Comment: Re:Benoit Schillings is the Chuck Norris of code. (Score 4, Informative) 132
You're thinking of Dominic Giampolo. Benoit wrote the App Kit and tons of good bits in the rest of BeOS but he didn't have much to do with the filesystem.
You're thinking of Dominic Giampolo. Benoit wrote the App Kit and tons of good bits in the rest of BeOS but he didn't have much to do with the filesystem.
A socket adds a significant amount of parasitic capacitance and inductance. At high frequencies, this can cost quite a bit of power. You're right about it also contributing to space and cooling issues.
BGA has among the lowest parasitics of all IC packages so it's not surprising to see it everywhere nowadays. Except for how hard it is to desolder, it is an awesome way connect ICs to PCBs.
Contact area (wire size) doesn't really matter... Your connection needs to be big enough to handle the worst case power and no bigger -- any extra metal is just wasted (to a point, but let's not get pathological discussing running 1 GHz signals down 0 gauge wire). The path of the signal tends to be much more important than the wire size.
Back in the old days, whenever our peering ratio started getting wobbly, we'd just set up NNTP servers and have them hammer away (either downloading or uploading, depending on what direction we needed to move the balance).
I assume ISPs are still doing this but they're probably using BitTorrent now instead of NNTP.
"So again, what was Linux hoping to achieve by dropping old "obsolete" OSS in favor of increasingly complex solutions?"
Linux deprecated OSS2, which everyone agrees sucks hard. It was a no-brainer.
OSS3 is significantly better but it was only recently open sourced. Frankly, if the OSS devs hadn't spent most of the last decade with their heads firmly wedged, audio on Linux would probably be in a much better state. Ah well.
"Blue Rubicon attests that 'if you do them right you can make a material impact on people's behaviour.'"
That will certainly make a material impact on Blue Rubicon's net profits. But change people's behavior? That's pretty unlikely. History is not on their side.
True, you need to "git config --global core.autocrlf false". But that's only an issue if your files have embedded \rs. Most don't.
Git has no concept whatsoever of file permissions either except for the execute bit. To control it on Windows, use "git update-index --chmod=blah". No big deal.
Nope. Visual Source Safe is a horrible hack -- VCS extensions on top of a file sharing protocol. It's such a disaster to use that Microsoft won't even use it internally.
They'll sell it but they won't use it. Gotta love Microsoft.
If someone else is in the same coffee shop as you, use gitjour: http://github.com/blog/75-git-over-bonjour
If not, just push your repo up to a machine that does have a static IP. Easy breezy.
Last I used Mercurial, I couldn't create a feature branch. I had to clone the whole damn repo. Well, either that or I could create a branch that lives forever, not a good idea if I'm doing a crazy experiment.
In Git, if I want to try something out, I create a feature branch (takes basically no time or disk space) and hack away. If it sucks, I just delete the branch. It's a very nice way to work once you get used to it.
Have they improved branching in hg?
Why stash / unstash before pulling master? Just commit your changes to a feature branch.
The only scary part of your workflow that I can see is the rebase. And, yes, rebase is a bit of an odd child. I hope that the Git team spends some time making it more more predictable and easier to use.
I request a weekend in Havana with Phil Silvers!