Well, given the pre-order price of the Gigabyte M528, which is to be released in 30 days, this is speculation, not a wishlist, per se. (The Gigabyte M528's price will be even lower than $700 USD--$689). I suspect Nokia, if it plans to keep it's price point on the NIT high, rather than dropping it down to a more realistic $250-$300 USD range, will have to implement beefed-up specs to compete with the MIDs, which are starting to come out. Personally, I think they could compete better by keeping their device lighter and less expensive.

A pipe symbol on the keyboard, please!

BlueTooth does the job for me. I don't see why IR is required, and stating 'IR does some things BT doesn't' is a useless remark because it doesn't state what IR does more/better/whatever. The N96 won't contain IR support either. I think IR support on gadgets is on the way out. I use IR to open my garage, and frankly, the distance is too big. With BlueTooth 2 or later it'd work fine over 100 meters which is more than enough time to open the garage before the car is there.

Zooming is awesome. The reason is that, like with fullscreen, the user is able to commit total focus to the application (or part of the application) the user wants to use. Besides being able to zoom in, the ability to zoom out is also useful. The first time I used it was on a WM called 3ddesktop, and I also used 3DWM back in the days. With software rendering though. Using the arrow keys one was able to zoom out from the desktop and zoom in to a different desktop (virtual screen) and continue other applications there. Nowadays, there are many research projects behind us implementing this, and also serious implementations.

Compiz (and QEdje (Evas is faster than Xorg)) provide a lot of interesting features. Like a cube task manager. I already gave some examples of minimizing and maximizing, but there are more examples.

I used Spotlight the other day, and instead of having desklets/screenlets/gadgets/widgets/applets/whateverets on the background on the desktop you bring them to the foreground with Spotlight, plus transparancy. You can see this kind of functionality back in the iPhone. It allows total focus to the specific task the user wants to execute. Also take a look at Expose. You can also see similar (and perhaps better) features like this in Compiz. Imagine using the cube as task switcher on the NIT.

Sun implemented with Looking Glass something similar as Compiz in Java, but AFAICT it never took off. I remember in the demo by Jonathan Schwartz in which he shows RealPlayer with the ability to turn RealPlayer 180 degrees and see the video in mirror mode (left is right, right is left). He also showed the ability to put notes on the back of the application (in that case, RealPlayer). OS X uses the same idea in Spotlight to configure the widgets (if that is what they're called). Every applet has an 'i' on the bottom right. Click on it, the applet turns around, and you can configure it.

If Nokia keeps the NIT simple (hardware-wise) to cut costs then competitors will be able to make a clone more easily. The advantage Nokia will have is experience in open source software development, Qt, Nokia Maps, multimedia integration, and years of experience with mobile phones. If Nokia keeps innovating they can use the previously mentioned advantages to earn more money while staying ahead in the technical sense provided the devices are ahead of the competitors. For this, the hardware has to be top notch, too. Ofcourse such comes with a price. That is normal. But new features are important. For example, a useful innovation in the iPhone 2G was the WiFi location positioning system. If you really want a simple device then you're at the wrong company (except for their phone line, which contains devices from very simple to very advanced). Many clone corporations are able to make a simple device for not many costs. The source for the maemo software is there as well.

I'm glad I'm not the only one who observed this.

My suggestion: replace $/GBP/EUR symbols with one generic symbol (I guess, given tradition and other usage of the symbol the $ would fit well) and use the symbol based on the setted LOCALE. Remove the other two symbols. Allow the user to change the LOCALE just like this is possible right now, and put the other symbols in the input menu. Because, using that menu, you can get a pipe, but its inconvenient if you use the CLI a lot.

I did in fact, later, give more detail (and I didn't give the detail in the first place because, frankly, I didn't think anyone would need me to elaborate -- I thought it was all pretty trivial information). It's basically that the fact that IR is line of sight, doesn't tend to have "trusted" devices, etc. causes IR to be more secure for exactly the reasons that it is less convenient.

Further, you may find yourself employed somewhere where BT isn't allowed to be used. And where they'll actively scan for its use (not just to counter external adversaries, as the Faraday cage idea would cover, but also from in-building adversaries). In that situation, IR is infinitely MORE convenient than BT ... because BT might cost you a job or security clearance, and IR wont.

