dlopen(3) and its friends allows a C developer to dynamically link to symbols at runtime, the ClassLoader.forName() method allows a Java developer to dynamically load classes at runtime, and other languages have similar constructs. Runtime linking is an extremely useful feature for implementing a module system so in-house developers or 3rd parties can create new implementations of well-defined interfaces; Apache HTTPD is a good example of how to use this effectively. Runtime linking has a number of benefits, not the least of which is security: a user (at runtime) can disable large chunks of code to limit the attack surface.

The May 2015 The State Of Go talk addressed this feature in the execmodes document, specifically:

• Go code as a shared library plugin with a Go style API
• Go code that uses a shared library plugin

It states that neither of these are to be available in 1.5 and indeed they do not appear to be. What's your take on runtime linking? Do you consider it a priority and do you think it'll be available in the next release?

And don't forget that because this is a brave new world, dozens of different companies are going to compete for it. One or two will succeed, most will fail, and a lot of houses will be littered with IoT devices with firmware that is no longer maintained because its creator is now out of business. This is so going to suck.

Those guys get upset when we don't spell write.

Really, dude? Really!?!

How combining learning algorithms with community moderation?

Seems like reasonable speculation to me. Valve may be hedging against the likelihood that if a significant portion of their user base moves toward thin clients and cloud services then those users will not have powerful PCs. But, if valve makes a console with high-end hardware then they have a better shot a staying alive.

This whole shift back to thin clients is something a friend and I have been trying to keep tabs on. Gamers have been the driving force behind the huge gains in CPU overclocking and GPU performance (which we're using for other purposes) and if PCs die (as predicted by many) then I'm worried that the powerful CPUs and/or GPUs that are so cheap and readily available today will become expensive "server" chips.

Not likely. http://www.cato.org/publications/commentary/telecom-amnesty

Then what does it use Wifi for? (Not intended to be snarky. I'm actually curious.)

I don't think I want my pin broadcast across a (potentially) insecure local Wifi network. I'd much rather the analog solution in this case (a signature). Yeah, my server could potentially swipe my credit card and steal the account info, but:

• * since it's credit and not debit, I don't have to fight to get my money back. I simply tell my CC company that the charge isn't mine and let them investigate.
• * it is a lot easier to catch an employee stealing card numbers than an anonymous geek who hacks the WLAN from across the street.

What's more -- Aardvark doesn't routinely collect information from its users. Apart from the Google Ads, this site is a cookie-free zone -- I probably know nothing at all about you anyway!

I went to your site and Firefox prompted me to accept the following cookie:
name: font_size
content: 0
path: /
domain: aardvark.co.nz
Expires: End Of Session

Not a big deal, really, but it doesn't match up with your policy.

You've done the research? Care to provide a citation for your claims?

The 50 (sometimes 47) million number that is often quoted comes from SAHIE. A quick glance at the about page will show you that:

The SAHIE program models health insurance coverage by combining survey data with population estimates and administrative records. Our estimates are based on data from the following sources:

* The Annual Social and Economic Supplement (ASEC) of the Current Population Survey (CPS);
* Demographic population estimates;
* Aggregated federal tax returns;
* Participation records for the Supplemental Nutrition Assistance Program (SNAP), formerly known as the Food Stamp program;
* County Business Patterns;
* Medicaid and Children's Health Insurance Program (CHIP) participation records; and
* Census 2000.

If you want to dig deaper then checkout the data inputs section.

BOTTOM LINE: it is not a "mailin" survey or anything of the sort. 47 million is the best, educated guess of the number of uninsured based on data from a wide variety of sources collected in 2005 and compiled by SAHIE; 50 million if you look at the 2006 data.

Don't you mean that it's harder to find stuff Kwickly?

How about Sir Dr. Stephen T. Colbert, D.F.A?

A court (the legislative branch).

Nope. That'd be a prosecutor (acting under the authority of the executive branch).

int main()
{
int fd, fd2;
struct flock fl;

fd = open(TEST_FILE, O_RDONLY);

if (fd < 0)
perror("open failed");

bzero(&fl, sizeof(fl));

fl.l_type = F_RDLCK;
fl.l_whence = SEEK_SET;
fl.l_len = 1;

if (fcntl(fd, F_SETLK, &fl) < 0)
perror("lock failed");

/* fd is now locked, whoohoo!!! */

fd2 = open(TEST_FILE, O_WRONLY);

if (fd2 < 0)
perror("open failed");

/* fd is now @^%$#ing unlocked!!!! For the love of God, why!?!?!? */

return EXIT_FAILURE;
}