Math Question:

I have two sets of data. One is for carbon emissions from various sources. The other is for net sequestration of carbon in forests and so on by biological systems.

The data points for the biological systems are, at best of equal precision and accuracy compared to the emissions data. Good arguments why they are worse, but they might be the same. Nobody thinks they are better.

OK but there is another difference. The emissions data for a year is meaningful in itself. If so much was emitted, that is what was added to the atmosphere for the year.

But forests, and the plant canopy of ecosystems are dynamic. Carbon content varies from year to year, from season to season, often even from day to day. So a snapshot, or even a series of snapshots over a year does not tell you very much. If you want to know now sequestration is affecting carbon concentration in the atmosphere you need a trend, constructed from data over multiple years - three or four years at minimum, probably much longer. If you wanted a meaningful net emissions number, emissions - sequestration, you would have to look at the trend over the past three or more years and allocate the current year's share of that result. My first intuition is that this is not a great thing to do if there a practical alternative way of figuring out whatever you expect to learn from this net number.

My second mathematical intuition is based on the fact that sequestration data is noisy. There are lots of factors that influence sequestration - temperature, water, pests (from the forests view humans are pests), fires, storms and other disasters. So my intuition is that given the noisiness of the data, even if the data points are as precise as those for emissions, picking the equation for the trend lines is going to less precise than the emissions data - either higher error bars or worse confidence levels.

So now my question: it this kind of qualitative information sufficient to draw the conclusion I'm drawing? Or do I need to do rigorous mathematical analysis of the data sets before saying that the sequestration trend is less precise than the emission data?

My backup drive just died. Pretty much to the day when the warranty expired. I'm groaning but I'd better get another. From what I've seen prices for .5-1 TB external drives cluster around $75, with about 20% significantly cheaper, 20% significantly pricer, and about 60% in that range. I don't want another drive that will last only a year. My computer only has USB2 not USB3. So: what do I get that will last. I don't want to pay more than needed, but would rather get one that is sturdy, reliable and will last than save $20. Anything in the $75 range that you would rec? Anything significantly cheaper. Reliability, sturdiness and will hold up are my first priority, Price is my second. speed would be my third except there is a limit to how fast UBSB2 can go.

This 1 TB drive is well-reviewed on both Amazon and Newegg and is only $70 at Amazon.

OK ordered. WD does last, and it is not as tiny as the last one, so hopefully that means sturdier. Thanks muchly

I am thinking of getting my mom an mp3 player and a subscription to audible. This would be her first one- does anyone have any reommendations that are super easy to use AND have larger print (my mom is 69)

This is a shot in the dark, but has anyone used cURL and/or openssl on a Windows machine? I have a cURL command that works fine on OS X but fails due to an SSL issue on XP.

The error I get is:

SSL certificate problem: self signed certificate in certificate chain.

[link]

That said, I'd be interested in why the XP machine has a self-signed certificate in the chain when the Mac doesn't.

Hmm, Time Capsule has needed a couple of reboots in the last six months for backups. Normal but random that it is more frequent? Or am I possibly getting to place where I might be needing a replacement?

Wireless, automagic backups are kinda brilliant.

But I don't want to need to replace it yet. The new ones are tall and the cat loves to curl up on the current squat one. It made of warms.

[link]

That said, I'd be interested in why the XP machine has a self-signed certificate in the chain when the Mac doesn't.

I tried that earlier and it still doesn't work. Using the -k or --insecure option makes it prompt me for a password and then I still get the same error.

I think there's something wrong with openssl on my XP computer. I was originally using Ruby to grab stuff from a webpage--this also works fine on OS X but fails with an SSL error on XP.

eta: Oops. Was using the -k switch at the wrong place. So I think I've gotten past the "self-signed certificate" problem but I'm still having another problem (that I didn't on OS X) which might be related.

eta²: OK, I tried a slightly different syntax on XP and got it working.

Does anyone have an online photo printing service they like that can import from G+?