Jump to content

Caption/supertag/tag test


Recommended Posts

As promised I've tested 3 images in bhz and they appear as follows:

 

Caption only, page 18, position 99/120

 

Caption plus tag, ditto 80/120

 

Caption plus tag plus supertag, page 17, 38/120.

 

No reason for anything else to affect the position, they're unsold.

Now I'm off to pile in some more tags.

 

Edit: aargh. Just realised I didn't test caption + supertag only. Watch this space.

Link to comment
Share on other sites

As promised I've tested 3 images in bhz and they appear as follows:

 

Caption only, page 18, position 99/120

 

Caption plus tag, ditto 80/120

 

Caption plus tag plus supertag, page 17, 38/120.

 

No reason for anything else to affect the position, they're unsold.

Now I'm off to pile in some more tags.

 

Edit: aargh. Just realised I didn't test caption + supertag only. Watch this space.

Hi Spacecadet - can you be more specific about exactly what you did/have done please?

 

Thanks

 

Kuma

Link to comment
Share on other sites

 

As promised I've tested 3 images in bhz and they appear as follows:

 

Caption only, page 18, position 99/120

 

Caption plus tag, ditto 80/120

 

Caption plus tag plus supertag, page 17, 38/120.

 

No reason for anything else to affect the position, they're unsold.

Now I'm off to pile in some more tags.

 

Edit: aargh. Just realised I didn't test caption + supertag only. Watch this space.

Hi Spacecadet - can you be more specific about exactly what you did/have done please?

 

Thanks

 

Kuma

 

Put bhz in caption and tags as specified in 3 separate images and waited 24 hours.

joseph, no, i haven't tried that. I've always repeated important words from the caption as keywords anyway.

Link to comment
Share on other sites

 

 

As promised I've tested 3 images in bhz and they appear as follows:

 

Caption only, page 18, position 99/120

 

Caption plus tag, ditto 80/120

 

Caption plus tag plus supertag, page 17, 38/120.

 

No reason for anything else to affect the position, they're unsold.

Now I'm off to pile in some more tags.

 

Edit: aargh. Just realised I didn't test caption + supertag only. Watch this space.

Hi Spacecadet - can you be more specific about exactly what you did/have done please?

 

Thanks

 

Kuma

Put bhz in caption and tags as specified in 3 separate images and waited 24 hours.

joseph, no, i haven't tried that. I've always repeated important words from the caption as keywords anyway.

I think you would have to make the changes to the same image not to 3 different images.
Link to comment
Share on other sites

Did anyone ever do these tests before?  If you have nothing to compare it to, it's hard to know if anything has changed from previous.

 

Sometimes I really do think we are all overthinking this.

 

Jill

Yesterday the image with bhz as tag only was on page 25.

Link to comment
Share on other sites

Doing them all at once Mark, you haven't taken account of the fact that there is separation between each of your own images. The only way to do such tests is to do what I've done, which is by making one change at a time where no other images will be affected due to the same tags/words in captions.

 

Geoff.

When I used to bother with bhz , my two images were always within a couple of rows of each other.

Link to comment
Share on other sites

 

Did anyone ever do these tests before?  If you have nothing to compare it to, it's hard to know if anything has changed from previous.

 

Sometimes I really do think we are all overthinking this.

 

Jill

 

I have done many tests. I've mentioned some results in the forums already but have been criticised by the usual moaners for some of the facts I've stated, so now I keep details of my testing to myself, or share only with a select few.

 

I suggest everyone does their own testing. It's not hard, it just takes time and you need to only change one thing at a time, and only one similarly tagged image at a time. The results, I have found, are then consistent (although rather illogical).

 

Geoff.

 

 

I was thinking 6 months or a year ago.  This would be a test against the old search engine as well as the old IM.  And comparing using bhz  is hard as most people only have bhz once (like me) in the keywords.  A better test would be in keywords actually used. This would compare against others who may have the keywords twice, once, caption and keywords. More the real world of searching. BHZ just wouldn't supply enough true results.  Not enough variables.

 

Jill

Link to comment
Share on other sites

 

 

 

Did anyone ever do these tests before?  If you have nothing to compare it to, it's hard to know if anything has changed from previous.

 

Sometimes I really do think we are all overthinking this.

