Matt Craven has written an interesting piece about Dave Winer over at The Blog Herald.

Along with a personal assessment, he brings up a recent Scripting News post which speculates, “What does an algorithm think?” — something I’ve often thought about, especially when falling foul of Google’s.

Dave’s post doesn’t actually answer the question, but has a little moan about posts on Techmeme : “Most of the authors don’t know the first thing about technology, never took a computer science class, have never written code, and don’t admit that understanding tech is a prerequisite for writing about it.”

In other words, only code writers need apply.

Now, I usually put myself in the ignorant category, unfairly as you’ll see — but being unfair to oneself is better than over-spicing the pudding.

Back in the 1980s when real computers were IBM mainframes or PDP 11 “mini” computers, and the hoi-polloi like me had to make do with “micros”, which really were micro then, I had a startup called, Earlgate Computers. It developed and produced software for the Sinclair Spectrum, the BBC (Acorn) computer and one or two others, like the Atari and Commodore 64. All utter relics now.

Yours truly wrote a series of programs titled, Fitness Software, which was aimed at the running and marathon craze of the period. The packages, on cassette tape, were written in Basic, and the series sold to two big retail chains in the UK, Boots and W.H. Smith.

Even so, I wouldn’t claim I’m a developer or a programmer by today’s standards, although I have written commercial code. I usually muddle through with the latest gizmos and avoid too much complexity where possible.

Nevertheless, I do get onto Techmeme regularly, so presumably fall into Dave’s “waste of space” class. I think he’s probably right.

However, a word of warning. Narrowly-based communities that talk to each other in jargon incomprehensible to even an intelligent audience, really belong in a social network niche, not on mainstream tools of the blogosphere. People with peripheral skills and general interests can often bring new perspectives and shine light into dark corners otherwise missed.

As Matt writes, Dave comes across as an irascible sort of fellow, forever banging on about RSS and outliners. Not quite “Hold The Front Page!” stuff.

His other strand, U.S. politics in election year, is much better, even for a Brit like me. I happen to be very interested in who or what the next President of the U.S. of A. will be.

Now for the unanswered question : what DOES an algorithm think?

It doesn’t.

Last Christmas I wrote a post with 10 predictions for blogs and blog networks over the coming year. It was meant to be humorous, so not many came true.

This season it would be difficult to do the same as most of the buzz has disappeared. Top bloggers have gone pro, and startup networks have become serious and less inclined to gossip — except incomprehensibly on Twitter, where it’s like listening to one side of a telephone conversation spoken in code. So I’m not going to do ten predictions this year.

2007 has been the year of unblogging. Blogging about blogging has hit the deadpool. The tedium just turned us off in the end, as it always does.

That’s good. When the medium becomes the message something is out of kilter. To say that A is B is gobbledigook since, if true, there would be no need for B. Now the message is clearly the message and we can all get back to simple basics, i.e. content, and use weblog software for what it was meant to be, an easy way to publish our thoughts online without giving an interminable running commentary.

Has that taken some steam out of the blogosphere? You bet it has. Google spotted this and is currently downgrading all but the very best examples of the blogging art. Money will be much tighter until the next boom, which may be some years away.

Advertisers too are reining in their drive into blogs, converting big downpayments into PayPerClick models. Who can blame them? Goldrushes never last.

So I’m going to make one prediction for 2008. It will be much harder financially for everyone, and only the best will survive. The credit crunch will squeeze out those who haven’t put something by for the hard times, and leave the dabblers stranded.

It’s going to be a long, hard haul, but hey, that’s where the fun is, isn’t it?

Chins up and mind how you grow.

I’ve just noticed that Techmeme has got a PR4. The site probably has hundreds of thousands of links in its archives. Its business is to link readers to the most relevant and authoritative technology news around the world.

Why doesn’t the new Google algorithm distinguish between that model and paid linkage? So far as I’m aware inclusion in the body of Techmeme is not based on any monetary transfer. Syntagma is often included and no payment is asked.

