Slashdot Mirror
Gnome.org Desktop Integration Bounty Hunt
tempest303 writes "In order to help improve integration between apps on the Gnome desktop, Gnome.org is offering bounties for the completion of a variety of integration tasks. Bounties range from $15, for submitting new .ical files for Evolution 2.0's multiple calendar view, to $2500 for allowing synchronization between Evolution's addressbook with Gaim's buddy list!"
Doesnt the open source model succeed by encouraging people to collaborate and work together? It seems to me that this bounty concept will only motivate people to hide information from each other and work against each other in the name of money.
It's all good and well to program for love and pride, but these rewards will help the project move over some of the less glamourous problems.
Consolidation is important in the Linux world; if coders spent more time on it instead of creating new competing apps (not that there's not a place for that), the world would be a better place.
Treehugger? Treehugger... Treehugger!
This is very interesting concept, image someone setting up a bounty server for free software (in general) where people could donate money to bounties on any free software project and hackers could claim the money ones they've solved it.
It's even worse with many GTK+ programs sticking to gtk1 instead of moving on to gtk2. Thus you end up with duplicate libraries. For instance, I use Gimp and Dia all the time. But Gimp (stable) is a gtk1 program while Dia is a gtk2 program. So I decided to go to the development version of Gimp to eliminate the redundancy, only to find that Xmms still used gtk1. Aaargh!
Don't blame me, I didn't vote for either of them!
damnit, im at the conference right now and i started on the gaim/evo bounty earlier, and of course its one of the 2 posted on the front page
;)
thanks slashdot
If there is one clear area where microsoft leads the field its application integration. Obviously the centralized control make this much more achievable.
In the long-term it may be more effective to build a high-level API to allow this integration. Perhaps some kind of built in RDBMS with a well defined schema for commonly shared application data. Several static tables to provide an area for common data (Contacts, Favourite websites/ftp servers etc) plus an extensible area for application specific data.
If the open source community had a well-defined process (shock horror!!) to request changes to the schema we could begin to provide the kind of application integration currently on offer by MS.
Integrating Gaim with Evolution is great but surely a strategy for integration email clients with IM clients in the general sense would be much more valuable.
Definatly a move in the right direction though!
Be vewwy, vewwy quiet. We're hunting integwation features!
According to IRC (ergo, its a rumour) Novell donated 25k to Gnome Foundation to setup this. Gnome Foundation then organized it and push it along with doing the screening and judging.
I didn't do this, now did I?
Slashdot's gnome Logo is outdated
g
The current gnome logo is more than 1 year old.
This is the new one:-
http://gnomedesktop.org/images/topics/gnomenew.pn
-- Hasbullah bin Pit (sebol)
Mod parent up.
I think this is a great idea. You think of something you really want, go to the bounty server and give it a price. If other people think it's worth kicking into, it'll add to the donation pot.
I think you've come up with another way to make money with free software.
The donators could also choose which licenses they'd accept the software to be released under.
This would also be interesting to try out with closed-source software. See how many donations are available.
I guess with the closed source option, you'd have to specify with whom you'd be willing to share the source. If I were donating 10,000 to get a closed source program that scours the stock market reports and lists the fast moving stocks, I don't think I'd like to share that particular program with the other two guys who bid +$3.28 each.
********* sig: If you don't like the law, get filthy stinking rich, and buy a better one.
Maybe the GNOME Foundation should offer a bounty for changing the old Slashdot GNOME icon.
Plueeeease, it can't be so difficult, can it?
rubinstein
Since you don't seem to mind trying development level applications, such as the Gtk2 version of Gimp; there's a fork of xmms called `beep' that uses Gtk2: http://linux-media.net/beep/
I've been wishing for better LDAP support in Evolution (and MUAs in general), and wrote up a page on my Wiki about Writable LDAP Addressbooks. Looks like they've got at least on covered.
Wil
wiki
...an adaptation of the Open Code Market idea. I'm glad the open source community is exploring more and more ways to make a living while creating free software.
I claim first use of "Error No. 0B" - or "No. 0B error." It'll be the new ID 10T!
While there is no doubt that Gnome is visually attractive, has there been any empirical evaluation of Gnome from a human-computer interaction perspective, i.e. a usability study? I've certainly never come across any such testing in relation to Gnome, which is worrying.
Cheers,
Ian
The bounty page says that they are trying to achieve integration between key components of the linux desktop. Evolution, OpenOffice.org, and gaim, are a few key components of the linux desktop. You may choose not to use them, but that doesn't make the statement false.
And since when did hackers need money to hack?
Hackers don't necessarily need money in order to hack. You miss the point of this contest. The point is, it's a win-win situation. If you are a hacker and could use some extra money, now you have some extra motivation to grab a task and work on it. And it's a win for Novell, who I hear donated the funding for this. So my question is, why are you so negative?
And where does this leaves the previous claims of GNOME being "integrated"?
There is always room for improvement, wouldn't you say?
