onHover is the new blink tag, HR and under construction graphic

I’m sorry, but this needs to be said.

Has anyone really visited TechCrunch lately through their browser? Lets run it down. The total requests made for the page is 56, and the total download size of all assets is half a megabyte. Of that, 43k is Javascript.

Techcrunch seems to have every “wow!” Javascript utility known to man included on their site. SnapShots on links, Sphere It, etc. Not to mention all the code for different ad networks, widgets, etc. I’m all for the use of cross-site scripting and handy tools, but its getting kind of ridiculous.

In the beginning stages of this “web 2.0″ shit, it used to be about a clear definition of what a site was, with the fludity of the data from that site to transgress the boundaries. It never was the aggregation of this fluditiy on any given site, but the ability for people to make their own tools which could mash these things up into wholly new things.

I don’t think the intention was to use the fluidity of data as window dressing on content for the sake of saying you could. I don’t think it meant laying tool over tool over tool for the sake of aesthetics and “just because.” Minimalism dictates one using the minimum amount of aesthetics to get a singular job done.

A blog such as TechCrunch is for the purveyance of information, not for the aggregation of randomness for the same of randomness. It seems that sites have lost this, and we’ve entered the arms race in terms of data heterogeneity. An arms race littered with Javascript, browser freezes, 6 second load times on an OC-3 and more cross-domain scripting than a script-kiddie vbulletin party.

Here is my proposal:

Stop.
Refocus
Think purpose outward.

Please.

This means: take your blog/homepage/etc. Remove anything that is referencing extra-domain Javascript, third party utilities or random cruft. Strip your blog down to its content and then sit back and go through these steps:

1) Given the context of the site, what do people need from it?

a) Meta-information about the author?
b) Meta-information about the subject?
c) Meta-information about the site?

Make a decision on that, and act.

2) Given the context of a post, what do people need from it?

a) A way to share it?
b) Inline contextual information?

Make a decision and act accordingly.

3) Given the context of the author, what do people need to know?

a) Status updates?
b) Meta-information about their physical context (location, photos, music?)
c) Biographic information?

4) Given the context of monetization, what is the least obtrusive means, and most congruent with the aim of the site?

I think if we all did this, our sites would get much better. I did it already to some extent, removing the Twitter flash thing. I stopped using Twitter, so it was rather stale. I’m also doing this on other sites I own and we’re applying these methods to Warner Bros. Records sites.

So my call for this week:

Strip your site down
Refocus
Reimplement.

  • Trackbacks are closed
  • Comments (2)
  1. I agree with you – as with the whole semantic markup thing – the idea of creating a website that has the presentation and information in separate files – all this ‘web2.0 bling’ is re-cluttering the information.

    I really hate the snap-shot links I am seeing on a lot of sites now. They are pointless. If the link text doesn’t inspire me to click the link, a picture of the site isn’t likely to either.

  2. Excellent post. Well reasoned, great conclusion.