Digital Lifestyle

yeah one of his recent email quips

Not sure where to put this, but:

On release day, I got an 17" i7 MBP.

It seemed fast, but still, some things were slow, especially *anything* while I was working in Parallels.

So, I upgraded my boot drive to a 256GB SSD (Crucial) and maxed out the RAM to 8GB.

Got the MCE Optidrive kit and plopped a 1TB media drive in.

My system is officially as fast as it wants to be now. By which, I mean that it was obviously bottlenecked on disk access, especially when in Parallels.

The thing is so fast now, it's unbelievable.

I will never go back to spinning disks for my boot drive, if I can help it.

Interesting. Thanks for the info, Tom.

Spinning disk is known to be the bottleneck in so many applications; what's surprising to me from this is the amount of reliance on it in the Mac OS. Almost like swap/mapping to disk in volume regardless of whether it's resource necessary or not.

I have firewire drives attached to my iMac. I just love the first thing in the morning (or any time after say 30 minutes of inactivity) when I check my email (or visit www.muppetporn.com or whatever) and then have the spinning beach ball of death for 15 seconds while the hard drives spin up (all the while I'm banging the side of the machine yelling "faster!")

You could turn off "Put the hard disk(s) to sleep when possible" in Energy Saver.

You need to flip them over, since you're in Australia. :)

New iMacs allegedly coming; I'll bet that the odd, bluetooth-compatible 3" touchscreen we've seen in the FCC filings is introduced in some form with the new iMacs.

That odd bluetooth multi-touch interface is gonna come in handy for my home theater.

Adam, I was thinking that about the Magic Trackpad as well except you would need a hard surface for it to sit on in order to click.

Quote
Cloudscout
except you would need a hard surface for it to sit on in order to click.

NTTAWWT

The Roku XR is on Woot today for $69.

Perfect timing. I wanted to buy a second one.

Netflix streaming to devices appears to be broken. It still works for computers but not for Blu-Ray players, Roku Boxes, Apple TV devices or PS3s.

Steve is probably going to have a fit since a lot of Apple TV users will complain to Apple about the outage.

Oops. Seems like IP lawyers looking into Android are finding a lot of Oracle's proprietary Java stuff.

Her entire argument is based on the following comment in the source:

* Copyright 2004 Sun Microsystems, Inc. All rights reserved. 
* SUN PROPRIETARY/CONFIDENTIAL. Use is subject to license terms.

She implies that this means it could not be distributed by anyone other than Sun. But that's not what it says. It just says that it is subject to license terms. It doesn't say what those license terms are.

I'm not saying her conclusions are wrong; I'm just saying she doesn't have enough information to draw the conclusions she presents. Someone would have to dig into the commit history of Harmony and determine who supplied that code and whether compatible licensing terms were obtained first.

Regardless, it seems like the Apache Software Foundation was a bit clumsy in their acceptance of code without maintaining clear copyright details along the way.

A little bit more research and it looks like my analysis was a bit off-track (the files in question were never a part of Harmony)... and it also looks like this isn't exactly news, either.

This is merely revisiting issues brought up back in November.

So, to amend my last post, it would be a good idea to figure out who supplied that code to the Android project... and when. If this is a case of decompiling a binary then mentioning the copyright warning isn't entirely relevant since that statement was from the source code, not the binary which was, ostensibly, decompiled. That doesn't make it any more legal but it does change the presentation of the issue.

It's also the one thing that open source advocates and opponents bring up... how do you guarantee the purity of community-supplied code? It's also a two-way street. Even Microsoft was found to have misappropriated some GPL code in their Windows USB/DVD Tool. The fact that the source code for most commercial software is not available for public inspection makes it exceedingly difficult to identify those kinds of violations.

It's why software patents are a horrible idea. I doubt any large open codebase is completely uncontaminated by "borrowed" source.

I really don't understand this argument at all. If all the Oracle-owned code is unit testing stuff (and assorted other never-linked libraries?) then how is there an argument that Oracle should be getting a payoff per Android unit shipped? Does using a pirated or unlicensed compiler or IDE or other toolset really mean that everything you compile with it becomes liable copy by copy, as though each product of an infringement were itself a new infringement? Patel says "the state of our current copyright law doesn't make exceptions for how source code trees work," but surely there's an obvious distinction between things that are part of the shipped product and things that are not.

I think it is more of a case of shaking the haystack and having 37 needles fall out. You can say that the files weren't included in shipping code, but there weren't supposed to be ANY. My guess is that Oracle has found at least a few that did ship, to be so aggressive about it. Who knows, though? SCO was/is aggressive with no basis at all.

I don't think it's that. Google distributed the code in violation of the license. They didn't distribute it with the devices themselves but they did distribute it. That means they would, presumably, owe Oracle something. I think the idea is that Oracle is going to try and strong-arm Google into a settlement. Oracle will try to pain the settlement as something that relates to actual devices shipped in order to get a sort of psychological "win" over Google.

Google won't settle on this one, though. I bet they take it to trial to let a jury decide what actual "damages" there may have been. Given the fact that the code itself was released under the GPL, it's unlikely they will be able to claim any significant monetary damages. Without a settlement, the best Oracle could likely hope for would be a symoblic win in the form of some kind of restraining order. This would only be symbolic since Google has already pulled the code.