There's a 3rd performance metric that I'd be more interested in seeing -

Set both clueless users and experienced (up to XP ideally) users some "typical" tasks to complete (copy a file, change some preference or other, configure network settings) and see how this compares to earlier windows versions/other OSes.

This means doing things like disabling a network adaptor for instance and then spending time trying to work out where the hell it disappeared to from the network connections window.

This is a far better indication as to the "worth" of the operating system imho than an utterly pointless "boot to desktop" or whatever. Those "boot time" stats are worthless as they're all (relatively) easy to overcome by the use of either SSD or (cache-derived) "instant on" solutions which will inevitably become mainstream in any case. What really matters is whether (for example wrt the UI) redesign actually makes the OS worse from a productivity perspective or better.

It'll take longer than you might think if the original code was written in AS2 and not AS3: There are several _critical_ shortcomings in AS3, most notably the inability to control timeline objects from code reliably as their creation order is somewhat haphazard, and often delayed by several frames from the creation request - this makes the ability to develop rich content extremely tricky using _flash_ and AS3 - the only real way is to manage the objects yourself, and that means forgetting about the timeline - in which case you're more accurately talking about _flex_ and not flash.

I, for one, would be much happier to have heard that Adobe had chosen to actually address some of the existing concerns with Flash 8/CS3/CS4 before rolling out yet another abomination incarnation of AS3 scattered across a wider range of devices :( [And no, they haven't addressed the issues afaik - they still exist with "No Priority" on their bugzilla database]

$10k? Surely it means $10k per CPU-C (Copy Published of Unverifiable Claim) ;)

I'll get my coat...

>> I fully support any foreign governments that choose to ignore the draconian DMCA.

I fully support any foreign government that chooses to acknowledge that US law is only pertinent within US jurisdiction, including and especially the draconian DMCA. I have nothing but total contempt for any foreign (or US) government which thinks that its laws apply globally.

DT

I care greatly about iTunes when considering an MP3 player - if iTunes is a pre-requisite, it's not the device for me!

I make no apology for not wanting a music player/downloader whose setup file alone is in excess of 70 MiB, nor do I make apology for not wanting to have to run rings around myself to purchase MP3 music without DRM encumberance at a fair price.

The straw that breaks the camel's back, however, is the issue of interconnectivity - Choose an Apple device because you want to sync to iTunes? Now you can't easily (yes, I know it can be done, it's just not as simple as it ought to be) transfer those tracks from the iPod back into a "conventional" player, not to mention the awful renaming, filing and other unhelpful ways the tracks are transmogrified by iTunes. Buy your tracks on iTunes? Now you can't easily sync them with A.N.Other-Device.

The arrogance of Apple here is staggering. Yes, you can buy your music from us.... but we'll tell you how to play it. Yes, you can buy our player.... but we'll decide what you can do with it. And regardless of whether you buy player or music.... you'll have to install our crapware on your computer, whether you like it or not! Totally unacceptable, and one of the driving reasons why I won't buy an iPod or iPhone (although for the iPhone, there's many far more pressing reasons why not!), and why I won't be herded/goaded into buying my music tracks from iTunes - I'll always choose a service that allows purchase of plain-old-MP3s at a fair price without the need for many megabytes of bloatware, and without feeling like I must buy a single brand of player on which to play standard files.

If that music isn't available elsewhere, then I won't be buying it, plain and simple.

DT.

For full-blown development, I use Embarcadero (formerly CodeGear/Inprise/Borland) RadStudio - There's a free version "Turbo C++ Explorer" which is a full version with the caveat that you can't create/install custom components for the Visual Component Library (VCL).

I've also extensively used Dev-C++ in the past - whilst I do like it, it often crashes, especially when performing Code Completion lookups on large projects. That said, the exception handling is fairly robust and I've never yet lost anything due to it crashing.

The C/C++ support in NetBeans is excellent - Code Completion works well, and the environment is very comfortable to work with. It also produces Makefiles for you enabling you to build your project easily outside of the IDE. I've never tried the Eclipse plugins for C++ as I find Eclipse too cumbersome to use - it has way too many features, often presented at the "top level" of the UI where the features you actually want are buried in nested menus/dialogs as opposed to pretty much all other IDEs I've used where the features you need are found comfortably to hand. YMMV of course.

Lately, I've taken to using Code::Blocks - It's by far my new preferred "small and simple" IDE. It's a fairly small footprint, available for both Linux and Window (possibly Mac too?), and most significantly for me doesn't require a Java install. I'm not convinced that it's easy to reliably generate Makefiles from it, but I haven't really tried all that hard. The Code Completion is good, as are the Symbol Browser (Class navigation) and GDB integration, and there's a number of plugins available to perform other tasks including profiling and suchlike.

Macs are the main competition to Windows, not Linux.

While this may be true, I would like to think that schools who "go down the Mac road" would be held accountable for that decision by those who finance them. I can't conceive of any justification for purchasing significantly over-priced hardware (in comparison to the relatively cheaper and virtually identical vanilla PC hardware).

Whether you like it or not, Microsoft pretty well dominates the commercial sector. Training people to use Mac products, whilst arguably an improvement over a Microsoft-lock-in, isn't going to be either the most appropriate use of funding and in one fell swoop limits the sheer quantity of available software to run on the chosen OS (although that's arguably a good point!)

The issue of cost (or more correctly value-for-money) leaves only one clear candidate, and that's Linux. As with Mac, the choice of software is limited - whilst Linux has (far too big) a selection of software available, I'd argue that the quality of much of the software (outside of the "key" mainstream apps) is somewhat limited: That said, Linux has more than enough (as does Mac) high-quality, feature-rich mainstream apps (such as Firefox for web browsing, a port of Adobe Flash, a plethora of Multimedia apps, OpenOffice, The GIMP etc.) but delivered in a value-for-money fashion (i.e. most of this software is free and well-supported by the community).

I'm not grumbling per se - I use Linux myself (Linux Mint if you must know), and am an advocate for Linux where possible. However, although Linux has matured over the years, it is still not quite (imho) "ready for the desktop". There are a number of design fragmentation issues that need to be resolved (choice of desktop manager, multimedia software, configuration toolset etc.) - most end users simply don't care what desktop manager or MP3 player they are using, as long as that software is stable and of good-quality both from a performance and appearance perspective, simple-to-use (i.e. intuitive), compatible with most (if not all) other software (in the case of desktop managers, for instance), and of sufficient caliber to insulate them from the intricacies of the underlying OS (eg by including adequate configuration applications, plugins, codecs or whatever).

For me, though, It's good to see each and every instance where Linux is gaining a foothold - not "just" because it's Linux, but because it represents an informed choice of a value-for-money OS, and also because it goes one step further to aiding in the catch-22 of "Linux for the desktop": Increased market penetration will (hopefully) lead to improvements, and improvements will lead to increased market penetration. As a parent in the UK, It would be nice to see Linux gain a significant share in schools here too: As a taxpayer, I am effectively contributing to their funding and take great exception at the amounts of money being spent on proprietary technologies that are, frankly, unnecessary. The money would be greatly appreciated and of use in other areas, and Linux is a credible solution to liberate these funds.

DT