Slashdot Mirror
German Government Commissions KDE Groupware System
tankengine writes "The German government has ordered a full-blown open-source groupware solution for KDE, to be delivered by the end of this year. It will consist of a server made of standard OSS components (Apache, Postfix, LDAP, etc) called Kolab, and a KDE client. The contractors are aiming for functional equivalence to MS Exchange and Outlook 2000."
They have been working on a similiar project called Aethera for awhile.
_ _
What has happened to this project?
I use Evolution everyday and found it very nice. The screenshots of Aethera look really nice and the interface from this screenshots look pretty damn intuitive.
Has anyone ever used Aethera?
How does it stack up to Evolution?
_______________________________________________
ACK
I think that's the sound of hundreds of M$ employees spitting their coffee all over their monitors.
A Linux-based Exchange-a-like would be a God-send to me, and I suspect to many others too. You can get some of the way there at the moment with IMAP and LDAP, but as has been gone over ad-infinitum on this site the calendar side is completely lacking at present.
Cheers,
Ian
The contractors are aiming for functional equivalence to MS Exchange and Outlook 2000."
At long last. KDE will have all the security issues of Exchange and Outlook. We'll see which OS has the most viruses now!!
Moderators: please read this.
Good quote, too many chars. Seriously, the slashdot 120 char limit sucks!
Since the groupware server will be assembled by existing OSS offerings, there is likely no requirement to use the supplied client if one does not want to. Indeed, if it's all open protocols, Evolution should be able to work just fine with it as is. Other (partial) clients should also be perfectly usable. This mix-n-match possibility is really one of the great strengths of OSS.
Trust the Computer. The Computer is your friend.
My first reaction what that they're reinventing the wheel. Then I saw that they're going to be using and extending many current KDE components. *IF* the KDE teams takes these changes/modifications and uses them to build a new base, great. If, however, this becomes essentially a fork of current Kmail, Korganizer, etc., I don't see this as a good thing.
And yeah, why not take Aethera and build on that - it's already more integrated with itself and other things, and I'm sure the Kompany could have used a nice gov't contract just as much as the team that got it (maybe moreso).
creation science book
My big chance at karma whoring:
More info available at kroupware.kde.org
I'll do it for cheesy poofs.
Yes, the KDE client will be GPL'd. It will re-use a lot of code/components from KMail, KOrganizer etc
The big news isn't the client. It's the server. Bringing all the available standards together into one server that a client like Outlook or Evolution can use is what's missing. Companies have no option for a good colaborative groupware server other than Exchange and Lotus Notes. Notes is a viable option, but myself and others do not like using it's client at all.
Point being, what they're creating (or bringing together) is the only server for Linux/Unix which will directly compete with Exchange.
Developers: We can use your help.
Look in :* Architecture Paper 1.0
Search for point "5. Windows Client".
Look at the history of cars and paved roads for example. Once there was enough cars on the road, the need for well paved roads became a public utility. Same is true for gas, water and electricity. Once it starts going that way, people's tolerance for non-standard ways becomes a huge issue.
then again, I could be smoking crack and this is just more PR bs.
It'd be great if this KDE effort were also compatible with Apple's current efforts (Address Book (LDAP, vCard), iCal (iCalendar), Mail (IMAP).
mbbac
Calendaring is the one add-on application that all Exchange sites use. It must be usable, well-thought -out, and provide full multi-calendar/multi-site functionality.
If they manange that - say goodbye to Exchange.
sPh
There's something about the Germans that makes them good at software engineering, in fact any type of engineering. Of the open source projects I've seen, some of the best ones have been German. They should make a good job of this.
I've been thinking that the Mozilla project should do something like this. They have the resources to handle an Exchange replacement. Imagine "Mozilla Server" which is a single-install replacement for Exchange/IIS; it uses existing OSS components like Apache but ties them together and simplifies configuration. The Mozilla client would be very well integrated into the server, able to access web pages, email, and newsgroups, as well as LDAP contacts, scheduling, and other groupware features.
Of course since the source and the standards are both open, many other clients would be able to access the data as well. But I think Mozilla/Netscape is enough of a force in the OSS world to set the standard for a project like this. I'm not sure KDE is.
From the kroupware koncept:
...
"ProFTPD offers good security features such as change root environment and a fine granular access configuration.
It's only functionality on the Kolab server is the legacy mode to enable Windows clients to publish their free-busy lists via anonymous FTP on the server."
Anonymous ftp access? Kiddies, start your pub-scanners!
You'd think komputer konsultants kould kode up a more sekure solution.
They do say its disabled by default, but since we all know there will be "legacy" systems around for years, they'll have plenty of wide-open boxes. Why FTP anyways?
I don't need no instructions to know how to rock!!!!
