Slashdot Mirror
Open Xchange Server Source-code Released
d3vi1 writes "Netline, the main developer of Open-Xchange, has just released the GPL licenced version to the masses. The product is mostly known by users because of SuSE's Open-Xchange Server, a product started from "comFire Groupware".
Open-Xchange is a groupware suite with WebDAV interface (XML), LDAP, iCal and HTTP(S) support. An Evolution plugin is on the way."
Open-Xchange is a very promising M$ Exchange replacement, but until they have an upgrade path from Exchange, we can't upgrade. If programs to convert away from Exchange were to exist, it would break our M$ lockin here.
This signature was left intentionally blank.
Before spending any time on this guy (as a corporate head, anyhow) I need to know how well it works with outlook. To wit: I need to know that the company users won't know the difference -- that they won't have a clue that anything has changed.
If that's the case, I'd be a bug on the ass of my LAN manager to convert us immediately, and he'd probably jump at it.
With Evolution, Connector and now Open Exchange the barriers to interoperability are breaking down.
Microsoft made a deft move by bundling together database and mail server technologies for Exchange (Outlook/Exchange gets used heavily at MyCorp).
It's good to see some opens source alternatives become available, not least because of the competitive pricing pressure it will put on those heavily used products.
"Provided by the management for your protection."
Or at least a good Evolution Windows port.
16,777,216 comments ought to be enough for any forum!
I have recently begun using a Linux PC in my office, and am using Mandrake 9.1, OpenOffice, and Evolution. There are a couple of things that Outlook has that Evolution does not (at least my version of it -- 1.4.6), but overall I have been very pleased. What is it that holds you back from using Evolution instead of Outlook?
A migration tool for openxchange that supports MS Exchange! It's commercial, but if you're already using Exchange, that shouldn't be TOO big a turn off. It might only support the commercial version of OpenXchange, not sure about that.
I completely agree. If this supported Outlook I could begin planning a migration today. I've gotten many people to move away from IE to Firefox, but I just don't see being able to get people to move from Outlook to Mozilla.
I should be more clear -- I use Evolution. But all the drones in the office use Windows desktops and Outlook.
Outlook bonds with Exchange (which we also use -- a slightly older version which works poorly with Evolution's calendaring) so that we cannot leave Outlook, and therefore cannot leave Windows. If we can get a work-alike for Exchange, we can slowly move people into a hetrogeneous (or even completely non-Windows) evironment.
We COULD upgrade our Exchange to allow Evolution to be more of a replacement for Outlook than it is, but that means spending money on a service, and our uppers would, after that, be unwilling to let us scrap it.
The optimal path would be to replace the Exchange server with something that plays well with Outlook, then migrate our people to Linux desktop, where those people don't need Windows speghettified apps.
Python would have also been a good choice. Being a modern object oriented language with a lot of the same functionality of Java without the closed source nature of it.
This product sounds good but out of personal experience it leaves quite a bit to be desired.
We recently installed this and tried using it in our office and found it to be very buggy and unstable. The first version we installed, 4.0, wouldn't even create user accounts properly. It would screw up the samba each time it tried to create the account.
The whole fact that it uses IMAP for its email made it clumsy to use in Outlook as you had your personal folders and then also had your IMAP folders and then also had your SLOX folders. Was quite confusing for most of the users. Also the calendar sharing wasn't that well designed at all. The user was forced to have two calendars, one in public folders and their own one. There was no way to share your normal calendar around the network.
SuSE have a good product here but it is still far from a proper Exchange replacement. We ended up sending the software back and getting Exchange instead.
(\(\
(^.^)
(")")
*This is the cute bunny virus, please copy this into your sig so it can spread
Just took a look at some of the code. Comments are in German! Not that there's anything wrong with German, but that might limit North American involvement in development.....
Other than that, the code seems organized well, from a quick glance.
All these different projects trying to come up with an end-to-end solution, and none of them really getting anywhere. We need a standard.