The only exceptions are the webclips from posts of Techmeme’s sponsor sites. Do they count as text link ads too? If so, Google is now attacking the basis of the citation system it set up in its early years and is ominously beginning to confirm a suspected lack of tolerance of any online monetizing scheme other than its own.

We sometimes forget that Google is a business like any other. To own both the largest search engine and the most extensive advertising program on the internet must offer great temptations for muscle flexing to its management.

However, the use of monopoly powers is always counter-productive in the end. Without keen competition a company quickly loses force in the marketplace and the respect of its customers.

One wonders how far this downgrading process will go. There are hints that the company miscalculated this move and is having to put back rankings to many sites. Not here yet though.

So what is the new gold standard? What now counts as a penalizable link in the Google dispensation? In the age of information, the facts of this case are not easy to come by.

We have fielded a lot of emails in the past three days about the still developing situation between internet giant Google and text-link buying agents, like TextLinkAds.com.

To recap, Google has conducted a manual search of mainly networked sites to penalize commercial publishers who sell links via the agencies, or who have hand-rolled ads without rel=”nofollow” in the code. The code stops Google’s robot from following the link to its source, thus avoiding clocking up a backlink for the buying site.

The more ranked backlinks a site has, the higher its PageRank. Google has been progressively lowering the rankings of sites on which perceived mal-dispositions are found.

For example, Syntagma — which we expected to rank 6 by now — has dropped to 3 in the latest round of carnage. Our site Fifty-Something Women — which has around 20 text links of various sorts — has fallen from 5 to 2. This gives the artificial impression that the site has low traffic and puts off some ad buyers who respect Google’s approval.

Google claims the ads are skewing the results of its search engine, replacing relevancy and authority with the ability to purchase links, i.e. buy authority and relevancy rather than earn them through merit.

There is some force in that argument, although the contrarian view is that Google itself is now skewing its own results by retreating from its pure search criteria.

Also, many observers believe Google is attempting to crush the businesses of the astonishingly successful agencies that sell the links. Its own text-based system Adsense, has undoubtedly taken a knock from TLAs in recent months, especially on low-to-medium trafficked sites, which do poorly on the Google system.

Whatever the facts in this case, Google has created the monster from which it is now suffering.

Background
A mere 15 years ago, at Silicon Valley’s Stanford University, founder Larry Page saw literary citations as a software opportunity for the web. Nowadays, it’s hard to see beyond the system he produced, first BackRub, a way of measuring backlinks to articles and sites, and then PageRank, the most addictive element in web oneupmanship. Has that system really served us well?

To generate Googlejuice you have to cite and cite regularly and relevantly. The blogosphere, in particular, is a madhouse of clickability. The genius of Google is that it didn’t just transfer the bane of academic publishing, the citation system, onto the web, but that it discovered how it could profit enormously from the process.
/Background

Is it now pulling the rug from under other companies who are using its innovative scheme to build businesses on the web?

The link, or citation, system is behind the present controversy in the search business. The wonder is that Google tolerated the SEO — or Google gaming — system for so long. The sudden rush to penalize publishers has a rather thuggish feel about it, particularly as the broken bit in this chain lies within the software of the companies that broker the ads.

Jason Calacanis, formerly chief of Weblogs, Inc, which sold to AOL for a reported $30m, has put up a good piece in which he explains how he created the interlinking system between blogs and also hosted the first text link ads.

Where are we now?
We are now at the point where we have to decide. Do we take the ads down, or leave them up and suffer the consequences, the extent of which is not yet clear?

At Syntagma, we have sites which are 90 percent plus dependent on text links. We are therefore going to sit this out and observe what happens. We will keep faith with our advertisers. Business as usual.

However, we strongly urge the brokerage companies of text links to do deals with Google on this matter. Not only is their business model in grave danger of being blown out of the water, but it may be that small tweaks in their software will be enough to allow Google room to compromise.

After all, there are a lot of enraged publishers out there, who feel they have been badly let down by the giant of the internet. Uncle Google has suddenly turned into the wicked stepfather.