But, the basic point is, BT has lots of security vulnerabilities that IR avoids, and therefore has it's own attractive aspect.

Oh, really. IrDA is more secure? How so? I've been able to open the garage of my neighbor using the remote control of my garage. There are devices to put off TVs of other people using a remote control. With BlueTooth, you can at least ut a password on it, and hide your device.

Besides that, one can easily use VPN over BlueTooth to establish a secure connection provided both endpoints support this. The endpoint doesn't have to be a phone. For example, you can use server (VPN) <-> 3G <-> mobile phone <-> BlueTooth <-> NIT (VPN). To transfer files, one could use e.g. GPG.

If your employer bans BT I think they don't want you to use IR either. Too bad. So you abide to the rule(s) by putting your BT chip (or device) off, put it in a Faraday, you use a different device without BT, you get a IrDA addon for your device, you put BT on invisible, or you find a smarter employer.

When was the last time someone broke into a PDA, or desktop for that matter, and stole files via IR? Doing it with IR requires at least some level of direct access to the system. BT doesn't.

And, I will admit that I don't know specifics on this one, but I know that "hidden" on a wifi network means next to nothing. What does "hidden" on in BT _really_ get you? It keeps out amateurs?

It's my understanding that, just like wifi, if you know what you're doing, you can easily get past bluetooth's rudimentary protections (being hidden, and the other aspects) and get access to its facilities.

With IR, last I checked, you don't trust any device automatically, when something wants to get access to your files, it _has_ to ask. And in order to get that far, it _has_ to establish line of sight with your IR port. In a closed office, or in your pocket, that's pretty difficult with IR. Not difficult at all with BT.

(and, no, not every site that cares about BT cares about IR, for exactly the reasons I'm stating)

Your discussion of access control is somewhat ambiguous; I can't tell if you're referring to a user level or a technical level.
Obviously, on a technical level, any device attempting to communicate with a device over any link has to "ask" in some way.

OTOH, if you mean on a user level, you'd have to be ******ed or running ******ed software to "trust any device automatically"; if irda implementations make it impossible to have trusted devices, that's a software detail, and it would be pretty straightforward to permit access automatically.

Line-of-sight is definitely an issue, but it seems rather limited in application, so I can't see the justification for including it in stock hardware. A mini/micro-USB OTG dongle should do handily, I'd think...

You have a point there John, but it depends on how the people _use_ BlueTooth (or WiFi).

BlueTooth currently simply has a bad name, but the same can be said about IrDA, RFID or DECT. IrDA and DECT are just less used these days while RFID does receive criticism.

In contrast to e.g. DECT, BlueTooth 2.1 has strong encryption. As soon as someone tries to pair with your BlueTooth device, you will notice.

Even if you don't trust it, you don't have to depend on it. For example, one can easily run a secure AP without using WPA2 + PSK using SSH, AuthPF, IPsec (or VPN), and PF. One can run SSH over 27MC, or GPRS. This is possible with BlueTooth as well (e.g. with PAN). I bet its possible with IrDA as well.

The big advantage of BlueTooth over say WiFi is that it draws a lot less power. You can also use BlueTooth to let multiple devices communicate with each other. You cannot do that with IrDA.

If your BlueTooth device is hidden they have to guess your MAC address in order to talk to you. This is not a trivial task.

And, whether you like it or not, Nokia is actively contributing to BlueTooth via Wibree, WiMedia, and BlueTooth 3 initiatives. So, I'm sorry, IrDA is on the way out.

One thing I sometimes do when I'm bored and sitting in the train is looking around for BlueTooth devices. You can derive the brand and type of someone's phone from this, but also their name. Its funny to walk to Jenny telling her her BlueTooth is on, without her knowing how the hell you figured out her name. This however, is a configuration issue on their side. Perhaps, by default, BlueTooth should be disabled on a device.

BTW, Bruce Schneier wrote an article about why convenience and security are not inherent mutually exclusive.

http://www.digitalworldtokyo.com/ind...d_in_a_second/