calendar sharing/groupware for hardy

Loye Young loye.young at
Thu Oct 11 17:54:03 UTC 2007

>You see the big picture here is that people
>need to stop using Windows on the desktop... and freeing them of Outlook
>is an important first step, as a whole swathe of users are tied to it.

This is the crux of the problem. 

Although you are right that the solution should allow users not to use 
Outlook, the solution needs to permit them to use it. IMNSHO, it is critical 
that clients are able to continue to use Outlook if they want to. (You can 
skip to the last two paragraphs if you are in a hurry and want to get to the 
bottom line.)

Let's say we have an enlightened IT manager at a company that's been addicted 
to a certain legacy proprietary software platform. The manager wants to wean 
the organization off of Outlook, Exchange Server, etc. What's the best 

1.  One way is to flip a switch and convert the servers and all the clients at 
the same time, but if the company is of any size, such a strategy would bring 
the company to a halt for at least a few days. The manager will be fired and 
a Windows-always-and-everywhere drone will be hired as a replacement. 

2.  Next, one could wean the clients off Outlook first and then switch the 
server from Exchange to a real groupware server later. That would in fact 
work, but the manager will spend many hours of aggrevation trying to get the 
Exchange server to play well with the clients. Also, since there are more 
clients than servers, the number of idiosyncratic issues for the manager to 
solve will be large. Then, when it comes time to convert the server, clients 
will need to be worked over again to connect to the new server. This will 
forever keep the IT manager and the pain of converting in the eyes of the 
entire organization. A career-limiting strategy.

3.  Finally, the manager could convert the server to open source first and 
then gradually migrate the users from Outlook to the open source client. This 
allows the manager to set up and test the server out of the sight of the 
client base. When everything is working hunky-dory ("smoothly" for those not 
from Texas), the manager can move clients over knowing that the server will 
behave well. 

My thesis is that converting the servers first (# 3) is the best way. I 
actually tried number 2 above one time. Huge headache, and I got blamed for 
all the problems resulting from having an Exchange server backend that was 
unstable and changed on every "update" from Redmond, which was almost weekly. 
Back in March of this year, my boss told me that he would be very supportive 
of me starting my own company (or anything else I wanted to do after my 
employment ceased in two weeks). 

But to make number 3 work requires that the clients be able to continue to use 
Outlook to connect to the server. Besides, there will undoubtedly be some 
beloved worker in the organization who always get what he or she wants. 
(Usually, someone in accounting.) The beloved servant of the Lord will say to 
the IT manager "I'll go along with what you want, but I cannot change my 
computer because: "the auditors are coming" | "our (suppliers | customers | 
bankers) require it" | "I'm too (old | young | important | ignorant) to 
change". Senior management's response is "Give the beloved whatever (he | 
she) wants. If that's a problem, perhaps you aren't happy working in this 
organization. We want you to be happy, so no hard feelings if you need to 
start looking for another job." Not every IT manager is as entrepreneurial as 
I am, so this will not seem like a good alternative. 

We can't force people to do what they need to do. Although they would all be 
happier if they did what we tell them to do, all we can do is give them a way 
to get from there to here if they want to. We gotta make it a smooth 
transition. Ensuring interoperability with Outlook is an important component 
of doing so. 

I can't say I know what the best solution is. I don't know that anyone has 
come up with the technique that will part the water and drown Pharaoh in the 
Red Sea. I've heard and read great things about Kolab, which has a web 
interface, depends on a reasonable and well tested base of applications, and 
has interoperability with many email clients. Unfortunately, the reference 
Outlook connector is proprietary (see, but on 
balance it's still cheaper to get rid of Exchange and migrate to open source, 
with a clean migration path for the clients to switch to open source in the 

Happy Trails, 

Loye Young
Isaac & Young Computer Company
Laredo, Texas

More information about the ubuntu-server mailing list