Slashdot Mirror
SUSE Openexchange Under GPL
Gustavo writes "'Netline Internet Service announced today that it would contribute its OPEN-XCHANGE Server, the core technology underlying the industry's top-selling Linux-based groupware, collaboration, and messaging application, under the GNU General Public License (GPL).' How does it compare to OpenGroupware.org which was open sourced a year ago?"
It probably has quite a bit to do with the Linux World Expo that is going on this week -- A venue for companies to make announcements about their Linux products.
Beware of he who would deny you access to information, for in his heart he dreams himself your master.
OpenGroupware sucks, eGroupware has many more features. I have had a difficult time with angel mail (egroupware mail component) handling courier imap CORRECTLY though. The integration of the fud forum, and ticketing agent is especially nice. Opengroupware would be nice, but if you compare other open groupware offerings it is lacking (i.e. phpGroupware/ eGroupware). Opengroupware also depends on the postgreSQl db backend, if you already have mysql running, tough. Although Opengroupware handles courier IMAP better, it seems to muck up the dates so sorting your mail sucks. Opengroupware also has limited functionality. It does like 5 things well. To be honest, openexchange will be a welcome suite to evaluate.
ymmv
They have an entire Demo-site to play with!
;-)
http://open-xchange.org
And it plays nicely with Firefox
Supported browsers
* Mozilla 1.0 or better
* Netscape 6 or better
* Konqueror 3 or better
* Opera 6 or better
* MS Internet Explorer 5 or better
Supported browsers
OPEN-XCHANGE(TM) is a collaboration platform that integrates open source and proprietary servers and clients. Accessible through a common web browser, OPEN-XCHANGE(TM) allows users to share e-mail, calendar, tasks, threaded discussions and documents originating from both proprietary and open source systems. For customers who need seamless integration with a Windows client, commercially available connectors will be released later in 2004.
Same problem as always, move along. Like the Bynari Insight connector, the magic bit is still closed. Interestingly SUSE have a connector called iSLOX for their OpenExchange product, which is a free download; perhaps these two added together will finally be the CAL-free-groupware-with-Outlook-as-the-client we've been looking for?
Opengroupware actually was a big hype, but they failed to deliver what everyone was hoping for: a complete open-source and free groupware server and client, with all popular features such as folder sharing, ACLs, free/busy, etc...
The problem with OpenGroupware, is that except for the web interface, there is no client. You can pay for an Outlook connector, yes, but it is rather expensive (no trial version available), and, more importantly, it did not prove very reliable in my testing. The web client, is not very impressive either. The community around OpenGroupware seems rather limited, I have the impression that all work is still done by one developer of Skyrix.
SuSE OpenExchange on the other hand, does have a very nice user interface. The Outlook connector works fine, and with KDE 3.3 coming out in a few weeks, we will have a free client under Linux. I have heard a connector for Evolution is currently in development. AFAIK Suse OpenExchange lacks ACL based folder sharing, hopefully this feature will be added soon.
And then there's Kolab, another competitor for this market. Currently, Kolab 2 is in development. It seems that it will offer a lot of features that people missed in Kolab 1, such as ACL based folder sharing, and server side generated free/busy. Problem with Kolab is currently also the lack of a native Linux client. Kontact 3.3 will finally have support for Kolab 1, but that's not very impressive, knowing that Kolab 2 will already be out at the end of this year.
Anyway, interesting times are coming!
Using a bunch of existing packages is better.
Linux already has a great e-mail server for example.
Why do we need another one.
It's better to have everything in parts, you have a calender server, you have a e-mail server, you have a whatever server.
That way you can have a choice in what exact server to use.
Each project has their own team of developers and specialists that know how to make a better service then somebody you that has to make the entire shabag up from scratch.
Development is faster, everything is comparmentalized. Then you just string it all together.
for instance I just installed Mythtv yesterday using Fedora Core2 and atrpms repositories.
