Some thought-tools

It seems to me that one of the problems with the kerfuffle about Google’s Toolbar AutoLink is the number of separate things being conflated into one “it’s teh evil!” argument, so that since we hates Google now, we does, anything that adds links to a page is one tiny slip down the slippery slope from turning your laptop into a Transformer that molests your pets. So, some thought tools that I don’t really expect will change anyone’s thinking:

OCLC is, in library terms at least, a Big(Nonprofit)Co. Built on the tax dollars of, among others, hardworking Amazon employees, they own a huge database of library holdings. With a bookmarklet from this page, whenever you are on an Amazon page, thinking about buying a book, you can instead click your bookmarklet, and find out if your local public library owns it, and reserve it right then and there. They use Amazon’s efforts to market books, leading you to the one you want through Amazon’s search results, or What’s Related links, or customer reviews, and then they strike right at the heart of their business, stealing away Amazon customers from their own page. Evil, or not?

Borrowing code from them, and Jesse’s Buglinkify bookmarklet, library-link is a bookmarklet that will convert unlinked ISBNs in the page you are viewing to links to see if your local library owns the book (if you are using Opera, or a Gecko-based browser like Firefox, or perhaps Safari, but not the hapless Internet Explorer, and with only moderately good recognition code, and only for things in the Open WorldCat pilot, and sometimes it will take several clicks to get them all). More evil, or not?

Since Amazon’s been bearing the brunt so far, how about Amazonify, a bookmarklet that will convert all unlinked ISBNs on the page into Amazon links. It’s just code, it doesn’t care whether you use it nicely on an unlinked bibliography, hoping to educate yourself, or on a more politically correct competitor’s site. Evil sometimes, evil always, or not?

But that leaves Amazon affiliate money on the table. If I could be bothered to sign up, I’d make it my own, but since I can’t: because I built that useful tool for you, I want to give some profit to my favorite web- and pod-casting station, Whole Wheat Radio, so AmazonifyWWR. Is that too much to ask, in return for my providing you with this handy tool? How about if I don’t tell you, and just put it in there?

Since I certainly don’t care how you remix my page, some ISBNs to play with: 0130923915, 032116332X, 1578050847.


Comment by Mark #
2005-02-27 18:41:03

There are already some Firefox extensions whose authors use/abuse Amazon links as a form of forced shareware payment. e.g. BetterSearch

Charming licensing terms, eh? At least he’s up front about it. (The inevitable upcoming Firefox-based spyware certainly won’t be.)

Comment by Phil Ringnalda #
2005-02-27 19:14:31

The version of the license on the web page seems to be missing a word, compared to the version in browserOverlay.js. Makes it seem positively un-Carlo-ish, in that blander version.

Then there’s the 67 Amazon search plugins available from, 66 of which are undisclosed affiliate links for the author. Or for that matter, the Amazon plugin that ships with Firefox, which includes an affiliate link for MoFo.

Comment by Carlo Zottmann #
2005-02-28 01:08:59

Ah yes, the licensing terms. Bad choice of words, I agree, so that’s why I ditched the bad F-word of it from it. (When I initially wrote it I was rather upset because some guy had just replaced my affiliate code with his and put the extension up on his website. We have come to terms since then. Anyways, that’s why I put it rather bluntly.)

Makes it seem positively un-Carlo-ish, in that blander version.

Could you explain that a bit, please? ;)

But in general: Is it wrong to build something to make a bit of money? It’s not like I make obscene amounts of money with it. I mean, it’s not like my extension is hitting the user over the head with popups or malicious redirects or whatnot. There’s nothing changed for the user; the Amazon links will get him/her where they would get him/her without the extension.

There’s always some uproar when 3rd party software is generating money, even if it’s just tiny amounts. Why is that?

Comment by Phil Ringnalda #
2005-02-28 01:32:41

Silly me, I didn’t allow <dl> in my comments. un-Carlo-ish: not like the Carlo I know, author of Referrer Spam Fucker 3000.

I don’t think making money from things you build is automatically bad, but I do think we’re headed toward a major culture clash in Firefox extensions, since they’ve traditionally been GPL, MIT, or Tri-Licensed, often just by assumption rather than explicitly, and some people are going to keep treating them that way even when they aren’t.