Really, if you think about it, this is the obvious step.
At some point, it becomes cheaper to just write your own version of the software than to pay the per-seat license fees that MS and other commercial software vendors charge. If you're a large organization (or a consortium of really large organizations), writing your office apps in-house is economically viable. It's even moreso if you've already got open-source components to work from.
And open-sourcing everything--even if you aren't legally obligated to--costs nothing and often means that you get free additions to the project.
Oh, but they have. This project is being pulled together by utilising existing projects, each of which have been running for considerably more than a few months. This seems to be a 'tying-up loose ends' affair, rather than a push to develop things from scratch.
True there's plenty of tying up to be done, but then that's why this project exists. The situation doesn't appear to be as bad as you believe it to be however.
Cheers,
Ian
Your post is a bit troll-ish but you raise some points worth considering.
I suspect that there is a market for a strong Exchange 5.5 replacement. There are a lot of midsized organizations out there (50-1000 people) who are running Exchange 5.5 and often NT 4 domains. They don't want to upgrade Exchange, because full implementation of Exchange 2000 requires Active Directory. And they are either satisfied with Novell eDirectory (NDS), or they just don't want Active Directory and the complexity it brings. And that is not to mention the Licensing 6.0 issues.
So, many of these sites are looking at Samba and other Open Source solutions when NT 4 goes off support. But the problem is not NT Server - it is Exchange. How do you replace that? Most sites only make light use of the groupware features, but they do make SOME use - particularly the calendar.
So, if an Open Source product is developed that can replace the core functionality of Exchange 5.5, I think you would see quite a bit of demand.
sPh
I'm not fully confident that stringing together Postfix, Cyrus, OpenLDAP, etc. is really going to produce a cohesive groupware server. Yes, it'll work, but it'll be difficult to install.
The real value here, though, is that the KDE project will now be defining a bunch of standard interfaces by which open groupware will access its back end services. Even if they don't get the back end perfect the first time around, by the time they're done they will have a very detailed set of specifications for the rules by which an open groupware client will talk with an open groupware server. Sure, there are standards for the basic protocols -- IMAP, SMTP, etc. -- but there are no standards for things like, which IMAP folder contains your task list? What's the URL to find another user's free/busy times?
I think this is a big step forward, but it can be done even better. (Full disclosure: I am a developer on the Citadel project, which aims to provide an easy-to-install groupware server; we're doing it as a single integrated server instead of stringing together multiple existing unintegrated packages. So my view on this is admittedly subjective.)
Tired of FB/Google censorship? Visit UNCENSORED!
Have you tried Evolution? The only thing remaining I'd consider useful would be new posting and shared calendars/busy-free schedules. Apart from that it interoperates with Exchange very well, though only via IMAP/SMTP. If you want the calendar etc. as well, Ximian have a (proprietary) solution for that.
I still see a problem. How many companies run plain generic Exchange? Even my small company doesn't. We use things like BlackBerry devices and other plugins. Without those most companies will still run in to issues when migrating.
what? it sounds like a national government doing public-good.
not everyone or every country believes that all needs should be met by private-for-profit entities..
only americans (like you probably are) would find this at all a problem...
personally - i think its * terrific * news.
(oh, btw, did you know that the DoD (your gummint) floats ALOT of subsidy to private USA companies..?)
With my government, I pay for corporate welfare and software patent clowns.
How can I redirect some of my tax money to go to Germany?!
When I worked in an industry that supplied steel makers with key components, most countries had a limit of the percentage of those components that could be imported. Once your reached that limit (say 40%) you were required to build a factory in that country to continue as a supplier.
Now, classical economists and super-free-traders will argue that such behaviour is inefficient and non-rational. And indeed, those policies are one of the reasons there is such a glut of capacity in the steel industry. But no country wants to be caught in a conflict and have its source of key {stuff} choked off. The same thing is playing out in military aircraft.
So perhaps the German government doesn't want to be held in thrall to a US supplier.
sPh
I dont fully expect KDE to deliver the entire product free to everyone
:)
You should.
It's going to be completely (L)GPL, just like KDE. Development is taking place on KDE CVS (in its own branch), which is publically available for anyone to see and participate in.
However, note that it isn't a volunteer project; Germany contracted a company to deliver the product by the end of the year (I'm pretty sure there are some KDE guys employed there). This is a great thing; some KDE devs get paid to deliver a really useful project for any and all to use! Bravo, Germany!
Oh, and they aren't starting from scratch; they're starting from kdelibs, kmail and korganizer.
Liberal (adj.): Free from bigotry; open to progress; tolerant of others.
Looking at the road map Aethera is due for release any moment
thank God the internet isn't a human right.
