[Gllug] Open Office, KOffice, Windows desktop migration, roadshows etc.

Jake Jellinek jj at positive-internet.com
Fri Dec 7 12:30:04 UTC 2001


Hi,

(just a quick note)

I'm still suprised though at suggestions to migrate companies from MS
Office / MS Word with applications like Koffice.

KOffice is without doubt a better application for many many reasons
however since it cannot (and nor can Abi Word or anything else I've
tried) open a word2000 document and retain its formatting in any way
remotely relating to the original document, it cannot seriously be
suggested to companies that they migrate. (BTW if you want to test this
I can send you a word2000 document I've been trying with)

The key to my argument is the migrational aspect. If we can assume that
a company has a number of documents, particularly those with tables etc.
which have been made in word2000 (and, I've not tested but I would guess
word97 also) then any migration to KOffice would mean re-creating all
these documents again. It would also mean that when CV's, tenders and
any other type of external document is received in MS formatting, it
cannot be viewed properly. Yes I know people shouldn't send things to
you in these proprietory MS formats, but in the real world they do. Yes,
I'd love to change that, but you can't change it without easy migration
in my opinion.

The real danger in suggesting too soon that companies/managers switch to
Linux / Open Source is that they are unlikely to try too many times and
give Linux too many chances. Pick a time and a set of software that is
going to give at least 90% functionality required otherwise people will
tire of it and take longer to migrate once they realise it doesn't
deliver quite what is promised.

This is why I am very excited by Open Office (NOT Star Office!) which
does truly open Word2000, Excel2000 etc. documents you already may have
and retain *almost* perfectly the formatting of the original. Certainly
close enough that any adjustment required is quick and simple. It also
runs seperate apps without the silly Star Office desktop taking over
your life, and it doesn't require or indeed isn't written in Java.

It isn't yet as mature an application as the others of course, because
it is still in development, but it's good enough to demonstrate and use
for those that only need occasional/light use of Office (which I think
is actually the case in most companies).

Open Office is surely the app that will allow the migration really begin
to happen, this is the app that should be watched closely for release,
and this combined I predict with KDE3 solving all kinds of problems
including quicker application start up times and more tightly integrated
printing systems will be when the road shows should start. 6 - 12 months
time I reckon. Now is the time to try out things, in 6 - 12 months it
should be the time to go out and try hard to convince others to make
major changes.

If you migrate people now it is possible, although slow to set up (none
of the distributions out of the box will do) and fiddly. The result is a
useable desktop system that takes time to perfect with still some
comprimises. You save money but you don't get something as polished or
as easy to live with...*YET*.

My suggestion is that people should indeed be trying this out on a
machine or two, getting used to the issues and showing management that
these things are well on their way to being viable, but don't go over
board and suggest that a complete company-wide migration is easy/viable,
since when you are proven wrong you will lose respect and more
dangerously possibly put people off the whole idea.

I am a techie/geek and also a company director. It puts me in an unusual
situation I think, I love using Linux at work and at home but I also
have to concentrate on the more everyday aspects of running a business
(staff, accounts, money etc.) so I think I have an interesting
perspective to this whole thing. As I've said many times, my aim is to
completely wipe out MS (or any other) licenses and closed source
(preferably even non-free) software from our company. We are about 90%
of the way there I reckon, but I don't expect to be able to fully
achieve this aim for at least another few months, and it certainly is
not going to be an easy thing to do.

Oh btw, never let anyone have dual boot systems with Linux and Windows
on if you are trying to get them to migrate. Windows will be booted more
often than Linux otherwise after a few weeks. I've considered subtly
sabotaging those windows installations that are under dual boot to make
them even less reliable, because the truth is, latest windows versions
are not as unstable as we'd like them to be!!. Stability is becoming a
slightly less easy reason to convince people to change OS for on the
desktop. Certainly I've found X freezes up fairly often (dependant on
graphics drivers and versions I guess), the main difference being
recovery (for those that know how to do it) to be possible without a
reboot.

What's most interesting of all about this whole Linux migration thing,
unlike any single company like Microsoft, Truly free software has no
deadlines, no shareholders and no one to answer to. If it something like
Gnu Linux 3 years or 30 years to achieve desktop supremecy, it is still
likely to acheive it. No one is going to be changing its course, taking
their money away or pulling out of the idea. It is actually unstoppable.

IBM might give up on it, DELL may go out of business and stop supplying
it pre-installed, SuSE might all pack up and go and do something else,
but Free software, Linux and Gnu will always be there, and someone else
will or can always step in and carry things on. It can be delayed, but
it can't be stopped. No single company however is or can be in such a
position.

This of course is why Microsoft tried to persuade governments that GPL
licensing should be made somehow illegal. Again this would only delay,
since laws can always be later overturned or changed. Companies like
IBM, and even in our small way our own company, can simply try to help
speed up the process.

I guess I can truly say I have a reputation now for long e-mails to this
list. (usually ones that don't get much of a reaction either). Sorry!!

Thanks,

Jake.
















-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug




More information about the GLLUG mailing list