Getty's New Watermark
gettyimages.com
gettyimages.com
If you're a designer at an agency (which would use watermarked photos in their comps, and then purchase proper rights once the client has approved) this is no worse than it used to be. You still have a watermarked image, it's just slicker looking. Having the URL in there is handy, but you would have recorded that anyways as part of your workflow. It's worth noting that the license behind these images still hasn't changed - you're only allowed to use them for 30 days, and not in a finished commercial product.
The other kind of client is the small-time blogger who would never pay for the image rights anyways, and would just use a watermarked comp shot in a post. According to the click-through license agreed to when downloading these images, this client type shouldn't exist - it's explicitly against what's permitted in the license - and yet it does, widely. For these clients, putting the URL in the image is a win for Getty; people who used these images will keep using them, but now those clients' viewers will be able to easily track the image back to its source, making it easy to actually purchase the image legally. This turns someone using a Getty image without permission from a problem (violating a license) to effectively free advertising.
The obvious next step - although I bet this would be difficult based on all the pre-existing contracts with the photographers - would be to accept that the second type of client exists, change the license on the comp shots, and try to bring them into the Getty fold. Change the license to say that you're allowed to use the images anywhere online you'd like for free, as long as the photo and watermark stays intact and your number of average monthly unique users is less than, say, 10,000. Doing this would take a sizable user base and turn them from license-violators into free advertising. Financially, this wouldn't affect 90% of the type-2 clients (they'd stay under the viewership cap), and would only impact the bottom line of mid-sized publications, who really should be paying for the rights anyways.
Maybe make gty.im a value-added free service - users could use the comp shots free for X users/month (where X is enough to cover only small/local sites), and then the next tier (X to 3X, say) would be hosted off gty.im, either for free or for a nominal fee. It'd provide analytics, a better-quality image, and would indirectly act as a sales funnel for moving up to a full Getty contract - and even if it didn't, it would start the relationship between Getty and small/independent publishers who have never thought about image rights.
I predict a greater use of watermarked images and greater traffic back to Getty Images.
I can assume that they fell out of favor because human memory favored the "3 digit-4 digit" chunking of the phone number rather than the "two letter-5 digit" chunking.
Having said that, the profusion of really really good modestly priced pro-am cameras has spawned an ugly race-to-the-bottom on stock photography. You can get ( often quite good ) stock photos for next to nothing from the low cost stock photography sites.
To put it in HN parlance, Getty's model has been thoroughly 'disrupted'.
Lets go one by one on the photos on that landing page in terms of minimum equipment and people behind them... short obviously of a camera and an achieved photographer, that is already hard enough to come by and expensive but is a common factor in all images.
Katana image: model, make up, big room with stupidly expensive lamps, costume, katana, watermelon, more than one 1k/2k+ flash head.
War image: ticket to war zone, bulletproof vest, AK-47, missile launcher, missile, model/combatant. Very likely many pro camera bodies, pro to resist dust, many to have spares. You won't find a pro body there if one breaks up, unless you can loot a dead photographer.
London image: helicopter, helicopter pilot, weather controlling machine (can be replaced by an insane amount of luck or many tries and a lot of time in each one).
Guy with car hat: model, car hat, not sure if only one flash head or camera mounted strobe. I'm not sure of how much a car hat costs but I think this is the cheapest photo of the lot.
I agree that the serious hobbyists that feed micro stock companies those 2 dollar pictures will not compete with warzone shots, shots requiring $$$ cost in travel, props, lots of models or whatever to set up unless they happen to be professionals "double-dipping" on a paid shoot. The cash just isn't in it.
The Micro-stock guys might surprise you though at how much cash they have invested in gear ( including lights ) Some of them take perfectly good photos of subjects that might previously have only been sourced more expensively through agency or in house.
NB I was curious what the low linear building in the lower center of the image is - it's the roof of Liverpool Street Station.
Or some reasonably talented kid with Blender (or a warezed copy of Maya) could knock them all out in a weekend, without spending a dime on anything not made by Domino's Pizza or the Coca-Cola Company.
If not now, then soon.