I have but one thing to say that even M$ hasn't really dealt with:
ALLOW THE FUNCTION OF CREATING CONFERENCE ROOMS AND LOCATIONS THAT CAN BE SCHEDULED AGAINST.
The only was I have seen Exchange pull this off is by having a user created for each account and keeping an outlooks session open for each conference room, then setting the conference room account to Auto-Accept invitations.
For the love of dear god please tell those german contractors to put this function in!! I am in a building right now with 67 conference rooms and I can't count how many times a room gets double, triple, and even Quad booked!
-=[ Who Is John Galt? ]=-
About a year ago, I decided not to use "closed" and the word "standard" together anymore.
"Closed specification" versus "Open Standard". That's what I use now. Sounds clearer, better, doesn't it?
I don't think it's out of line, or uniquely American, to fear the government entering into direct competition with known economic paradigms--and that's what we're talking about. When the US subsidizes a company, its intention is not to displace an entire sector with whatever it's ordered, ie. we take a bid for a jet, which is a standard product that can be produced by a number of US companies. The production of the jet hurts no other sector of industry.
On the other hand, with this paradigm, there is the *risk* that the government starts displacing companies by releasing a free product. On a bigger scale, it would be like the US paying a lump sum for a technology that creates free cars for everyone--sure, this would be pretty cool, until the big 3 go out of business and the economy crashes down behind it. This might not (probably not?) happen, but it's worth thinking about--even if you're Canadian.
I'm sure the original poster, despite being _such_ a _typical_ ignorant American, is aware that the US subsidizes private industry. I just think the original poster was just exploring an idea--you don't have to get all offensive and anti-American about it.
> ..it costs.
Well...it's that kind of an attitude that helps to put good companies out of business (Easel readily comes to mind.) I *bought* Ximian Desktop Professional in order to SUPPORT Ximian in their efforts at providing Open Source to the community. I didn't have to. In fact, I was already using it before I purchased it.
Since the company I work for uses Exchange (puke) I also bought the Exchange Connector, two licenses in fact, so I wouldn't be forced to use Outlook. I don't care that the Connector is closed source. What I care about is having the ability to use Linux in a Windows-dominated environment and having seemless integration without pestering IT about it (IMAP/SMTP does not give the full range of Exchange/Outlook functionality.)
I also care about providing financial support to a company that gives a LOT back to the community (Ximian GNOME, Mono, Evolution as well as their work with OpenOffice and their evangelicism.)
But that's just my opinion.
"Those who would sacrifice liberty for security deserve neither!"
If it wasn't for governments subsidizing software development, we wouldn't have the internet, the web, HTML, MP3, JPG, MPEG...
All of these things came from government funded projects. I know that many people find it hard to believe, or don't want to believe it, but it is actually government funded projects which drives most of the innovation in the software industry, not Microsoft, Oracle, etc. This is why I believe all the recent government interest in Linux and Open Source is really the death knell for most "off-the-shelf" software.
If they can pull this off, it will be an impressive success for the open source model. For a set of contractors to go from getting the problem description to a complete implementation in 3 1/2 months, due to the existence of a good set of tools, would really show the strength of the model.
Getting a custom installation generally takes far longer than that. If this project works, it will start to look reasonable for companies who are planning to get a proprietary solution to get an open source one at the same time to use until the proprietary one is ready.
They only need to make the source available if they distribute the binary. If it was kept for internal use, they wouldn't have to make anything available to anyone.
Right?
But there's really no downside for them to allow distribution of their custom code, as long as someone else pays for the bandwidth. In fact, the wider testing/scrutiny of the code would be a plus.
Can you imagine if all governments started doing things like this? The rate of useful development for open source software would skyrocket. Not only would it let more coders work on projects full time, but maybe a tighter focus and clear specs would improve the usability of the resulting software. And even if, say, the Ruritanian government's groupware project failed, the successful Armenian groupware project would step up to fill the gap.
It amazes me that, in my country, individual local councils hire incompetent companies to screw up important services like benefit distribution when they should be clubbing together to develop a GPL'd local government suite. Sure, you wouldn't get widespread use of such software by a big pool of users, but it still makes sense to have 50 councils funding something they can all use rather than each one getting a bespoke solution. I suppose they each have different legacy tardware, but even so...
<Sigh.> What was my point again?
>>The Mozilla client would be very well integrated into the server, able to access web pages, email, and newsgroups, as well as LDAP contacts, scheduling, and other groupware features.
>Isn't this how we got into this mess in the first place? The OS needs, IE, which you can't remove, and Outlook requires IE, and everything is integrated into these two applications and their support subsystem. Look at the trouble it's given the Windows users.
True enough, but is that arguement supposed to support the KDE alternative approach? That comment applies equally to KDE, does it not?
Just TRY decoupling Konqueror from KDE... you can't even (EASILY) build KDE without all the theme stuff.
