| |
This page was last updated on November 23, 1998, and reflects the
information available at that time. While it's possible the circumstances
have changed since the publication of this page, the techniques described
are still valid.
What is AltaVista's Photo Finder?
On October 13, 1998, the same day the Senate passed the Digital
Millennium Act to protect content creators on the Internet, Alta Vista
launched a new "service" called Photo
Finder (click the "images" tab)
Upon entering a word or phrase, this search engine will return
dozens if not hundreds of images of that item found on the World
Wide Web. The pages it returns feature thumbnails of all the images,
and links to their source. Originally, the pages contained no copyright
information or disclaimer whatsoever, either general, or specific
to any image. Due to the uproar their launch created, Alta Vista
has since added the following disclaimer:
"Anyone who would like to use any of the artwork, photographs,
or images found through or linked to by AV Photo Finder should contact
the owner of the images and ask for the appropriate permission or
authorization."
It should be noted that this disclaimer currently appears only
on the opening page, where the search word is first entered. On
the returned pages that actually contain the copyrighted images
of others, there is no notice of any kind.
It's a shame AltaVista did not see any reason to follow the above
advice themselves before they placed the copyrighted images of others
on their servers. They claim it is illegal to reproduce their
copyrighted logo and other intellectual property without their permission,
but they apparently see no need to respect the intellectual property
of others.
Despite the fact that searches have returned the trademarked logo
of CNN, copyrighted images of Disney, Inc., and Star Trek
graphics (all properties that their owners have zealously defended
on the Internet), AltaVista has stated they have no intention of
suspending this "service," regardless of the freely offered
opinions of lawyers specializing in intellectual property rights
that their use of 3rd party images in this way represents a breach
of copyright law.
Most web authors make use of the ALT attribute of the <IMG>
tag to provide a short accurate text description of the graphic's
content, for the benefit of those surfing the web with image loading
turned off. AltaVista uses the robot "vscooter" (as opposed
to their text robot, scooter) to index this ALT text, and the image
it describes. It would appear some weight is also given to text nearby
the image on the page. I do not know if the actual file name matters.
This robot was developed in conjunction with the company Virage, who
list some info in their press
release.
They currently claim to have over ten million images in their
database.
What can I do to prevent the indexing of my site's graphics?
If you are a content creator who is concerned about protecting
your copyrighted works, I'd advise complaining directly to AltaVista
via e-mail.
Of course, I did this the day after their launch, and have yet to
hear a word from them. I'm hoping this is because they've been inundated
with a pile of complaints. Please add to it.
AltaVista does provide a couple of ways to opt out of their indexing.
In my opinion, the onus should not be on those whose rights are
infringed to take action, it should be on AltaVista to obey the
law. Despite that fact, we must use whatever methods we can. They
offer two:
- Robots Exclusion .txt
This solution will only work for those who have their own domain.
If you have a robots.txt file in the root of your domain, all
robots that come to your site to index it will read this file
first, and follow its directives. If you're not familiar with
this, you should probably read the Robots
Exclusion Protocol. If you already have a robots.txt file,
add one blank line after your current declarations, then add the
following two lines, followed by a carriage return (hit "Enter"):
User-agent: vscooter
Disallow: /
If you do not have a robots.txt file, open a plain text editor,
like Notepad, create the above two lines, end the file with
a carriage return, save it as "robots.txt", and load
it into the root directory of your site (the same one with your
"home" page). This will cause "vscooter"
to go away without indexing your images, but will allow AltaVista's
normal text robot, "scooter", to index your site as
it always has.
- <META> tags
Robots can be also be excluded on a page by page basis using a
<META> tag. Unlike the robots.txt file, this solution will
work whether or not you have your own domain. For example, adding
the <META> tag below within the <HEAD> of your document
will prevent a robot from adding a page to its index and links
on the page will not be followed:
<META NAME="robots" CONTENT="noindex, nofollow">
You can find more details on <META> tag robot exclusion
on this
page, which points out that only a few robots support this
tag, and the above example will affect all of them. However,
AltaVista's robots do support this <META> exclusion,
and their image robot supports two additional options not included
in the standard:
- NOIMAGEINDEX prevents that page from appearing in the AltaVista
Image Search index but the page can still show up in the main
AltaVista index.
