Recently Google announced an update to its Google Image Search product to improve its UI and make it more useful.

However, the UI seems to have been greeted with mixed reviews. In addition to some noting the similarities between the new search and Bing’s offering, the response to it has been split with some users hating it and others loving it.

Webmasters, however, have also been chiming in and their opinion seems much more negative. The reason is that, while all the changes were to the user interface, at least one of those changes has the potential to actually take audience attention away from the original sites.

It’s a problem that has a growing number of webmasters upset and, now that the new format appears to be here to stay, the chorus seems likely to grow.

The Changes

The most visible changes to Google Image Search are on the image search page itself. Gone is the “grid” layout and it is replaced with a more varied one that uses larger thumbnails with less whitespace. Gone are individual pages as it is replaced with a nearly-infinite scroll that will let the user go through 1,000 images at once.

One thing that has been removed is the image information, including dimensions and relevant text. Previously that was displayed below the image but now that information is only shown after the user hovers over the image itself, one point of contention among many who use the service.

Finally, clicking the image also now produces a different result. Rather than going to the original site loaded in a Google frame (with the Google portion on top), Google loads the site in a frame, with their panel to the right and adds a “lightbox” effect to the image, placing it over the original site and graying out the page.

It is this last change that has been the subject of most of the controversy among webmasters.

Not only does this add an extra click between the image result and the original site (forcing the user to click on the site if they want to see it), but it also loads the entire page in the user’s browser, using the site’s bandwidth, and keeps the ads from being clickable until the site is activated.

In short, many feel that Google is de-emphasizing the original site and, instead, putting even more focus on the image itself and their own search panel. This has the potential to hurt webmasters who earn revenue from advertising and see a lot of traffic from Google Image Search and that has some upset about the changes.

Time to Leave Google Image Search?

Previously webmasters, especially artists, have had something of a love/hate relationship with Google Image Search. Though it brought in some well-targeted traffic, many who did not know better used Google Image Search as if it were a stock photo library.

Google, for its part, hasn’t done a great deal to dispel this idea. It offers a vague warning that “This image may be subject to copyright”, a warning that is still present but has been moved to the right-hand side of the results page, and it did introduced a Creative Commons search that is very difficult to find for those who aren’t actively looking for it.

However, since Google Image Search as a good driver of traffic, many tolerated fact some users were abusing the search but it may be time to reevaluate that. The reason is that the quality of traffic from Google Image Search is almost certainly going to decline as sites are hidden behind the image, grayed out and are unclickable until activated.

Many, are going to wonder if Google Image Search is worth participating in and, since you can remove your site from Google Image Search without removing it from the main index, now may be the time to do just that.

Bottom Line

In the end, this is a choice up to the individual webmaster. Personally, I won’t be taking this step at this time. The reason is that I don’t see a great deal of traffic from Google Image Search now and most of the images on this site are logo and screenshots. If I were an artist or photographer where my images were more valuable, I would have to reconsider but, then again, I would also be more dependent on Google Image Search for traffic.

So I ask you the questions. What do you think about the new changes? If you are unhappy with them, do you plan on removing your site from Google Image Search?

All in all, there are no easy answers to this one but I can definitely see why many webmasters are upset.

When I work with visual artists, one of the most common complaints that I hear is that too many people treat Google Image Search as a clip art or stock photography library, taking and using works without permission, assuming that it is ok.

Yesterday, I was beta testing the ImageExchange Firefox plugin (full report should be tomorrow) and part of it was entering a lot of queries into Google Images to see how well the system works.

At one point I had an issue with trying to determine if “clipart” or “clip art” would produce more results so I punched both into Google’s regular search engine, below is the ad that came up.

In short, Google was advertising its image search feature as a means to find “Clipart Pictures”, even above Clipart.com.

In Google’s defense, it appears to have been a one-time thing. I haven’t been able to replicate it today. Also, it is almost certainly an automatically-generated ad based on keywords. Still, the fact that Google could connect Google Image Search with clip art, even though it is made up of images from all over the Web, most of which they do not have the copyright in, is disconcerting.

Update 12/16: Even though I haven’t been able to replicate it today, artist Meredith Dillman was able to get it to work last night following my original tweet on the topic. It appears that it wasn’t just me.

Granted, the ImageExchange extension, which I am testing aims to make Google Images (as well as general Web browsing) useful for stock photo searches, by itself Google Image Search has no means to license images, only point to sites that can (such as iStockPhoto, etc.)

To be clear, I don’t think that Google thinks of its image search product as a clip art gallery and it seems that this was a mistake and one that has already been fixed. But I would also argue that Google hasn’t done a great deal to dispel the idea either. Other than a vague “Image may be subject to copyright” warning in the frame above the clicked link, there isn’t much notice that the images aren’t for reuse without permission, especially for a layperson.

