Hallo,
I would like to build a centralized address book for my SOHO server. Is that possible without the use of openLDAP (I would like to avoid that)?
My clients will be:
Thunderbird, KMail (for KDE 3.5) and webmail.
If so, do you have any suggestions/Howtos?
Web mail packages/vendors have their own ways to maintain address books. This is where you need to begin your investigation. Find out what mechanism they use/provide. For example, Google users can use the “Google Contacts” extension to access Google database.
Thank you for your answer. I actually try to achieve the “impossible”. I want to have a centralized address book in my home server that will combine the address book I have in the Google and other places. Then, I want to use that address book with my KMail (1.9.x), the Thunderbird of my wife and a web client (and I have not decide which one yet)!
So, the obvious choice would be openLDAP. But I really hate that beast. So, I try to find out if there are alternatives, or not. An idea would be to use the MySQL I already run on the server. But, I find it a too much for that job.
I guess you would have to write something, to move data from and to various clients in different formats to a standardized format in a central database. KMail f.e. imports Thunderbird data, not the other way 'round.
Another solution would be to start using Opera. It provides syncing of almost all data. You could all use the same shared addressbook I bet.
Hello tpe,
There are generally two application approaches to what you want to do… The “traditional” and generally most scalable solution is to utilize a “Server” based solution… In this architecture the common addressbook is in a single location accessed by all.
So, for instance you only list mail clients but not what mail/workgroup server you might consider. If you wish to setup a Server in your network, you can look at Server based solutions and I think all your listed clients will work with the mail servers I’m familiar with(Just look at the documentation of any mailserver you’d consider).
On very small networks, typically but not always less than 10-15 Users, ad-hoc solutions can be considered where the addressbook instead resides locally on each and every member of the workgroup. The advantage of this architecture is that members can disconnect and go “on the road” with a local copy but in any case each and every addressbook’s currency is subject to the ability to replicate with all others in the group, and multiple issues can affect reliable replication.
Note though that ad hoc replication technology has been advancing very fast over the past 5 years or so, and some technologies can support very large networks (at least hundreds of members). Typically these new technologies utilize their own custom apps, but will usually be packed with workgroup collaboration features so you don’t have to support multiple apps for different uses.
Lastly, even if you’re very small don’t overlook “cloud-based” groupware where the solution may be server-based but it’s running on someone else’ server (Google is a big example but there are small solutions, too like 37signals).
I’m sorry to say this is one area I haven’t updated my personal knowlege for more than a year (I’ve mainly dealt with Server based networks running enterprise software).
Tony