Thread: Ads dont connect
Hi , Clicking on the advertisements dont take you to the clients's site or more information page it simply redirects to the postgresql.org server. I dont have any ads but people paying for it would never like to have it. -- With Best Regards, Vishal Kashyap. Lead Software Developer, http://saihertz.com, http://vishalkashyap.tk
Just tested from here, and the redirects to the client sites work fine ... do you have cookies disabled in your browser?/ On Wed, 1 Dec 2004, Vishal Kashyap @ [SaiHertz] wrote: > Hi , > > Clicking on the advertisements dont take you to the clients's site or > more information page it simply redirects to the postgresql.org > server. > > I dont have any ads but people paying for it would never like to have it. > > -- > With Best Regards, > Vishal Kashyap. > Lead Software Developer, > http://saihertz.com, > http://vishalkashyap.tk > > ---------------------------(end of broadcast)--------------------------- > TIP 4: Don't 'kill -9' the postmaster > ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Just tested from here, and the redirects to the client sites work fine ... > do you have cookies disabled in your browser?/ I don't have cookies enabled, and I always go right back to www.postgresql.org, if I click on anything. Why not just encode the destination directly in the URL? e.g. http://ads.postgresql.org/redirect.pl?id=42 - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200412021007 -----BEGIN PGP SIGNATURE----- iD8DBQFBry/IvJuQZxSWSsgRAhVNAKDqLh8fYPlGHpQClI7yqhPXbFpUfwCfR/sI v+eQIA5TF5o9FjCfXn/AtlE= =L4Yd -----END PGP SIGNATURE-----
On Fri, 3 Dec 2004, Greg Sabino Mullane wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > >> Just tested from here, and the redirects to the client sites work fine ... >> do you have cookies disabled in your browser?/ > > I don't have cookies enabled, and I always go right back to www.postgresql.org, > if I click on anything. Why not just encode the destination directly in the > URL? e.g. http://ads.postgresql.org/redirect.pl?id=42 if we were only running on central site, that would be possible, but not possibl with static pages on several dozen mirrors ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >> I don't have cookies enabled, and I always go right back to www.postgresql.org, >> if I click on anything. Why not just encode the destination directly in the >> URL? e.g. http://ads.postgresql.org/redirect.pl?id=42 > if we were only running on central site, that would be possible, but not > possibl with static pages on several dozen mirrors ... But could the default action for the script (in the absence of a detectable cookie) be to go to a random ad? Seems that would make the most sense. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200412080954 -----BEGIN PGP SIGNATURE----- iD8DBQFBtxWXvJuQZxSWSsgRAr1rAKD4A3U8Wtk4jR4QFc0w/0fQy0XCxwCaA8r9 gwtah649jf4b9viz6EDR+A8= =p9vI -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Greg wrote in December 2004 about the ads on postgresql.org: === >>> I don't have cookies enabled, and I always go right back to www.postgresql.org, >>> if I click on anything. Why not just encode the destination directly in the >>> URL? e.g. http://ads.postgresql.org/redirect.pl?id=42 >> if we were only running on central site, that would be possible, but not >> possibl with static pages on several dozen mirrors ... > But could the default action for the script (in the absence of a detectable > cookie) be to go to a random ad? Seems that would make the most sense. === Any update on this? I actually saw an ad I was interested in last week, but clicking on it bounces me back to the main page. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200509061022 https://www.biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iEYEARECAAYFAkMdplUACgkQvJuQZxSWSsiEWwCdHAR8nLOJWg5B6cOr+oewa8oH oygAninz/fCML+vsc7bhPcmAljo+ftru =4fCY -----END PGP SIGNATURE-----
On Tue, 6 Sep 2005, greg@turnstep.com wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > Greg wrote in December 2004 about the ads on postgresql.org: > > === >>>> I don't have cookies enabled, and I always go right back to www.postgresql.org, >>>> if I click on anything. Why not just encode the destination directly in the >>>> URL? e.g. http://ads.postgresql.org/redirect.pl?id=42 > >>> if we were only running on central site, that would be possible, but not >>> possibl with static pages on several dozen mirrors ... > >> But could the default action for the script (in the absence of a detectable >> cookie) be to go to a random ad? Seems that would make the most sense. > > === > > Any update on this? I actually saw an ad I was interested in last week, > but clicking on it bounces me back to the main page. Any update on what? Without cookies enabled, the click through will go to http://www.postgresql.org ... having it to go 'some random site' wouldn't be particularly appropriate ... We could put up a page on http://www.postgresql.org that explains that cookies are required for the banner ads, and have it