Ancestry Sharing Links not working [closed]

Question

I’m horrified to see all of my Ancestry sharing links going to a spammy site. I’ve cleared cache and do so regularly. No help there. I use the WikiTree Sourcer. Safari on iPad. Help!

Comments

Hi Jana. This is a known problem affecting all Ancestry templates. For example see

https://www.wikitree.com/g2g/1569003/problem-with-ancestry-com-links

---

Jim, just a note - this was affecting all Ancestry links, not just templates. I have posted an update.

---

Are you sure that is correct, Steven? As I pointed out here, direct links to Ancestry were working normally.

The fix introduced seems to be a change in the templates to use linksynergy instead of Partnerize. Under the circumstances, moving away from the latter seems a good idea.

---

Jim, I am quite sure. For clarification, I do not mean all internet links to Ancestry - but links to Ancestry from WikiTree. Of course modifying the URL will work, and was one of the workarounds.

WikiTree has a partnership with Ancestry. We get a small credit for clicks over their to site, purchases made, etc. This is one of the ways that WikiTree is able to remain free, and that is where the URL prefixes come in to play.

---

I don't understand what mechanism would have caused this, Steven. Why would a link to Ancestry from WikiTree have been affected if it was not in a template and didn't have the prefix? Not everyone uses the template. If they didn't, wouldn't those direct links to Ancestry have continued to work? Are you saying something more than the prefix affects how Ancestry treats incoming requests? For example, might HTTP referer headers do it?

---

@Jim, the mechanism and global understanding is not important (and going off topic). That is what the Tech Team is here for. :)

---

Steven, it certainly isn't under WikiTree's purview or control, but something that is important is that profiles and FSPs for which members have used the internet Wayback machine as an archival method--of whom I'm one of many--probably need to be aware that all prior archival snapshots will now contain invalid URLs wherever links to Ancestry.com were used.

And as you note, not just instances where the templates were used: all links to Ancestry were previously made active with a prefaced code used by the Partnerize tracking system. Links that did not use the MediaWiki external templates are still intact and copyable as the full URL to Ancestry, but templated links are not. They won't be usable any longer unless people know how to search Ancestry using the numeric database and record IDs. This is somewhat more complicated to reconstruct when templates like the one for Ancestry shared images are used.

I've checked about a dozen Waybacked profiles and FSPs from 2022 through last March 14, and they show all links to Ancestry.com rendered as a URL with a prefix using the "prf.hn" domain, a name registered and used by Partnerize for tracking purposes. These are now invalid URLs. All links to Ancestry, presumably as of the fix of the recent problem, now point to "linksynergy.com," registered and used by Rakuten Advertising.

It's certainly no concern of ours if WikiTree chooses to change affiliate aggregator services, but I don't believe a global change that impacts all Waybacked archived data is off topic. In fact, I believe WikiTree should offer a general communication about the change.

GEDCOM exports really aren't a very effective means of backing up our own work since everything in the biography box, including all source citations, come through as plain text in notes fields, and there the use of the recommended external-link templates don't resolve to a link at all: they stay as plain text. To create a "live," formatted backup of a profile or FSP, there aren't many options.

---

Edison, are you saying that you were archiving links using the links generated by WikiTree, and not the physical ancestry URLs?

Edit: Read over your message again, and I think you are saying you were archiving WikiTree pages, and those pages contained the previous referral URL's and would not longer work.

In that case, that is correct. However, I am not sure the reason for using the archive to store information, when change histories are readily available, and the core has been updated to handle those referral URL's.

---

Steve, I'm talking about archiving an entire profile or FSP. As one recent example: https://web.archive.org/web/20230314171252/https://www.wikitree.com/wiki/Space:Walker-15949_Research.

If you scroll down to the entry identified as source item #4 you see use of a templated Ancestry link to an indexed record, and a direct, non-templated link to an Ancestry sharable image. In both, a preface is appended to the rendered link: "ancestry.prf.hn/click/camref:1011l4xx5/type:cpc," which is used by Parternize to identify the referer. The links themselves are now invalid. If you know how, you can still extract the original URL, but that may escape many WikiTree members.

Too, were the Partnerize account still active, I would assume WikiTree would still get any referral activity from those links. The Wayback Machine will only summarize to a count of 10,000 for a particular domain name, but at the very least over 10,000 archive snapshots have been made of WikiTree pages.

The new Rakuten system generates links that are likely to be even more challenging because they replace "special" characters in the URL encoding. The result of that cited indexed record at Ancestry is now, on the WT live page: https://click.linksynergy.com/deeplink?id=Xib7NfnK11s&mid=50138&murl=https%3A%2F%2Fsearch.ancestry.com%2Fcgi-bin%2Fsse.dll%3Findiv%3D1%26db%3D5063%26h%3D6829.

---

Crosspost; sorry.

However, I am not sure the reason for using the archive to store information, when change histories are readily available, and the core has been updated to handle those referral URL's.

 I registered my first domain name and created my first websites in 1993. I have long, first-hand experience with the ephemerality of websites. Some WT members put in literally thousands of hours working on profiles. You'd no more choose not to backup that work than you would choose never to backup your PC's or Mac's hard drive(s).

Link-rot is the same reason that Wikipedia strongly encourages contributors to use Internet Archive links for all source citations rather than a live link. A) The reference can be viewed in the same format as it was when the citation was written, and, B) there's a reasonable chance that the Waybacked URL will outlive the live the URL.

Personally, I add a Waybacked URL of my WikiTree profiles to all--well, most...many--of the profiles on my tree at Ancestry in the public "Comments" area.