Jump to content

I sense a re-rank


Recommended Posts

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

Link to comment
Share on other sites

I've moved up a page to near the top of page 4, but i have BHZ as a supertag, not in the caption.  When I changed it from a tag to a supertag, it jumped from page 25 to page 5.

 

Hard to tell with the results now if Alamy is going by caption, as most of the images in BHZ are probably supertags as opposed to being in the caption.  As people change them, placement will keep moving around.

 

Jill

Link to comment
Share on other sites

 

 

... 

 

I've seen recently that pseudos are linked together rather than being independent as they used to appear to be. That doesn't mean that ranking doesn't come into things at all, but it does mean that something regarding rank and image order has changed. I have a feeling (with possible evidence but nothing definite) that there might be a new way of ranking that uses an average CTR/sales of all our pseudos.

 

Of course we only supply the images and do all the keywording, so why would we be told anything about the way the search engine works?  :rolleyes:

 

Geoff.

 

Interesting. Member services told me only a few days ago that pseudos are still completely independent of each other. It would be interesting to see their response to evidence to the contrary. 

Link to comment
Share on other sites

 

 

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

 

 

That's interesting that you see it that way. So you made no other changes at all, to either that image or any others with BHZ, in any of your pseudos (which now appear to be linked)? So you just changed the BHZ tag to a supertag, and the placement increased, without using other filters and with "relevant" selected, and not changing the caption?

 

My own tests show that supertags are still giving lower placement than tags, in 4 different tests, and I've just checked that one of those is still in the same place as the other day, and it is.

 

Geoff.

 

 

Yes Geoff.  No other changes to anything, just the BHZ tag changed to a supertag.  I can't think of any other explanation than that supertags are working.

 

Pearl

Link to comment
Share on other sites

 

 

Interesting. Member services told me only a few days ago that pseudos are still completely independent of each other. It would be interesting to see their response to evidence to the contrary. 

 

 

They did? That's very odd. I can't remember how I proved it and I'd have to look back at other threads and test results, but my conclusion without any doubt was that they are not independent.

 

I'll look into that and may post again soon, if I find anything.

 

[Edit] This is a link to where I proved it. Post 11 in that thread shows the test results.....

 

http://discussion.alamy.com/index.php?/topic/7237-pseudonyms-per-account-appear-to-be-connected/?hl=%2Bindependent+%2Bpseudos

 

I'm sorry Alamy if it seems as though I am disagreeing with you again, but those tests prove to me that pseudos are connected. Anyone else can take it as they choose but I'm simply reporting results of my tests.

[/edit]

 

Geoff.

 

Thanks, Geoff. I missed that as I was away on a trip at the time. I'll put it on my to-do list to have a look at what people did there and see if I can replicate it with my own images. I have had my suspicions for a number of years, because some of my little-used pseudos tended to bunch up together - and they weren't median ranked. Hmm. 

Link to comment
Share on other sites

 

 

 

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

 

 

That's interesting that you see it that way. So you made no other changes at all, to either that image or any others with BHZ, in any of your pseudos (which now appear to be linked)? So you just changed the BHZ tag to a supertag, and the placement increased, without using other filters and with "relevant" selected, and not changing the caption?

 

My own tests show that supertags are still giving lower placement than tags, in 4 different tests, and I've just checked that one of those is still in the same place as the other day, and it is.

 

Geoff.

 

 

Yes Geoff.  No other changes to anything, just the BHZ tag changed to a supertag.  I can't think of any other explanation than that supertags are working.

 

Pearl

 

Your finding is indeed mysterious Pearl.

Over the next few days, could you toggle the BHZ between tag and supertag (no duplicates) to make sure the effect is repeatable? Plus, as you know, no BHZ in the title.

I'm going to recheck things too, but I don't see supertags moving my images up.

Link to comment
Share on other sites

 

 

 

 

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

 

 

That's interesting that you see it that way. So you made no other changes at all, to either that image or any others with BHZ, in any of your pseudos (which now appear to be linked)? So you just changed the BHZ tag to a supertag, and the placement increased, without using other filters and with "relevant" selected, and not changing the caption?

 

My own tests show that supertags are still giving lower placement than tags, in 4 different tests, and I've just checked that one of those is still in the same place as the other day, and it is.

 

Geoff.

 

 

Yes Geoff.  No other changes to anything, just the BHZ tag changed to a supertag.  I can't think of any other explanation than that supertags are working.

 

Pearl

 

Your finding is indeed mysterious Pearl.

Over the next few days, could you toggle the BHZ between tag and supertag (no duplicates) to make sure the effect is repeatable? Plus, as you know, no BHZ in the title.