The micro stock model has worked because pricing was high. It's swung the other way. Fair value is not received in many cases.
If you want Art, you can go put up a gallery, and whine there about how the little people who aren't paying you money don't properly appreciate your vision.
If you want business.... Economics 150, friend: In an efficient market, the marginal cost of a new unit of something is the same as its price. If that's less than what you think is "fair", perhaps that's a signal that (a) there are too many people in this market, don't bother, use your skills and/or equipment elsewhere, or (b) someone else can get snazzy photos to people-who-buy-snazzy-photos more efficiently than you.
The price of photographic equipment, the price of introduction to stock-photo companies, the price of accepting-stock-photo-company-terms-and-conditions, and the price of browsing-stock-photo-catalogs have all fallen dramatically thanks to technology and the Internet. That run-of-the-mill stock photos are now much cheaper ought to be expected.
Surely if "the art" is actually a quality that objectively exists and has value, it will resist attack from those that lack it.
Now we are living exceptional times in that cost of hardware to do good work is so cheap that anyone can have a go at photography / programming / ...
Getty makes its money on both ends. They're also slowly rising the purchase prices for both user submission sites to be around $70/photo average compared to $10-20 two years back.
Let's take the katana image, f'rinstance. You want to use it for your French-language extreme cookery magazine? Cool. That leaves the image available for the Karate-gi supplier's worldwide site and catalog, as well as for publication in languages other than French (unless you wanted a worldwide print publication exclusive). If you want first publication rights and exclusivity for a year, you've tied up all other rights in the image.
The rights you want in the image may or may not be available, and if they are available, they will determine the licensing fee because they affect the remaining rights. If you go to a place like iStockphoto, you don't get any kind of exclusivity, and can find that every one of your competitors are using exactly the same images you are using. Oh, and so is the weekend "contractor" down the street—you know, the guy with the Skilsaw, cordless drill/driver and Workmate who ruins people's bathrooms and kitchens (but is so affordable). You get what you pay for.
If you'd like something cheaper (which generally won't look as good, won't have the metadata searching capabilities you needs as a professional photo editor, or may not go with the story you're trying to illustrate), the market will provide - in the form of iStockPhoto and others.
Getty's main client base doesn't ever purchase just one photo, will rarely look at the price on the website (they know what they need, and already know roughly what it will cost), and more often than not will have their own separate pricing contract with Getty that looks nothing like that series of drop-down boxes.
If you're searching for an individual photo, looking at the price, and considering if you need it - you're just icing on the cake for Getty, as opposed to the target of their business model.
Even Getty has such a section: http://www.gettyimages.ch/creativeimages/royaltyfree
Of course, the really great looking photos often are not royalty free, which is why Getty features those on that page.
Saying "$160? What a joke." puts you in the same category as the guys on rentacoder asking for $100 Photoshop clones.
There're a lot of cheaper alternatives with a very direct pricing. See for instance: http://photodune.net/
However Getty does have some really nice images!
A lot of these folks are keenly interested not just in the images themselves, but also in the contexts and markets where they may have previously appeared. Often, the fewer associations an image has, the more likely they are to buy it. And if they want to ensure that no one else buys something they're working with, they'll pay very substantial amounts for full exclusivity.
However those aren't ever going to be paying customers anyway. When you license stock, you get more than just the high resolution file, but also 1. exclusivity if you've requested it and the necessary waivers to use the image in the way intended.
For example, a night time shot of the Eiffel Tower requires a waiver, as the light show is subject to copyright. (Believe it or not!) While a day shot won't require a waiver.
Not having a legitimate waiver is asking for a damages case by the owner, those depicted or other entity. This is covered in most countries via copyright & publicity laws.
If you're asking about metadata, I can say that on the Getty images I've used in the past (not for years though, so it's old data,) they were pretty diligent about ensuring Authorship information in the metadata. I can't remember whether or not there were any specific Getty attributions in there, like a link back to a page it could be purchased from, but in other comments here I see that they have a 'creative ID', which I didn't know about, so it's possible I just glossed over it.
