Having spent some time evaluating the Structured Blogging plug-in, and discussing it at length with students, I have become less and less certain that what it offers is worth the price. At first sight the ideas behind it, and its implementation, are impressive and thought-provoking. The fact, however, is that in fifteen months the plug-in has signally failed to set the world on fire.

In December 14, 2005, Phillip Pearson wrote a post called Structured Blogging is official! This announced that the plug-ins were now publicly available, and linked to the Structured Blogging site. He was optimistic, and indeed most of the commenters were too. The next day Joshua Porter wrote a favourable essay explained how the process would benefit bloggers and aggregators.

Since then the developers have moved onto other, mainly philosophically related, projects and Structured Blogging has languished. As I noted earlier, for example, I had to do a fair amount of work to get the WordPress plug-in to work with WordPress 2.2. So, what went wrong? Why has the project failed to set the world on fire?

Reasons for slow growth

One possibility is that the idea was fine, the implementation was fine, but the maintenance took too much time and so the developers went elsewhere in search of food and water. I think this unlikely.

A variation of this would be to say that the idea was fine, the implementation was fine, but the developers thought of better ways of achieving the same goals and went off to implement those. I find this even less likely because if this were the case then they would have surely replaced the web site with one linking to the new improved models.

A second possibility is that the idea was sound but the implementation was flawed. There is indeed a sense in which this might be true. Although the plug-ins for WordPress and Moveable Type worked perfectly when they were introduced some important features of the whole infrastructure were missing. Deliberately there was no aggregation software written capable of using the system, because the developers’ intention was to show that people would use the plug-ins even before the benefits of aggregation were introduced. The result, perhaps, was that some people did indeed use it for its layout features, but they were not a large enough mass to encourage the development of the rest of the infrastructure.

A variation of this is the claim that the implementation was flawed because authors were too lazy to use the form-based entry pages. As someone who has started using them I find this difficult to believe. I think that if WordPress had these forms built into the admin system then people would use them to write reviews without even noticing.

A third possibility is that the idea was never sound in the first place. It seemed plausible upon first hearing but, after further thought, it simply fell apart.

Initially the idea of structured blogging, as implemented here, falls into three different parts, because the plug-in performs three very separate, although related, tasks. Firstly, it offers a set of new design options for a WordPress-run site that enable reviews, advance notices of events, job listings, and so on, to stand out from the general run of posts. Secondly, it offers a neat way of having selected posts automatically forwarded to other sites. I can have the PS3 games reviews I write automatically forwarded to a games review site at the same time that they are posted onto my site. Thirdly, it hides a bunch of XML in the page source that can be harvested by aggregators that I have no prior knowledge of.

The first function is immediately visible and, depending on your needs, enhances your site quickly and easily. It does this by providing templates for new entry pages and corresponding templates for the posts when they are displayed. However, if this is viewed as an internal enhancement, then it could be achieved by a much lighter plug-in that didn’t bother creating the XML-based micro-content.

The second function is server-based and could simply exist on its own. If there is not already a plug-in to enable you to forward posts automatically to other sites then there ought to be!

The third function problematic in theory and in practice. Greg Yardley wrote that:

“Profiting off user-generated content is Web 2.0 colonialism.” That sums up how I feel about the much-praised (and widely backed) Structured Blogging initiative, which makes it easy for bloggers to use microformats to mark-up specific genres of blog posts – reviews, classified listings, and so on. Microformats make blog posts machine-readable, which in turn allow them to be used by applications. Jeff Clavier sees Structured Blogging “eventually pushing blogging into richer types of applications – and enabling new types of aggregation.” Indeed – if adopted, it will. Which is what irks me. Structured user-generated content, especially aggregated reviews, is very valuable. Case in point – the del.icio.us purchase. Since del.icio.us’ functionality is easily replicable, the deal was all about the value of user-generated content. You’d think with content being worth so much, the Structured Blogging initiative would contain a way for the content providers to indicate, in a machine-readable fashion, just how they would like to be reimbursed for the commercial use of the content they’re providing. Not so – at least not anywhere I can see.

