Stenis
(Sten-à ke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
41
@JoJu
No matter which profile you have as a photographer there are a few things I think most photographers ought to be able to agree on when it comes to using some kind of metadata editing software and that is that it has to be future proof. It has to be able to migrate, scale up or scale down between different systems without getting locked into som proprietary back yard. Scaling up can be to a multiuser system or a system that is more flexible, that can handle more than one database simultaneously or a system being able not just to handle images but all sorts of documents.
Fortunately, so far IPTC/XMP seems to prevail and it also seems that it works pretty well to migrate between the most common converters and PM for example. I don´t deny there are issues. I first had problems with which location data I should use between PM and Photolab. There are three to choose between and first I picked the wrong one. Issues like that are common and I think it´s strange that the Description element in IPTC wasn´t earlier read by Windows for example but in Win 11 even Microsoft seems to have finally got it right.
Usually, a mismatch between two systems where the receiving system lacks a field from the sending system in itâs IPTC or XMP schema doesnât necessarily stops the data interchange, it just wouldnât display that data but differences in data types can create problems. So one has to test carefully.
Metadata maintenance can be hard work and I also think most people can agree on that it might not be worth it for all. So, these new, at least partly AI-driven metadata workflows might leave some things to be wished but they might be a better solution for some than no metadata at all. I don´t think it will be something for me personally but I am interested to see where it will take us.
Today there is an article in one of Swedens biggest evening papers Expressen about an AI-bot (from OpenAI) that writes better than most pupils in the age between 13 and 15 do. As an old teacher under nine years, I would not be surprised. The message is that the school has to talk about this because this might develop to one of the school systems biggest challenges when pupils start to use it to solve their different tasks and home works with the help of this bot instead of study and read by themselves. Culture journalists writes that a bot never can replace a human but the writer of the article responds that he thinks these journalists has missed the point that he means is that this âfirst-generationâ bot already writes as good or better than most of his pupils both when it comes to the language itself and the content. He also reminds these journalists about that 50% of the population has an IQ under the average IQ-level.
Stenis
(Sten-à ke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
42
I think it has to be the high res. previews used when postprocessing the images that makes the difference, donât you? The images developed in Photolab hasnât got any C1 masks yet.
Stenis
(Sten-à ke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
43
An example of the difference image metadata or an illustrated blog story can make for the searchability. Usually, the images in stories like that are indexed by Google
We werre talking earlier about whether it would be any idea to add metadata to ones lmages. Here a little story from today.
I suddenly received an SMS today from a girl in Sweden who had seen some images i took of a Fiat 500 outside the Town Hall in Vaxholm (yes, our little village with less than 5000 inhibitants has called itself âTownâ since the fifteen houndreds and in swedish it is even funnier because the word âStadâ is used by both by this little shit hole and by the capital Stockholm).
She is converting Fiat 500 cars to electrified versions of that car and asked if she could use my images on her web and I usually give that permission if people respect the Creative Commons Licence BY which means they have to write who took the images and they are not used in large scale businesses.
With two electrical wheel engines it takes you from 0t to 100 km/h in 10 seconds.
Below the illustrated blog story that this girl found (translated quickly by Google Translate)
Postprocessing of the images makes some bytes difference per image, plus eventually bigger preview file sizes, depending if I increase structure sharpness, leading to less compression.
I treed with an old catalog, 2010 Iâve been in Liverpool with only a Canon G11 (I loved to have it with me everywhere, never let me down, Nikon P7800 is equally cool, just without OVF), and I had so much fun with that little beast.
First the catalog showed 730 MB (178 pictures). Preview is set to 5160 pix long side, but the G11 only has 3648, so I could get away with 1:1 previews and still save space.
But I wanted to know how big the catalog would be with half size previews (2560): 188.5 MB after generating previews.
And in 5160? 557 MB after generating previews
And in 3840? 555 MB, so that means, C1 is not blowing up the previews beyond what the camera can deliver. But what about the 730 from the beginning? The old catalog was bigger than the new now is and I still like some pics:
In Switzerland is also a Fiat 500 electrified project going on, I read. Good idea, to give old cars in good shape and more individual design a second future.
Now, nowhere you wrote that she found your images because of the metatag Fiat500 or because she googled and Google checked your metadata - but naming the picture âFiat 500 outside of Vaxholm Town Hallâ could have done the finding job equally, right? Just teasing, your metatagging is cosmic, out of this world and simply superb.
And you know, Google image search is rather powerful. Without your metadata workâŚ
Stenis
(Sten-à ke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
46
I really like the extreme feeling of depth in the first image and the boat ⌠just lovely. A real architecture contrast in the third. Reminds me about a place here too.
Iâm now leaving the Mac to let it re-create 17.000 previews of the second oldest catalog. Again, the numbers are decreasing. I suspect, C1 uses a better compression for the previews.
Edit: After creating new previews and although those are the maximum size, catalog size went from 130.48 GB to 112.16 GB. Now I wonder what the current main catalog will differ
Stenis
(Sten-à ke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
48
An example of the difference image metadata or an illustrated blog story can make for the searchability., I wrote and this example is absolutely an example of the fact that Google indexes all images that is a part of the blog stories I write. I have a lot of images in portfolios too (and those do not carry any baked in metadata but since there is text written to those images on that webside it´s indexed too but earlier they were invisible for Google.
I also have a small static web exhibition on another site that was the first I published and those were indexed by Google from the very beginning. They are examples of early repro images that don´t carry metadata inside the images from what I recall but since there is text content written to each image on the webside Google finds them too.
Below a Google Translate translated link to the site I mentioned above
@Stenis , I use PM and PL together much like you. You mention that integration with C1 and PM is problematic, care to expand? I note there is a full sync XMP option in C1, much like always in sync in PL
But only via XMP sidecars. This works for RAW files, though not perfectly, but PM does not recognize sidecars for tif, jpg, dng, etc.
Iâve given up trying to even sync the few items I had wanted to sync from C1, star ratings and color labels mostly, because of this. And I was running into issues with, if I remember correctly, color labels being linked with Urgency (I think) which caused issues in the other direction.
