The Enlightenment Foundation Libraries are a collection of libraries that make up the foundation of the future enlightenment-0.17 window manager, and of several applications:evidence – the file manager (screenshots: 1, 2, 3, 4, 5, 6),
entice – image browser (screenshots: 1, 2),
entrance – display manager (screenshots: 1, 2, 3, 4),
euphoria – xmms2 frontend (screenshots: 1, 2)
and many more.
Here is an announcement, an interesting example “How to make a DVD player in 17 lines of code” and the Unofficial Enlightenment User Forum.
thx for info!
I forgot to mention of
envision — video player
http://www.cuddletech.com/img/envision_frontpage.jpg
engage — an OSX docker clone
http://xcomputerman.com/gallery/screenshots/engage2?full=1
http://atmos.org/gallery/sshots/atmos-2004-06-06.png
and many more..
Also have to add, that thanks to imlib2/evas everything acts smoothly and is animated.
All I have to say about Enlightenment is that it’s unusable. Visual Usability is really thrown out of the window in favor of eye candy. Now, if the target of these developers is to simply research the “coolness” of things, then E is a great, fun project. If their target is real life end users though, E –the way it looks and behaves today– is useless.
is Rasterman the only one who develops Enlightenment?
I assume Enlightenment is one of those projects like Black Box that will never reach 1.0 because it will never be perfect.
You are probably right about it’s usability. I don’t know I haven’t tried it, but it looks sooo great, and it doesn’t try to copy any other existing DE.
Well, there is no Enlightenment (0.17) at the moment,
so in fact it’is useless, the wm’s code will be completely rewriten, when the libraries are ready.
Quite complex and responsible solution in my opinion.
We can not talk about usability, when there is no window manager.
@Eugenia
Unuseable for you maybe, but for a long time Enlightenment was my desktop and it was perfeclty useable as a Window manager and very basic desktop. Perhaps you should clarify by saying “unuseable for my purposes or in my opinion unuseable for the average user.”
This is exactly what I said. Read more carefully when I talk about “end users” and “my take”.
And please use the same subject header when replying.
The problem with E, regarding usability, is that it is very different from the common DEs.
However – it took about five minuits before my father and sister managed to get a grip on it (this was DR 16). That says quite a lot. Ok – I had set it up with names such as webbrowser for mozilla and so forth, but still.
The window tumbnails in the desktopswitcher hooked right into my fathers spinal cord wich is odd since he cannot grasp the multiple desktop concept in gnome.
E has paved the way for lot’s of niceties in the OSS world and will probably continue to do so. It’s pretty. It’s great, but it is not gnome wich sort of just works.
That said – I play around with E a lot – it really talks to the geek in me and allways make my friends working at m$ to drool…
If their target is real life end users though, E –the way it looks and behaves today– is useless.
This is where you are very wrong. Fact is, every user is different, and some users will be very comfortable in this environment. I don’t know if I will but I’m sure some people will. Hell, people are still using E16 even though its pretty ancient.
I agree that it’s not a desktop that you will want to give to a new convert that just wants to get their work done. This is good, there’s already 2 mayor DE’s aimed at that user. E, however, may be perfect for those users that like to customize their computers, those that may have been running Litestep or something on Windows.
Just because these people don’t conform to your picture of the “real life end user” doesn’t mean they aren’t using their computers to get work done.
Is all this eyecandy still drawn in software by the way? If so, does anyone know if there’s plans to make it hardware accelerated? I’m a sucker for eye-candy but not if it slows down my computer significantly.
U can choose software or hardware acceleration.
It’s up to you (i.e. GPU)
All I have to say about Enlightenment is that it’s unusable. Visual Usability is really thrown out of the window in favor of eye candy. Now, if the target of these developers is to simply research the “coolness” of things, then E is a great, fun project. If their target is real life end users though, E –the way it looks and behaves today– is useless.
If I remember correctly, rasterman gave an interview a few years back, where he said that the “linux desktop is dead”. His main goal was developing those libraries for embedded use. (that’s where he saw the future of linux, on embedded devices). So I doubt that this project has real life end users as their target market.
I can’t find the article itself (it seems to be taken down), but there are still some references on google.
all the e libs are ’embedded use’ ready, true, but the linux desktop is on top again
All I have to say about Enlightenment is that it’s unusable. Visual Usability is really thrown out of the window in favor of eye candy
Well, believe it or not you’re right with your comment when looking from your viewpoint, except your viewpoint has missed the point completely. At least two reasonos why:
1. Enlightment never intended to be end user or Joe Sixpack friendly. Enlighment was always highly teamable, extremly fast and as geekish as possible. 0.16 was still the most explosive and attractive desktop I’ve ever seen. I expect 0.17 will be also. You’re comparing goals between Gnome and Enlighment. WRONG.