(He also noted that the phrase at the beginning of his post is something that Paul Mooney wrote in a comment on another blog.)

He is correct, and there are terms for these kinds of transaction. They are examples of prosumption or conduction. Stores have loyalty cards to acknowledge that when you shop with them, and they aggregate the information you provide, you have produced value in the very act of consuming. Loyalty cards are a way of paying you for the work you have done. (Whether they are a sufficient reward is another argument entirely. All that matters here is that the transaction is being acknowledged.)

On the Structured Blogging site, one of the proposed benefits is that various kinds of places will start to use my data, including “applications that run in web browsers, like a Firefox plugin for comparison shopping that also reads product reviews”.

It is difficult for me to see why I would want this to happen, unless there was some way in which this was necessarily a two-way transaction that benefited me for publishing my data.

There is a practical difficulty too. The micro-formatting means that the entire page is, in effect, included in the page source twice. This poses no difficulties if you write short reviews like this, but a 3,000 word review will become a 6,000 word page, and I am not convinced that I am being completely delusional when I suspect that my reviews load noticeably more slowly than the other pages.

The underlying model

Having thought about this from as many angles as I can, I am less and less convinced about working free for other people who, by definition I do not know or interact with.

I suspect that the fundamental model of networking is simply wrong at this point.

By way of an example, Bryan Alexander has recently started a wiki for restaurants in airports. The idea is simple. He is fed up running around strange airports looking to find somewhere to decent to eat only to be told later that he missed a fantastic little place opposite runway 3. I think it is a good idea, and I will make an effort to contribute. I think it would benefit from standardised entries, structured in the sense of having a form-based entry form that guarantees reviews can easily be compared and correlated. I think I might benefit marginally by being able to post a review on this site and have it forwarded automatically to Bryan’s.

It is quite clear how “payment” happens in this system. It is payment by mutual benefit. However, the networking model is also clear. Bryan had an idea and blogged about it. Some people will read about it. If enough people contribute and find it useful, they will tell more people, and it will grow. If it grows sufficiently maybe it will need its own domain, and it will become a recognised tool.

The networking model that is implied in the structured blogging approach is subtly, but crucially, different. It implies that there are people already blogging about restaurants in airports, and that they can be aggregated by a third party into a monetized site. If this is successful the ostensible end result will be the same, but the relationship between the contributors and the site will most definitely not be the same. Many of the contributors may not even know they have contributed.

In the current model the starting point was Bryan’s observation of his personal need, and subsequent public initiative to call content into existence. In the other model the starting point is an observation by an aggregator that certain content already exists and can be put to other use.

The New Zealand Coffee Review mentioned in the Structured Blogging site provides another example. It is described as a “review server” that might aggregate relevant content. The problem with this is that anyone in New Zealand interested in posting reviews of coffee shops would surely already know of the site and post there directly, especially since direct posting gets you a user profile. If they did know then their content would not need aggregating, since they could simply have it forwarded automatically.

In other words the model at work here is very similar to the one that Bryan has just begun with the airport restaurants wiki; and the first two functions of the plug-in are all that are required to operate such a system.

Conclusions

I like the enhancements to the pages and will continue to use them. I like the idea of automatically forwarding posts in principle, although I cannot think of a personal need for it at the moment. I think the underlying assumptions of making micro-content freely available need to be firmly placed inside a larger, mutually beneficial ecology.

I do no mean that they need to be centrally planned or, indeed, planned at all. I simply suspect that the ecology will grow better if aggregators are somehow encouraged to step up and announce what they want and how they will ensure the benefits are shared.

Aggregation cannot simply be a matter of adding numbers, because not all data – not all comparative data – is numerical. Stefan Rimaila has written a short piece about this.

Even after questions of origination and renumeration are addressed, there are still issues remaining concerned with levels of granularity. How micro does micro-content need to be? How structured should structured blogging be?

I suspect we are looking at the beginning of a long and bumpy road.