For this reason I truly appreciated the way that Halo 3 handles this. It keeps track of your completion and difficulty on a level-by-level basis, including playing single vs. co-op. That way, you still get game completion credit for having played every level, but only the the "hard difficulty completion" achievement if you played all of the levels on hard difficulty. So at the end of the game if you played two levels on normal, you can go back and just play those two levels on hard to wrap things up.

I completely agree with you and have often thought the same thing. Difficulty in multi-player games is "technically" a different challenge than dynamic difficulty in a single player game. As previously mentioned, single player dynamics can be easily, if not poorly, adjusted by increasing enemy count, etc. But multiplayer adds the dimension that you have to adjust the ratio of difficulty between two players. Something should be done to put both teams on even ground. Whether that's through ramping up health or biasing weapon spawns to the losing team, or just making the best players stand out more by glowing or something, there are things that can and should be done to make it easier to have a fair fight when a group of players get together to game. We've already seen this sort of thing starting and we have no farther to look than Rock Band / Guitar Hero that allow players of multiple skill levels to play at the same time.

GCD and OpenMP have very little in common. OpenMP is a language extension. It requires the programmer to understand what environment their program is going to run in, what variables can be shared and how, etc. GCD merely asks you to identify blocks of code that are independent and it handles parsing them out to threads, variable replication, etc. It's the difference between providing detailed blueprints of a car (the OpenMP way) and just saying "I want a car" (the GCD way). You can *almost* think of GCD as a user-friendly frontend for OpenMP.

Why is everyone so caught up on the hot swappable feature. I mean, I get it, it's convenient to hot swap, but that's gotta be a terrible price / performance penalty. First off, convenient hot swap requires real estate, i.e., you have to be able to access the drive from the front of the rack which means you won't fit 45 drives in a 4U space (not 3.5 inch drives anyway). Assuming they have massive redundancy (which someone using a system like this would), it's not that big of a pain to power off the entire node, slide it out, swap a drive, slide it in. That just took you what, 4 minutes? Hot swapping requires a lot of faith in your OS (raid and filesystem subsystems) and controller cards to handle that situation gracefully and reliably (which is why you pay a lot for a box from Sun/HP that "guarantees" you have that ability).

That's what I was thinking. In many retail businesses (clothing I know for sure), there is almost always a 100% markup over wholesale. Perform the same analysis on a Monster RCA cable and tell me what their margin is ...

You left out the #1 reason to shop online which is what the article relates to most: there are no sales people when you shop online. Amazon has never given me an obtrusive pop-up about an extended warranty or lied to me about being in stock.

The only real downsides are 1) if you want to return it during the return period you have to ship it (pain as far as I'm concerned), and many online stores now don't have a price guarantee for the first 30 days like most brick and mortar stores (Amazon cancelled their program almost a year ago but will still honor it once per user if you email them).

I, too, vote for spotlight!

Agreed, though there have been several instances where OpenOffice would not start until I removed the entire preferences directory and started over.

While you are correct, this thread was specifically talking about the command line apt-cache search and the like. The GUI's are an improvement, especially recently, though I believe Synaptic is still just as ignorant, but I don't have an up to date Ubuntu system to try it on.

Out of curiosity, do you know if most Windows applications also remove your personal preferences from the hidden directories in your home folder? It seems to me that they wouldn't though I guess I remember a few specifically asking me if they should do so ... just curious.

The problem of storing user preferences has always been a pain but I do prefer their non-hidden state on the Mac to the hidden "Local Settings" folder on Windows (or even worse, in the registry).

I concede, my information was based on my impression from a few years ago. I just checked my latest Fedora install and noticed that their default menus have dropped the names of the actual applications entirely, instead just saying "Spreadsheets", "Video Player" and the like.

I was just remember my earlier Ubuntu installs which (and maybe my memory is wrong) presented me with menus containing things like "File Roller", "Pidgin", etc. whereas a fresh install of Windows comes with fairly simple but clearly titled applications like "Paint", "Internet Explorer", "Notepad". You get the idea. But it looks like that area has been addressed by at least the most popular distributions so kudos!

Your point is certainly valid and I admit that I didn't put enough thought into it before I wrote. Though you do admit that the names we see in Windows are much more familiar to us as many of them have been around almost as long as Windows has. Though I have to think that there is plenty of room for improvement both to icons and names to help linux adopters along. SOME distributions do add the extra description to the menu items. I seem to recall Mandrake used to do that with their KDE menus back in the day (using the optional program description field so that they could be turned off once you were familiar with them).

I guess I'm just stuck in a wishful thinking pattern of saying, if you're willing to concede that Amarok is to be officially bundled with KDE, then can't we just rename it to KDE Media Player so I'm not left wondering what a blue fox icon will do when I click it and people don't laugh at me when I pronounce it wrong ;-)

As a side note, I personally don't suffer from these problems cause I've been around for a while, since RedHat 6 or thereabouts. I believe that GAIM made more sense than Pidgin since it at least retained a common Windows name inside of its own.

The problem is that by and large the "search" functions of package managers are completely useless. Suppose I'm searching for an instant messenger client. Coming from a Windows machine I'd use my "tech saviness" to search for "im" or maybe "im client". The result (performed using yum search on a fedora machine) returns 4523 programs. I'll give you an example entry:

"ftplib.i386 : Library of FTP routines"

That's odd, I searched on "im client" and it returned an ftp library, along with 4500 other useless entries and are seemingly unrelated to my search. It turns out it does that cause it breaks my two words up into two separate searches so it was actually "client" that returned the ftp result, even though "client" doesn't appear in the package name or short description.

  Now I know that I really want to install pidgin but what on earth would I ever search for that would return that? It turns out that even searching on the quoted string "instant messenger" won't return pidgin because the description uses the term "messenging". Sigh.

What bothers me about this situation is exactly what the author says in the article, recommending that people don't install updates for packages they don't recognize. So a major security hole gets patched in X11/whatever and the end-user reads this article and says, "shit, I don't know what the hell X11 is so I better not let it install that on my computer".