Your free mouse is waiting

Here in Australia, Atomic: Maximum Power Computing is a foolishly-named but actually-rather-good magazine for which I write.

I do occasional features, and two regular columns. One of those is the "I/O" question-and-answer letters column, which I reprint six months later on Dan's Data.

Allow me to share with you a secret about that letters column.

Hardly anybody sends in any bloody letters.

This certainly isn't the result of an actual lack of people with questions to ask. My interminable series of Dan's Data letters columns is evidence enough of that.

People just don't send questions in any more.

Seriously - there are, like, two letters a month to io@atomicmpc.com.au, and on average about 1.2 of those letters fall into the HELO WUT IZ DA BEST GFX CRD PLZ??!? category that gets them swiftly despatched to my "Atomic lousy letters" folder.

About 0.6 of the remaining 0.8 letters will describe, often at great and depressing length, some computer problem about which all I can think to say is "yep, your PC is really badly screwed up, all right".

This falls somewhat short of the high standard my readers have come to expect.

You might wonder, at this point, how it is that the I/O column keeps running in Atomic.

It keeps running because

1: People like reading it, and

2: I get tons of questions sent to me @ dansdata.com.

So I just skim the most "Atomic-y" of those letters off the top - yes if it's someone asking about using mercury as CPU coolant, no if it's someone asking about cheesecake - and use them to bulk out the I/O column.

I still e-mail my reply to the questioner as well as include it in the Atomic column, of course. So it's not as if a question from some guy in California falls silently into a magazine in Australia which he will never read.

Since the only misrepresentation involved in this is that it gives the impression that people who sent questions to one e-mail address actually sent them to another, I don't feel very guilty about it. Many of the people whose letters are repurposed probably do not actually read the magazine in which their letters and my answers will then be printed, but that's no big deal. Certainly not by the standards of letters columns in general.

The dirty secret of the letters pages of local - and not-so-local - newspapers the world over is that it's perfectly normal for them, like me, to not get enough interesting letters to fill the space.

Since they, unlike me, have no other source of material, they then just make stuff up.

Completely.

Sometimes there's something juicier, like one of the newspaper bosses forcing his partisan letter in under an assumed name or something.

But usually they just have a few beers and then make up something entertaining to fill the empty space.

I have never been driven to doing anything like that for the Atomic column. But one problem does remain.

The best I/O letter each month wins a prize. For ages now, it's been a shiny new Logitech G5 mouse.

But only "genuine" letters to the I/O address, from people who help to pay for the whole enterprise by buying the darn magazine, are eligible for the prize.

And, worse yet, I've a personal rule that if you've already won the prize once, you can't win it again for a very carefully calibrated period of time most elegantly describable as "however long it takes for me to forget your name".

Since so very few decent letters are coming from actual readers of the magazine, the I/O prize is at this moment very easy to win. Just send a good question to io@atomicmpc.com.au, and you'll be in the running.

"Aha", I now hear many devious people saying who live on the other side of the planet from Australia and have no interest at all in buying Atomic. "I shall send a letter, and win the prize, and nobody will know I'm not really a reader!"

Well, no, nobody will. But you won't know you've won, either, because (a) I won't tell you (I suggest a winner to the Atomic editorial team, but it's not up to me to actually definitely decide who gets the prize, and my Dan's Data column reprints won't tell you either), and (b) the Atomic prize-sending department tends to be a bit on the slack side. They often only actually send out the prize when the lucky winner reads the mag, lets out a happy cry and then sends an e-mail with his or her address in it.

(This means that some poor sucker has probably failed to buy that month's magazine, or something, and missed out entirely.)

If you go to all of the trouble of setting up a devious scheme involving an Australian friend who subscribes to the magazine and will tell you if you win and, I don't know, on-ship your mouse to you so nobody gets suspicious about some guy in Anchorage who isn't on the subscriber list winning the prize, then I say you deserve your plunder as much as any Australian.

Actual Antipodeans who actually read Atomic, though, should bear in mind that asking me things at io@atomicmpc.com.au rather than dan@dansdata.com may be unexpectedly profitable.

The amazing $125 sitemap

Reply-To: peter.kramer@mplw.com
From: "Peter Kramer" <peter.kramer@mplw.com@gt;
To: dan@dansdata.com
Subject: Sitemap File missing - http://www.dansdata.com
Date: Sat, 15 Sep 2007 03:24:25 +0200

As I was on http://www.dansdata.com this morning, I was unable to locate a "Google Sitemap file" on your website.

