Kiran Jonnalagadda’s Blog

A critical examination of the service’s teething troubles, with suggestions for improvements.

In the heydays of the Linux User Groups, the communities served as more than Linux advocacy and assistance forums. Each LUG was also a travel advisory forum. It used to be such that when visiting a new city, you could look up the local LUG and ask if anyone was interested in meeting up, or if they could help you with accommodation, and usually be granted both. The LUGs were a great way to meet like-minded people in unfamiliar places.

The LUGs were of course just the latest in a long tradition of such mutual assistance communities, and as newer forms of getting together emerge online, it is inevitable that there will be newer attempts at helping get familiar.

Thus we have Metroblogging (Metblogs for short), an umbrella site that hosts a blog each for cities around the world. Metblogs is interesting because it makes the geographic region itself the focus, instead of having it incidental to another activity. LUGs, in contrast, formed around cities for no other reason than that cities facilitate easy gathering. Metblogs also limits posting access to those who were invited in (or in case of a new city blog, those who specifically applied and were vetted by an unidentified approver), and requires posters to maintain a minimum frequency of over three posts a week.

While these procedures undoubtedly are meant to ensure quality and vitality, they also suggest that Metblogs is not the organic outgrowth of another activity (presumably blogging), but one forced to fit a specific role. From a cursory examination of some metblogs and critical examination of one over a few weeks, and the resulting disappointment at the quality, it appears the role itself may not be well defined. Exactly what purpose does Metblogs hope to serve?

At this point, it is worth examining the structure of the operation and its visible shortcomings. This is not so much to criticise the operation as to suggest improvements.

The term “Metroblogging” embodies a brand, and the brand promises consistency in delivering on expectations. Metblogs is also an umbrella, granting authority to each team via the distinct cityname.metblogs.com domain and visual style, but not guaranteeing quality since posts are unedited, and denying authority to anyone else. The team’s composition is essentially by undefined criteria.

With small teams such as at BoingBoing, the individual personas collectively make the brand identity, but Metblogs is too large for that to work. When providing an umbrella and limiting access, Metblogs is also expected to define consistency via a style guide, just like any traditional print publication. Since there are no editors, compliance must be at the level of the individual poster and must be mandatory.

In the absence of a style guide, the result is akin to putting the team in a glass cage, where they make an exhibit of themselves along with their words. That it rained in your favourite city may be welcome news, but that your poster lost his umbrella in the process of arriving home to file the report is unwanted clutter. It may serve to strengthen the poster’s identity, but that identity is swamped by the overall bulk of Metblogs, and as such, is only accessible to dedicated readers, the kind who regularly visit the comments section.

Comments are critical to the blogging process, for they provide the feedback that dictates direction and builds the dialogue that puts the poster at ease talking to readers, but herein lies a trap. Because comments lie one level deep, they are hidden from the reader who browses only the front page or reads via a feed, or even to one who clicked through before the comments arrived. The poster, however, sees all. Should the poster choose to continue conversing via the next post, most readers now see only one side of the conversation. This creates the unintended effect of being talked at, instead of talked to.

This is why being informed that your poster lost his umbrella is irritating. It is of little consequence to the supposed topic at hand, the city itself. Such detail belongs in the poster’s personal journal, where readers gather more for who the poster is than for what is being said.

It may perhaps help if more readers tracked the commentary, but this is hindered by the fact that comments are hidden from top level view, that the technology and tools for feed syndication focus on posts, ignoring comments, that the user interface for comments stunts the commentator’s identity, and that there’s no intuitive way to engage in conversation in the comments instead of just responding to the post. A commentator with something significant to say would rather say it in place where their words will receive greater prominence, leaving the comment space for general murmur of agreement, thereby furthering the cult of the poster.

One workaround is to structure posts in a manner that encourages commentary. When a post rounds off its statement, there’s little space for anyone else to speak, but an open-ended post provides the necessary gap. Tacks such as “what do people think?” don’t work as well as infusing an element of self-doubt into the very post.

This form of reader engagement becomes critical when there is no editorial control. If a poster reiterates a popular but fallacious opinion, where is the space for dissent? If the dissenter is not also an authorised poster, the opinion is lost in the comments, where the poster may choose to not bring it to the top level for fear of embarrassment, personal disagreement, or otherwise, as a result of which, Metblogs itself appears to be propagating the fallacious opinion. Even if another poster chooses to refute it, the question remains, does Metblogs want to be a space for such debate, or was that an unnecessary distraction from the main course?

Such a cult of the poster is unfortunately antithetical to the promise of the umbrella Metblogs brand. The choices appear to be to (a) adopt a posting style that brings commentary to the fore, (b) ease access to top level posting, or (c) loosen the association between city teams so that their reputations are not hindered or undeservedly boosted by others. The latter may be achieved by allowing more than one blog per city, so that any given team is no longer automatically authoritative for that city but must work for their reputation.

Having multiple blogs will also help teams focus on particular audiences. The current target audience is ambiguous beyond the generic definition of anyone interested in a particular city. Is the target the long term resident, who may be interested in civic issues and on how the city is changing, or perhaps the recent immigrant, looking for unexplored avenues, or is it the short-term visitor, expecting to be informed of highlights and activities for a spare evening? Maybe there is another way to partition the demography?

The observed team appears uncertain too. Their current strategy is to post as often as possible about anything at all, perhaps in the expectation that there will be enough for everyone, but with the result being a flood of inanity with an occasional interesting post drifting by.

To summarise, the issues are that the current system of one blog per city grants undeserved authority to teams while preventing them from focusing on specific audiences, that the system of deciding who is allowed to post coupled with the interface for commentary brings more attention to the poster than is appropriate, while stymying discussion, and that the minimum frequency requirement introduces too much noise in the signal.

None of these are hard to solve, given willing site administration and city teams, on the assumption that this analysis is valid. Despite the teething troubles, Metroblogging is an interesting initiative; its growth is worth watching.

Update (Aug 23): Metroblogging’s publisher Bode Media responds. Comments on that later.