Rejection reasons for GAJ:
-
The original design idea is one we want to see happen during the GNOME 3 development.
And what is the current idea… Sorry but it is EXACTLY the same features. Browsing Activity through time and finding associations using Zeitgeist and Searching, Tagging and Bookmarking using Tracker. The reasoning here is to vague…
-
needs more integration with the rest of the desktop and the overall.
You mean integrating Zeitgeist not GAJ… GAJ is a UI so do you want us to provide GAJ widgets like Empathy does with its widgets?
- GNOME design; right now, it feels too much like a standalone application.
It took us a lot of effort to get it looking the way it does now and I believe it Truly blends in with the Desktop. Hylke Bons did some amazing designs. And sorry it is not like we came out with a surprise design and implementation. We worked in the open. We blogged about the development. You could have told us it looks to “standalone”… Empathy is a standalone app AFAIK. I somewhat agree that GAJ might feel out of place in GS, it feels more like a G2. That is partly because of the highly themed environment GS sports, anything that is not native to GS will feel out of place there actually.
- The team might need to work a bit more closely with other GNOME teams for the integration.
Integrate a UI??? Again this makes no sense at all… Again you seem to not be able to differentiate between Zeitgeist and GAJ…
Rejection reasons of Zeitgeist:
- Not needed since GNOME Activity Journal is not approved. If another module will need it for 3.0, this decision will get revisited.
I don’t know any official module that use Tracker yet. However it is a blessed dependency, which makes A LOT of sense to me, since it encouraged us to write Tracker extensions and make it a GAJ dependency. How do you expect people to depend on something that is not blessed. We will be shipping out shit loads of plugins soon so maybe you will revisit this decision (Thank you GSoC for Michal Hruby).
These being official reason, I know there are a lot of other reasons that could have played a role:
- Zeitgeist in Launchpad.
A lot of people hold a grudge against us for that but hey its free, open source and wont bite you. Sorry but GNOME’s infrastructure from my point of view just doesn’t match launchpad. I am not talking Bazaar vs Git. But the whole Management System.
- Not integrating with Shell.
We want to but there were many problems in the way that we managed to put aside. Yet we need to find the time. If anyone wants to integrate into Shell contact us. But as long as we are not blessed it makes it hard for us. It make sense to integrate Zeitgeist into Shell and give it functionality but it makes no sense to reject GAJ for that. GAJ will offer more Zeitgeist functionality than Shell… More detailed browsing is one example.
- Our high affiliation with Ubuntu and Canonical.
Their mentality about encouraging development and innovation ACROSS the desktop. Yes *apparently* Canonical did its fair share of *mean* stuff but hey lately they have been amazingly nice and supportive. Plus all our devs use Ubuntu.
Canonical hired our lead architect and payed him to write libzeitgeist a C Wrapper around our DBus interface. It would make perfect sense for other GNOME projects to look at it… Thank you Canonical
- Our professional image
Yes I am over enthusiastic but why should my team suffer from that. Zeitgeist has 5 core devs and 3 contributing devs. We all make sure that we are having fun and every1 established his role in the team to offer the best possible combination. Our code is reliable. We hunt all bugs out and are very stable and efficient even though written in python.
—
For now we will live with the downstream, ISV and community projects that embraced us, ask GNOME people to help us get things right, with the Zeitgeist and GAJ inclusion for 3.2…
You are right. The arguments for the decision are strange. +1 for inclusion.
Here. Have a cookie.
I ranted about the decision from a much less informed perspective on my site the other day. The decision stumped me because, to an end user like me no party to the semantics of GNOME politics etc, GAJ was a perfect fit for the aims GNOME 3 laid out in their plans.
I want to applaud the work you do – and have already done – because, without sound cliche, it is the fuuuutuuureee!!
Vince needs to post a concrete list of things you should do to get into Gnome. That stupid talk about feeling is just dumb.
I would love to see this included! Do let me know what I can do, email somebody, sign a petition, or anything. I am not a programmer but would like to help. Am donating now.