If you're asking about the second, I believe the answer is no, but I say so without having tried it, even cursorily.
I was going to question the usefulness of an invisible watermark, as you could otherwise just compare a hash of the image, but I suppose an invisible watermark would likely persist through re-editing / resizing?
Apparently they expect people to use their javascript widget to select a home country, and then send you to e.g. http://www.gettyimages.com/Creative/Frontdoor/NewWatermark?L...
FWIW http://www.gettyimages.co.uk/Creative/Frontdoor apparently gives some details about the server, the directory structure and the auth system in place that might not be worth exposing to external IP addresses.
Such a temporary picture MUST be neutral because you need it to fit your composition/layout. This rectangle is the opposite of neutral: it's huge and very visible. This big square is going to clash more often than not with other shapes around the image which won't be aligned with it provoking all kinds of troubles for the designers who use them. Clients will think that the square is part of the designer's intent and, when it will be explained to them that no, this big ass square isn't part of the design they will rightfully say "Well, get rid of it!". But it won't be possible without paying large buckets of money to Getty.
Or using another service.
This may seem like an intractable tension, but I don't think it necessarily has to be--at least not in the following situation:
1) The content creator cares more about attribution than monetization.
2) The content sharer cares more about sharing unobstructed images than about passing off the content as their own original creation.
In this case, there could be some sort of standard metadata that accompanies the image and is only displayed upon some user interaction, such as a mouseover. We could even use steganography[1] to do this right now, without needing to settle upon a new image format.
I realize this is a bit too idealistic to be taken seriously. Just interesting to think about, is all.
I wouldn't bet on that, if you're talking about a professional photographer.
Apart from the pet-peeve of the missing http://, is splitting URLs across lines a thing now?
If you're smart enough to find the image, download it, embed it in your website, you're smart enough to figure out the rest.
It's not ment to be used on a production/deployed website to drive traffic to getty. It's just a reminder to you.
So, in a sense, you're right. It is both a URL and a 'product code'.
Also, it looks fairly easy for an "unfair use"-er to "heal" the watermark away. They should consider a watermark with higher entropy. (like a stipple-shaded box or one with subtle random variation in shade around the box edge.)
http://www.gettyimages.com/detail/photo/chiufen-village-high...
http://www.gettyimages.com/detail/news-photo/singer-and-song...
Pretty easy to crop out the right shot and have a perfectly fine, watermark-less photo.
On the other hand, like most DRMish solutions, I guess if you're dedicated enough to not paying for works like these, you will.
As someone who works with stock photos constantly, I can tell you that clients hate seeing watermarks, they're usually right in the way of the useful part of the image and are often distracting. As a result the art department photoshop them out which is painstaking and a waste of time for a low resolution image that is only going to be used once.
I'm pleased that this new watermark is very easy to undo/disguise. Getty seem to understand that stock-pirates are fine with using low-resolution, highly compressed preview images without a license, while their actual customers need images with watermarks that are unobtrusive and easier to sell to the paying-client.
The truth is that the image need only a very minor watermark, so the creative can track down the original. Creative agencies aren't stupid and are well versed in the laws that relate to stock images. They don't use random photos they find on the internet, and larger or risk-absolving clients will often require a copy of the license and waiver for the stock imagery.
The watermark obscures vital detail in the image, kind of destroys the feeling of the original photo. So those of you who can only seem to muster up negative comments, go back to 4chan. The new watermark is innovative in an industry that feels like its been the same forever.
Maybe there are details I'm missing, but you don't mention any RIAA-like practices or gross inflation of the cost/damages.
I don't think the "fee" is unreasonable if a few hundred dollars is the going rate for licensing the image (although I personally would never pay that much), I just find the automated process of hunting these people down and asking for money right away (no option to just take it down immediately) a little distasteful, and in a similar vein to the RIAA's tactics.
It's a slippery path from there to charging arbitrarily ridiculous amounts like [$15K][1] for a violation, leading to insane lump sum damage amounts like [$75 trillion][2].
[1]: http://arbornet.org/~danpeng/ [2]: http://www.pcworld.com/article/223431/riaa_thinks_limewire_o...
