bonus features, Mylio will help you sync you automatically sync photos between your phone/table/desktop and external hard drives/nas all without going through the cloud.
(Disclaimer: I work there, but my position is supported by many other reviews.)
I don't want ever want to "sync" to local storage, nor rely in anyway on local storage. Local storage is temporary. I want durable "cloud" storage.
But note, I want to be able to safely backup my entire digital collection wherever I choose. Perhaps backing up the Amazon Glacier, or something like that. I don't trust that a new dotcom is going to necessarily survive.
Thanks for the pointer, going to check it out.
I completely agree. One of the features that I love about Picturelife is the ability to setup your own storage. Instead of paying for the service, you can connect it to an Amazon S3 bucket. All the photos you upload are stored in a bucket that you own and you only pay the S3 bills.
Then, because you have complete control and access to the bucket itself, you could setup a backup to Glacier every week or something.
I like how picturelife: - has location based searching - shows "very rough" thumbnails if you browse quickly - Can handle my 9000+ pics on my iphone easily - doesn't REQUIRE me to backup to the cloud (but has that option)
Like others, i'm still looking for a good desktop solution to keep track of my RAW photos from my SLR that is in multiple LR catalogs - it all ends up on my NAS bu t unbrowsable really. Mosaic has been a great fit to help see my LR photos in iOS - better than LR mobile. http://www.mosaicarchive.com/
1. First of its kind to group photos by moments.
2. Easy to organize your pics into awesome albums.
3. SafeView, brilliantly hide your personal photos, and only show relevant photos to your friends
Get from Google Play: http://goo.gl/L1myRJ
- Moments just seems to be organized by day instead of by month as in Timeline view.
- No easy way to bulk-select photos to add to an album.
- Curious overuse of radio buttons. Did not like.
While not strictly a photo organizer app, it does have a photo gallery, timeline, and raw file format support.
- please allow multiple folder selection for volume source folder
- the analysis phase is really slow and it looks CPU-bound on the ImageScaler.resize() method; try to find a faster resize library or at least allow user setting for scaling (smooth is quite slow) and multithread the resize - right now I have 7 idle CPU cores and you have a really slow application. This is bordering on unusable. I have ~2TB of jpg and raw photos, it would take at least half a month to resize them - and I didn't even ask for a thumbnail.
- A snapshot cannot be stopped (there is no "stop" button/menu entry).
- the persistent mode is called "Persistent" in one place and "Archival" in another
- Not sure I understand the "persistent/archival" setting; I would expect that I can access the old/deleted files as long as they are in the repository files even if the source is deleted/unavailable. I now have a 2GB repository "backup" but I cannot access anything from it because I removed the volumes (the repository did not shrink). At least make it the default with a big warning if you disable it.
- 64MB file chunk for the repository is quite small for something containing photos that cannot ever be deleted (a raw file can easily have 10 or 20MB); file access in folders with a lot of files is slow
- please read EXIF data - it's way more important than Hex views; the metadata-extractor library[1] is stable and fast (I'm using it for my photo organizing application)
- please rotate the thumbnails by EXIF data
- corrupting/deleting a backup file doesn't give any warning (just that the backups don't work anymore and I cannot see the photos)
- not providing a name for the volume in step 1 gives a warning when you want to leave step 2 (and not when you leave step 1)
- raw image decoding is painfully slow; processing an 8MP Canon RAW takes 40 seconds; it estimates 6 hours for 480 photos.
- try to buffer the DBInput/Output as RAF can be quite slow; check MappedByteBuffer
So it looks like a good idea but it's really slow and it can fill up disk space with files that cannot be recovered.
Persistent mode idea is for situations where the user wants to accumulate a large repository piecemeal - for example, a photographer can dump a day worth of files in a folder, upload to the repository and then delete the files on the local machine. This way the next sync will not delete the files dring the next snapshot.
Can you email me an example of an image file that needs to be rotated (all mine have been rotated using windows preview)?
Again, thanks for your comments! Very helpful.
Xoopit search was better than gmail search. Significantly.
Xoopit created a "personal media," [1] not social media, out of my inbox. This felt better than Facebook to me; when I had FB that is.
[1] I heard the term personal media from Tyler Crowley, which helped to change how I thought about this space.
The project is still in its infancy (as in "not widespread yet") but it's advancing at a good pace. For the moment the focus is put on how to run it in the cloud, so that people can install a simple instance somewhere for them, for their family or even for business (so you can run a public instance). But if you want to see it for yourself, you'll see that a good amount of work has been put on images (the golang's stdlib for image was improved from camlistore work) as you can see in the videos.