redirect to there though ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Any update on what? Without cookies enabled, the click through will go to > http://www.postgresql.org ... having it to go 'some random site' wouldn't > be particularly appropriate ... > > We could put up a page on http://www.postgresql.org that explains that > cookies are required for the banner ads, and have it redirect to there > though ... Any update as in "any fix". I suspect more and more people are blocking cookies except for known sites, and ones with "ads" in the name are particularly likely to be shunned. Perhaps, if no cookies are detected, ads.postgresql.org opens to a simple page that states that since cookies are not enabled, it cannot tell which ad you clicked on, but presents a clickable list of all sponsors/ads for the site, so the user can find the one in question they were interested in. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200509061114 https://www.biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iEYEARECAAYFAkMdsnUACgkQvJuQZxSWSshiTwCg1aZRmrIKEYf4eNLLDcux4hi4 6o0AoOEzYaJrpUPYPZZqdq43S8r3XJUg =Rwfj -----END PGP SIGNATURE-----
Would it be possible for you to use session ids thus avoiding the whole issue of cookies? On Tuesday 06 September 2005 11:15, Greg Sabino Mullane wrote: > > Any update on what? Without cookies enabled, the click through will go > > to http://www.postgresql.org ... having it to go 'some random site' > > wouldn't be particularly appropriate ... > > > > We could put up a page on http://www.postgresql.org that explains that > > cookies are required for the banner ads, and have it redirect to there > > though ... > > Any update as in "any fix". I suspect more and more people are blocking > cookies except for known sites, and ones with "ads" in the name are > particularly likely to be shunned. > > Perhaps, if no cookies are detected, ads.postgresql.org opens to a simple > page that states that since cookies are not enabled, it cannot tell which > ad you clicked on, but presents a clickable list of all sponsors/ads for > the site, so the user can find the one in question they were interested in.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Would it be possible for you to use session ids thus avoiding > the whole issue of cookies? The problem is that we have static mirrors. The ads needs to rotate dynamically though, so (I presume) ads.postgresql.org sends an image and a cookie. When someone clicks on the link, it uses the cookie to know which image they clicked on. So, sessions would not help. There are some more advanced tricks that could work (e.g. AJAX) but probably overkill. Although ads.postgresql.org could in theory also track the IPs that they send the image too and try and match that way if the cookie was not found. That should fix it 99% of the time, I would think. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200509061432 https://www.biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iEYEARECAAYFAkMd4VsACgkQvJuQZxSWSshylgCdFLPcZSvPne2W4uBk5yDIVd1T TPcAoOcMXKdxqDcedMDbukQvQ4eTlPu/ =+amj -----END PGP SIGNATURE-----
On Tue, 6 Sep 2005, Greg Sabino Mullane wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > >> Would it be possible for you to use session ids thus avoiding >> the whole issue of cookies? > > The problem is that we have static mirrors. The ads needs to rotate > dynamically though, so (I presume) ads.postgresql.org sends an image and > a cookie. When someone clicks on the link, it uses the cookie to know > which image they clicked on. So, sessions would not help. There are some > more advanced tricks that could work (e.g. AJAX) but probably overkill. > Although ads.postgresql.org could in theory also track the IPs that they > send the image too and try and match that way if the cookie was not > found. That should fix it 99% of the time, I would think. Except then it has to save the IPs for every hit to every page *in case* someone happens to click through ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >> Although ads.postgresql.org could in theory also track the IPs that they >> send the image too and try and match that way if the cookie was not >> found. That should fix it 99% of the time, I would think. > Except then it has to save the IPs for every hit to every page *in case* > someone happens to click through ... But aren't they already saved in the apache logs? It would be expensive to parse them, true, but this would be a rare case for those who are not accepting cookies. Or at the very least, what if the non-mirror site (e.g. www.postgresql.org) used a slightly different system that did not rely on the cookie/image scheme? Just trying to drum up revenue for our advertisers, after all.. :) - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200510032206 http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iD8DBQFDQeQYvJuQZxSWSsgRAgFiAKDoxrvef7NLzkjzuTfmTvoUjLDCUACfUsWd RuOeOeytmGt+nS6eDqaDRW0= =ZrTt -----END PGP SIGNATURE-----