Why can't those making the money, i.e. Ximian / Novell / SUN do the work for their profits?
They do. But now those outside Novell, etc. have a chance to make some money and contribute to our favourite desktop.
Celebrate the finer things in life
I know this is likely to be modded as a troll, but...
It sure looks like Gnome 2 has been basically trying to turn into OS X. I remember asking on gnome-devel and gnome-list, back in the pre-Gnome2 days, why things like button order were changing between 1.4 and 2. After a lot of hemming and hawing the final answer seemed to be "because Apple does it this way, and they're known for user friendly design". The hoped-for Evolution + Gaim interoperation looks to be a clone of the way iChat and Mail.app work together. Looking through the bounty list, a lot of the UI stuff strikes me the same way.
Don't get me wrong; I think OS X is the best user interface available. But if they are not trying to do anything original, why don't they just close up shop and tell folks to "just buy a Mac"?
#DeleteChrome
Debian has no default. You have to choose between GNOME and KDE.
Debian's KDE support today is superb, only rivaled by SuSE.
(I think the misunderstandment that Debian is a GNOME-dist comes from the the 2.2-time when they didnt distribute KDE at all, because of license issues)
Wouldn't have been more easy for Evolution, Gaim and other programs to share a single addressbook?!
Learn from KDE, where Kopete is now dumping it's own contact list in order to share the same Kaddressbook with Kontact, KMail and any other KDE program.
Why making things more complicated instead of making them simplier?
The new file selector for GNOME 2.6 is already being worked on by Federico and others.
Celebrate the finer things in life
FINALLY, someone out there is getting the idea! You can't eat a "thank you", and while the amounts paid may or may not be worth the effort put in, it's something! Great incentive.
With even a little bit of cash out there for developers to earn, projects like gnome can go a lot farther, a lot faster IMO. My hat is off to you guys at gnome.org!
BeauHD. Worst editor since kdawson.
Ximian Connector: $69 personal use license.
http://www.ximian.com/products/connector/
Of course, I'll sell you one for $100 and pocket the $31 if it makes you feel better.
I'm against picketing, but I don't know how to show it.
The stupid thing about it is that whilst the button order Apple uses *may* be better in their usability tests, this is in the Mac environment where Apple can enforce consistency.
In a typical Linux desktop however, all of the non-GTK2 apps have the Windows button order. This includes things like all KDE programs, all GTK1 programs, Mozilla, OpenOffice, closed-source programs for Linux etc.
What is more confusing, a slightly less intuitive button order which is consistent across all apps (and incidentally the one that most new converts are probably used to already from their Windows days), or one where the button order changes all the time no matter how "intuitive" one of those button orders is? I find it difficult to believe any usability tests will find the 2nd option easier for Mr and Mrs Average to use.
I do admire Gnome's attempts to have HIG guidelines and I think it is a good thing, but sometimes I have to wonder if they really understand usability or if they are just blindly following Apple's guidelines without thought of how it works in the real world of the Linux desktop.
Evolution's contact editor allows you to annotate a contact with the dates of their birthday and anniversary. However, these dates don't automatically copy themselves into your calendar...you won't see them when you glance through your schedule, and an alarm won't fire to warn you of a friend's upcoming birthday...Clearly, this is a travesty."
Indeed it is a travesty. And a travesty that exists between Apple's Address Book and iCal apps as well. You can get round it using software like Birthday Shifter, but this really ought to be in the main app's functionality.
Cheers,
Ian
You made your post sound like a troll, but I would like to support some of the stuff that you said.
When I tried out KDE, one thing that I found to be very impressive was the integration among the core applications and those non-core applications that are built as 'KDE applications'.
However, here is the catch, I am a GNOMEish person, (personal preference mainly) and when I use a KDE application (not a QT application), a whole load of stuff happens, and it is as if half of KDE is starting.
Your fears about something like this happening with GNOME is justified, and would like to run XFCE now and then and it would be great if GTK only apps were available.
In Debian packages, I have always found many gtk programs distributed as two packages, one for GNOME-ized and the other gtk-only, and I have found gtk-only versions to be faster and (since I am using GNOME 1.4, I do not get much benefit of integration).
You might have been modded down as a Troll, but your point is worth thinking about.
Thanks.
GrimReality
2003-11-23 01:52:46 UTC (2003-11-22 20:52:46 EST)
P.S.: Please don't call GNOME 'bloatware', I use GNOME 1.4 daily and has found it to be not-bloated. I have not used GNOME-2 much, but it seems to be a lot more simplified (and faster) than GNOME 1.4, so I don't think GNOME deserves to be called 'bloatware'.
My time isn't free. I've had to spend a lot of money to get my education, as well as a lot of personal time I could've been using towards something else. As any economics student could tell you, there's no such thing as a free lunch.
This provides a nice feedback mechanism that allows non-programmers to reward programmers for "filling in" and doing what the non-programmers wanted to be done. It's a natural balance, and I consider it progress in how opensource is developed. One of the few sustainable ways we could keep Linus Torvalds working on the kernel 40 hours a week is by having IBM, Red Hat, et all contributing to his work, just like Red Hat employs Alan Cox, or any number of other examples.