I'm going to recheck things too, but I don't see supertags moving my images up.

 

 

In my test image I found BHZ supertag beat BHZ tag (but not by much), but BHZ in caption beats both (all tests on same image).

 

But, I'm also finding if I swap my BHZ supertag to a different image, the results change dramatically, suggesting other factors are at play.

Link to comment
Share on other sites

 

 

 

 

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

 

 

That's interesting that you see it that way. So you made no other changes at all, to either that image or any others with BHZ, in any of your pseudos (which now appear to be linked)? So you just changed the BHZ tag to a supertag, and the placement increased, without using other filters and with "relevant" selected, and not changing the caption?

 

My own tests show that supertags are still giving lower placement than tags, in 4 different tests, and I've just checked that one of those is still in the same place as the other day, and it is.

 

Geoff.

 

 

Yes Geoff.  No other changes to anything, just the BHZ tag changed to a supertag.  I can't think of any other explanation than that supertags are working.

 

Pearl

 

Your finding is indeed mysterious Pearl.

Over the next few days, could you toggle the BHZ between tag and supertag (no duplicates) to make sure the effect is repeatable? Plus, as you know, no BHZ in the title.

I'm going to recheck things too, but I don't see supertags moving my images up.

 

 

I have converted the supertag back to a tag so we'll see where it ends up after the update tomorrow. No other changes made.

 

Pearl

Link to comment
Share on other sites

 

 

 

 

 

Well my BHZ image for my best pseudo has gone from p 1 (pre 05/12) to p8 (post 05/12 to p 23 today!!!  And yet all the normal real life test searches show no change for that pseudo.  I don't know what is going on.

 

Pearl

 

On closer inspection BHZ was the very last tag and isn't in the caption at all (as it shouldn't be) so I changed it to a supertag.  The site has now updated and my image is on p7 leaving me to conclude that supertags are working.  I still don't know why my best selling pseudo is on p7 whilst my second best is on p2 but hey ho!!

 

Pearl

 

 

That's interesting that you see it that way. So you made no other changes at all, to either that image or any others with BHZ, in any of your pseudos (which now appear to be linked)? So you just changed the BHZ tag to a supertag, and the placement increased, without using other filters and with "relevant" selected, and not changing the caption?

 

My own tests show that supertags are still giving lower placement than tags, in 4 different tests, and I've just checked that one of those is still in the same place as the other day, and it is.

 

Geoff.

 

 

Yes Geoff.  No other changes to anything, just the BHZ tag changed to a supertag.  I can't think of any other explanation than that supertags are working.

 

Pearl

 

Your finding is indeed mysterious Pearl.

Over the next few days, could you toggle the BHZ between tag and supertag (no duplicates) to make sure the effect is repeatable? Plus, as you know, no BHZ in the title.

I'm going to recheck things too, but I don't see supertags moving my images up.

 

 

I have converted the supertag back to a tag so we'll see where it ends up after the update tomorrow. No other changes made.

 

Pearl

 

 

It's ended up on p24, near the top.  Seems to confirm that supertags are working in my opinion.

 

Pearl

Link to comment
Share on other sites

I almost feel like some contributors are treated different than others, there are those that have images move up when they add a supertag and there are those who get images demoted when they add a supertag.  Would it make sense to take let's say one of Pearl's images and apply some random tags and then take the let's say Geoff's image add the same tags and play with supertags and see what happens???

Link to comment
Share on other sites

I almost feel like some contributors are treated different than others, there are those that have images move up when they add a supertag and there are those who get images demoted when they add a supertag.  Would it make sense to take let's say one of Pearl's images and apply some random tags and then take the let's say Geoff's image add the same tags and play with supertags and see what happens???

You mean, Image Manager is back, and this time it's personal?

Link to comment
Share on other sites

 

I almost feel like some contributors are treated different than others, there are those that have images move up when they add a supertag and there are those who get images demoted when they add a supertag.  Would it make sense to take let's say one of Pearl's images and apply some random tags and then take the let's say Geoff's image add the same tags and play with supertags and see what happens???

You mean, Image Manager is back, and this time it's personal?

 

I'm sorry but I don't understand.  What do you mean by Image Manager is back? 

Link to comment
Share on other sites

I've never been a 'game player' or had the inclination to delve into the wizardry of the search engine, but focused this year on building up the numbers on my port. I've entered the details on my images in the priority of supertag>tag>caption>optional in the hope that logic would eventually prevail. I've randomly used real term searches and BHZ just out of interest rather than as a source to take further action, and having variously bombed and rocketed my positioning now is almost exactly as it was before the new IM. I'll live with that and keep beavering away

Link to comment
Share on other sites

Just to add some more irrationality to this thing, I tried the tag to supertag experiment on my three pseudonyms and the one on page 9 dropped to page 21, page 27 dropped to page 28, and page 30 moved up to page 28. I'm taking an Easter break!!