Get in touch if you need anything and/or have feedback :) shai@fomo.fm
Given that I've been doing this ever since I started amassing digital photos, and that the author doesn't, and that her NY Times post is on HN, I have to wonder what others do with their photos.
Each year has a folder: YYYY. Inside that folder is an Aperture library per category (e.g., Weddings, Portraits, Personal, etc.) Inside each Aperture library there are projects for each grouping of photos (e.g., Valley of Fire). Every file has a unique name that includes the date photographed and the frame number.
I do allow Aperture to keep the image files internally because it simplifies backups. Aperture does not support XMP sidecar files like Lightroom and others do, so the only benefit I feel I'm losing out on by not storing them separately is having a file structure that matches the library structure. The libraries themselves are OS X packages, so it is possible to extract the files relatively easily if I'm no longer able to open Aperture in the future.
I also do not care to have most photos on multiple devices. I have the ability to remote into the computer with the photos from anywhere I have an internet connection, so I'm able to grab a file if needed. With multiple terabytes of photos it's impractical to make them universally available to all of my devices anyway.
It's not open source, but in the past I've used picassa to do recognition. You let it run, interact with it to resolve ambiguities, and then pull the face assignments/rectangles out of the .picasa.ini files that it drops next to your photos. (There was also a contacts.xml file somewhere that provided the mapping from a guid to name/email.)
archive/people/[location]/Bob archive/people/[location]/Multi/Bob & [x]
archive/people/[location]/Multi/Bob & [y]
I understand that my answer to the problem is a crappy one, namely "enjoy curating your photos and occasionally culling those who have become redundant or obsolete". But for what it's worth, it's how I do it :)
year\year-month location
So for example 2010\2010-11 Paris\2010-11-27\Nikon\Louvre
Is where all the pictures of the Louvre I took on that day are. If I need to find pictures of the Louvre, I just use Everything on the top level folder and search for "Louvre".
It works well enough, and I just sort everything back at the hotel while I'm copying off of my camera onto my laptop. I'm not super strict about the specific locations, sometimes under "Nikon" will just be a pile of pictures. Depends on how motivated I am to sort. I'm usually really good about filtering out pictures of people from stuff though, and I post pictures of people to my G+ account so my friends and wife and can enjoy them, and I use my flickr for stuff. The best quality stuff I then publish through Getty. Maybe less than 1% gets professionally published, but I have almost 120,000 photos organized this way. I spend lots of time editing metadata in flickr and even more for the Getty material since they have very high quality control.
I don't really make much money at it, but it's a hobby and I get to call myself a "professional photographer" in some settings.
The ones I publish, I go through, pick the best shots from the group and when I go through and edit the levels and crops and such I usually do a little pruning of that set along the way and the edited versions go in yet another folder deeper (so I have a non-destructive copy). It's useful for me to publish several variations (like different zoom levels or crops) so they buyers can buy one that fits their need the best.
It's not easy to make ends meet in this market: the percentage of users that are willing to pay a monthly/yearly subscription fee is too low to make up for the cost of storing the photos.
The big players provide online photo storage for free, so the user's position is understandable. But it doesn't make it any easier for a small business that wants to, well, make business.
We started pushing some cool features like auto-magic clustering of photos into moments. Now Google and a few others do it too, so I know it doesn't sound all that impressive now.
* By default the photos can be re-sold and the user gets part of the profit.
* Services such as photo restoration
* Selling physical goods such as mugs, tshirts, postcards, posters, puzzles, Christmas cards, photo books, etc
* Ads on top of a chat system based around photos ;)
* A social network that uses the photos to get your friends to join the network
* Now that you are backing up your photos for $ we could back up all your data
* Daily advertising games such as 1 random users that uploads a photo today featuring Taco Bell gets $1000 (aka advertisers pay for everything).
In fact any company that derives revenue secondary from having photos, but not from the users storing the photos is incentivized to let the users store the photos for as cheap as possible if not free (when possible). This is probably the number one problem with any company that is trying to sell storage for photos and will no doubt force them to pivot or close down eventually.
Alternatively the other option is to only cater to users that are willing to pay and have no free option at all. You would want to interview those users and make sure the problem they have is being solved.
It has automated cataloguing, organization, and offline backup of large number of photos all synchronized across multiple devices.
Disclaimer - I work for the company, but I really feel that the product can help people manage their photos.
From online services I like (and fear) Google+ which let you search your photos via context without any prior tagging (e.g “show my photos of cats”).