Webomatica - Latest Comments in 1,000 Spam Comments

Re: 1,000 Spam Comments

webomatica — Tue, 23 Jan 2007 18:59:24 -0000

Sheeez.... 100,000 spams a day? I'm surprised the internet doesn't collapse under the weight!

Thanks for the nice comment regarding technorati. Keeping track of the progress gives me something to watch over, kind of like magic rocks growing. I think there's something about people interested in tech where it's fun to track proress of stuff that makes others' eyes glaze over.

Re: 1,000 Spam Comments

engtech — Tue, 23 Jan 2007 18:07:16 -0000

I deal with the spam by using Akismet and outright blocking .info and .biz. Have to check comments once a day, but not that bad. Real popular sites get like 100,000s a day.

I have no doubt that you'll do well on Technorati. You've got a really good blog going on here.

Re: 1,000 Spam Comments

webomatica — Mon, 22 Jan 2007 16:34:43 -0000

Pretty sweet, engtech. If this blog ever gets close to your position I'll chalk that up as evidence we're in a Web 2.0 bubble :) But if that means 4000 spams a day - sheez, how do you deal with it? Delete delete delete?

Anyhow, thanks for the MyBlogLog info - I need to fix that.

Re: 1,000 Spam Comments

engtech — Mon, 22 Jan 2007 16:24:45 -0000

Technorati Rank is really easy to acquire until the sub-10,000 range. Once you get beyond 5,000 you pretty much stop checking since it changes in a positive fashion so infrequently.

I have some old stat pages from when I started. Kind of funny, looking back. http://engtech.wordpress.com/tag/stats

I'm up to 4000 akismet spams a month now. The !@#$ers.

One solution to slow MyBlogLog loads is to use the image mechanism instead of javascript... since it will just show empty space until they're loaded. http://engtech.wordpress.com/2007/01/15/myblogl...

Re: 1,000 Spam Comments

webomatica — Mon, 22 Jan 2007 14:19:03 -0000

Hey dave... yeah... the alexa thing is wacked... as with you guys, the Google Analytics and my own server logs had me scratching my head.

In an ideal world, sites would move away from page views and something more tangible like user satisfaction, unique users, user loyalty, or time spent on a site.

Then there is the issue of more sites getting into the ajaxy type stuff which can make a site more useable but reduces page views. I'm all for usability at the expense of pageviews. But for some dudes wanting to keep the page views up, they might decide against usability.

The best example: sites that break up "top ten" lists into ten web pages so you have to click on each one to see the next entry! That's annoying. So they might get more page views out of me but as a user I have a bad impression of the site.

Re: 1,000 Spam Comments

David — Mon, 22 Jan 2007 14:05:06 -0000

Oh and congratulations on the 1000 spams and the techcrunch ranking :-)

Re: 1,000 Spam Comments

David — Mon, 22 Jan 2007 14:04:30 -0000

Yeah Analytics is pretty nice. That's what we're using for metrics tracking.

I totally concur on Alexa. It's a joke. Bottom line. Maybe it's valid for the top 1000 sites, but I honestly don't see how it can begin to track the breadth of sites after that using a panel-based system. There's just no way. Smaller sites are just never going to register properly, especially when the panel is self-selecting (i.e. folks who download the alexa toolbar).

Our best Alexa rank is still from the day we got techcrunched back in August. But per analytics, we've seen days with 20x the traffic (uniques and pvs) of that day since, but haven't even come close to the alexa rank since. I'm sure others can corroborate.

But suffice to say, when I see folks quoting alexa numbers, I usually just roll my eyes. For most of us not in the top 1000, these numbers are pretty much meaningless.

Of course, this problem isn't unique to alexa. When I used to use comscore/mediametrix/nielsen etc. at Yahoo or Ziff the same problem existed. Bottom line is these types of companies are still using an old-school network TV solution (small survey group/panel reporting) for a long-tail web (which kinda defies small sample groups). I don't know how you fix it, but surely some upstart company out there must be saying...wtf? Why do we keep trying to track web usage in this fashion?