pors

Yeah fitbit is great. If you follow a calorie counting diet it helps you by adjusting the allowed amount of calories based on how much you move. It is also an amazing motivator to get off your butt.

The problem with visualization of traceroute results is that it needs a very reliable GEO-IP database. They exist, but are very expensive. Hops that are not in cities (but in some ocean) are often placed in the city of the owner of the router. With the rise of IPv6 this only becomes worst.

Well, there is also something as "state". You don't want "method=delete" to work as POST either, unless the client is authorized. Same with GET. I don't see why POST would be better than GET really.

How do CDN's solve bad connectivity of your app server?

Linux isn't a language either, but they have Linus since forever.

Yeah one of the founders is a legendary hacker: http://en.wikipedia.org/wiki/Rop_Gonggrijp

I think the reason for this story not being picked up is that the tech blog writers understand that there is no demand for such an app. 99% of the people don't care to sign up for a service like MSN in the old days and WhatsApp now, and everyone is on Facebook right?

Very long article about 99 designs, with exactly 0 links to their site. Why? Some outdated SEO advice?

About dropbox: use it in combination with this and google can't index anything: http://getsecretsync.com/ss/

But yeah, I agree with your letter. We used jot.com (paying customer) and after the Google acquisition they left the product rotting with zero support, and turned it into "Google Sites" (yuk) after a couple of years.

I agree, to-do lists don't work, and the proposed agenda thing won't work either (not for me at least). I am hooked with GTD (https://en.wikipedia.org/wiki/Getting_Things_Done) and use http://activeinboxhq.com/ with gmail. It sounds like an advertisement, but this method really works for me and gives me control over my workflow (100s of emails per day).

Well, that is what the article claims, they compare their own stats with comScore and Alexa stats. But I think the conclusion is indeed flawed as the Flurry stats include people with smart phones with apps on it that use Flurry, which is just a subset of all people using the Internet. They should include the absolute numbers to make the story complete. But yeah, there is a strong trend we can't deny :)

And this: http://www.joelonsoftware.com/articles/fog0000000069.html

It's very clear what I'm supposed to do: +1

I will never ever enter my email address without knowing what it is for and what you are going to do with it.

I don't say "persist without looking where you are going", of course, pursuing a bad idea for 20 years won't work out. I can't back it up here on the spot, but if you study a bit you'll find that most success stories were no instant success. It became a meme for a reason, same thing about having a good idea. Just a good idea is not enough, but it certainly helps.

Plancast: Two years is way to short to draw conclusions IMO. Why don't they address the user feedback? If they want to reach the broader audience, why not change the definition of "Event" into something we do more often? Etc.

Skills are highly overrated I think, it needs persistence above all to succeed with your startup. E.g. when I read this http://techcrunch.com/2012/01/22/post-mortem-for-plancast/ my first thought was: he gives up too fast (very interesting comment by Scoble also). I don't say you don't need skills at all, but if you never give up you will pick up the skills along the way.

AGPL was made to also catch "hosted software". It still needs you to modify the OSS before the license kicks in. It has not been tested in any legal case as far as I know, which is why large corporations avoid using AGPL licensed software (like MongoDB).

cool, thx!

With such a long list of investors (http://www.crunchbase.com/company/readfu) it must have been quite a large sum of money to acquire Summify. So IMHO not just a talent acquisition.

#6 Go Parallel

Uhm yes, that's the whole idea of using node isn't it? Handle each job that takes a bit of time asynchronously. Pretty obvious.

> Recently, a couple of Harvard graduates launched a program called Gym Pact based on the simple principle that if skipping the gym is a broken contract with ourselves, we ought to pay a penalty for slacking.

Hmm, this site does this for many years already: http://www.stickk.com/

Tried it once with some friends and it worked perfect, even without a financial penalty.

Wow, just scrolling to the bottom if this article took ages. #sent2kindle

Cool logo, how did you make that? Is it a font?

A standard is a good thing, because people build tools around standards, that will inter-operate.

It starts exactly at 1/1/2012 it seems, looks like a measurement problem to me

Very nice! You could improve it by filtering out accounts that follow a large number of others. Many of the real VIPs have many followers, but don't follow that many themselves.

Yeah this is only about office software.

I find this advice way too generic, misses the real insight how to get coverage etc.

I think this guide needs some general explanation of what git is, and what the thinking behind it is (like here: http://progit.org/book/ch1-3.html).

Of course you can get started with this guide, but you have no clue what you are doing. It's learning by following the recipe versus learning by understanding.

Interesting idea! I wonder how they think to address performance (one data query needs many requests to different nodes if I understand it well) and availability (if one node is down, I can't access my data?).

Probably true, but there's always exceptions (think Steve Jobs)