To make matters even worse, their CC-licensed search, which might actually be useful for finding clip art and stock photos, is buried at the bottom of their “Advanced Options“.

While it isn’t Google’s responsibility to make sure that everyone who uses its image search does so properly, it would be nice to see Google make it easier and more effective for what is almost certainly one of its most commons uses. This would help both artists and those seeking to use art at the same time.

That and not advertising it as a source for clip art would also be good idea…

One of the bigger headaches many visual artists have talked to me about is that many, often misguided, think that Google Image search is the same thing as a stock photo gallery and that anything they see or can find is right for the taking.

The problem, however, is that most of the photos in Google Image Search are copyright protected, legal for Google to use in the way it does under the Perfect 10 v Google ruling. Others, using the full-sized images in blog posts or site designs, often run afoul of the law.

However, today Google announced that it was integrating Creative Commons into its image search. The new feature promises to find images available for a variety of reuse scenarios, including commercial use, and actually make Google Image Search an effective way to find images for repurposing.

As promising as this sounds, the system isn’t quite perfect, though it is a great first step.

How it Works

If you’ve ever used the basic Creative Commons Search under Google, you are probably already familiar with how to use the one for image search. First, visit the Google Image Search home page and click the “Advanced Search” link.

There, toward the bottom of the options, you’ll find a dropdown box for filtering the results by license.

Once you select your desired license terms, just perform the search as usual and you should see the familiar results, but with a notice like this one above the thumbnails.

From there, as with regular image searches, you can click the thumbnails to view the Web pages they are on.

All in all, the search seems to work very well. The results seem to come mostly from environments with clear licensing and there’s a good variety of images on most of the queries that I tried. That being said though, there were still a few hiccups and concerns that I saw.

Some Limitations

The biggest limitation that I saw when using Google Image Search’s CC functionality was that it, understandably, can not detect licenses perfectly. In the original blog post, Google warns that users should verify the licenses independently, which is a very good idea.

A big reason is because Google is detecting image licenses on the page, not within the post. So, for example, if a blog post has its text in under a CC license and but uses an image as a fair use example, the image, at least in some occasions, winds up in the search. It can also happen to where Google places unneeded restrictions on an image if, for example, a public domain work is used on a CC-licensed blog.

The other, and less avoidable, problem is that many people are placing CC licenses on works they don’t own. A simple search for “Marilyn Monroe” found many commercial images that were, almost certainly, not licensed for commercial use for free. Other celebrities produce similar results.

The user has to make the final decision about whether they have the correct license for an image and, on that front, I urge caution. Remember, all the search does is point you to pages where an image you like and a CC license exist at the same time.

The other problem is that, despite Google’s impressive breadth on the Web, there seems to be a limited number of sources that it pulls from for these searches. Flickr is by far the most common source, which isn’t shocking as it is the largest repository of CC-licensed images, but it and Wikimedia seem to make up well over 90% of the results I checked. There were a few Blogspot blogs and at least one Typepad blog as well as a few other domains, but they were the extreme minority.

In the end, Google’s search did not seem significantly more robust than Zemanta’s image search, even if they have far more images on paper.

Still, it is a tremendous step forward for Google Image Search, even if it also a catchup to Google’s main search product, and means that, with just an extra step, those used to using Google Image Search to find photos for their blog posts and templates will now be able to do so legally and freely.

Bottom Line

In the end, this new feature is nice, but doesn’t blow me away. Though I’m a big supporter of Creative Commons, the implementation of this is understandably flawed and limited. Much of the benefit is already available via Flickr, Zemanta and Photo Dropper.

Furthermore, as with its main search product, Google works to bury its CC search functionality as deep as it can. Though it isn’t as well hidden as its counterpart, its unlikely anyone is going to stumble on it by accident.

This is definitely a step in the right direction for Google Image Search but there is still more that could and should be done with it. The question is whether or not Google has the dedication to this project to make it happen.

As part of the second approach, I make it a point to seek out and highlight services that assist and encourage the legal reuse of content. In fact, on Monday I wrote a column for the Blog Herald detailing five sites/services that can help bloggers get images for their posts.

However, when I found out later that day about a new service called Lizzer, which supposedly “is the easiest way to add links and embed content on the Web” I was initially very excited. But as I used the service and saw first hand how it works, my optimism turned to horror as I realized that Lizzer does not enable legitimate reuse of images or content, but rather, encourages both copyright infringement and image hotlinking.

Lizzer, sadly, shares very little in common with the services that I featured in my Blog Herald article and, instead, gives Webmasters, especially those who work with images, a lot of reason to worry.

All Rights Reserved

When I first logged in to my new Lizzer account, I instantly noticed an issue with the service. From the very first page load I could start searching for content. But, while most of the content was innocent, YouTube embeds and links to articles, one of the options was to use a Flickr search.

