🤯 60% OFF CYBER MONDAY SALE 650+ books, courses & coding assessments!

The case against web 2.0

    Harry Fuecks

    Via Simon, Gina Trapani asks some very relevant questions in the anti web 2.0

    I still prefer my text files and my own MySQL databases to anyone else’s, and I’m not sure why I’m so alone in that sentiment.

    You’re certainly not alone but blogs, generally, only reinforce positive buzz, often uncritically. You’re also stepping into politics here, which doesn’t travel well on tech-blogs. Breaking my own habits for once though, I’ll put a toe in, by way of anecdote.

    Back when I was at University doing a sandwich course, took a year working at IBM in London. One of my managers back then was something of an IBM high-flyer – interesting guy but I never knew quite what to say to him – generation gap, too much respect for elders and so on tied my tongue. Anyway – keeping this story short – we’d been at some IBM / customer conference all day, for the product we were involved with, and he gave me a lift home. Most of the way it was embarrassed silence, on my part, not knowing how to strike up a sustainable conversation. Then, seemingly out of the blue;

    You know Harry; after three years, power corrupts. I’ve seen the same thing again and again – it doesn’t matter who’s got the power or how much they’ve got – after three years, power always corrupts.

    …which left me at even more of a loss for words, but has stuck with me ever since. Whether three years is the magic number here, I can’t say but I do think the principle is correct. Which leads to the conclusion: power itself is the problem, irrespective of who owns it, and we should be extremely cautious when we dish it out, if at all.

    OK – if you’re not running for the political escape hatch – so how does that relate to web 2.0? Well, simply, in doling out all our data to central repositories, not in our control and not open for us to inspect fully , we’re placing massive amounts of power in the hands of a minority. There is some discussion of the consequences happening, but nowhere near enough.

    Consider del.icio.us for example (not that I think there’s anything wrong with del.icio.us). In relation to individual users, del.icio.us has a pretty good policy, to the point of having implemented features that allow you to export your bookmarks should you wish to leave. But what they’re not giving away (partly because it would be very hard to implement in a form that scales) is the “big picture” – there’s only so much you can query as a normal user. But let’s say I’m an employer and I’m looking for a hot Javascript developer – one way I could (in theory) find one would be by tracking down del.icio.us users who’ve consistently been the amongst first to bookmark popular links on del.icio.us – that’s likely to be someone who’s at least ahead of the curve in Javascript (perhaps not an ideal employee but still). Technically you could perform this query, given it’s only a single tag, but you’d be skirting the border of API abuse – from the del.icio.us terms;

    4. Feeds and API

    […] You may not use these or any other features or the Site itself to allow the display of a substantial portion of the del.icio.us database or reproduce, duplicate or copy the del.icio.us Site. […]

    If I wanted to make this kind of search across multiple del.icio.us tags, or start analysing some of the social networks that have built up in del.icio.us, I’d certainly be abusing the API and generating some significant HTTP traffic. But del.icio.us have these queries at their disposal, potentially giving them opportunity to sell them as a premium service – which raises some significant questions about who’s buying…

    OK – personally I’m not of the view that this type of view of data should be somehow protected and kept in the dark. Instead think it’s about being it being Open for anyone that wants it plus (ideally) I want be able to see how my data is being used. It’s a free data thing.

    Anyway enough rambling. Specific to Gina’s interest and futureproofing data, Mark Pilgrim’s post on long term backups (plus all the comments) is worth a read, but not conclusive. From a wider perspective this is all about (hard to solve) implementation detail – how do we really do a decentralized web, while preserving all that’s good about web 2.0?