A few months ago, the folks at the Citadel project took notice of the specs for the Kolab project, and began promoting its storage and network formats as a proposed standard for open source groupware. It was a nice, simple, elegant design, using vCard and vCalendar formats. Others shared the same view: for example, the Aethera people joined in, and made their client Kolab-compatible. We at the Citadel project made our server Kolab-compatible. This was shaping up to be something good.
So what did the Kolab people do? They designed "Kolab 2" which uses data formats that are neither forward nor backward compatible with Kolab 1. They completely disregarded not only their installed base, but other projects that were working towards compatibility. The new format is proprietary (documented and unencumbered, but proprietary) and gratuitously abuses XML instead of following the industry-standard vCard and vCalendar formats.
The Aethera and Citadel projects are currently in discussions to work together to create a true. open, standards-compliant, cross-platform, end-to-end groupware solution. We invite others to participate as well -- we won't ignore you the way the Kolab people have.
As for OpenXchange? As others have suggested, this is really just a couple of bells and whistles glued onto someone else's IMAP server. It's not really a true solution.
Tired of FB/Google censorship? Visit UNCENSORED!
export your stuff to psts, then import it to your (imap) mailbox. i still wouldn't use the product though.
The "outlook" ;) for upgrades is less daunting. There is an upgrade *path*, though it's not as automated as we'd like. Publishing and supporting a migration tool sounds like a good way to make a living, capitalizing on the migration from Exchange to Open-Xchange. Especially if it were a plugin installed when O-X is installed, which led the installer through the upgrade path with data import and a tutorial. A "Call for Help" button connected to a live support desk could also clock some ducats, while ensuring a graceful migration. Corporate IT departments love that kind of organizational reliability when risking any platform change.
--
make install -not war
We've been using SuSE's SLOX for a while and while Outlook connectors exists (OSLOX, ISLOX), they are basically syncing tools. It would be a semi-transparent cutover, but I don't think this newly GPLd groupware (similar, but different) supports such a connector....
[ insert your own witty .sig here ]
To make it transparent to Outlook users, you would need a set of MAPI service provider DLLs for Open-Xchange, just like what Lotus did for its Domino Server (http://www-306.ibm.com/software/lotus/products/ac cessoutlook/).
Anybody know of any MAPI service provider development project for Open-Xchange?
What's the "closed source nature" of Java? Any Java program can include its source code (or a pointer to it), under GPL or any other license. If you're referring to the source code for the Java Virtual Machine that runs the executable Java programs, you'd have to compare that to the Windows OS and x86 microcode that Python programs run on, which are also proprietary and closed.
--
make install -not war
Corporate users will almost certainly notice a replacement of their Outlook with *any* Evolution port. Even if it's better - that's just more of a difference. The most important consideration in corporate desktop upgrades is that users not spend *any* time coping with the change, as that's work not producing the product or service that is the company's actual business. And any noticeable change adds to training time, and all that adds to the risk that the users won't successfully cope, and that is an unrecoverable disaster in an IT rollout.
--
make install -not war
If the program runs on a free Java implentation there is no problem. Python runs on free operating systems.
Analogies don't equal equalities, they are merely somewhat analogous.
Dependency on other, proprietary, software is not relevant to whether source code is open or closed. You're talking about whether Python/Java programs are sold, not whether their source is open/closed. The fact is that Java is no more "closed source" than is Python, by "nature".
--
make install -not war
Can you please back this up with links to discussions you have had with Kolab project? Specifically where you noted on the mailing lists that you had started working on their format with a 3rd party and wanted to keep synce with Kolab? Did they not inform you then what plans were? I find it very hard to believe that once told that two other OSS projets wanted to work with them that they just ignored you or didn't advise on the new changes to come.
Also since you just noticed the specs a few months ago it is quite likely that the Kolab changes were already in the process of being implemented and you simply jumped in at the wrong time. Is that not a possibility? Fresh projects sometimes break compatibility and while one would like to never see that happen there isn't always much you can do about it. I assume they did it for a very good reason and not just to torpedo your efforts.