I am not referring to a regular "site map" for people to visit online, but rather to a script called "Google Sitemap file" which helps Google to read and index your website overall content. I advise you to visit us online where we explain clearly what is a "Google Sitemap file" and what you need to do to get one: http://www.sitemapfile.net

A Sitemap file is a "script/code" placed in the root directory of your website which captures all the crucial information about your website, thus facilitating the crawling and indexing process for Google. We can set up your Google Sitemap file for $125 should you need help to do so.

If Google takes the time to publish a page titled "What is a Sitemap file and why do I need one?", it is obvious that every responsible online marketer should take action accordingly.
Read what Google says about Sitemap file and why you need one: http://www.google.com/support/webmasters/bin/answer.py?answer=40318

Regards,

Peter Kramer, Ph.D.
peter.kramer@mplw.com

GLOBAL VIBRATION INC.
1250 Connecticut Ave N.W. Suite 200
Washington, DC 20036 USA
TEL: 1 (202)-787-3989 - FAX: 1 (202)-318-4779
http://www.mplw.com:
Multilingual Search Engine Promotion Services since 1999.

Even if I didn't regard being described as an "online marketer" as a deep personal insult, it would still be my considered opinion that this service is a rip-off.

As other people have observed (after getting this same offer for sites that apparently already have a sitemap...), making a sitemap is likely to be a semi-automated process that takes about 15 minutes.

That makes $US125 for making one a pretty good hourly rate. Even before you notice that they're apparently offering four different and separate kinds of $125 sitemap - Google, MSN, Yahoo and "General". Only $375 if you order all four!

I can see nothing in the sitemap format that actually requires those files to be different for different search services. And since April 2007, Google, Yahoo and MSN have supported automatic "discovery" of sitemap files via a simple robots.txt entry. So you don't even have to manually submit your sitemap URL to get it noticed. Not that the submission process was ever difficult enough to justify a separate fee.

And there's more.

Global Vibration (insert joke here...) aren't even selling you an automatic-updating sitemap service.

As far as I can tell after reading their mildly illiterate FAQ, they'll just make one lousy XML file and then, I guess, charge you another $125 if you want more addresses added to it and aren't smart enough to twig to the fact that you can edit the thing yourself.

And, furthermore, dansdata.com has no need for a sitemap file, as a cursory examination of the site reveals.

The basic purpose of a sitemap is to make it easier for search engine spiders to find dynamically created pages that can't easily be located by just "clicking on links".

Web forums, for instance, are difficult to effectively spider. If you've for some reason decided to use a Flash interface for your site navigation, that'll also stymie spiders.

Google spiders all of the pages on Dan's Data with no trouble whatsoever, though. Google also discovers new pages on my site within hours, if not minutes. I used to manually submit new pages to Google just to make sure, but they show up in searches just as quickly if I don't.

Dan's Data also has zillions of incoming links from other sites. Even if I deleted my huge full index page and all of my intra-site links, most if not all of my pages would still be regularly spidered.

And I don't have any "dynamic" pages at all. Dansdata.com is a good old fashioned flat-file site.

That makes it painful if I want to change an element on every page - I have to re-upload the entire site, which at the moment means about 36Mb of HTML - but it reduces the load on my server. And it also makes the site trivially easy to spider, since every URL is simple and static and there's no half-baked Content Management System shuffling stuff around.

Dansdata.com has been around since 1998, and has a PageRank of 6. Oddly enough, despite the fact that Global Vibration claim to have been providing "Multilingual Search Engine Promotion Services since 1999" (http://www.mseo.com/ and http://www.globalvibration.com/ have apparently only existed since 2001...), their own site currently has a PageRank of... zero!

I would also like to propose a General Rule of Credibility: Anybody who puts "Ph.D." after their name whe they're trying to get you to buy something is less likely to be on the level than someone with no letters after their name.

If I were uncharitable, I might wonder where Peter Kramer got his doctorate. I might also wonder what discipline it was in.

Dare you enter... the Nostalgia Pit?

Herewith, a site with a reasonably complete archive of scans of old Australian Commodore and Amiga Review (back to the Commodore Review days, up to the Amiga Review days) and Professional Amiga User magazines.

