Categories
Social Media Specs

I broke Nofollow

I’m still trying to write something on Technorati Tags. What’s slowing me up is there’s been such a great deal of interesting writing on the topic that I keep wanting to add to what I write. And, well, the weather warmed up to the 60’s again today, and who am I to reject an excuse to go for a nice walk. Plus I also watched Japanese Story tonight, so there goes yet, even more, opportunity to write to this weblog.

Thin excuses for sloth and neglect aside, it is interesting that a formerly obscure and rarely used attribute in X(HTML), rel, has been featured in two major technology rollouts this week: Technorati Tags and the new Google “nofollow” approach to dealing with comment spam. Well, as long as they don’t bring back blink.

Speaking of the new spam buster, after much thought, I’ve decided not to add support for rel=”nofollow” to my weblogs. I agree with Phil and believe that, if anything, there’s going to be an increase of comment spam, as spammers look to make up whatever pagerank is lost from this effort. And they’re not going to be testing whether this is implemented — why should they?

But I am particularly disturbed by the conversations at Scoble’s weblog as regard to ‘withholding’ page rank. Here’s a man who for one reason or another has been linked to by many people, and now ranks highly because of it: in Google, Technorati, and other sites. I imagine that among those that link, there was many who disagree with him at one time or another, but they’re going to link anyway. Why? Because they’re not thinking of Google and ‘juice’ and the withholding or granting of page rank when they write their response. They’re focusing on what Scoble said and how they felt about it, and they’re providing the link and the writing to their readers so that they can form their own opinion. Probably the last thing they’re thinking on is the impact of the link of Scoble’s rank.

Phil hit it right on the head when he talked about nofollow’s impact, but not its impact on the spammers — the impact on us:

But, again, it’s not so much the effects I’m interested in as the effects on us. Will comments wither where the owner shows that he finds you no more trustworthy than a Texas Hold’em purveyor, or will they blossom again without the competition from spammers? Will we do the right thing, and try to find something to link to in a post by someone new who leaves a comment we deem not worthy of a real link, or will new bloggers find it that much harder to gain any traction?

That Phil, he always goes right to the heart within the technology–but blinking, lime green? That’s cruel.

No, no. I don’t know about anyone else, but I’ve spent too much time worrying about Google and pageranks and comment spammers. A few additions to my software, and comment spam hasn’t been much of a problem, not anymore. I spend less than a minute a day cleaning out the spam that’s collected in my moderated queue. It’s become routine, like clearing the lint out of the dryer after I finish drying my clothes.

Of course, if I, and others like me, don’t implement “nofollow” we are, in effect, breaking it. The only way for this to be effective as a spam prevention technique is if everyone uses the modification. I suppose that eventually we could fall into “nofollow” and “no-nofollow” camps, with those of us in the latter added to the new white lists, and every link to our weblogs annotated with “nofollow”, as a form of community pressure.

Maybe obscurity isn’t such a bad thing, though; look what all that page rank power does to people. But I do feel bad for those of you who looked to this as a solution to comment spam. What can I say but…

Categories
Specs Weblogging

The other shoe on nofollow

Recovered from the Wayback Machine.

I expected this reason to use nofollow would take a few weeks at least, but not the first day. Scoble is happy about the other reason for nofollow: being able to link to something in your writing and not give ‘google juice’ to the linked.

Now, he says, I can link to something I dislike and use the power of my link in order to punish the linked, but it won’t push them into a higher search result status.

Dave Winer started this, in a way. He would give sly hints about what people have said and done, leaving you knowing that an interesting conversation was going on elsewhere, but you’re only hearing one side of it. When you’d ask him for a link so you could see other viewpoints, he would reply that “…he didn’t want to give the other party Google juice.” Now I imagine that *he’ll link with impunity–other than the fact that Technorati and Blogdex still follow the links. For now, of course. I imagine within a week, Technorati will stop counting links with nofollow implemented. Blogdex will soon follow, I’m sure.

Is this so bad? In a way, yes it is. It’s an abuse of the purpose of the tag, which was agreed on to discourage comment spammers. More than that, though, it’s an abuse of the the core nature of this environment, where our criticism of another party, such as a weblogger, came with the price of a link. Now, even that price is gone.

*or not

Categories
Specs

Too late solutions

Recovered from the Wayback Machine.

Flash at 6: Google calls Dave Winer. Ooo. The suspense.

Per Sam Ruby:

Robert Sayre: I noticed that the links his comment form have an interesting rel attribute.

Implemented. Prediction: that wouldn’t solve the problem.

I agree with Sam — this isn’t going to solve the problem. Gas station cash registers have signs saying that the attendants only keep 20.00 in cash on hand, but they’re still robbed.

Still, I remember something like this being discussed before.

Waiting for more. I love surprises.

Categories
RDF Specs

What is FOAF and why do you need it?

I thought I would break in with a little tech talk and discuss FOAF, or Friend of a Friend. If you hang around weblogging for any length of time, you’ll probably come across this term. Might be nice to know that it’s not some kind of new goverment regulation.

FOAF is XML created using a specific RDF (Resource Description Framework) vocabulary that allows you to provide a file with information about yourself, and basically, who you know. It’s the brainchild of Dan Brickley and I believe Libby Miller, and has its own support site and blog, though I’m not sure if the weblog is still being updated.