Seth
who ever is running GNOME in the ground needs to step down. seriously. The shell is utter crap (let’s be honest, it looks like, and is, utter crap), no real new technology is being accepted. What ever happended to using svg like plasma did? or embracing python like GAJ and zeitgeist?
seriously who ever it is accepting a payout (yeah we know it’s not true) from microsoft (oops there it is again) needs to quit with it. (to stop these accusations, press 1, or in other words, drop the dictatorship and listen to people).
We’ve told you, repeatedly, what is required.
You are just not listening because you are on a mission to sell your technology.
I love you, Seif! Remember, Do was never part of GNOME and it still brought joy to many users!
So you’ve changed your stance on Canonical since yesterday when apparently you believed “RH and Canonical are killing GNOME and its sad…”?
http://www.omgubuntu.co.uk/2010/06/i-know-im-in-no-position-to-criticise.html#comment-55609233
Just what I thought!
I love Zeitgeist!
“like Empathy does with its widgets”
Just to get things straight here, libempathy is no more so emoathy is not providing any widgets to other apps.
About the rest… if you quote, link to the source. If you do not really quote, don’t make it look like you quote.
Personal feeling about zeitgeist: nice idea but just nowhere near “ready”. Sorry but waiting 20 seconds for stuff to show up even in the newest GAJ is not what I call “usable”…
“I don’t know any official module that use Tracker yet.” — Totem’s had a Tracker plugin since late 2007, though it’s optional.
It would be really sad to see such a promising piece of technology as Zeitgeist go unused. I hope this won’t put off other applications that would like to use it.
@Matthias please list them down in a comment. And how is selling my idea interfere with gnome inclusion.
but that makes no sense!
what does it mean “standalone application”?!
All applications of gnome are “standalone”!
Empathy is standalone, firefox/epiphany are standalone, tomboy is standalone, rythmbox is standalone!
What do they mean?? I really can’t understand it!
And Zeitgeist is a fantastic technology, it’s soooo stupid to let it down!
It’s also they’re fault if it isn’t integrated: if I’m not wrong, they never really worked with you!
If things are going this way, W Ubuntu and M Gnome3!
In fact, now Gnome3 is going to be Gnome2 + Mutter + Shell!
Where’s the big innovation?! The only innovative thing is the Shell, but without Zeitgeist and Clutter it’s so.. annoying!
I’m happy of Ubuntu’s team using Zeitgeist (or am I wrong? I believe Zeitgeist will be part of Unity..)
@Michael http://permalink.gmane.org/gmane.comp.gnome.devel.announce/101
Also we are updating the UI… trunk is much faster now…
I was right!
“Unity and Gnome Shell are complementary for the Gnome Project. While Gnome Shell presents an expansive view of how people work in complex environments with multiple simultaneous activities, Unity is designed to address the other end of the spectrum, where people are focused on doing one thing at any given time.
Unity does embrace the key technologies of Gnome 3: Mutter, for window management, and Zeitgeist will be an anchor component of our file management approach. The interface itself is built in Clutter.” [Mark Shuttleworth]
Funny..
Unity is more Gnome than Gnome itself
Seif, I think many people have been put off by the speed of the latest release. Any this is the release zeitgeist/GAJ are judged by, atm. If trunk code is much faster that’s very nice, but decisions have to be made at some point, based on some version of the code.
That said, if the fast code is ready, release it and try to push it into the latest distributions, so that people can experience how well it works! Ubuntu, Fedora etc will ship zeitgeist and GAJ even if they are not part of GNOME (Fedora 13 has zeitgeist 0.3.3.1 and GAJ 0.3.4). If the techology works for most people I am sure it will be adopted by GNOME very quickly.
@Leolas Firefox may be standalone, but it’s not part of GNOME. Epiphany is part of GNOME and is deeply integrated with the rest of the platform.
Seif, when they say integration they mean that GAJ as a separate browser doesn’t make sense. What r-t wants (and the wider community would probably agree) is to have GAJ’s functionality spread all around GNOME.
Having a timeline based search in Nautilus menus would be one thing, for example. That’s totally different from launching GAJ separately. What they expect you to do is to push for such integration. Whether you drop GAJ as a standalone app or turn it into a library, the goal is not to collect lots of individual apps.