I can't remember when I started writing for ACAR. January '92 might have been my first issue (sound sampler review, page 16), but I suspect I did a piece or two before that. After a while, I was the Assistant Editor, and stayed in that job until the publishing company went broke.

(Entertainingly, I was listening to this, one of the few MODs lurking in my large MP3 collection, when I turned up my review of ProTracker in the March '93 ACAR.)

My new fake address

Nerds the world over often find themselves filling out online address forms with nonsense data.

Perhaps you're making a free account at some newspaper that doesn't have any entries in Bugmenot. Perhaps you're trying to download drivers for your old scanner. Perhaps you're registering some product you actually did just buy, but not on the understanding that you'd agree to sign up for junk mail.

A large subset of the sites dumb enough to request personal information from people who'd rather not give it are also dumb enough that they only accept US addresses, with five-digit postal codes. And, heck, if you're entering nonsense data anyway, you might as well leave everything set to defaults, which almost always means a US address.

If you actually live in the USA, you can just declare yourself to live at 123 Foo Street in the suburb and zip code where you actually reside. If you don't have such information at your fingertips, though, you have to come up with something.

The result of all this is that the raw data pouring into marketing databases the world over contains a disproportionately large number of people who say they live in Beverly Hills, California. Because even if you never watched the soap opera, you probably still know its name. And that name, complete with handy zip code, constitutes the only US postcode that most people around the world, from New Zealand to Iceland, can think of.

You can just type random digits into the postcode box instead, of course. But then you hit those fabulous sites that check to see if the postcode is valid, or even whether it's valid and matches the string from the "town" box.

So 90210 it is, for nonsense-enterers the world over.

But I, for one, am moving my database-polluting 97-year-old Ecuadorian-born Jewish grandmother alter ego out of boring old Beverly Hills, and into Compton. Compton's zip code is easy to remember: 90222. And "Compton" is faster to type, on those tiresome occasions when you have to.

And, of course, Compton is cooler.

Alert: eMate now actually useful

eMate screen

I am typing this on my little green computer.

Well, actually I'm adding these words to the top of a half-written block of text that I composed on my PC.

Which, yes, means that I'm able to move text from the PC to the eMate.

And if you're reading this, it means I'm also able to move it back. Which is a nice bonus.

(Here's my post about transferring data to and from the eMate. It took me a while to edit all of the cursing out of it. Precis: Use the serial cable, Luke. Don't bother with anything else.)

The eMate keyboard's only about 90% of normal size, but I can still type on it much faster than the poor little computer can squeeze small-font-size words onto its 480-by-320 bitmapped screen.

As an example, I'm hammering out this paragraph at the best speed I can manage, and when I stop typing and look at the screen NOW, the eMate has only actually gotten around to printing the "when I stop" part of this sentence to the screen. It took another seven whole seconds before it made it to the "NOW".

When you're starting a new document, the eMate Notes application is much faster. Then, it nearly keeps up with my roughly 80 word per minute typing (slowed a bit by the smaller keyboard). Once there's a significant amount of text in the document, though, things slow down.

(The dotted handwriting recognition guide lines don't make any difference to screen drawing speed, but they're unnecessary if you're only going to use the keyboard. This stationery file lets you create new notes without lines.)

Fortunately, the eMate keyboard buffer is big enough that the slow update speed isn't a problem. It's not as if the thing just sits there and beeps at you when you've typed 16 characters ahead of what it's gotten around to displaying.

I suppose a lightning typer could freak the eMate out if they really tried. But the small keyboard means this isn't really a computer for that kind of user anyway. As long as you pause for thought now and then, and don't often decide to delete the last word ("How many times did I just press backspace? Dammit, now I have to wait and see.") you ought to be fine.

This slowness also means that the eMate isn't the greatest place to do your editing, let alone HTML markup. 153600 pixels sounds like quite a lot - why, an old greyscale Palm has only 25600! - but it really only gives you about 100 characters by 23 lines of small-font text, plus menu stuff above and below.

That sounds perfectly decent, by ancient-word-processor standards. It's not as if people didn't get lots of work done on 80-by-24 text mode programs like WordStar or AppleWorks.

But text mode was fast, and the eMate's bitmapped graphics are slow.

