With the latest version 3.5, Google’s Picasa for Mac has finally emerged from Beta. As Google is renowned for visionary solutions and as the program very much mirrors the functionality of iPhoto, it is worth having a closer look at its geotagging capacity. At first glance, it looks very well integrated, but – as with our previous posts – we are looking for a proper workflow support and future-proof metadata handling.
Google Picasa
Test Setup
(See my “Test Scenario” post for details.)
Picasa scans the hard disk and external drives and reads most image formats, including RAW. The CR2 test files are presented in a tidy interface which is a definite plus – Google have done a very good job with the frontend.
Side-By-Side View
From the “Places” tab in the lower right corner, Google Maps appears with the usual display options “Street Map”, “Satellite”, “Hybrid” and “Terrain”. A starting point for tagging is the search field.
Geotagging starts with the search button or with zooming and dragging the map.
The usual pin action (drag & drop) determines the location of the image or images. That’s it. Compared to iPhoto – where fine-tuning takes some extra clicks – this is a lot more comfortable. But just like iPhoto, Picasa for Mac offers no other way to geotag images: Google Earth integration is not yet available, and batch tagging with track log files is not supported.
As far as a tagging workflow for a bigger number of images is concerned, Picasa does not match the effortless way this is done with our reference application HoudahGeo. Each image requires a minimum of three clicks (next image, drag pin, confirm) rather than one or two. That is between 50 and 200 % more clicks, a lot of extra time.
Also, the map panel can only be resized within certain boundaries which is not ideal for covering larger field trips.
Once they are tagged, the selected images show up as pins on the map.
When one or more images are selected, the map shows all images as pins on the map. Unlike iPhoto, images and map are shown simultaneously and there is a pop-up preview for each individual pin. In my personal opinion, this is a more convenient and sensible layout. (Think about the amount of images you are likely to take in your hometown – by default, iPhoto would display them all as a single pin in the town center that opens a gallery with potentially 10 years worth of local shots.)
The M Question – Future-Proof Tags?
Now there is the vital question about metadata integrity and compliance with the travelling standards paradigm.
The first discovery is that editing of IPTC location fields is not provided for. You may attribute a location to a folder, but this is not particularly clever: what would you call it if it contained the images of your last day trip around the countryside?
The only place Picasa accepts descriptive location data is on folder level. Simple, but not a good metadata concept.
The only location data that is stored on a per-image basis are coordinates. The minimalist approach is not necessarily bad as it means less hustle with typing while preserving the most important information – longitude and latitude – from which everything else can be deducted.
Now let us have a look under the hood: does Picasa keep the geo information to itself or will it travel with the files?
The answer is ambiguous: it depends on the file type. If you work with JPG images, GPS data is instantly written into the files’ EXIF header. But if you work with RAW files, you will have to export or upload the files first – just like with iPhoto. (Existing geotags are read from RAW files, though.)
Quite obviously, Picasa has no file independent metadata engine like Phil Harvey’s ExifTool – which is a severe flaw in Picasa’s design.
Picasa does not cripple the metadata headers, but transforms values and shifts tags.
There is another reason why they should have used ExifTool: even though Picasa does not cripple the image’s metadata like iPhoto, it transforms the values and shifts values to other tags. For many entries this may not do any harm, but there are side-effects with symbols – like the copyright sign or even something as innocent as a forward slash – and with language-specific characters.
I have not found a pattern in this behaviour, but here is an example: my IPTC Source entry “Klaus Meßlinger / Klaus Messlinger Photography” was changed to “Klaus Me§linger / Klaus Messling”. This would not have happened with Phil Harvey’s tool.
And there is one more issue with regards to our geotagging purpose: the location’s altitude is not being determined, but both a zero elevation value and the altitude reference value are created. So, consequently, Picasa will set any existing altitude value to zero once you move the pin even slightly.
Other Metadata Issues
Another – in fact much bigger – metadata problem outside the geotagging realm is that existing IPTC keywords – “tags” in Picasa’s terminology – are not read from RAW images, including DNG. This is a pretty strange behaviour considering that EXIF geotags are read. So for me, using Picasa with my RAW file collection means either to enter keywords twice or to abandon them altogether.
Again, JPG files are doing OK. And again: ExifTool has offered filetype-independent metadata modifications for years, so there is no excuse for doing a worse job.
Conclusion: Smoothness and Reliability
Just like iPhoto, Picasa fails our test because of its lack of metadata integrity, its half-hearted support for RAW files and weak compliance with the travelling standards paradigm. However, it is doing a much better job than iPhoto in terms of usability and added funtionality while offering a near-identical feature set.
If your images are all JPGs, if a caption is all the metadata you ever need and if manual geotagging is alright with you – look no further. However, if you shoot RAW files and use a track logger, do your geotagging elsewhere. You may use Picasa simultaneously, though: a geotagging workflow does not have to be done beforehand as image folders are rescanned continuously. This is another advantage over iPhoto.
Finally, Picasa may take up quite a bit of system resources, probably due to indexing processes. Whenever I have it running, other apps tend to react sluggishly, even though I am on a fairly new Mac.
Test Results Overview
CR2, DNG, JPG etc., reading restrictions for metadata in RAW/DNG files
Reads file system, updates on changes in Lightroom etc.
No track log support, only manual tagging or re-tagging(Batch) adjusts EXIF time, but no time zone supportNoNoGood integration with Google Maps, various views, but requires 3 clicks per imageJPG metadata is preserved, but transformed; no support for RAW/DNG files metadataNo; Picasa online platform can be used for map viewSee my introductory post for more details on the Nine Requirements.
Documentation
No on-board user manual, but fairly self-explanatoryApplication Details
Related Posts
- All posts in “Geotagging” category
- Geotagging and the Mac (1) – Basics
- Geotagging and the Mac (2) – Test Scenario
- Geotagging and the Mac (3) – iPhoto ’09 “Places”
- Geotagging and the Mac (4) – HoudahGeo
PS: Beyond Geotagging – Faces & Co.
Remember, we were looking at Picasa from a geotagging perspective only and our rating refers to only that.
I personally find Picasa the better iPhoto and I like using it alongside Lightroom for images that have already been through my workflow – but not for editing.
It offers a very reliable face recognition and a good integration into Google accounts: faces can be matched with people from your Google contacts – a clever move as it reduces typing time and updates on name changes. Another cool idea: for any given person, Picasa shows which other people were on images with them, which is the visual counterpart of social networking.
(Being myself, I had to take another look under the hood: the faces information is stored in a hidden INI file on a per-directory basis. A strict interpretation of the travelling standards paradigm would require a per-file basis, but at least the metadata file travels nearby and outside the application’s proprietary database. And as long as the overdue metadata standard for faces is not officially defined, this will do.)
Finally, Picasa offers sensible editing features and a smooth integration with its counterpart web galleries. With an advantage in integration, usability and features, and the fact that it is free software, it beats iPhoto. And it is a good companion to more ambitious image processors like Aperture or Lightroom, as it adds the map view and faces features the pro apps are lacking.
Tags: Geotagging, Review, Software, Workflow