Now Mythtv is a full featured entertainment system for the linux-based home theater. It provides a database that keeps track of settings and tv listings so that you can see discriptions of tv shows and movies, search thru stuff, and set recording times.
It supports running arcade emulations and snes emulations among others to make a nice little gaming box.
It supports having a seperate front end, and a seperate back end, so you can have a powerfull server with lots of disk space for all those recordings in a completely different place the house.
I supports weather lookup, web browsing, and bunches of other stuff.
Now with all this it's basicly a GUI front end using all these differetn services other linux products provide.
Mysql database. Mencoder and mplayer for the player and encoding. Xmame for the emulation, other stuff for the browser. XMLTV for the tv listings. IVTV/V4l for the drivers. And dozens of other packages.
Probably over 30 individual programs and packages I needed to have installed on my fedora box.
How hard was it?
apt-get install mythtv-suite and then they were all installed.
Not a big deal for the end user, although Mythtv isn't the easiest thing to setup it's easy to use, stable and works correctly.
Now if you built it up from scratch you'd have one gigantic program that would take for ever to make and probably wouldn't work as well. The costs of developement would be enormious.
A waste.
Much better to use functionality already aviable to Linux then recreate everything as one monolythic program. Much better.
The challenge faced when trying to move from Exchange is the tight integration between Outlook, Exchange and Active Directory. As the user authenticates to AD when accessing Exchange over RPC(enc. in HTTPS if desired), defined in the Outlook MAPI profile it is a single user instance. The competitive products do not this ability for obvious reasons, and for anyone looking at large scale Exchange support environments that is daunting. If the site does not have Exchange/AD then over alternatives are more attractive, but when faced with AD/Exchange 2003/Outlook 2003 and throw in Sharepoint Services (free Win2K3 service) then it's an interesting beast to try and tackle. Competing on a FOSS vs $$$ argument becomes very hard when the support/value proposition is in favour of the $$$ solution.
Squirrel Mail is often overlooked, it's plugins give it shared calendars with (some) outlook compatability, todo lists, and tons of other stuff. The calendaring system is very simplistic ( no auto-repeat of an event, events are limited to 6 hour intervals ), but depending on what you use it for, it's very nice.
eGroupware is a fork of phpGroupware. Neither will scale beyond, say, 10 simultaneous calendar users.
open source release conspiracy going on.
The Linuxworld Expo is taking place in San Francisco this week.
IBM and Novell probably just wanted to time their news releases with Linuxworld.
"Can of worms? The can is open... the worms are everywhere."
Not to echo half of your previous replies, but it depends on what your CTO wants. Most manager-types I've consulted for have this idea that there's this magical technology somewhere that makes them suddenly understand their business like they did when it had 20 employees. And for whatever reasons they think that software is a groupware suite.
In my experience, Exchange commits some design sins that are so grevious that there are almost no good situations in which to use it:
I've consulted for quite a few managers who really really wanted Exchange. In each case I told them they didn't need it. It can be a real blow to a manager's ego to have to accept that he doesn't run an "enterprise", but in 99% of the cases that's true.
Exchange is a mediocre MTA, a slightly sub-mediocre contact manager, and a slightly better-than-mediocre calendaring system with some glue scripting that sometimes works to tie them all together but often doesn't. Its sweet spot performance-wise is from about 100 to about 300 users broken into 10-15 organizational groups, working on a single VLAN, transporting no more than about 20,000 messages a day total. If your organization fits those criteria, Exchange may well be a good solution for you. If not, I can tell you from my clients' bitter experience and my very expensive overtime cleaning up after it that Exchange WILL end up costing you a lot more money than almost any other solution.
Most managers who want to use Exchange want a public calendar, a big contact database, and IMAP email. That's not rocket science. There are outstanding mail transport agents, contact databases, and public calendars; if you can't get a developer to pipe them together for *much* less than an exchange license, you're looking in the wrong places for developers. Plus, your support costs will be much less with that solution since you don't have the single, concentrated point of frequent failure that Exchange becomes.