Amigas had no text mode and some lightning-fast text editors, but that was because of their coprocessors. I think the principal strategy used by the early Macintoshes to deal with the same problem was (a) only having 1.14 times as many pixels as an eMate, and (b) encouraging patience in their users. MacWrite was a great success, but it bogged down just like the eMate when a document was more than trivially long.

(Perhaps I should add an overclocking switch to my eMate - though I'd need at least three toggle switches sticking out of the casing for the eMate to match my old Amiga 500. You didn't have a proper A500 if the RF shield inside hadn't been removed so many times that all of the little tabs had broken off, resulting in metallic boinging noises while you typed. Maximum resolution? Well, that depended on how much overscan you could cram onto your monitor, didn't it?)

Absence of usable editing features can, of course, itself be a feature. This sort of thing is at the very core of the Write Without Interference philosophy, in which the elimination of distractions like editing or looking up hyperlinks allows you to get to the core of your thoughts both faster and better. If you know you're going to have to edit what you just wrote, put in a couple of asterisks or something and keep on going with your brain still afire with the magnificent creativity that only you can, uh, create.

See? See what I mean? I'm back on my PC, now, and I just had to take time out to find that funny link, which broke my train of thought and left me writing this paragraph instead of filling in the perfectly judged words that I intended to come after "magnificent creativity", above.

And now I've finished that paragraph, saying something else because I forgot what I intended to say, and look what a hash I made of it.

Playing the triangle

In the olden days, you used to get spam from people running link farms (groups of many-paged sites full of useless "directory" pages with hundreds of links to each other), telling you that they'd added a link to your site from one of their dreadful pages and unless you linked back, they'd DELETE THE LINK OMG.

Back in the mists of time this may actually have worked - if, by "worked", you mean "artificially inflated the value of these sites so their worthless pollution floated up into people's search results and they got some ad-viewing traffic".

It doesn't work any more, though. Anybody who joins in these scams by linking back now connects themselves to the "bad neighbourhood" mojo that's applied to all known link farms by the search engines. This achieves the exact opposite of the ranking-boosting traffic bonanza promised by the spammers.

So, nowadays, the spammers have moved on to trapezoidal triangular linking.

You used to get spam from someone who runs www.creativeusesforsnot.com and is apparently convinced that some random page on your site where the word "snot" appears is a perfect match for his very important directory of links, but not if you don't link back. Now, you get spam from the same guy, except he's telling you that if you link to creativeusesforsnot.com, he'll link to you from elephant-toenail-trimmers.com.

Because the search engines don't know his two sites are connected, all of these links look like perfectly kosher "one way" propositions, and everybody wins. Eh? Buddy? Buddy buddy buddy?

A few of these e-mails a day have been leaking through to me. Here's a typically moronic example:

From: "Shamim"
To:
Subject: Link Exchange Request
Date: Tue, 26 Jun 2007 10:30:44 +0530

Dear Webmaster

I handle online marketing for my client's site http://www.petwellbeing.com/dog-kidney-disease-p91.cfm

As you all know about the Google's new algorithym and the improtance of oneway linking. I am also looking for triangular linking ( New Virsion of Oneway linking ) to increase the linkpopularity of my site as well the ranking in major search engines.

I will also add your site on to my directory http://www.rainforests.com.au/ within24 hours of your positive reply.

please add my site at least page rank(2) page.

I request you to do have a look on to my website and add it on your website and reply me with your site's details.

Here is my linking details :-

URL : http://www.petwellbeing.com/dog-kidney-disease-p91.cfm

Title :Canine Kidney Disease Treatments

Description :Effective natural pet medication for canine kidney disease treatments to reduce irritation and pain.

Link will be added at: http://www.rainforests.com.au/rainforests/Rainforest_Birds.htm
: http://www.petidtags.org/linkmachine/resources/resources.html

You can also paste the code given below :

Canine Kidney Disease Treatments Effective natural pet medication for canine kidney disease treatments to reduce irritation and pain.

Your link will be added on my site within 24 hours. So if you are interested for link exchange with my site please let me know and we can do a better work for our sites.

Thanks and Regards

Shamim

shamim124@gmail.com

There's no connection between the three domains mentioned in this e-mail as far as whois records go (although I was amused to note that the registrant of rainforests.com.au put what appears to be his real Australian Business Number in his registration!), but Google would have to be pretty stupid to be unable to connect them. Google are all about seeing patterns in links, and triangular linking creates repetitive patterns. A links to X, then X links to C. A links to Y, then Y links to C. Et cetera. This sort of thing seldom happens for valid reasons.

