Posts by Christian Berger

In a few years...

...we will see this as just a big mistake as desktop PCs.

So what I would recommend people to do is to simply use such devices as "dumb" terminals. For example run Web-services or even VNC/RDP to a local server running the software you need.

Of course my favourite one has to be the 1967 version of Casino Royale

you cannot make movies like this one any more.

http://www.youtube.com/watch?v=SSEmKXzpQMQ

2004 scanners are not vintage

Scanners typically last for ages, and the specifications haven't changed a lot in the last 15 years.

Different is good... because...

We can now, in situations where speed isn't important, compute both SHA-2 and SHA-3 and use both. If they are different, the combined strength of both algorithms is at least as good as the weakest. If they are similar, someone might find a way to break both at the same time.

Horray for UTF-8

Without which this discussion wouldn't have been possible. :)

(for our younger readers, there used to be a time when computers had problems with non-english characters, you couldn't write a text file containing Umlauts and expect it to arrive correctly at the other end)

It's the 1990s all over again

When companies binned their Unix servers and Terminals for unconnected PCs, then built little Novel then Windows servers with separate networks. That's _still_ the main cause of headaches for sysadmins today.

So remember, the most important part of IT is to not follow short term trends. Whenever a new technology comes along look at how you can use it wisely and choose only the technologies which are beneficial.

Has applications for viewfinders and head-up displays

There you need a high resolution (>Full-HD) in a limited space. Unless of course you are working with 2 displays per eye and use one for the "background" and one, with variable focus, for the point of sharpest seeing. You'd need some optics to move the image of the "sharp" display around.

For a while it was even more popular than LPs...

... back in the 1990s there were actually more Compact Disks sold than LPs. This might have had something to do with the high price LPs had back then. Of course today you can get an LP for an Euro while CDs are still much more expensive. The LP has reached a price point now where you can just buy one without knowing what's on there.

It might attract the usual crowd

There are some Microsoft Fanboys, the people who gladly write and even buy software written in VBA, the people who, when they are not writing VBA use OLE and DCOM and since then have used every new technology coming out. Such software products then require you to install everything from VBX to .net and Silverlight, as each one of those technologies has been shoehorned into the product.

That's the market Microsoft will most likely get, however that's also the market they alienate by not supporting Win32 on ARM.

Certainly a bad thing for anything security releated

A normal chip card has to be inserted to have a contact. It's impossible to contact that card without my consent. Now imagine some malevolent person building NFC readers (e.g. card terminals) into park benches. It would be trivial for them to authenticate non-PIN transactions.

And even with PINs it's a nightmare waiting to happen. Imagine kids putting an NFC reader into their back packs connected to a little computer (i.e. Raspberry PI) and writing a little script to constantly enter random PINs. Once they contact your card and are able to enter 3 wrong PINs you have a problem. Guessing wrong PINs is easy, and it's a prank many people can do.

It's normal

I mean for example I have to pay for German commercial TV via advertisement money, even though I don't watch it. There's little difference between such a fee being collected via advertisements or via some sort of enforcement agency.

Nobody will do a re-write for Windows

The whole point of Windows is that you can run age old stuff on it. Windows itself is long obsolete by now.

Most software in productive use for Windows is just barely maintained. Such packages started as a project to learn about the development system (e.g. Delphi or VB) with the developer trying out every feature they can. This sometimes leads to new software which needs Win32 to use VBX components because some piece of software needs it, and that piece of software won't compile on the "new" 32-Bit compiler

On those systems you have layer upon layer of cruft, you have ever fashionable technology from the last 20-30 years in there.

And that's not just in applications where it doesn't matter. For example there are SCADA systems controlling power grids running on OPC, that's OLE for Process Control. If you are a long-time Windows user, you might know OLE from Corel Draw 4 crashes, or from errors resulting in opening one document with embedded OLE objects on another computer. Networking works with DCOM, a technology so bad that Microsoft stopped supporting it.

Nobody will rewrite such things, and even when they do, they surely will try to avoid those problems.

The problems with touch interfaces in a nutshell

Touch interfaces are hard to discover, there are many ways to move your hand, and unless you try them all, it's hard to find them.

Touch interfaces also aren't very expressive. You need a lot of gestures to do things you would otherwise do with a few commands.

ZX8(1)

I had an ZX80 with the ZX81 ROM, as well as a TI99/4A. And I have to tell you, the ZX80 was _way_ faster. It's mind-boggling how fast the Ace must have been then.

In a nutshell

Every company believes in something called "growth". It means that everything will be more and more. So it's only rational to expect there to be more and more data transmitted on the network.

That is why I, as a customer pay money. I pay it not only for the ISPs and mobile companies to run and maintain their networks, but also so they can constantly upgrade the networks to suit the growing demand.

Ideally the usual peak utilization should stay at about 50% so if one of the redundant links goes down, nobody will notice.

It used to be that way when engineers were in charge. Their main objective was to provide as good as possible service given budget constraints. However now it seems the ISPs strive to reduce the spending, while keeping the cost up, hoping that most ISPs will do the same and blame users on the problem. This works great for big ISPs and in countries without any real competition. Their misinformation has worked so well people believe that regular bottlenecks in the ISPs network are normal, and that data somehow costs significant amounts of money.

It's getting harder for them to find good engineers

Apple, particularly since the second coming of Jobs, has gradually alienated engineers. Back then, you could buy an Apple and you always had the sense of it being a bit over-engineered and well made, with decisions being taken (like the PPC) for rational reasons, or at least because of a certain element of beauty. At least the hardware architecture was a thing of beauty.

Now with Jobs, we get locked down boot loaders and operating systems. We get devices which are so badly designed you cannot even replace a broken battery. You can only install one operating system, the one that shipped with it. And that operating system was dumbed down to a point I cannot do anything useful with it as an engineer.

For engineers Apple has just become one of the worse consumer electronics companies. It feels like, at Apple you'll only be able to design products deliberately made bad by management decisions and you are not allowed to make it as good as you could. Working at a company which is like that, I know how frustrating that can be. All the interesting stuff inside that device, like for example the GSM/UMTS/LTE baseband has been outsourced to other companies. (there's only a tiny hand-full of baseband stacks, made by very few companies)

So Apple is no longer a place I'd like to work as an engineer. Luckily for Apple, since they have transcended into a religion, product quality doesn't matter anymore.

Now let's imagine this is fixed and Secure Boot is on

The attacker had to have physical access (or root access) to that machine to install it.

Now if he had root access, he could do anything in user-space, regardless of the boot process, so that's not the interesting case.

If he had physical access he could do one of the following:

Install a hardware keylogger/replace the keyboard with a bugged one of the same model

re-flash the service MPU (if it's a good laptop) to log keystrokes.

change the service-mode firmware (x86 code running along with your OS, but with higher privileges)

change the firmware of your north- and south bridge (yes, there are processors in there with full RAM access)

There are many ways to manipulate a system once you have physical access. Secure Boot only pretends to close one of them.