You post seems entirely one sided and I'd really like someone from Kolab to comment on dicussions you have had with them.
btw going back three months on the Kolab Development and Kolab Format Discussion lists your project doesn't seem to show up once. Perhaps you can clear this up if I'm looking in the wrong place?
Analogies don't equal equalities, they are merely somewhat analogous.
Maybe, maybe not. You've introduced "feature creep" into the consideration of the openness of source code in a given language, here Java vs. Python. Or you're coopting the benefits of open source code into the benefits of payless software distribution. They're separate considerations. Regardless of the price of the software, its source code can be open or closed.
--
make install -not war
What? I'm talking about freedom in the free speech sense. And no idea where you got the "feature creep" problem from, it's not part of the argument.
Analogies don't equal equalities, they are merely somewhat analogous.
I'm still looking for a free/cheap host for these servers. I can sync bookmarks and calendar via webdav from a free/cheap web host, but I can't find anyone that provides ldap, so the only way I can sync my contacts between home and work is to run my own ldap server. Does anyone know of a place providing ldap/open exchange services to the public?
You know I've been wondering this ever since the whole "lets replace Exchange" movement started. What is so unique about exchange that no one else (commercial or otherwise) has come out with a viable replacement? And no I don't consider "lockin" unique.
I suppose I misread your sense of "free" because I'm used to hearing the "purity" argument, for all software running under the same terms of distribution, applied to consistently paying nothing for software. A hallmark of Stallman's FSF, to which you referred. That would be "feature creep", applying price freedom to expression freedom, which I accept that you are not asserting.
To recap:
rute20740 posted:
"Python would have also been a good choice. Being a modern object oriented language with a lot of the same functionality of Java without the closed source nature of it."
I replied:
" What's the "closed source nature" of Java?"
You replied:
"If the program runs on a free Java implentation there is no problem."
I maintain that Java is, by "nature" (its inherent properties), as open or closed in source as, say, Python. Python is more "free" by association, and perhaps in nature - its interpreted source is more "natural" to distribute than a binary, while Java's executable binary is more natural to distribute than its source. But it's natural enough to distribute Java source, and there's nothing preventing that. So Java's "nature" isn't closed. Compile the published Java source and run it on any platform, open or closed, that includes its VM.
As a developer, I prefer all execution states to be observable, for my own productivity and that of other developers on which I depend. But I can distinguish between the openness of one component, and another with which it interoperates, without applying the openness of one to the openness of another.
--
make install -not war
Is why we didn't get some open source Outlook Connector yet. Some companies (OpenExchange) won't even provide an eval of the connector to see if it really works in Outlook.
Bynari - too buggy.
SLOX - not tried yet.
We ended up using CommuniGate Pro from stalker.com.
I don't like it.. server is too closed source and inflexible.
BUT it has _excellent_ OL interoperability... My boss just forced me to use it because of this.
When we'll see some open source or at least free Outlook Connector to these exch-replacing systems, we effectively killed exchange.
- Arwen, I'm your father, Agent Smith.
- Well, you're just Smith, but my father is Aerosmith!
--dave
davecb@spamcop.net
It's like if someone says I've been using my cell phone for organizer and sending messages and email and overall I've been very pleased.
Companies have forms, applications, scheduling, shared files, shared contact lists and a whole bunch of things built on top of Exchange Server and a big part of it simply can't be migrated. It's completely different from an individual user's experience and satisfaction with 3rd party MUA.
I'm curious if Apple could port this to compile on Darwin and then include it in the XServe offering as a email server application... if so they would do an awesome job of re-outfitting it with a great front end for admins... and since XServe can be purchased with unlimited seat license it would be a serious conversion tool both to XServe for email as well as away from Windows Desktops.
A fool throws a stone into a well and a thousand sages can not remove it.
I eagerly await all the agonizing, cliche OSS moments that will come from this product.
:-) I agree the name sucks ass but it's better than its other name: BILL Open Workgroup Project.