2. Choosing the right theme makes E’s Visual Usability at least as good as Gnome. It’s just that there is some kind of standard that all E themes are very dark and uber-geek futuristic. Make a theme contest as you did with Gnome, you might be impressed
And again:)
#!/bin/sh
while [ “1” = “1” ]; do
Gnome=”Joe User can figure it out” && E=”Uber Geek wanna impress geeks”
echo “repeat after me: “”$Gnome”” != “”$E”
done
Personally, I’m dropping gnome as soon as I find time to install E17. p.s. You can still run gnome applications over E.
Evas has had a Qtopia backend for a couple of years now (in addition to linuxfb and directfb backends), so the EFL libraries are already quite embedded-ready. In fact, I believe there is already some work going into writing input drivers for Ecore so that ecore_fb/evas can be used on embedded devices with full input capabilities.
But the EFL was created to break new frontiers in GUI capabilities on the UNIX desktop, and the actual window manager is still in early development and not even at a usable stage yet. Unlike some who are quick to dismiss the work of the Enlightenment project based on first impressions of an older-generation product, I would encourage you to actually check out these libraries and see what they do. Emotion, for example, allows you to write a DVD player in just 56 lines of code, no kidding. And it’s got full canvas capabilities including alpha-blending etc. — see this screenshot for example: http://xcomputerman.com/files/emotion_trans.png
Argh!!! Tiny fonts!!! 😉
I agree with Eugenia that Enlightenment seems to never really have focused on usability. Scifi like futuristic “cool” looks seem to have been a major (main?) goal for them.
Just take a look at screenshots like this from the official E homepage: http://enlightenment.org/pages/img/efl_shots/az-desktop.png and tell me if you can take that sort of window manager looks seriously… Sorry, but I really can’t. Ok, it is only a theme, but why should any grown up person prefer such crappy toylike interface anyway? For some reason or another even most E themes seem to prefer such futuristic but unclear interface and buttons and other such things that are pure crap IMHO…
Enlightenment has some good ideas though, but I haven’t yet found any good reason why I should use and prefer it instead of GNOME, KDE, IceWM, Fluxbox, Openbox, Waimea etc. If somebody disagrees and thinks that E16 is very cool, ok, but personally I cannot see why…?
I think that E has been a good example of open source projects where young enthusiatic developers have lots of ambitious ideas but lack expertise on fields like usability design. Other similar examples are, for example, some *nix media player GUIs.
However, a few preliminary screenshots of E17 that I’ve seen promise that E could get better. So, I say nothing about E17 yet. Time will tell when the devs get E17 finished. Anyway, I sincerely hope that the E developers could spend some more time studying usability. The project could only greatly benefit from that, I’m sure.
Can’t u post pics that are LESS THAN 2 MEGS for the poor 56k users ?
like hm.. 200K ?
Have you even used it for more then a day, a week, a month ?
now that the libraries are nearing completion anyone have any idea when the DE will be fully complete?
Anyone know the font in “evidence” screenshot 1 used in the eterms?
engage does definitely look great
http://xcomputerman.com/files/engage.mpeg
I’m sorry, but am I the only one here surprised at what passes for a bleeding edge “explosive” and “attractive” (Eugenia’s words) DE these days?
Obvious usability issues aside, I unfortunately cannot find this anything other than ugly. Which sucks because I WANT to like Enlightenment – I WANT eye candy, but this just doesn’t seem to look that good for me – fonts that aren’t anti-aliased and too small to read, colours that clash, *really* nasty looking terminals, ideas that just don’t seem fresh when compared to the latest offerings from Metacity and KDM, and just the whole Amiga desktop feel of the thing doesn’t really do it for me in the 21st Century.
I found this wich may be of good use to those with rpm based distros.
http://sps.nus.edu.sg/~didierbe/
2 screenshots:
http://netart.eu.org/data/pic/g27-etools.jpeg
http://netart.eu.org/data/pic/g27-etools1.jpeg
@Metic
The screenshot you referenced is clearly a developer’s desktop – that’s E17. You’d have to seriously *try* to run it. Between compiling the libs (in development) and then the actual window manager, file manager, etc.. Take that, plus the fact that there are only a handful ( http://themes.freshmeat.net/browse/978/ ) of themes for it, you’d wind up with something like that.
If you’re looking for something usable – e16 will get you there. Between the multiple desktops and the way e handles them, window snapshots, and all the flexibility of themes, eye-candy, interoperability with every other DE (Gnome, KDE, wmaker, etc) it’s like a skeleton that you can arrange any way you see fit – the only catch is – *you* have to do the arranging.