I don't think it's got anything to do with devaluing the image by posting it on a micro website, so much as recognizing potential commercial value on someone else's work, without giving them their due.
If this had not been a business (like folks downloading mp3s, and not reselling them), I'd probably think about it differently, but in this case, there was specifically a commercial intent.
I agree the $15k / $75T thing is beyond ridiculous, especially given the non-commercial nature of that infringement. We're not arguing there.
I'm arguing 'oops, I'm sorry I made money using your work without paying you' isn't good enough. Who knows, maybe it could have been $5 - the sum isn't the important part, it's the principle.
So, that way, if someone, say, shares the photo on Pinterest, it isn't a total loss for them. I can see the theory behind not giving it so much noise.
A few years back I joined HN and was so intimidated by all the smart people here that it took a long time before I started commenting. Now, if you're not paying attention it's easy to mistakenly assume that the smart people "know better" and will comment with their opinions on how they'd do XYZ better. A lot of the threads I saw back then had people giving contrarian opinions and getting lots of attention for it. So for the newbie it's most likely easy to assume that's how things are done around here and that's how you prove your intelligence/expertise. But the thing is, what often looks like a contrarian view is often a question or an alternative way to do XYZ that isn't really saying XYZ is bad or wrong at all.
So I think it's a case of monkey see monkey do mixed with a misunderstanding of popular comments, who their authors are, and missing context. Everyone wants to be the smartest person in the room. HN can be an intimidating place because of how seriously exceptional a lot of people around here are. Hell, I still feel like a moron half the time when I comment. Egos get in the way of accepting that instead of being the smartest you should learn from the smartest and seeing that you don't have to say anything at all. In fact, saying nothing is often the smartest thing to do. How does that saying go? Something like "Shut up and let people assume you're stupid instead of opening your mouth and confirming it"?
It's not necessarily a negative reaction, and it doesn't mean I'm not going to buy into the idea eventually. It just means I don't take it at face value.
I used to get frustrated that people would see that kind of reaction as negative, but I've gotten better at mitigating that kind of reception over the years. And the people I've worked with long term have come to find ways to use it to all of our advantages.
For example, on a couple of development teams, I've ended up as that "unofficial lead debugger" for everyone's code. One mistake I work to avoid in debugging is thinking "Well, the problem can't be here, I know this part works." Cause, you know, that's inevitably where the problem is. :-)
But there's a big difference between that and between people who are just bashing.
I don't consider myself a fool, or ignorant, but choose to remain silent most of the time because of the exceptional insight many have, and I love HN for it. i just skip over the negative stuff and look for those nuggets of insight so prevelent here.
[1] http://bible.cc/proverbs/17-28.htm
[2] http://www.brainyquote.com/quotes/quotes/m/marktwain122650.h...
First, it seems very useless to say “This is cool, I like it”. We know these types of comments, which often remain around the bottom.
Much easier is to write a useful objection that would grow into a thread and is likely to add to the discussion. You can point out a factual mistake, doubt something—you just need to stay more or less on topic.
The problem is often, though, in how the objection is phrased. One can say something like: “This is very good work, but [objection phrased as a question]. But still, [x] is a nice idea.” I always try to object like that. However, it takes time. If the person appropriate for the feedback is unlikely to read HN, politeness doesn't seem to be worth it. We also need to account for cultural differences. Then, many of more technical people here seem to be not very sensitive to politeness level, but very sensitive to factual mistakes.
So we end up with comments that look frank, but are useful and start discussions.
New people probably want to comment in the same spirit to attract attention and build up karma, but they may assume that frankness is the key. (What for one is a plain simple comment, for another may look like a blunt statement.)
So we end up with comments that often look frank and aren't very useful.
It's a shame, because it never used to be that way. And if ever it was, it was in a positive "look, I made this!" way.
To me, the fact that a story like this would even get to #1 on HN says more about how HN has morphed. Would we have seen this 4 or 5 years ago? Unlikely IMHO.