 

Jim. ;)

Link to comment
Share on other sites

It's a mystery and a mess.  I don't think logic comes into the equation at all.

 

Pearl

 

There could be quite a complex equation involving numerous terms e.g.

 

A= How many tags match search term

B = How many tags don's match the search term

C = How many supertags match search term

D = How many supertags don't match the search term

E = How many words in cation match search term

F = Some sort of keyword order/proximity factor from Supertags 

G = Some sort of keyword order/proximity factor from Tags

H = Some sort of order/proximity factor from Caption

I = Contributors ranking

J = Age of image

K = Previous sales or zooms of image

 

And maybe a few more?

 

All the terms A,B, C etc. are combined using individual weighting factors (e.g. to make supertags more weighty than tag) to calculate the final rating of the image.

 

The trouble maybe that equation or the calculations behind some of the terms are messed up, giving strange results. The behaviour may actually be very predictable, but from the outside (without knowing the equation), it looks like rubbish.

 

Given the lack of fixes and response from Alamy (and the advert for a Search Engine software engineer), I wonder if the software engineer that wrote the equation/code has have left? 

Link to comment
Share on other sites

I re-did my test: same old, same old. My good pseudo moves up from page 21 to page 6 when I change the supertag to a tag.

My lower ranked pseudos show the same trend, but not so severe: page 23 to 22 for the supertag to tag switch.

Link to comment
Share on other sites

Nothing random that I can see about using BHZ as a search term.

For me it has some benefits:

 

3000 results make finding an image manageable while still providing a good field.

Changes to the BHZ collection are few.  The static collection avoids problems of someone dumping a lot of new images in.

I can easily spot my image among the others.  Searching for my CN Tower pictures among 3000 other CN tower pictures is like finding a needle in a haystack.

 

I do others searches too.  BHZ is consistent with my other findings, and I use it as my benchmark.

Link to comment
Share on other sites

 

It's a mystery and a mess.  I don't think logic comes into the equation at all.

 

Pearl

 

There could be quite a complex equation involving numerous terms e.g.

 

A= How many tags match search term

B = How many tags don's match the search term

C = How many supertags match search term

D = How many supertags don't match the search term

E = How many words in cation match search term

F = Some sort of keyword order/proximity factor from Supertags 

G = Some sort of keyword order/proximity factor from Tags

H = Some sort of order/proximity factor from Caption

I = Contributors ranking

J = Age of image

K = Previous sales or zooms of image

 

And maybe a few more?

 

All the terms A,B, C etc. are combined using individual weighting factors (e.g. to make supertags more weighty than tag) to calculate the final rating of the image.

 

The trouble maybe that equation or the calculations behind some of the terms are messed up, giving strange results. The behaviour may actually be very predictable, but from the outside (without knowing the equation), it looks like rubbish.

 

Given the lack of fixes and response from Alamy (and the advert for a Search Engine software engineer), I wonder if the software engineer that wrote the equation/code has have left? 

 

 

Click Through Rate - CTR.

 

Views/Sales or Sales/Views used to be really important.

Now with pseudos with no sales overtaking pseudos with lots of sales, that seems out of the window.

 

The fact that Creative and Relevant now mostly give the same result is a very bad sign for everybody who was not in Creative.

Of course that greatly benefits contributors that have come/are coming from a certain kind of agencies.

 

wim

Link to comment
Share on other sites

 

 

I almost feel like some contributors are treated different than others, there are those that have images move up when they add a supertag and there are those who get images demoted when they add a supertag.  Would it make sense to take let's say one of Pearl's images and apply some random tags and then take the let's say Geoff's image add the same tags and play with supertags and see what happens???

You mean, Image Manager is back, and this time it's personal?

 

I'm sorry but I don't understand.  What do you mean by Image Manager is back? 

 

It's a joke, I'm sorry if it only works in English.

Link to comment
Share on other sites

 

 

 

I almost feel like some contributors are treated different than others, there are those that have images move up when they add a supertag and there are those who get images demoted when they add a supertag.  Would it make sense to take let's say one of Pearl's images and apply some random tags and then take the let's say Geoff's image add the same tags and play with supertags and see what happens???

You mean, Image Manager is back, and this time it's personal?

 

I'm sorry but I don't understand.  What do you mean by Image Manager is back? 

 

It's a joke, I'm sorry if it only works in English.

 

 

Nope, but one has to think twice - I had  :)

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.