Think about why we don’t have a standalone “GNOME Calendar” app, but an integrated solution in e-d-s which shows in Evolution, the panel, Tasks, Tasque, Hamster, … such app would have been rejected on the same basis as GAJ.
You should not feel this as plain rejection, look deeper into the reasons, think why would you have rejected GAJ if you had gave those reasons.
Keep the hacking, with patience.
@diegoe
I’m not involved in all this, but isn’t Zeitgeist the thing you’re are talking about?
I’d integrate with the Shell “search tool”, for example, or with Nautilus..
And he’s already doing it, in Nautilus-Elementary.. but actually, Nautilus-non-Elementary didn’t seem to be being much developed, in the last times.. but again, I’m not a programmer and I’m not really so involved in this
Can’t you people collaborate more?
@ReinoutS
Yeah, Firefox isn’t a Gnome project, but it’s full of project that are standalone software, but are pre-installed on gnome..
@Michael
GAJ e Zeitgeist sono semplicemente troppo ubuntu-based.
Sono sicuro che se Seif creasse un branch su GIT e rimuovesse Ubuntu, Canonical e Mark Shuttleworth dal suo vocabolario (e dal blog) i suoi progetti sarebbero accettati senza tanti problemi……
Gnome-shell senza l’integrazione in Zeitgeist ha una INUTILE E ORRIBILE implementazione delle funzionalità di ricerca. Perchè nessuno degli sviluppatori ha mai proposto a Seif un’aiuto? Perchè a soli 5 mesi di distanza dal rilascio di Gnome3 mancano ancora tante idee su come risolvere gli evidenti limiti di usabilità di gnome-shell e le persone che propongono soluzioni intelligenti vengono messe alla porta?
Sembra ormai evidente come RH cerchi di limitare il potere decisionale di Canonical in Gnome
@Seif
Continue on your way, if you do not GNOME Foundation considers other at will provide recognition for your work (as David Siegel says just above!)
“But I don’t care what they say, I’m in love with you.
They try to pull me away but they don’t know the truth. My heart’s crippled by the vein that I keep on closing.”
http://www.youtube.com/watch?v=EAMUIBKGZfAd.
Hi Seif,
Tracker had the same issues and it took a long time with multiple rejections before it was allowed to be an approved dependency
An initial rejection is not uncommon and no need to take offense
Lastly, I would very much welcome zeitgiest functionality in tracker – that route would solve all the problems and be a win-win for everyone
Anyway whatever happens Im sure things will work out in time…
jamie
I’m actually quite happy to see the rejection, it means that the process is working well and that some folks out there are still committed to bringing a coherent and quality desktop.
@Jamie
Older versions of Tracker and the current Zeitgeist / GAJ are not comparable IMHO.
Tracker has long been an experimental software, slow and incomplete. It also requires many of the resources of the system that was installed.
Zeigeist is lightweight (well, not exactly a featherweight but runs just fine on my eee701 cmq), efficient (much more of the technologies underlying the gnome-shell) and FLEXIBLE.
I think that the only fault is the rapid evolution of Zeitgeist!
@Barra
You are badly misinformed about tracker
Tracker is worked on by Nokia and runs well on N900 mobile devices and should have no problems with any netbook out there . The storage component (metadata db) is independent of the indexer and is well optimised. Its written in C and has some of the smartest hackers working on it
It also will be used in Shell and its fairly easy to add a miner to index recent files or any of the other zeitgeist functionality with it. Again an easy route from zeitgeist into shell land is via tracker.
Just for correctness, Empathy does not provide its widgets as a library any more. Really it never should have either.
Personally I’m sad not to see GAJ being a part of GNOME 3, it feels like the wrong choice.
@Jamie
I ma glad tracker got in… We will surprise you next week with some work I did with Rob Taylor… We have a Zeitgeist Extension now that pushes its data into Tracker Store… However it needs a custom branch of Tracker that supports our ontology.
Again it is a matter of time but we are releasing the extension soon with a couple of others as an official Zeitgeist supported extension.
@diegoe you are talking about Zeitgeist not GAJ… It is like saying. You can use Zeitgeist everywhere, we already have extensions for rhythmbox and other community projects.
