Evolution Exchange Connector goes OSS

This is huge news. No I’m really not joking, but you need to be a IT futurist like me to appreciate it fully: Novell have just announced that the Exchange connector for the Ximian Evolution client will be FREE from May 14th.

I kid you not, this could well be the final straw needed to convince businesses to move to Linux desktops in droves.

I know that you’re all sitting there thinking ‘TurboTas, you are mad as a hatter’, but perhaps think again.

Evolution is a rock solid mail client with all the features you expect from Outlook except that you needed to pay to talk to an exchange server with the OWA engine running.

Lets be honest here: Exchange is pretty good at what it does: not many (FOSS) systems come close to it’s functionality (Yet!). Now that the platform requirement has gone, it’s just become even more useful.

The next big step would be evolution compiled for Windows. Then we really would have a killer app. Alas, that won’t happen in the near future: Evolution has core gnome dependencies. All is not lost though: in 2000 a small team did succeed in getting gnome running on Windows.

Of course, the next big step after that would be a FOSS Messaging Server with all the nice features you’ve come to expect from Exchange: It’s on the way!

Watch this space and remember that you heard it here first.

If you want the lowdown on what’s hot and what’s not is the world of FOSS, TurboTas is available at pretty good day rates and has been advising Large corporations and Goverment contractors on the use of Free/Open Source Software since 1994. Just get in touch

TurboTas 2004

One Time Pad generation

The problem with one time pads is that they are a bit hard to produce and distribute. One of the hardest parts of the process is to make a pseudo random stream of 0’s and 1’s. I’ve had an idea about how to do this.

Why not simulate a homogonizing device, such as say,a food mixer. Start off with a 3d cell model of a circular container, much as you would with a CFD model. Lets say take a million cells.

Lets set eachj of the cells either on or off. The initial method is relevant but lets’forget that for a moment.

Next we introduce the ‘paddles of the mixer. These paddles operate on the bits in the bowl in the same way as they would in a real mixer: we’ll simulate viscosity and a distance/velocity algorithm to affect each of the cells during each iteration of the mixer.

The ‘whicking head’ can be moved around by the user to form a unique patter or the heads could be induced into ‘chaotic’ motion.

We’ll have some kind of distribution check to show when the process is complete. I’ve a clever plan for this based on voume shading, but ignore than for now.

Once the mix is complete, you have (hopefully) a stream of bits which is as random as you could possibly hope for. these can be read out in any order and used for the pads.

A sensible precaution would be a trim to remove data from the ‘edges’ as this may be less likely to move.