All's true that is mistrusted
Many /. geeks just don't get that people use exchange not for email, but for all the other stuff it does - group calendars+meeting appointments, resource reservations, shared address book etc.You know, there is a reason why OpenExchange is called a groupware. And BTW, people also do use Exchange for E-mail, too.
1f u c4n r34d th1s u r34lly n33d t0 g37 l41d
Actually, the lastest patch adds spell checking in the web email client.
It's lovely and intuitive software with a few odd bits here and there, mostly related to the fact that non-native English speakers designed the interface. I even heard a Novell exec refer to it as "Germanglish," which is about right.
Then again, with it open sourced, we can all get in and fix that little stuff, right?
I'm rolling two slox servers out right now, and most users have been very impressed.
GPL'ing it should accelerate the development cycle, which you can see here:
http://devel.slox.info/
Thats an interesting observation. I too have seen stuff happen that was like stated in both posts. It seems that the average msce flunky can't really admin exchange right and some shops try to not have an admin there for it at all.
exchange needs to be played with from time to time. In my experience an exchange system just left there to do its job will have all sorts of isues. You can't treat it like a postfix or sendmail system and set it up and be done with it untill your change users. (well maybe if you not using any of the features it has outside mail) the funny thing is that an organization big enough to use it should have someone on staff running it. I just saw a law firm get toasrted out of 150+ thousand to install an exchange server and upgrade all thier workstations (about 30). To date knowbody is using the shared contacts, calendering or anythign. I stongly feel they would have had a better return on investment if they would have scaped thinking about productivity and bought some rental property somewere. This doesn't count for the now 7 times in 3 motnhs the system has went down for no explainable reason with at least 3 different consulting firms looking into it while a samba box with postfix running jumps to the rescue durring downtimes (that was in place before the upgrade).
unlike that place i have seen exchange run without hickups at other sites too. it seems to be about a 60-40 split in who will have problems.
I tried an early version of DBMail and found sufficient problems to build my own - it's not very hard to hack a postfix delivery agent to store data in a database and not much harder to hack a POP3 or IMAP server to get it out again. I haven't tried a recent version.
Experience, though, is that storing e-mails in PostgreSQL isn't a particularly wonderful idea. There's a high ratio of insert/delete operations to read operations and this causes rapid growth in both database and index sizes. It's a bit of a slow operation to vacuum and reindex the databases regularly. Also, archiving and restoring mail for individual users is a problem unless you write software to do it.
In retrospect, the file system is a better candidate for storing e-mail than a database!
Also, I generally tend towards the opinion that the last thing the world needs is *another* implementation of the IMAP protocol. IMAP is badly designed, badly documented and almost impossible to implement in an interoperable fashion. If you stick to a "mainstream" implementation, it's more likely to have been tested against a variety of clients.
We're in the opposite boat. 110 user medical practice. We use Exchange 2000, which required us to upgrade our site from a Domain to Active Directory.
1) Exchange is expensive per chair. In our situation, Medicare cuts have tightened our budget enough I'd rather spend the money somewhere else.
2) Exchange is hardware intensive. While that server is also a home directory server, I really wouldn't care to run many more users on a single server, while with alternatives (postfix for example), I'd feel comfortable running a much higher user-to-server ratio.
3) There are just certain things that require mucking about with the Active Directory internals. It's unsupported by MS, but the only way to do certain things. I'll grant some of our issues here are due to having to run in a mixed domain/ADS mode for a while.
4) We'd still require running Exchange behind a mail filtering incoming/outgoing server. Take our bias with a bit of salt as you will, we also don't allow any MS product to touch the outside world without filters/firewalls/scanning in place.
Very few companies use all the functionality that Exchange provides. Better to look at alternatives. Do remember that for most cases, you are also tying yourself into Outlook and all the problems that entails, so that should factor in as well.
In any case, expect a heck of a lot more handholding of the service than you have with sendmail.
Good luck!