While that sounds great, without knowing what licensing terms I wanted, the service had no means of telling which images I could use legally. As such, I knew that there was a decent chance it could return images not appropriate for my sites.

Still, I performed a search for the term “plagiarism” and opened up the Flickr dropdown. Below is a snapshot of what I saw, notice the first image, which is highlighted.

It seemed odd to me because I’ve done a similar photo search using Photo Dropper, which filters for my CC licenses, and have never seen this image in the results. After clicking the “Preview” button followed by the “Source” link, I was able to see why. I was taken to this page, where, at the bottom, I saw the following copyright notice.

The image is “All Rights Reserved” and is marked so clearly in the Flickr API. This image should never have appeared in these search results as embedding it could land a Webmaster in trouble should the owner of the image decide to complain.

However, Lizzer wasn’t done. There were still other features to explore and the worst was yet to come.

Hotlinking from Google Image Search

Though the Flickr issue seemed serious to me, it only affected a small number of photographers and artists, namely those who post all rights reserved images to Flickr, and, since Yahoo! was paying the bandwidth tab, it seemed unlikely many would complain.

However, as I looked through the service’s options, I noticed that I had the ability to add Google image search. I immediately cringed because I know well that the vast majority of images displayed in Google Image Search are protected by copyright. Though Google itself is protected by fair use, it is unlikely that any full-sized use of the content by a user of Lizzer would have the same benefit.

I performed a similar search for “plagiarism” and opened up the Google Search list. The results I saw are below. Again, pay attention to the first image, the highlighted image of Bart Simpson.

I once again followed the image to the source and was then taken to this page. Once again, there was no mention of Creative Commons or any other license to use the image. From a legal standpoint, copying and pasting the whole image is potentially very dangerous (Note: I’m completely ignoring the use of the copyrighted character in this work and treating it as a wholly original image for the sake of originality).

However, the real surprise came when I used the fake blog post to try and embed the image. Clicking the link caused the image to appear in the post, exactly as planned, but then I looked at the code that the embed feature created. I saw the following:

The image, as you see below, is hotlinked straight from the original server, in this case, pandemiclabs.com.

In addition to raising a series of issues about copyright infringement and content misuse, we now also have to worry about abuse of resources. If I were to actually use this image code, not only would I be using the image without permission, but I would be using their bandwidth and server resources to do it.

Obviously, a lot of Webmasters are not going to be happy about this. Not only is Lizzer making their content available for use when no license was granted but, in many cases, is encouraging bloggers to hotlink images straight from the original server without permission. The worst part is that, someone using Lizzer, might think that the service is like Photo Dropper or Zemanta and that they have permission to use everything they see.

Clearly though, that is not the case.

Defending Lizzer

In Lizzer’s Defense, the service is currently in alpha and the version that you see today will likely change before the final release. However, a bookmarklet is already available for wide use and works well with all major blogging platforms, including WordPress, Blogger, MovableType, Tumblr, etc. This means that Lizzer is currently in use by at least some bloggers, many of whom are likely unaware of the copyright issues.

Fortunately though, most of Lizzer’s functionality is centered around finding links and videos to embed. Due to the way YouTube sets up their service, disabling embedding is easy for those that chose to do so, an issue that Flickr has not been able to address.

Out of all the search features they offer, only the Flickr search and the Google Image search seem to have any serious issues. However, they are pretty big problems that are both very easy to see and even easier to avoid. Flickr offers an API that allows you to look for CC-licensed work and there are many ways to search for licensed images.

I’m not going to be so bold as to say that finding licensed images is a “solved problem” but it is clear to me that Lizzer is a step backwards from well-established solutions.

However, what is most disconcerting to me is that I emailed Lizzer on Monday to raise these issues. At that time, the Google Images search feature was not working correctly and the hotlinking issues were not present. Unfortunately though, as of this writing, no one has written me back and the Google Image search function has come back online.

Though I grant that Lizzer is currently very buggy, I experienced multiple issues trying to take these screenshots, it is more than usable and that is bad news for content creators, especially those who work primarily with images.

Conclusions

The simple truth is that Lizzer, in its current form, is very dangerous both for Webmasters who publish content and its members that use Lizzer to find content to embed. With so many great tools already available for finding and embedding licensed content, Lizzer not only seems risky to use, but also dated.

Hopefully Lizzer can correct these issues and make itself a respectable content location tool. However, in its current form it not only misses many of the best sources of licensed material, but leads users to embed things that they do not have the right to.

There simply is no reason to use Lizzer right now, there are much better services available. Granted, Lizzer is in an alpha state, but it has already shown a great deal of disregard for Webmaster content and the legal safety of their service.

Unfortunately, in my experience, these aren’t so much “bugs” as they are flaws in the thinking of the developers. Thus, they are much harder to fix.

Hopefully though, they will be able to resolve these issues, for their sake and ours.