TrackBack as an Internet Standard

I resisted the urge to comment on the process of submitting trackback as an internet standard the other day because I didn’t want to just completely flame the idea. I knew there would be folks who would want to push pingback instead and I’m doubt that would be much better.

The biggest problem with both trackback and pingback is spam. Plain and simple, spam. Neither of them do enough to even put a dent in slowing down spam, let alone preventing it. Because of this I think that go through the effort to make a standard is pretty much a waste of time. Who wants to enshrine one more way for people to send spam as a standard? Want to do something helpful, come up with a completely new trackback/pingback/pick-a-new-name mechanism that is designed to actively slow down/prevent spam.

For background checkout trackback and pingback sites and specs. There is also a pingback vs trackback paper (via Matt).

UPDATE Thu 23 Feb 2006 @ 2:45pm: I should add that I like the general concept of trackback. I like the site to site (S2S?) interaction that it provides. More than once I’ve discovered a pointer to my site with additional feedback that didn’t show up in any of the blog search engines for several days (and sometimes never showed up at all). Keeping it decentralized means I don’t have to worry if Technorati isn’t responding to searches right now. I hope that Adam is right and that they do find a way to deal with the huge potential for abuse via spam.

5 thoughts on “TrackBack as an Internet Standard”

  1. One of the things we’re trying to do with the new spec is to solve the spam problem. Trackback spam exists because it’s an easy way to insert data onto your page anonymously. So the solution to that is to provide a way for trackback clients and servers to provide authentication and identity.

  2. Is trackback spam REALLY that much of a problem though? I mean, I have had my fair share of trackbacks that weren’t really relevant but it pales in comparison to those automated programs that put a comment on your blog.

    But, hey if you could come up with a way to do trackbaks without any spam I’d be all for it.

  3. Trackback is quite vulnerable to spam, but I haven’t seen much Pingback spam around. I believe the reason for this is that the Pingback specification recommends that Pingback servers check the page being submitted for links to the article in question. This raises the bar for spammers – not by a large amount, but enough that it’s just easier to spam Trackback instead.

    Spam was in fact the explicit reason for this spec text, as you can see from the very first version of the spec:

    It was discussed more on the mailing list here:

    A host whitelist with manual approval of the rest seems like it would solve the problem once and for all. You could even build it semi-automatically by hooking in blogrolls, articles you link to, etc.

    In hindsight, there’s a lot to be improved with both Trackback and Pingback, so neither make a good candidate for a standard. This recent turn of events seems centred around “standardising Trackback”, but in essense, it’s really building a successor to both Trackback and Pingback.

  4. Hmmmm… seems something in my last comment caused a couple of paragraphs to be lost. If this proves to be the case, if you keep an eye out on my blog, you will find more details regarding what the above links refer to.

Leave a Reply

Your email address will not be published. Required fields are marked *