DISQUS

Two notes...

Even with relatime the atime data needs to be in another section of the disk so it will still require a write.

Also, you might want to consider the noop scheduler as IO schedulers on Linux (at least the current ones) don't make a heck of a lot of sense in the SSD world.

Kevin

You missed websitepulse...

http://www.websitepulse.com/

They do a FREAKING AMAZING job!

They can even physically call you if there's a problem with your servers.

An amazing number of really powerful features.

Kevin

Yeah.. I agree with Scoble.

First, due no harm.

It's one thing to write a negative opinion of a product to prove you're independent. It's quite another to actively sabotage presentations during a conference.

At this point you're not independent; you're a pathogen.

Good thing we have an immune system for these sort of things.

Markus.

I'm not sure if that comment was directed at me but for the record I'm very happy with the company I run :)

My point is, that if other dating sites our floundering, don't chime up and clue them in. Stay silent and let them crash around you.

Google only talked about their ad revenue when the had to go public.

Reddknight,

I agree, it might take some of the fun away :)

Of course we usually charge about 1/5th the cost of running your own spider so this can really help startup costs.

If we can help out in any way let me know.

I might also mention that we provide free access to researchers - which is WAY cheaper than writing your own crawler :)

Thanks for the feedback Aviv.

Yeah. We could have been to early for you back then. We're probably going to be expanding our crawler to do some back filing of archives to solve just this purpose.

Archival access isn't requested that often honestly.

Let me know if there's any other way we could help. Our content extraction and/or language categorization might help.

PS. This was the third time posting this. Mashable's comment for is crazy.

I want a DMCA takedown notice too!!!

http://www.feedblog.org/2006/11/michael_crook_p...

Are you kidding? I can't imagine having an acquisition discussion at 8AM........ Some people just aren't morning people. If I were to wake up at 4AM I might be able to function correctly by 8AM.

MS people are SMART and you don't want to be talking to them when you're in prime form.

Robert.....

It would look a bit like this:

http://tailrank.com/2006-09-04

Steve Irwin dying was huge news..... the biggest meme since we've launched Tailrank..... I suspect that another 9/11 would look the same.

Kevin

The problem is that the cure is worse than the disease. The leaker probably won't be asked to resign and even if he is he can just use this as distraction.

Dunn should be kicked off the board and sued...... what she did was a crime (literally).

Kevin

"... Are struggling to figure out how to get outside the TechCrunch/TailRank/TechMeme/Reddit/Digg/Slashdot/Om/Scoble bubble."

Whoah...... Tailrank is part of the bubble! Sweet! :-P

Tailrank picked it up......

http://tailrank.com/posts/562949953851330/Downtime

I think we were a bit behind that day as we had a cascading memcached failure which slowed our robots down.. ug...

Kevin

I'll betcha userplane was bought because AOL reads your blog ;) ... hint hint future sponsors! :-P

Yeah...... I've always been a big fan of the river of news which is why Rojo was/is a river of news (with folders) aggregator.

I still think its the best product out there but I'm obviously biased. What I find REALLY strange is why people still like bloglines. I think its the "it just works" philosophy which is very important to a lot of people. I'm starting to think this means more than feature set.

Tailrank also has a river of news style view ...... you can also adjust how much data it will show simply changing the minimum ranking at the top of the page. Pretty cool :)

BTW.... I should note that all of this stuff has a RESTian API so RSS aggregator developers could add memettracking directly into their products if they wanted to.

Contact me if you want to talk about it... I haven't had a ton of time to writeup or document the API.

Kevin

Just use Tailrank's custom memetracker support.

Create an account, upload your OPML, and Tailrank will only show items within the OPML file you uploaded (we also support additional mechanims like direct import from Newsgator or Bloglines).

We can then use this list to recommend even more blogs we think you'd like...

Kevin

is it me or did you not give them a link?

Yeah.. targeting the geek market is a classic mistake. I've certainly make it and continue to make it :)

That said I want Tailrank to be easier ot use for the mass market... Hopefully more updated and advanced UI will help that... which we are always working on improving.

There's also the advantage of "doing it well" ... while Google does a good job in China the local guys (assuing they can build a decent product) always have an edge up...

Language of course makes it a step harder.

There's also focus. Having someone KNOW the space you're in really helps.

That of course doesn't mean that Technorati couldn't compete in this space it just means that the ball is still in play.

Yeah..... plenty of puns to be had by all :)

Planet is pretty nice too... you give it a bunch of feeds and it generates a blog from the output.

Robert.

You're a GREAT guy and all... don't get me wrong - but you're beating a dead horse.

Its dead. Microsoft will have to go through a LOT of bad days before it needs to change.

Take all this energy and start your own company. You'll make WAY more wealth than you'd ever make at Microsoft and I think consumers would be happier.

Hell.... you already know what MS is bad at.

Start a startup that will end up getting BOUGHT by Microsoft.

Anyway...

A note about feature creep and everythingitis.

Enter the market with something THIN... do one thing right an do it WELL.

I def could be doing a better job with this regard in Tailrank but for right not it seems to be doing a pretty good job.

There are other reasons as well. You're engineers can specialize and get good at ONE THING.

Your users will also appreciate it as they don't have to learn a complicated app.

Kevin

I think its time to stop quoting Alexa numbers. They're all over the map. They're tracking no where NEAR what Tailrank's real numbers are. I mean its nice to see them going up but as Tailrank's Firefox market share approaches 100% my Alexa numbers will approach 0% ...

Oh man... this would have been a PERFECT joke... if only he could have kept a straight face.

Hey Robert...

The nofollow rel attribute would actually be fine for this. I'm not sure if Memeorandum supports it.

The problem I have with it is that within comments it might be nice to follow a nofollow attribute.

I've blogged about this in the past:

http://www.feedblog.org/2005/08/nofollow_consid...

"At the time it really bothered me since while it would in fact punish spammers it would also publish bloggers since legitimate links wouldn't be included within ranking algorithms (which yielded to nofollow). The problem was that at lot of people went storming through the gate with their support for nofollow."


...... I have a better idea though. Maybe we could have a central repository for bloggers so that thhey can link to an MSM site and note how reputable it is.

I've had problems with MSM rags in the past (including the print version of Wired (not the online one)) so I can feel your pain.

Onward!