I also tend to have more of a problem with affiliate links than other ways of making money, especially when they aren’t clearly disclosed. Partly it’s the way they feel sort of sneaky, where with an ad there’s no question why it’s there, but an affiliate link (outside a search tool, obviously) always makes me wonder whether it’s there because the person wanted to link, or wanted the money from the link, and partly it just doesn’t seem like something deserving of payment. Getting paid for a link? Hell, I throw off a dozen or two of those a day!

You write an extension people want to use, and they don’t mind having a third party give you a few pennies if they buy something through it, fine by me. A Sherlock plugin, though, where it’s a matter of copying a dozen-line file and changing one URL and a variable name or two, and then using someone else to market and distribute it? That seems over the line to me.

Comment by Carlo Zottmann #
2005-02-28 01:48:11

Ah yes, the RSF… an hommage to Penny Arcade’s Fruit Fucker 3000, you know. *g*

I agree on the Sherlock plugin. It is a bit over the line, yes.

Comment by Carlo Zottmann #
2005-02-28 01:49:37

PS: One day I’m going to use the ”Reply to this” link of the comments instead of just typing away in the comments box that’s there, I promise.

Comment by Rijk #
2005-02-28 02:18:13

Hm… Only one of those three ISBNs is converted to a link for me. Tried in Firefox and Opera 8b2.

BTW, instead of having to have lots of bookmarklets, I’d prefer to simply double-click the number and get a ’Look up at Amazon’ option in the context menu. Installing a mycroft plugin for ISBN and using the Context search extension is more comfortable IMHO. This is default behavior in Opera BTW.

This doesn’t visually change the page, and still makes it terribly easy to ’steal’ webpage info. Ergo, ’Context search’ is just as evil as the new Google toolbar :)

Comment by Phil Ringnalda #
2005-02-28 09:01:42

I think I blew it while I was messing with Jesse’s code, and left it only taking the first ISBN in a particular text node: hit the bookmarklet again and again, and you’ll get the others, use it on a more real page, and you’ll get better results. But that’s why they are thought-tools, not real-tools. For real, I’d probably do it in an extension, where I had more script-room, and validate the number, and call OCLC’s xISBN service myself to find the ISBNs for other editions of the same work, and call the Amazon API before putting in a link that would fail.

Suggesting that it be a context-menu option is the most reasonable suggestion I’ve heard from people who object to the links (most people aren’t suggesting that it work with selection, since that would be more of a pain for addresses than for the various numbers), though it’s still much less useful to the user, who can’t see what is and isn’t hot without trying. I still need to find an ISBN that Amazon doesn’t know about, to make sure that the toolbar doesn’t link unless the link has a target, but I suspect it doesn’t, and having to right-click, inspect the right-click menu for the presence or absence of a menu item, wouldn’t be a satisfactory alternative to a link.

But while I wouldn’t want to use them as an alternative (even though I already do), the myriad other ways that the numbers and addresses in a page can already be used to steal someone’s customers away right off their page make that a very weak argument against Google’s toolbar. There’s nothing new about harvesting data from a page and allowing for navigator to other sites about that data, including competitors’ sites, so for this to be excitingly different it needs to be either because adding links actually does make some difference in perception, or because everybody else is allowed to, but Google isn’t.

Comment by Phil Ringnalda #
2005-02-28 09:13:30

Hmm. Consider this: if links are evil, but context menu options are not, and selecting a whole address is a pain, so right-clicking anywhere within the target data is better, then probably the best way to implement the context menu is by adding links with text-decoration: none; color: inherit; and onclick="return false;". Otherwise, you have to keep parsing the text in both directions every single time the user right-clicks, to see whether you are in something you can link, and what sort of menu option to include. That would be sloppy and slow, much better to parse the whole thing once, and keep track of what’s hot, and what sort of thing it is, by wrapping a DOM node around it and leaving the destination in an attribute of that node.

Is it okay to insert links in a page, as long as the user doesn’t know you’ve done it?

Name (required)
E-mail (required - never shown publicly)
Your Comment (smaller size | larger size)
You may use <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <del datetime="" cite=""> <dd> <dl> <dt> <em> <i> <ins datetime="" cite=""> <kbd> <li> <ol> <p> <pre> <q cite=""> <samp> <strong> <sub> <sup> <ul> in your comment.