Another <Google> loser heard from

Because it lacked the all-important feature of being about me, me, me, I’d forgotten that I wasn’t the only one getting unpersoned by Google: Matt Mower doesn’t exist either.

The situation isn’t quite as bad as he thinks: in fact, he shows up in [“Matt Mower”], it’s just that when your weblog is result 151 for your own name as a phrase it feels like you’ve been disappeared.

Hmm. Moved from one URL to another (and worse yet, the old one doesn’t redirect to the new), content republished by at least one aggregator which doesn’t decline to be indexed by search engines. Sounds pretty familiar to me.

I’d say the one difference is that Matt cares and I don’t, but if (MS employee) Duncan Mackenzie is going to have trouble finding me again because I’m not indexed in MSN Search unless you already know my URL… well, now you’re starting to get into “actually does matter to me” territory.


Comment by Mark #
2006-01-09 16:27:59

And then, on the extreme end, is Scoble (it’s amazing how many sentences start that way, but never mind that), who seems to be determined to dominate the top 10 results for his name by moving to yet-another-domain-that-isn’t-his every year. (NEVER (under any circumstances) publish a weblog to a domain that you don’t control., but never mind that either.)

To which we apparently need to add ”and NEVER (under any circumstances) publish a full-content feed.” Or reorganize your site by employing fundamental web standards. I wonder whether you would have been better off using META refreshes on all your old pages…

Comment by Phil Ringnalda #
2006-01-09 17:07:49

The nice thing about having changed my prefs to get 100 results per page at Google (so it’s quicker to get down to where we losers hang out) is that you get to see even more fun: I love the result for The Economist’s profile of Scoble down at result 22: title ”Skip” description ”Skip.”

I guess some people don’t worry about how search engines see them.

Comment by Kafkaesquí #
2006-01-09 23:22:11

So let’s see…I can remember your name, but I can’t recall your url? Yeah, sounds like a technical problem.

Comment by Phil Ringnalda #
2006-01-09 23:56:20

Heh. No, that one’s BKC. This one was ”I thought I remembered your name, but the URL isn’t working: was it .org, or am I misspelling it, or did you move? Eh, search engine ought to tell me.” And, indirectly, even MSN Search will if you’re close with the spelling, since you’ll get plenty of links to things that Phil Ringalda did or said (heh, I’d forgotten that I’d once done some good in the world, turning someone on to Kris Delmhorst (in a post that’s now a poster-post for why I’ll never use a non-HTML markup)). I’m not quite sure what my takeaway from that should be: that I don’t need to be indexed, only linked? During the blog-clog arguments, my thesis tended to be that if you’re unsure what result to return for a search, returning a weblog is a fair bet, gambling that it will link to the proper result, even if it takes an extra click or two. Maybe I should adopt that for my own weblog, figuring that if you don’t get something linking to me in your results, then I wasn’t good enough to inspire good enough linking.

Dunno, but having not ever had quite enough ego to stalk misspellings of my name, I’m off to see what people said about eight out of nine letters over the years that I missed seeing at the time.

Comment by Sam Ruby #
2006-01-10 05:02:32

I get that all the time.

Comment by Phil Ringnalda #
2006-01-10 09:28:18

I think of your URL as being ”inter rats no autocomplete Sam Ruby.” Now that I’m not Feeling Lucky with addressbar searches, I may have to remember that it’s not ley.

Comment by Phil Ringnalda #
2006-01-10 00:01:06

Note to Google: your results for [how many cows in Texas] now completely blow. Yahoo eats your lunch on that ((in)famous) query.

Comment by Kafkaesquí #
2006-01-10 00:12:23
Comment by Phil Ringnalda #
2006-01-10 01:18:01

And has, I fear, utterly Jeevish results: first a crap result, then me with a URL that’s been redirecting for a couple of months, then the utterly worthless ”RSS feed presented as though it’s something you should click,” then a result that’s far better than what I bombed, assuming it will remain up-to-date, since it neatly encapsulates a lot of data about the variation in numbers both between wintered cows and cow-calf pairs and year-to-year as markets and weather and subsidies and whatnot push the numbers around. That’s my impression of Jeeves in a nutshell: some hidden gems, yes, but they are hidden in a giant steaming heap of manure.

Comment by Kafkaesquí #
2006-01-10 08:46:08

What do you want from a butler? You know he’s always spitting in the soup.

Comment by Aristotle Pagaltzis #
2006-01-10 16:08:09

In case you haven’t seen it, you really need to read the Interview with the Search Engine.

Comment by Phil Ringnalda #
2006-01-12 17:44:16

It’s tempting to say that MSN Search’s response to [how many cows in Texas] is ”fuller plumper lips and Tofu” (in case they churn: number one result is a militant vegan site, number ten is a collagen injection ”information” (and text ad) site), but how can you not love a search engine that returns Uncyclopedia’s ’You have two cows’ page for that query?

Comment by Doug #
2006-01-11 12:35:08

Please excuse me for co-opting your comment form, but I was unable to find another method to contact you. No contact form, no e-mail address, no ”about” page. Google was no help (surprised?). I suppose I could have picked a random username to attach to your domain name, but then I would have looked like a spammer if I didn’t guess just right.

Anyway, I was just going to suggest you look at this bug report I filed for the SimplePie feed parser. No idea if the problem is on your end or theirs, but I’m sure the makers of this fine product would appreciate your input. I’m not associated, I just use their code on my site.

Comment by Phil Ringnalda #
2006-01-11 15:25:59

Though actually [email phil ringnalda] plus some interpretation would work pretty well, since that gets you the feedvalidator mailing list at Sourceforge, where addresses are ”hidden” by only including the first two letters of the domain, which, when they are ”gm” are quite likely to be Gmail.

Other than the obvious, big fat bug in SimplePie, I don’t have any ideas: my feed is UTF-8, and it says so in as many ways as I know to say it: in the content-type header, in the XML decl, and by virtue of being served as application/*+xml even if you ignore both other places that it already said it was UTF-8.

Comment by Phil Ringnalda #
2006-01-11 16:46:55

Ah, misuse of htmlentities(): SimplePie tries to turn CDATA sections into escaped character data before the XML parser gets ahold of the feed, so it can do some liberal fixup of potential parsing errors (a really bad idea, you should always parse first, and only try error recovery when there’s an error), but it uses htmlentities() when it should only use htmlspecialchars(), and compounds the problem by not passing a charset to the function, so that UTF-8 is treated as ISO-8859-1, and the first couple of bytes of a multibyte UTF-8 character are turned into an &acirc; entity, leaving the remaining bytes of the character as just invalid garbage.

Comment by Mark Kenny #
2006-01-23 10:23:10

Stick your name in every title tag on your site. That’ll bump you up from 151 :D

Comment by Phil Ringnalda #
2006-01-23 10:27:15

Or run you down off the bottom end for over-optimization, one or the other.


Sorry, the comment form is closed at this time.