The magazine of the Melbourne PC User Group Open Source: Migrating From Windows Major Keary |
People who are nervous about migrating from Windows to Linux remind me of the arrival of Windows 95; more than one large enterprise paid the money for upgrade software, and then decided not to migrate from Windows 3.x. The cost of retraining, employees getting used to the idea of a new operating system, .... Eventually the transition took place, possibly because employees had picked up Win95 skills elsewhere. Moving to Linux is something like that: apprehension about the unknown. These days the differences are not hard to deal with; one Linux distro, Xandros, even provides users with the option to choose a Windows or Mac look-alike interface.
There is a learning curve, but it is mostly to do with terminology (especially in respect of naming drives) and file systems. Where Windows has prevailed in a business setting, whether very small or very large, it is likely that - at the very least - different versions of Windows are in use, and Networks may have grown with demand rather than by forward planning. Untangling such situations is a problem that has to be addressed, and it has nothing to do with the supposed complexity of Linux.
Another issue is that of software, which has two forks: applications used to run the business (database, word processing, etc.), and applications used to run the Network and communications. In order to make the best software decisions - in terms of cost, support, ease of use, and suitability - it is necessary to research the options.
Windows to Linux Migration
This book is designed for system administrators responsible for the delivery of Network services that satisfy three criteria: reliability, low cost, and ease of use. It is the only title I have seen that focuses on Win2Lin migration for business users, and will be of interest to system and Network service providers - particularly those who specialise in Linux systems. The book covers migration issues for Windows 95/98/98SE/ME/NT4/2000/XP. Windows Linux Migration Toolkit is not a guide for those who want to move over to Linux from a single-user desktop system running Windows.
Apart from people who do the technical work involved in a system migration, this title is a valuable resource for executives with a responsibility for ensuring that computer-related services meet the criteria mentioned above. The Migration Toolkit provides an appreciation of the issues, such as the differences between file systems, and administrative matters such as the need for an inventory of computer-related assets and documentation of functional requirements. Executives can skip much of the technical content and still get the big picture.
A Network can't simply be scrapped while a new one is constructed in its place. The migration process has to be planned, which requires an assessment of current infrastructure and an understanding of how Linux handles the various services.
For the purpose of illustration the book uses two fictitious companies (one small and the other medium-large) with different needs and infrastructure, and tracks their respective migration processes. That approach enables the introduction of real-world "scenarios, benefits, challenges, and estimated costs that an administrator can identify with".
There is a helpful note on creating infrastructure diagrams; it mentions that Visio (Windows-based) and Dia (open source and available from http://www.gnome.com/projects/dia can be used to produce diagrams of the kind used throughout the text. A professional looking schematic diagram will always improve a proposal's chances of acceptance.
Migration processes are laid out in a logical order with clear language explanations of those topics requiring a sound understanding, such as the various Windows file systems vis-à-vis Linux file systems, backup issues, and messaging services. At the end of each chapter there is a summary, a solutions fast track with bulleted items that succinctly state the main points (handy for the system administrator who needs an aide mémoire when reporting to management), and an FAQ.
A chapter, Desktop Migration Roadmap, discusses the migration to Linux in the context of desktop (workstation) users within an organisation. It discusses the different kinds of user and how to prepare a desktop asset list that identifies each machine with its capacity, peripherals, operating system, and so on. Also recommended is a catalogue of file formats with their respective Windows applications and Linux counterparts. Training, designing a Linux desktop, testing the desktop, and related issues are all covered.
A group of appendices address security issues: Network analysis; intrusion detection systems; and vulnerability assessments. A companion CD contains automated Linux migration scripts and other material.
The book does not recommend any particular distribution of Linux.
A good source of information is to be found at http://distrowatch.hostcentral.net.au, which mirrors the main DistroWatch site. All existing distributions are listed with links to reviews, comments, and download sites.
The Migration Toolkit provides a remarkably thorough treatment of Windows to Linux migration. It is not a promotion of Linux over Windows, but a practical manual that explains the way a makeover has to be managed.
|
David Allen et al.: Windows Linux Migration Toolkit ISBN 1-931836-39-6 Published by Syngress, 490 pp., RRP $89.95 incl. GST |
 |
Software Issues
Reprinted from the May 2005 issue of PC Update, the magazine of Melbourne PC User Group, Australia
A factor that has to be considered is initial purchase cost and the ongoing cost
of running any given application. An interesting observation is made by Woody
Leonhard, a well-known author, in WinXP all-in-one Desk Reference for Dummies
[Wiley, 2005]:
".... Microsoft is moving out of the business of selling software into the business
of renting it - and charging for the 'glue' that binds companies, individuals,
buyers, and sellers together".
Such a situation has the potential to increase the cost of software, and to make
it more costly to distribute customised versions of software to clients. Whether
one is a single user or chief financial officer of a large enterprise, the
medium to long term cost implications of becoming locked into proprietary
software have to be considered.
It is for that reason we are now seeing literature on the subject of licensing
and the implementation and management of open source software.
For a thorough discussion of software migration Paul Kavanagh's Open Source
Software: Implementation and Management is the best resource. It examines a
number of issues, especially "lock-in to the Windows platform". The author is
remarkably forthright in his comments about Microsoft, but he cannot be accused
of not being objective.
Anyone who wants to understand all facets of choosing and implementing open
source software, using proprietary software with open source software (not every
Windows-based program has to go), and licensing should have this book.
Developers who find themselves wanting to use both open and closed code in a
program, or who want to find open source alternatives, will find it a valuable
resource.
A remarkable text for its scope, lucid discussions, and technical detail. For
many books the back-cover blurb is an exercise in hyperbole, but in this
instance the listed features are a true reflection of the content:
A chapter on the cost of open source systems gets down to real money terms; any
dollar figure, of course, has to be taken as an approximation - it is impossible
to establish a realistic exchange rate, so to speak, that enables a computation
of today's price - but "$0" will mean just that regardless of which currency at
what date. The costings enable readers to apply a measure, which is typical of
this book's realistic approach.
Paul Kavanagh:
Open Source Software: Implementation and Management
ISBN 1-55558-320-2
Published by Elsevier,
366 pp.,
RRP $77.00 incl. GST.