You don’t usually make a FOAF file by hand–either it’s created for you by your tools, or you can use the FOAF-a-matic, a handy forms-based tool that generates valid FOAF XML for you. You can then copy the contents into a file, named something like foaf.rdf, and put this file into the same location as your weblog. Some weblogging tools can do this for you, and you’ll need to check with yours to see if it does, or doesn’t manage FOAF files for you.

To enable people to autodiscover your FOAF file, you can then add the following into your primarily web page, in the HEAD section:

(Note, I had to remove the example because it was not showing up in the page, even with angle brackets being escaped. This most likely is a bug with the underlying tool implementation. The link to autodiscovery also shows the code.)

To see a badly outdated version of a FOAF file, you can check out mine.

Now that you have an idea of what it is, you might be wondering what’s it good for.

Some weblogging tools use FOAF files to auto-generate blogrolls for a weblog. Some people might consider this a goodness, but I’m not one of them. The reason why is that just because you know someone doesn’t mean you want to recommend to the world that they read them.

Others build libraries of photos and friends’ photos using FOAF and it’s image capability, which could be particularly useful for managing photos across many different web sites but based on the same event.

There has been talk of using FOAF to build a Web of Trust — to be able to state who you trust in FOAF and then a person knows based on this that they can also trust them. Though there has been a great deal of work on FOAF and privacy and accountability, the vocabulary isn’t there yet, and even the creators would be hesitant about recommending FOAF as it stands now to be used for a basis of trust.

There has also been discussion about extending the FOAF vocabulary to expand the types of relationships available. However the concern on this is the impact something like this could have, if one person considers you a friend and you put into your FOAF file that they’re just someone you know. And do we really want people to know whom we love, have birthed, work for, and so on? Sometimes. Sometimes not.

However, using relationships internally in an application, something such as an address book, could be very useful — but then why use FOAF, which is primarily used to create networks based on publicly accessible FOAF files.

Regardless of use and opinions about FOAF, it is now the second most widely dessiminated example of RDF/XML in use today, after RSS 1.0. And if you hang around weblogs, you’ll be stumbling across it at one time or another. More than that, the tools you use may be asking you whether you want a FOAF file or not.

Categories
RDF Specs

Critical Mass

Recovered from the Wayback Machine.

When I read about the RDF Data organization, I was reminded that the difficulties inherent with deriving a new vocabulary and associated functionality isn’t found in the bits of XML or the bytes of code: it’s generating enough interest and uses thereof for the vocabulary to reach critical mass; making it into a viable component of the semantic web.

By critical mass, I mean that there is enough meaningful data to inspire applications that mine the data and that in turn, generate processes that couldn’t be done without the data: similar in concept to the critical mass that HTML received and the subsequent spawning of both browsers and search bots. Private or commercial applications that use RDF/XML for their internal systems are all well and good and provide needed exposure–but they aren’t a component of the semantic web if the data is not publicly available and with enough critical mass to make it useful.

You may have noticed that I used the small ’s’ and small ‘w’ semantic web; the reason is that I see the Semantic Web, the uppercase version, as a top-down approach to building an intelligent web. A bottom up approach is just us folks, doing whatever it is that interests us and gets us excited–and, I hasten to add, that can be translated into RDF/XML. For some the exciting bits would be FOAF, others RSS, others DOAP, and so on. These are the vocabularies that need a critical mass, as the Uppercase bugger has knights and other nobility to do its promotion.

My own interests in semantic web data that can be defined with OWL/RDF lies in two areas: poetry and web object history. These have been represented by my work on two systems: The RDF Poetry Finder and PostCon. Yes, the two perpetual motion systems, always in development. Both of which would meaningless in and of themselves, unless they reach critical mass.

For instance, I use PostCon to manage some of my redirects and provide intelligent responses when a page has been pulled. I’ve also generated PostCon RDF/XML files for all of my weblog entries and placed them on the server. I believe at one point, I could even semi-search them in Google, but only when I’ve linked them from my HTML pages.

As for the Poetry Finder, well I’ve tried to interest two major poetry sites in this but to no avail, and am either looking at supporting a centralized repository of data for the nonce, or trying to get webloggers to generate RDF/XML files to go with their poetry discussions. (More on this later.) A simple enough form that can generate the RDF/XML, just as with FOAF, should work. It’s getting people to use it – demonstrating an advantage. FOAF adopters adopted FOAF because they’re basically tech tinkerers. Poets are not known to be tech tinkerers.

Regarding the data jewels of others, DOAP, the brainchild of Edd Dumbill hasn’t reached critical mass yet, but should. I think the key would be incorporation into a hugely popular site like SourceForge.

The growth of RSS has reached critical mass and way beyond at this point, though the differing formats still cause confusion. It was helped with its early promotion by major companies , but the real key was it’s support by aggressive individuals who have all the zeal of a fresh missionary among bad sinners. Even if the support was for plain vanilla XML rather than semantically intelligent XML (ooo, did I say a bad thing ooo). FOAF’s growth has also reached critical mass, helped primarily by the happy and gentle persistence of it’s creators, as well as adoption by some high profile people and applications.

Both vocabularies were also helped, quite significantly, by weblogging. In fact, I see weblogging as the leading agent of change for the semantic web–the tool/technique/genre/thing most effective in helping a vocabulary reach critical mass; and I’m not even wearing any pajamas as I make this statement (sorry, bad joke). The only problem is trying to get enough of a critical mass in weblogging to be heard above the competing noise, and then enough webloggers interested to jump start the generation of the data to reach the semantic web critical mass–all without having to have the zeal of a fresh missionary among very bad sinners.