If you are new to MEG/EEG data analysis...
You may want to start with a basic [http://www.mrc-cbu.cam.ac.uk/research/eeg/eeg_intro.html introduction to EEG and MEG analysis], and then look at the description of our [http://www.elekta.com/healthcare_international_elekta_neuromag.php Elekta Neuromag MEG system]. If you are about to run an MEG experiment yourself, you should read the [http://imaging.mrc-cbu.cam.ac.uk/meg/StandardOperatingProcedures Standard Operating Procedures].
You should get an idea of [http://imaging.mrc-cbu.cam.ac.uk/imaging/ScientificComputing how computing is organised at the CBU].
Then you should familiarise yourself with a [http://imaging.mrc-cbu.cam.ac.uk/imaging/UsingVNC VNC viewer], or you won't be able to use our Linux cluster from your desktop PC (or from home).
You may find the [http://imaging.mrc-cbu.cam.ac.uk/meg/MEGday29Jan10 presentations from our "MEG day"] (29 Jan 2010) useful, illustrating data acquisition, pre-processing, source estimation and other issues.
There is also page about [http://imaging.mrc-cbu.cam.ac.uk/meg/MEGpapers MEG literature].
Your life as a neuroimager will be much easier if you have a good grasp on Linux (Unix) and Matlab. The following links will get you on your way, and may serve as useful references when you get stuck.
You can look at [http://www.mathworks.com/academia/student_center/tutorials/launchpad.html tutorials and demos] offered by Matlab itself. This site also contains links to Matlab tutorials at some universities. You may want to start with the [http://www.mathworks.com/access/helpdesk/help/pdf_doc/matlab/getstart.pdf Getting Started] section. The short introduction to [http://www.antoniahamilton.com/matlab_for_psychologists.pdf Matlab for psychologists] from the University of Nottingham is also useful. You can start at the very beginning with this [http://imaging.mrc-cbu.cam.ac.uk/pdfs/matrices.pdf primer on matrices].
This is a systematic and well-illustrated [http://www.ee.surrey.ac.uk/Teaching/Unix/ introduction to linux].
If you've never used shell scripting before, this [http://imaging.mrc-cbu.cam.ac.uk/meg/AnalyzingData/Primer_ShellScripting primer on shell scripting] or this [http://www.freeos.com/guides/lsst/ tutorial] would be a start.
Not essential for "normal" users, the scripting language [http://www.python.org/ Python] is increasingly becoming popular for its flexibility and simplicity. You could start with the [http://wiki.python.org/moin/BeginnersGuide Beginners' Guide].
MEG analysis software:
Most software packages will offer tutorials and demos. They will familiarise you with basic analysis principles, visualisation options, and give you an overview of what's available to you. It makes sense to do the tutorials of software packages even if you are not planning to use them routinely. For example, MNE includes an example data set (tutorial in the [http://www.nmr.mgh.harvard.edu/meg/manuals/MNE-manual-2.7.pdf manual]), and there is a [http://www.fil.ion.ucl.ac.uk/spm/data/mmfaces/ general SPM5 demo for MEG analysis], and a [http://imaging.mrc-cbu.cam.ac.uk/meg/SpmDemo tailor-made SPM5 demo for data acquired at the CBU].
And finally one last piece of advice: Talk to people! Ask questions! Try things out! And never give up...