Oh, and these guys keep sending out these brain-hurtingly stupid e-mails to zillions of recipients, who then post them to the Web and Usenet, where the world can see the scheme exposed. Sometimes the spammers cut out the middleman here, by spamming mailing-list addresses and getting their messages archived online automatically.

So even if these dorks don't accidentally spam people (or spam-trap addresses...) that actually lead directly to the search engines, they can still be discovered very quickly.

Are they, though?

Well, the root pages for the sites mentioned in the above spam are all still sitting pretty at Google PageRank 4, which is quite good. The sub-pages actually mentioned in the spam are down around PR3, but it's normal for sub-pages on valid sites to have a slightly lower rank than the root page.

When I looked at some sites promoted in previous spams that've been visible online for months - here, for instance - I found that their root pages still had OK PageRanks - well, PR3 at least. More interestingly, the sub-pages that the spam tells people to link inward to are also still doing OK, but the sub-pages that link outward in return are now down on PR0 with the rest of the hoi polloi.

So it does seem that Google is somewhat wise to this scam. If you do what a triangular-link spammer asks you to do, your site's PageRank mojo will indeed contribute to the PageRank of the page you link to, but as soon as Google notices the pattern, the spammer's return-link page will plummet to PR0 and so his link will do you no good.

This isn't an optimal solution, since it means the triangular-link scam will still work just fine for the spammer, if people do what he says. It'd be better if triangular link beneficiaries were being classed as "bad neighbourhoods" just like the old-style link farms.

But if it becomes common knowledge that these schemes are as fishy as they sound, at least fewer people will fall for them.

Blog payola update!

A reader, on seeing my post from the other day about the wretched hive of scum and villainy (and blatant payola) that is PayPerPost, wrote to observe that Text Link Ads appear, from his experience, to be eager to move in on the same area. Only more so.

Text Link Ads' usual stock in trade is straightforward, non-annoying ads of the type described in their name. But my correspondent, let's call him Harry, says they contacted him to try something a little bolder.

This new scheme apparently pays per undisclosed sponsored link. Include a link from a blog post to whatever dodgy dealer likes the idea of buying these kinds of ads, absolutely do not reveal to anybody in any way that you're getting paid for it... and get paid for it. Genius!

Harry says he's now seeing... unusual... links popping up on a number of blogs that also run Text Link Ads. When a blog that's usually about WordPress plugins and such (plus, tellingly, tips on how to make money with your blog...) suddenly runs a post about this k00l new site y0 that totally letz you watch Spider-Man 3 for free d00d, I believe it is not excessively cynical to suspect that something is up.

(I note that the above-linked blog runs, as well as Text Link Ads, those god-awful Kontera ads. As I write this Kontera are upholding their golden reputation for ad relevance by linking the word "movie" from the fishy Spider-Man post to an ad that says "Find Leading Lease Resources for rent movie online Here." Excelsior!)

Oh noes! I am rejected!

A reader noticed my lengthy and profitable career with ReviewMe (i.e. this one review), and suggested I check out PayPerPost instead.

Both are services that allow Web writers to sell their services to people who want someone to write about something. One big difference, though, is that ReviewMe take half of everything you're paid, while PayPerPost take a much smaller cut.

So that was interesting.

As soon as I looked at PayPerPost, though, I saw the other big difference. PayPerPost lets people list "opportunities" that have the condition that whatever you write has to be complimentary.

ReviewMe specifically forbid that requirement. And that is, of course, their downfall.

It turns out that the kinds of companies that have to pay for blog attention are, by and large, not deserving of positive attention. And so they vastly prefer more... ethically flexible... services like PayPerPost, over ReviewMe.

PayPerPost claim that "open-tone opps" (i.e. "opportunities" to write either positively or negatively about an advertiser) are "the majority", but this is obviously some strange usage of the word "majority" that I wasn't previously aware of.

PayPerPost is, in brief, full of hopeful corporate johns trolling for a blog-whore to write something complimentary. There's really almost nothing there but solicitations to journalistic prostitution, as far as I can see.