> >> Although ads.postgresql.org could in theory also track the > IPs that > >> they send the image too and try and match that way if the > cookie was > >> not found. That should fix it 99% of the time, I would think. > > > Except then it has to save the IPs for every hit to every page *in > > case* someone happens to click through ... > > But aren't they already saved in the apache logs? It would be > expensive to parse them, true, but this would be a rare case > for those who are not accepting cookies. Or at the very > least, what if the non-mirror site (e.g. www.postgresql.org) > used a slightly different system that did not rely on the > cookie/image scheme? Can't talk about the ads.pg server, but al the static servers have logging turned off to handle the performance requirements. And what do you mean with the "non-mirror site www.postgresql.org"? www.postgresql.org is the mirrored site ;-) //Magnus
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Can't talk about the ads.pg server, but al the static servers have > logging turned off to handle the performance requirements. Right, but all we need is the logs from ads.pg, which hopefully can tell us which ad was just served to the requesting IP via this: <img src="http://ads.postgresql.org/display_image.php?lvl=sponsor"> > And what do you mean with the "non-mirror site www.postgresql.org"? > www.postgresql.org is the mirrored site ;-) Oh yes, I forgot we have some round-robin DNS action going on now (which is a good thing!) so nevermind all that. :) On second though, the first half probably won't work either, if the script that sends out the image cannot log IPs. I'll fall back to my original plan B then: simply call up a "list of all sponsors" page. Right now it opens a new window, and puts the main postgresql page on it, no matter which page you came from. We should do better than that. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200510042110 http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iD8DBQFDQykTvJuQZxSWSsgRAjpxAKCMqjyjLK+9Uld0U7EBzx9P7PxS+QCgxM1H QqQl7lYip4rxR+2QQauPLhc= =kMaK -----END PGP SIGNATURE-----
On Wed, 5 Oct 2005, Greg Sabino Mullane wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > >> Can't talk about the ads.pg server, but al the static servers have >> logging turned off to handle the performance requirements. > > Right, but all we need is the logs from ads.pg, which hopefully can > tell us which ad was just served to the requesting IP via this: > > <img src="http://ads.postgresql.org/display_image.php?lvl=sponsor"> how? when display_image runs, it sets a cookie ... all that the logs show is "http://ads.postgresql.org/display_image.php?lvl=sponsor", there is nothing that says which ad was displayed for that IP ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
Greg, let me know if recent changes to the software (basically, replaced the inhouse with phpAdsNew) fixes these problems for you ... On Wed, 5 Oct 2005, Greg Sabino Mullane wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > >> Can't talk about the ads.pg server, but al the static servers have >> logging turned off to handle the performance requirements. > > Right, but all we need is the logs from ads.pg, which hopefully can > tell us which ad was just served to the requesting IP via this: > > <img src="http://ads.postgresql.org/display_image.php?lvl=sponsor"> > >> And what do you mean with the "non-mirror site www.postgresql.org"? >> www.postgresql.org is the mirrored site ;-) > > Oh yes, I forgot we have some round-robin DNS action going on now > (which is a good thing!) so nevermind all that. :) > > On second though, the first half probably won't work either, if the script > that sends out the image cannot log IPs. I'll fall back to my original > plan B then: simply call up a "list of all sponsors" page. Right now > it opens a new window, and puts the main postgresql page on it, no matter > which page you came from. We should do better than that. > > - -- > Greg Sabino Mullane greg@turnstep.com > PGP Key: 0x14964AC8 200510042110 > http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 > -----BEGIN PGP SIGNATURE----- > > iD8DBQFDQykTvJuQZxSWSsgRAjpxAKCMqjyjLK+9Uld0U7EBzx9P7PxS+QCgxM1H > QqQl7lYip4rxR+2QQauPLhc= > =kMaK > -----END PGP SIGNATURE----- > > > > ---------------------------(end of broadcast)--------------------------- > TIP 9: In versions below 8.0, the planner will ignore your desire to > choose an index scan if your joining column's datatypes do not > match > ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > Greg, let me know if recent changes to the software (basically, replaced > the inhouse with phpAdsNew) fixes these problems for you ... Bingo. Seems to work great now, even when I reject cookies. (which are now coming from a numeric IP, not a domain name). Thanks for following up on this. - -- Greg Sabino Mullane greg@turnstep.com PGP Key: 0x14964AC8 200510231651 http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iD8DBQFDW/fbvJuQZxSWSsgRAr5tAJwIHd0nzv1UkwSTvXmCJolwyfO8gwCg0uTH e2g4+H1qqBFTu2amPiSiBJw= =z5nS -----END PGP SIGNATURE-----