Personally, I think 70% of this is backend, desktop/OS-agnostic stuff... and should be worked out in a desktop-neutral manner. I'm not saying everything would be built with KDE dependencies, but it would not surprise me either.
We really need some cooperation and coordination between KDE & GNOME... I'm not putting forth the "one desktop" concept, but surely we can get cooperation on things like calendaring. If GNOME has a calendaring concept (Ximian's commercial Exchange plugin doesn't count), then all the open source folks should leverage what they have in common.
I managed to implement quite a lot of public folder functionality into IMAP4 folders that different kerberos user groups could subscribe to... it's not undoable, maybe not as easy though.
EOM
tankengine
has ordered
contractors
functional equivalence
Now, sequencing them and adding a single apostrophe, you get:
tankengine has ordered contractor's functional equivalence
I think this entire story is a front for hiding secret messages in the link texts themselves. So we may want to start poking through other Slashdot stories and look for other secret messages.
(Note: Yes, this is a joke.) ;o)
-----
"You spilled my egg... I needed that egg."
Many posters have argued that government intervention into private software markets is bad, and that Europeans are foolish not to see how bad this really is
We already have government intervention into US-produced software. Europeans know full well about this, and are wise to push open source solutions.
Having another country's government spy on your citizens IS a proper concern of one's own government.
"dope will get you through times of no money better than money will get you through times of no dope"
An open source replacement for Exchange's calendar store could eliminate a lot of Windows Server installations. Thousands of businesses are tied to Windows Server because Exchange works exclusively with Windows Server and Outlook works (almost) exclusively with Exchange.
Exchange calendaring replacements have been developed by HP and Steltor, and acquired by Samsung and Oracle, respectively. Those products generally don't integrate with Outlook's calendar as well as Exchange does, but they prove the viability of the Exchange-replacement market, and an open source product would have a big pricing advantage over those commercial alternatives.
The tough part is persuading the end-users to switch from Outlook to a new calendar client. If IT can do this, the odds are good that IT could convince the users to switch from Microsoft Office to Star Office.
Maybe it's premature to short-sell MSFT, but this initiative could be a crack in the wall.
Likewise, some people don't fully understand the GPL. They think that if someone is selling GPL'd software, then they must give the source code away to everyone for free. Really, all GPL requires the seller to do is provide a copy of source code to their customers. Sure, the customer can then turn around and give that source code away to anyone for free, but the seller is under no obligation to do so, because they're only providing software to the buyer. The GPL is not about giving away all your rights as a software manufacturer or retailer, it's about preserving the rights of the buyer.
If the German government is the sole customer of the Kroupware program, then the developers of that program are under no obligation to put up an anonymous access FTP site and say, "Free Downloads for Everyone!" They are only obligated to provide the source code to their buyer. The German government could then distribute it for free to all German citizens, but the citizens could then likewise distribute it for free to the rest of the world. The GPL is not about restricting rights; it's about preserving them.
I'm not sorry to see that many /.ers can see the economic reasons for one government to stop paying money hand over fist on a continuing basis to a foreign company - a company which, as a result, exerts some degree of power over that government's electronic and information policies. That money isn't on loan - it's gone from that country forever, into the coffers of another country.
In this case, you've got a nation-state building and using something for and by itself - something that will compete with products that private market workers and investors in another country far away are, in some cases, making large fortunes from.
If this is a problem, maybe the American government shouldn't have produced the Internet in the first place. Imagine how much better-off we'd all have been if that had never happened. Or maybe not.
I'm not fully confident that stringing together Postfix, Cyrus, OpenLDAP, etc. is really going to produce a cohesive groupware server. Yes, it'll work, but it'll be difficult to install.
That would seem to be a problem for vendors, not users. If Debian can make installing the maze of dependencies that is gnucash as easy as "apt-get install gnucash", then they can probably handle some groupware suite as well.
Now, it's true that DIYers may have some extra headaches. But, quite frankly, people who say, "I want to do it the hard way, 'cause it's more fun," and then turn around and whine, "this way's too hard!" don't get much sympathy from me.
(And before you start moaning about those poor Debian/RH/Suse folks who have no choice but to wrestle with these dependencies, note that it's a Debian developer saying these things. We revel in the challenge.)
think this is a big step forward, but it can be done even better. (Full disclosure: I am a developer on the Citadel [citadel.org] project [...]
Well, good, competition is always good, even with free software. I'd like to wish both projects the best of luck, and hope that neither one falters in their goal to bring us high-quality groupware software.
The government defines the software markets by introducing the concept of copyright. Without that, the dollar value of any particular copy of a piece of software would be almost zero.
How can a government possibly "distort" something that is its own creation? There is no "natural" state of a software market independent from government control.