| 12:35 am on Jan 23, 2001 (gmt 0)|
Only those se's that also spider images rayjam. That would mainly be Fast/lycos and Altavista. Even then, it has to be a juicy filename for them to pull it.
| 10:26 am on Jan 23, 2001 (gmt 0)|
Juicy meaning a heavily searched on term? My image names are the same as my alt tags etc. Does this give that page a uniqueness amongst many other pages of similar file size within a site. The question for me at this time, is what makes one page different enough from another for the SE to consider it unique. For instance two pages with simialr content and file size but with differences as mentioned above will the image reference be enough. And if I do change the file size by actually adding a real image into the page, by what percentage do the pages need to differ to make them appear worthy of indexing. My theory is that se's like diversity. they want to see varying content on a theme but to what extent are they willing to go to determine this. Has any research been conducted into determining uniqueness?
| 11:25 am on Jan 23, 2001 (gmt 0)|
I don't know if that it is enough. The working theory that I've put together from reading here is that the page needs at least 10-15%% difference between another page to be considered unique. That extends from the text to the html itself.
Google is excellent at detecting like pages. I think they are indexing the html in one step and comparing it to other pages from your site to detect and strip templates.
Yes, juicy meaning a quality keyword. Determining what a quality keyword is for a image filename is very tricky. Think in terms of what people will search for when they do image searches. Also, if you are trying to get into the image search db's, go look at their format for images and see which one they like - it pays to do your homework on that aspect before submitting.
| 11:49 am on Jan 23, 2001 (gmt 0)|
What is the advantage to the website of an image-only hit? In other words, why would someone want to be in the images database for a search engine?
| 11:57 am on Jan 23, 2001 (gmt 0)|
For me the image is there just to add variety. It is interesting that you say the html is stripped but most pages on a normal site would have distinct similarities. So do you think it comes down to something as simple as line breaks?
| 12:11 pm on Jan 23, 2001 (gmt 0)|
I guess If you were running a fan site or maybe images from Hubble! that sortofthing IMO
| 12:58 pm on Jan 23, 2001 (gmt 0)|
Yes, rayjam, I guess I can see a value if the pictures are both highly desirable and also characteristic of what your site offers. That way you stand a chance of an image picker wanting to take a full look at the site.
Sometimes I think the most common use of an image search is plagiarism. For instance, I've written before about a needlepoint site one of my clients runs -- and the copyright problems that the internet has thrown into the mix.
Artists and designers who were barely making enough to keep going before the internet now get ripped off on a regular basis.
So, I guess my attention has been more on avoiding getting images indexed and I couldn't imagine how anyone would want to get in the database.
| 1:46 pm on Jan 23, 2001 (gmt 0)|
Depends on how the search engine lists you Tedster. The most desireable being those that offer a link to the page in addition to the images. I have pages that pull regularly and routinely from image searches. Those are good pages to do self advertising on. Those lessons now go into my both my design phase and naming phase when creating graphics.