GAJ is a UI for itself that shows up info that nautilus wont be able to do…
[...] to be a board member, I highly recommend communicating in a different way than what you’ve done now. Anyway, just read the comment by Diegoe, I think it is pretty clear. In addition to that comment, [...]
Well, make it standalone then.
I am a KDE user but I like Zeitgeist, it’s lightweight and provides acceptable results even without tracker running, and GAJ is a pretty nice interface.
If you made it a “worldwide” (AKA working properly on any DE) standalone app, everyone would benefit from it, and it will possibly become more important on its own.
I’ll chime in that I’m with @diegoe on this one.
@Sief, you’re making the assumption that your app fits with Gnome’s principles for fallaciously tautological reasons (i.e. it fits in because you think it fits in).
Let’s look at a use case that exposes the weaknesses of GAJ: A new file.
How do I find/create a file that Zeitgeist hasn’t seen before (and thus, can’t expose to GAJ? I can’t! I have to open up Nautilus or something else to access the file first, then Zeitgeist will be aware of it and able to publish it to GAJ.
This is ridiculous and breaks the tight integration that Gnome is going for. This also means that I will *never* use GAJ for serious work. Because Nautilus already exists and is more fully featured, GAJ is a redundant app.
It makes a whole lot more sense to put GAJ’s feature set into Nautilus and other existing apps. And no, I’m not confusing GAJ with Zeitgeist here. You keep repeating this and I think it’s contributing to your difficulty in understanding the criticism that caused GAJ to be rejected. Stop trying to correct people, and start listening to what they’re saying even if they’re saying it imperfectly.
I think Zeitgeist and the features that GAJ demonstrates are fantastic and something worth getting into Gnome. I think GAJ is a great proof of concept, but viewing it as more than a proof of concept exposes many weaknesses and missing functionality.
It’s time to start listening to the feedback you’re getting. Even if it stings a bit to encounter different viewpoints, the feedback *is* valuable. These are both your (potential and actual) users and also engineers with a lot of experience in how to build a cohesive user interface. Don’t make the mistake of assuming these people are wrong simply because they don’t agree with you. We all want to see Zeitgeist and GAJ succeed.
And now nothing about GNOME 3.0 holds my interest whatsoever.
@ Jamie: Today tracker is a great product but you can not say the same of the old release. Today is part of gnome in the past it was not for this reason
@Brett:
First let me thank you for an amazing constructive mail. Let me elaborate on some issues.
Yes I never said GAJ does fit in. I just think the that the reasoning is to vague… I do get What Dieogo says and I agree with him in a lot of points as well as I agree with you. I could even list more points why GAJ should not be included in GNOME.
Yes Zeitgeist has a little weakness with created files but we r fixing it.
Why should GAJ replace nautilus. We never assumed that. For me GAJ is to keep an overview of my workflow and quick access to stuff. It does not compare to nautilus in any way.
We do have Zeitgeist features in some apps but again maybe not enough apps.
You assume that we do not work with community while most of GAJ was designed and set up by other GNOME members.
Again I am sorry Diego if I it comes over that I took things personally, and maybe I did a little, but the reasoning for me as well as others as you can read here is very vague.
I 100% agree with Diego here, he clearly explains the point of most of the long-standing community here. Having GAJ standalone does not make any sense to me, I’d rather have contextual, nice, usage of Zeitgeist to harness its full potential.
Really, I so don’t care about MacOS X, but still, I could not even think they’d actually provide such an “independant” app.
Really, I’m 100% behind Zeitgeist project, but I’d like to see the Nautilus, Rhythmbox, etc. patches flowing up the git repo. The projects can always have a soft dependency on Zeitgest.
If you manage to have 1 or 2 modules with Z integration, this will be a huge success, and this will encourage everybody to better integrate with your technology.
It’s great to see those blog posts, but I want them upstream and in my distro
Congrats for all the hard work, and go on, you’ll make it next release, you’ll certainly be included in the next release, so you’ve got a couple months to create nice patches and work with the main projects
“…talking about Zeitgeist not GAJ…”
Let’s make this very clear: the rejection was about GAJ and the rejection means zeitgeist was not pulled in as a dependency.