- NOIMAGECLICK prevents the use of links directly to the images.
Instead, there will only be a link to the web page.
My advice would be to place the following <META> tag
in the <HEAD> of any document linked from an outside source,
as well as the index page ("home" page) of your site,
and the index page of each of your sub-directories:
<META NAME="robots" CONTENT="noimageindex">
Note, the content is not case sensitive. This should prevent
their image robot from indexing your site, although there is
no reason you couldn't put it on each page you have.
What if my site has already been indexed?
You can certainly still apply the above solutions, and hope their
robot re-indexes your site in the near future. But in the meantime,
your images have already been saved, resized, and recompressed by
AltaVista for display on search returns. They claim that they will
remove your images from the database only if you place an appropriate
robots.txt in your site, and/or proper <META> exclusion tags,
and then submit your site's URL at http://image.altavista.com/remove.html.
This page is not very clear to me, but I believe this submission triggers
a new visit by their spider at some later date, who will then find
your new exclusions. One would assume they would then remove existing
images from that site that were previously indexed. But these are
just my beliefs and assumptions, as AltaVista is far from clear in
the few words they offer on this subject.
You can still try to prevent the link from the AltaVista thumbnail
to the full sized graphic on your server, but even then,
your options are limited. If you do not have your own domain, I'm
unaware of any option other than changing the file names of all
indexed images so that AltaVista's link to them returns a "404
- File Not Found" error, but this is obviously a pain if you
have a lot of images.
If you do have a domain, ask your web host about the possibility
of using a .htaccess file with a <Limit GET> section to deny
referrals from specific domains. Depending on the software your
server runs, this may or may not be effective, so you must contact
your web host for details. Another option is to purchase a commercial
CGI script that limits referrals from specified domains, if your
account supports the use of custom CGI scripts.
These are the options of which I'm aware, but if you have any
other ideas, or comments about this page, please e-mail
reid@photodude.com.
This page was placed online on October 17, 1998, and that same day
I implemented the robots.txt and <META> exclusions on my site.
At that time approximately 50 of my graphics had been indexed. On
November 2, I found that all of my images had been removed from their
database. I'd like to think it was due to the measures detailed above,
but I have been unable to find any trace of anything calling itself
"vscooter" making a single server request in the previous
three weeks.
During the second and third weeks of November, "vscooter"
made two visits to my site. Each time it requested my robots.txt
file, read the exclusion, and went away, as advertised.
AV Photo Finder has made a few slight modifications, as all thumbnail
images now link to the web page on which they appear, rather than
the free standing image as once was the case. But there's a text
link to the image alone that offers the view the thumbnail once
did. They've also added more prominent links on their home page
to previously existing pages that explain how you can have your
site added to their index, and the steps you must take if you'd
like to be excluded.
However, a spokesperson stated (see the first article linked below),
"The thumbnails are not the issue (they are an acceptable 'citation'),"
so they clearly plan to continue displaying copyrighted works, and
will do so until the content creator takes active steps to stop
them from doing so. Interestingly, if you click on the word copyright
at the bottom of the home page of AV Photo Finder, it takes you
to a page that says, "No logo, graphic, sound or image from
this website may be copied or retransmitted unless expressly permitted
by Compaq." They seem to understand how the law protects them,
so it's hard to for me to see why they can't extend that understanding
to others.
AltaVista Photo Finder Has Artists Concerned
http://www.searchenginewatch.com/sereport/9811-photofinder.html
Copyright Battle Looms For AltaVista
http://7am.com/cgi-bin/twires.cgi?1000_t9810140.htm
Compaq Accused Of Copyright Infringement
http://www.newsbytes.com/pubNews/120384.html
Photographer Claims AltaVista Breaches His Copyright & Trademark
http://7am.com/cgi-bin/twires.cgi?1000_t9810290.htm
AltaVista Disclaimer Not Enough For Copyright - Kelly
http://www.newsbytes.com/pubNews/121019.html
Lpce : bloquer vscooter (in French)
http://www.lpce.com/lpce/conseil/
The URL of this
page is http://www.photodude.com/av.htm
|
|