They pay lip service to journalistic integrity, saying that they "will not accept Opportunities that require our bloggers to be dishonest in any way". But the "Opportunity" list in reality is a long litany of companies that want bloggers to "promote" or "create buzz" or write in a "positive tone only" about their Web sites or products.

And oh, the products on offer.

Beachfront real estate in northwestern Mexico - about which I, here in Australia, am welcome to hold forth at great length, for pay, as long as I keep it complimentary!

When I'm finished with that, someone wants me to "Review our new free Thai dating site", a job which I'm sure will take no more than five minutes. Thailand's practically in my back yard! Why not?

Oh, and I mustn't forget the doctors offering "labia reduction surgery", "anal bleaching", and every other kind of genital-related plastic surgery that's ever made you say "eew". Positive tone only, guys!

And let's not forget all the representatives of the fine and upstanding payday loan industry, who're eager to get positive coverage for their super-competitive 500% APRs!

Needless to say, all this made me eager beyond words to sell myself out and become part of the burgeoning online payola scene.

Regrettably, though, PayPerPost rejected Dan's Data as a suitable place for their priceless Opportunities.

Fortunately, the PayPerPost rejecting-dude told me what the problem was:

While the content of your blog is fine, I would like to give some advice on your format. When advertisers create opportunities, they are expecting that the post in it's entirety will be shown on the blog. If the posts have the "...Read full post" or just one line with a link to the post, that is a major deterrence for the advertiser. If possible, please format your blog to show the full posts and re-submit."

No problem, man - I'll get right on that!

The fact that the average length of an article on Dan's Data is more than 3500 words, with an average page weight of, I dunno, 200 kilobytes at least, would in no way impede putting the last five articles on the front page in full!

Hell, let's make it the last twenty articles! I could have the biggest page on the Web! Take that, Gene!

The nature of the rejection, of course, is something of a giveaway. PayPerPost don't want real writers to apply. Hell, they don't even require your site to have an archive; all they ask is that your paid-for posts remain "active" for 30 days. Then I suppose you can delete them and swear up and down that you never said a word about how that Psychic Development Course was the best thing ever.

I'm not making that up, by the way. "Promote Psychic Development Course" is one of the current Opportunities, along with another bunch of courses and tutorials from the same people, all of which I'll go out on a limb and say are just as bogus. And the quacks and shysters just keep on coming.

PayPerPost also say that "Nobody wants to hear how much you got paid for your sponsored post."

Actually, y'know, I think I really rather would like to know whether someone's getting a five buck tip or a $500 windfall in return for his post about how he sincerely believes some dudes have found the cure for dyslexia.

Be aware that people with a Google PageRank as high as mine can make three hundred US dollars, bang, just by answering the call to "Introduce free Spyware Terminator to your readers!"

Oh, and there are lots of YouTube videos about PayPerPost, too. Because PayPerPost listed their own "Opportunity" a while ago, asking for video testimonials. Awesome!

All of this is pretty much the outside scoop in the blogging community, where PayPerPost has been making friends since the middle of last year. But since I don't hang out with the dudes whose benchmark is wringing a dollar out of every one of their RSS subscribers every month, and since I also missed the Slashdot story, it was all new to me.

Fortunately, it looks as if the network is (to coin a phrase) routing around this crap. Payola isn't new, and it's not going away, but it's not making any great impact I can see, either. There's a pretty sharp line between the sell-sell-sell scumbags with exactly one value who "monetize" their readers and the people like me who have, I dunno, maybe 2500 RSS subscribers specifically because I don't keep trying to screw cash out of you all.

I'm sure there are readers out there who can't tell the difference, but I don't think blog payola's going to make those readers any worse off than they already are. Everyone else ought to be able to detect the subtle signs of PayPerPost-ish bulldust.

All this isn't to say I won't put text ads or something in the Dan's Data RSS feed (update: I've done it now!) at some point. The damn thing's getting more than six hundred thousand pageloads a month all by itself. If y'all want to download it that many freakin' times, I do not feel it's unfair to ask you to occasionally punch a monkey or something in return.

Unfortunately, I haven't yet found a reputable feed ad agency - or a disreputable one, for that matter - that can handle my incredibly obscure "text file I upload via FTP" RSS creation technology. They all want to hook into blogging systems that I don't use for good old flat-file Dan's Data.

So my feed remains pristine. Dammit.