Deleted comment
Not true since the days of TIFF. JPEG has application-specific tag and Exif data. PNG files are often abused to contain non-image data (i.e. Adobe Fireworks stores vector data after the end of the image). Even GIF files have a comment field!
So, stick the hyperlinks links inside JPEG tags. Use Exif so others can read them. Or make a Photoshop plugin that will display them and let the user click on them.
You could also stick the tags in using steganography, which is more in line with how a "watermark" should work.
A QR-code is very close, though...
As far as I know you'd need at least two devices (the one whose screen is displaying the image with the QR code, and the one you're using to snap a photo of it). And then you'd probably want send the decoded URL back to the original device to load in the original browser.
Edit: Maybe you were thinking about the analog world, but watermarked images tend to show up in the digital space (in mockups especially).
There are three purposes mixed in this "watermark":
a) Give the innocent viewer a way to know the name of the photograph and go the original source of the image, maybe to buy the picture.
b) Make it harder for less-than-innocent people who want to take these images and use them for illustrative purposes without people noticing they did not actually take these pictures.
c) Inject a fingerprint in these images so that can serve as a proof of ownership in case of legal issue, and that may also help crawling the web searching for them.
For a), it seems much preferable to have a classic caption, where more useful information may fit (date, place, etc)
For b), with my limited gimp skills, I guess I would need 30 minutes to remove the watermark the first time, and then it would take 2 minutes.
For c) I suppose a real invisible watermark is actually much harder to erase and a better proof of ownership, but maybe they do add such a fingerprint also?
Anyway, I see here yet another contradiction of DRMs: content provider want more and more people accessing their content for free, but they also want, at the same time, this access to be unpleasant enough so that these users would buy the paid content. I don't think this scheme will survive in the long-term.
Example: http://dl.dropbox.com/u/58607934/stuff/getty.jpeg
So in other words, option b and option c has always been on the table. However, we're in an era where images are becoming much easier to share on other platforms. All someone has to do to share a picture these days is to pin it or post it on Tumblr. All that metadata gets stripped when people do that. At least with this, people know where to find it.
Considering the changes in sharing images, why wouldn't Getty want to take advantage of this? That way, if people are going to share them anyway, at least it isn't a total loss to them — and hey, people might even be more likely to share them because it doesn't have a supremely ugly watermark.
Media outlets pay $100 — each — for many of these photos. This a big business for Getty, so they need to protect it. Now, Getty could do things to make it easier for smaller entities — say, bloggers — to take advantage of this. This feels like a good, if imperfect, compromise.
Now you might be willing to use GIMP to modify a photo, but most people won't. It's why Adobe can charge $600 for Photoshop. Too expensive? You're not their target market.
This isn't to say there's not a better way, but it just feels like we're wringing our hands over what seems to be a good-faith effort to improve something which previously hindered the experience.
Anyone can add a watermark to an image, regardless if they are the owner, creator, copyright holder or whatever.
Anyone can even add a getty watermark to an image. This is called a copy attack and it's the reason why the watermark is not very useful for identification by a crawler. A getty watermarked image could be a real getty image or any image someone inserted a getty watermark. In other words you could sort of DoS the getty crawler very easily.
Digimarc had an interesting research paper on their website about this topic but unfortunately it had been removed. In 2011 I wrote an email to Digimarc customer support asking if their newest watermarks are still vulnerable to a copy attack and the answer was: "There are hundreds of applications and specific use cases for digital watermarking as I'm sure you found in your research. Some of our published papers describe theories and some describe applications and solutions that are available for license today as end-user products. "
EDIT: Found the Digimarc paper http://www.google.com/url?sa=t&rct=j&q=&esrc=s&#....
Since these new watermarks include a URL that identifies the Getty stock number of the photo, someone trying to fake that would end up putting a bogus stock number on the watermark, which makes it even easier to ignore fakes. The fake watermark would either use a number that doesn't correspond to a Getty picture at all, or a number that points at a different image.
You are right with your objection that watermark contains the stock number. I don't see however how this can help a crawler to identify a fake stock number. To compare the (possible altered) image with the original we have to use an image hash and then we could have used the image hash in the first place.