 

Jill

 

I have done many tests. I've mentioned some results in the forums already but have been criticised by the usual moaners for some of the facts I've stated, so now I keep details of my testing to myself, or share only with a select few.

 

I suggest everyone does their own testing. It's not hard, it just takes time and you need to only change one thing at a time, and only one similarly tagged image at a time. The results, I have found, are then consistent (although rather illogical).

 

Geoff.

 

 

I was thinking 6 months or a year ago.  This would be a test against the old search engine as well as the old IM.  And comparing using bhz  is hard as most people only have bhz once (like me) in the keywords.  A better test would be in keywords actually used. This would compare against others who may have the keywords twice, once, caption and keywords. More the real world of searching. BHZ just wouldn't supply enough true results.  Not enough variables.

 

Jill

 

 

I agree that using BHZ isn't a good way to do it. Neither is comparing with the old search engine and MI, as so much has changed now. The way to do it is to take one image, note it's position for a specific search, make just one change, then after a database update check it's position again. Also it's worth noting the position of a few other random images for the same search just to make sure nothing else major has happened. That way, you know it's only your image position that's changed. The next day you can make one more change, and see how things change again.

 

It doesn't honestly matter if you do use BHZ or not as you aren't comparing against others, and BHZ is completely useless now for checking comparative rank. The only way of doing these tests is to check your own images against the same image after making a change. Nothing else is consistent or predictable enough.

 

Geoff.

 

Presumably using an image which has keyword(s) you are going to search for that aren't in any other of your images, so that the (random?) dispersal algorithm that spreads results from a single pseudo doesn't affect results?

Link to comment
Share on other sites

 

 

 

Did anyone ever do these tests before?  If you have nothing to compare it to, it's hard to know if anything has changed from previous.

 

Sometimes I really do think we are all overthinking this.

 

Jill

 

I have done many tests. I've mentioned some results in the forums already but have been criticised by the usual moaners for some of the facts I've stated, so now I keep details of my testing to myself, or share only with a select few.

 

I suggest everyone does their own testing. It's not hard, it just takes time and you need to only change one thing at a time, and only one similarly tagged image at a time. The results, I have found, are then consistent (although rather illogical).

 

Geoff.

 

 

I was thinking 6 months or a year ago.  This would be a test against the old search engine as well as the old IM.  And comparing using bhz  is hard as most people only have bhz once (like me) in the keywords.  A better test would be in keywords actually used. This would compare against others who may have the keywords twice, once, caption and keywords. More the real world of searching. BHZ just wouldn't supply enough true results.  Not enough variables.

 

Jill

 

 

I agree that using BHZ isn't a good way to do it. Neither is comparing with the old search engine and MI, as so much has changed now. The way to do it is to take one image, note it's position for a specific search, make just one change, then after a database update check it's position again. Also it's worth noting the position of a few other random images for the same search just to make sure nothing else major has happened. That way, you know it's only your image position that's changed. The next day you can make one more change, and see how things change again.

 

It doesn't honestly matter if you do use BHZ or not as you aren't comparing against others, and BHZ is completely useless now for checking comparative rank. The only way of doing these tests is to check your own images against the same image after making a change. Nothing else is consistent or predictable enough.

 

Geoff.

 

 

 

I have an image on line 2, first image of a search with 32,000 images.  It only has 8 keywords, but does have the search in both the caption and the keywords. (Shows you how that green bar means squat). I will alter the caption and remove the search term.  I believe it will fall greatly as it does seem caption has more weight than before (which it should).  I will play with this image over the next week and see where it lands.  My final will be giving it all the keywords I can think of to see if it is better off with just a few.

 

Jill

Link to comment
Share on other sites

 

Did anyone ever do these tests before?  If you have nothing to compare it to, it's hard to know if anything has changed from previous.

 

Sometimes I really do think we are all overthinking this.

 

Jill

 

I have done many tests. I've mentioned some results in the forums already but have been criticised by the usual moaners for some of the facts I've stated, so now I keep details of my testing to myself, or share only with a select few.

 

I suggest everyone does their own testing. It's not hard, it just takes time and you need to only change one thing at a time, and only one similarly tagged image at a time. The results, I have found, are then consistent (although rather illogical).

 

Geoff.

 

 

I think this is rather important, and am presently doing a similar test with ten of my images which I will report back on after the weekend once I know the database has definitely updated

 

Kumar (the Doc one)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.