Exchange 5.5
Paying taxes to buy civilization is like paying a hooker to buy love.
Why don't you give them a chance?
Yes, they should change their name ASAP (as in NOW you morons !).
But other than that it seems like a reasonable approach to me.
Python allows for rapid developement, performance critical parts can be done in C. Postgres, while no performance wonder, seems like a reasonable stable foundation. I have not seen their code (in fact all I know is from a glimpse at their website) but with some skilled and dedicated coders they should be able to get somewhere in a couple months.
Ofcourse there's always the possibility of a premature death but why are you bashing them like that? Only for the name?
Admittedly, if they don't change that they'll be crushed by MS-legal quicker than I can bunzip that tarball...
Does OpenExchange run on any open source Java implementation (gcj, Kaffe, etc.)? If not, you have an open source project running on a closed-source runtime only--less than ideal.
Analogies don't equal equalities, they are merely somewhat analogous.
Does anyone know of a reasonably objective review of MS Exchange/Outlook replacements, running on Linux/BSD? I'm looking for categories such as:
- feature list compared with Exchange / Outlook (calendar, public folders)
- plays well with Outlook (many sites just want to replace Exchange, but still use Outlook)
I've got several small business customers who are well informed and don't want to get caught up in MS dependency. They're either running demo Exchange (with the built in time bomb), or an email-only server and wishing they had calendaring. In general, they'd prefer to use Outlook as long as they have the ability to dump it and replace it with something else with little / no business impact.
Any pointers / URLs?
Why isn't their a Win32 port of Evolution?
Groupwise is very nice but still the client is a few light years behind Outlook 2003. But the Server has certain features that make it light years ahead of Exchange. So you end up with a few Outlook junky grumblers and a majority of satisfied users. The problem is when some of the grumblers are at the top of the company comparing notes with Outlook users at other companies. They don't understand the technical justifications for Groupwise over Exchange and the herd mentality dominates. Then there's always of Groupwise's high cost and of course it's closed.
)))))))))))))))))))))
Vote for the liberal. It's important.
Modded to -1! Who ever said OSS fanboys don't have a sense of humor?!?!?!
I'm sure it's a fine product, honest.
Now before I get modded down, I be to remind whoever might read this that what I am saying is FACT. - bogaboga
You know what will happen, don't you? You will replace Exchange on the backend. Your replacement will interface with Outlook like shit on a stick. Everyone will be happy, birds will be singing, small animals will skip and play in the sunny meadow.
Then Microsoft will upgrade their Office Suite. Your PHB will insist on upgrading. And suddenly the clouds roll in. It just won't work right anymore. And who's to blame? Microsoft? Nah, they're just being a good little feudal kingdom, and behaving in the best interests of their money grubbing shareholders. One of whom is your PHB. No, it's you, you stooge. You broke Microsoft! And you're outta here!
The right answer to everyone's need to integrate mail and calendaring is to use a different system altogether. One built on open standards. No Microsoft in the front, no Microsoft in your rear. And everyone will live happily ever after.
I suspect the future is not MS Outlook or evolution, but webmail that will work on telephones - just like what some people have been using for a while.
There is also Scalix, which does have a migration tool from Exchange, and can also live side-by-side with Exchange servers, making a migration easier. It's not Open Source, but it's an easier migration path than other commercial (and Open Source) options for those looking to get away from Exchange.
Real programmers don't comment their code. If it was hard to write, it should be hard to understand.
Evolution Connector for Microsoft Exchange Server 2000/2003 (formerly Ximian Connector)
If someone is interested in creating a list/feature overview of opensource groupware with me, please mail me (adress in my journal).
Couldn't Microsoft 'Test', or 'Windows Scripting' software 'scrape' the information from its Exchange product?
I haven't looked at the ' Open-Xchange' product yet, does it have an 'import' option?
What's wrong with the name? 'Exchange' is a generic term and not trademark-able.(at least, not in my country) 'MS-Exchange', however, is.