In its current form, users can install GAJ on their distribution as they please, distributions can even decided to ship it by default and people how don’t like it can remove it. GAJ is stanalone and just happens to depend on zeitgeist, but nothing else does. This may change until the next round of distributions, but only slightly: I really don’t expect you will be able to provide patches for DEEP integration of zeitgeist into GNOME in 2-3 month.
So, the rejection was both correct and not hindering you in any way.
Have a look at the conduit project for example. Anyone using it? It was pushed into GNOME and excepted, without any integration into GNOME. Distributions never picked it up because it was not mandatory (= standalone) and didn’t really work that nice.
Anyway, I think coding a peace of software and then trying to figure out how it fits into GNOME is always a bad thing. The process should be the other way around: identify a number of issues in GNOME, try to improve the workflow by designing a better GUI, write the software to solve the problem.
I encourage you to apply earlier in the process next time to give more time for discussing and working out any issues.
I thought Zeitgeist was one of the “blessed” technologies in GNOME 3? Not GAJ necessarily, I see that mostly as a testbed and demonstration of Zeitgeist’s strengths, a standalone application which provides a more powerful or involved way of working with these new technologies.
This is a classic chicken-and-egg problem. GNOME wants to see Zeitgeist used in more applications in order to accept it as a dependancy, but Zeitgeist needs GNOME to make a commitment in order to attract developers. As Seif noted, there have been some small steps in making Zeitgeist work with applications like Nautilus or Rhythmbox. But ther won’t be any serious work implementing these technologies and/or replacing old ones (GtkRecent) if Zeitgeist is not considered as a dependancy, at least for some future release after 3.0.
Your post is interesting, but a little cryptic for those not familiar to three-letter-acronyms starting by ‘G’…
You might have reached a somehow larger audience by replacing GAJ by Gnome Activity Journal (which I myself had too search/guess to decrypt).
Thanks for giving everyone on the Linux Haters Blog a good laugh at the pathetic state of politics in the GNOMECentralCommunistCommittee™
Every post like this finds our position vindicated.
Thought I’d better clarify: I’m not the hater himself, just one of the people who frequent the flamefest on his blog’s comments.
Yelp has a tracker back-end. Yelp will be much faster for Gnome 3.
Hi, I’m very interested in Linux but Im a Super Newbie and I’m having trouble deciding on the right distribution for me (Havent you heard this a million times?) anyway here is my problem, I need a distribution that can switch between reading and writing in English and Japanese (Japanese Language Support) with out restarting the operating system.
My 2 cents:
Shell is terrible, ugly, in early stage of development, based on hypothesis not confirmed by experiments yet and still the main thing in GNOME 3. Zeitgeist is working thing, based on very powerful idea and as result making a lot of buzz and getting a lot of attention. Many people seeing it (obviously wrong) as the main advantage of GNOME 3. But that being said, GAJ is really not needed as stand-alone application, because its main purpose at the end is still to locate some files (among other things). But thinking in terms of user scenarios what do we do when we find the files we want? We want to manipulate them somehow. Open them in some application, send them, copy them, rename them, delete them, move them, apply batch procedure… And therefore we have a file manager. So having both Nautilus and GAJ as different applications basically means that we have to use GAJ to find some files and then somehow transfer/select this group of files in nautilus (i have no idea how to do that if there are in different folders/devices) and do whatever we want to do with them. This makes no sense and is the reason for the “no need for standalone application” argument for GAJ, not Zeitgeist. As I(and some other users) see it, GAJ functionality should be just another view mode for nautilus.
Keep up the excellent work, Seif (and team), I am waiting for the geo-location part of zeitgeist as well
I’m really happy that it wasn’t accepted since the idea behind it is plain stupid.
[...] is very picky when it comes to accepting commits upstream into the core GNOME modules, for example, Seif Lotfy’s work on Zeitgeist was denied. It’s also definitely worth reading the points of view from Ubuntu legend, Jono Bacon, who is [...]
just integrate this piece of shit with Nautilus already. thx