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
We use this method of automatically publishing our calendars via SSH to a web server at my office and, thus far, it has worked flawlessly. Perhaps somebody can use JiCal as the backend for this bounty?
If I had the money, I would offer a bounty to anyone who could come up with an equivalent of EndNote that works seamlessly with OpenOffice (I would happily pay for the package if they produced a version for Linux) or a bounty to anyone who can come up with an interface as easy-to-use as SPSS for any of the existing (powerful but hard-to-use) statistical programs...
That's my wishlist, anyway, but as I'm a student I can't afford to do it alone, and don't have the time to code these interfaces...
As others mentionned, this sounds a lot like the Open Code Market idea that has been discussed here previously.
:)
For $25k, Novell just bought amazing publicity. Perhaps an Open Code Market could attract such financing?
Big companies could even offer matching funds to any/certain types of OS software, letting users direct where the money goes. This would not only help finance and promote projects, but publicize the company and the Open Code Market.
And since I'm giving away business advice... it seems to me trade associations would also be a good funder for many targetted projects (I imagine that would be a good way to get funding for things like accounting systems, specialized database packages, etc...)
Someone please try those ideas out. I'd much rather make a living selling code that will be open
Information: "I want to be anthropomorphized"
Unless your clients can specify the requirements formally (and if they can do that, why don't they just write the program themselves in a functional language?), there's always room for a lawsuit. Thankfully, society has developed a mechanism for extremely low-cost, high-speed lawsuits: binding arbitration.
So what happens is the Client submits their signed spec (possibly after refining it with the Developer) and payment to a knowledgable (capable of understanding the spec) and trustworth neutral party: the Arbitrator. The Arbitrator signs the spec and then passes it on to the Developer. When the Developer thinks the project is done, they demonstrate it (possibly using signed code) to the Arbitrator, who then decides whether it satisfies the spec or not. If so, the Arbitrator pays the Developer and passes the code onto the Client, otherwise the Arbitrator corresponds with the Developer to work towards completion. If the Developer gets hit by a bus before the project is complete, the Arbitrator gives the payment back to the Client.
(Obviously the Arbitrator can get a percentage of the payment, a fixed fee from either party, or do the work pro bono.)
I understand that Gaim supports XMPP, but it does as for a secondary one. For example, when it starts it request you to login to AIM.
Why not support Gossip or even Tkabber instead? Why Gaim?
Well, if you think it's just a flame war about IM clients, then take this:
In addition to three GAIM-related bounties, the fourth one is "purely" related to AIM protocol: Handle aim: links in Epiphany. When I read this I begin thinking that Gnome management team has been sold out to AOL. Otherwise why wouldn't the include also Handle JID: links in Epiphany.
What's wrong with Gnome team?
Less is more !
The free software idea imo works perfectly... to a certain extent.
What's different now is that specific things are expected in order to fit in a clearly outlined project plan.
Meaning: you have to walk in line. That's something you generally don't do for free. Hence the bounty.
Still I think this doesn't show a deficit, neither does it say anything about socialism - or in what sense the OSS is or should be socialist in nature. If it wasn't for all those "egotistic" flags already planted, we wouldn't have this discussion in the first place.
It just shows a clear limit to what you can expect people to do for free and when you have to offer incentives.
And that is something every government that is to an extent socialist in nature (not talking about failed communist/totalitarian experiments, talking about most western countries) takes into account as well. You're on the wrong track. It's not a weakness, it's a strength.
We should be glad there's a limit to what people do freely. Even if that makes running projects and governments more of a hassle.
I think, therefore I am...I think.
Not only is this important -- I think it would really make a whole lot more sense if KDE, Gnome, and everything else shared a lot of common file locations. My mail in Evolution and KMail should be stored in the same place by default without me resorting to strange mbox/maildir symlink hacks. My Evolution and KAddressBook should use the same files, so I don't need to manually sync them. It doesn't make any sense that they aren't.
I get the sense that 2004 is going to be an extremely important year from a usability perspective, and it will determine whether or not Linux succeeds on the desktop. 2003 brought us the great applications that caused me to switch from Windows to Linux in the first place. Now we need to bring them together for Joe User.
I switched from Gnome to KDE recently. Why? Consistency. It sounds silly, since Gnome prides itself upon the consistency of the user interface and the comprehensive Human Interface Guidelines. But KDE has a very nice predefined widget set. While I hate to dredge up the file selector, it's consistent whether it's embedded in K3B, in my file manager, or an open/save dialog; while some people have gripes with Qt, it's an extremely elegant toolkit that makes it ridiculously easy to derive new widgets. I can right-click an image in Konqueror, and save it directly to my webspace via FTP or WebDAV because all the file selectors take advantage of my kioslaves. They're both powerful in different ways, and if they could find ways to leverage both environments' strengths in one another, Linux would be absolutely unstoppable.