Try this: Install e16, and click on every meny option. Use the keyboard (alt,ctrl,shft) combinations, look through the neat-o e help thingy.. Try a few themes. It’s much more simple than KDE/Gnome, but it’s also pretty well right on target. I bet you’ll be at least a fan, if not a fanatic.
Long live Enlightenment!
P.S.
There’s no closer functional equivalent to the way OS X works than e16 in my book. That is, with the addition of rox and a few other things.. e16 + SuperKaramba + rox-filer… Huzzah!
Maybe someone else can take it and make something less ugly out of it. Rasterman and other E fans seem to have rather ‘unique’ taste
I wouldn’t put too much value in some abstract definition of usability. The only authority on usability is the user, not some abstract studies or theories.
Yes, this means that the enlightenment desktop can be just as useable as MacOS for someone that’s used to how it works.
As I wrote so critically above, here are a few good points…:
I think that the basic idea that E devs have of a desktop shell might be quite a good one. Something smaller and less bloated than GNOME and KDE (but more than a plain window manager) is needed more and more.
Also, it doesn’t hurt if someone tries to develop some brand new ways to handle window, work space etc. management etc. like E devs have tried to do. There’s no denying that the devs have ideas. Also E doesn’t need to serve everybody.
The basic tools to build a good competitive DE/WM and user interface seem to be there. E just might need some more serious usability design.
Less scifi toys, more serious tools, please.
One more thing.
The look is good,
but THE FEEEEL is just fantastic..
@those who say “oh no! no!”
This is alpha stuff, guys.
@Eugenia:
I agree with you. Looking at screenshots, the Enlightenment WM looks very cool in the eye candy department. But if you just want a WM, then Enlightenment is software bloat in its purest form (and not finished either, after much development effort). It looks more like a mini operating system on top of Linux to me. If a mini-OS is what you want, then why not configure a customised kernel/C library/WM combination, and make it just that? (So that it doesn’t need Linux anymore, and can go its own way).
Some would call it a full-blown desktop environment, but if that includes a programming language, X replacement and more, that’s not a good description either, and still a lot of bloat/duplicated functionality. For me, that puts it in the same corner as KDE & Gnome: avoid it if you prefer a lean & mean system.
Its developers are clearly capable, put in a lot of work, and yet, what have they got? You tell me what exactly it is. Maybe they just don’t know what they want?
Its too bad that despite 30 different WM’s for Linux, you still can’t share themes between them, so that a background walpaper, common theme elements and (maybe) keyboard shortcuts are used by whatever WM you choose.
If a mini-OS is what you want, then why not configure a customised kernel/C library/WM combination, and make it just that? (So that it doesn’t need Linux anymore, and can go its own way).
What are you talking about? Mini-OS? Just because it has its own little scripting language? Enlightenment is not an X replacement nor does it aim to be.
For me, that puts it in the same corner as KDE & Gnome: avoid it if you prefer a lean & mean system.
Haven’t actually tried it have you?
Its too bad that despite 30 different WM’s for Linux, you still can’t share themes between them, so that a background walpaper, common theme elements and (maybe) keyboard shortcuts are used by whatever WM you choose.
Ridiculous amount of work for as good as zero gain. Who cares if you have a different theme for each WM. While it’s nice from a design perspective to say, “let’s get a common base for all this similar software”, in reality it’s usually not worth the effort.
Two points:
1) You miss the fundemental point of Enlightenment, especially E17. A key goal of E17 is that Enlightenment should provide only mechanism, not policy. A UI theme should be able to completely change the look & feel of the interface, from sci-fi and unusuable, to drap and newbie-friendly.
2) EFL is more about technology than UI. It’s up to others to make a good UI out of it. What the Enlightenment project is providing is some very powerful libraries to make that task easier. From a technology point of view, the EFL is extremely promising.
Enlightenment is the only desktop that has made me drop my jaw unconsciously. Beat that! So recently I decided to try the new “Entrance” in cvs. I installed it, and I just kept starring in awe at the screen for what seemed like eternity.
My peeves:
1). Usable desktop configurable applications:-
E16 is just annoying to configure. Some of configuration options don’t clearly explain the functions they serve making configuring the desktop and experimentation or trial and error game. No, I don’t even understand how to hack the config files. Admittedly, I have tried to.
2). Fonts:-
How do you adjust the damn thing in E16? They are tiny. My eyes begin to hurt after 15 mins after using E, only to realize that it’s because I am squinting.
3). Apps:-
E in general needs more apps. Using my GTK+/GNOME apps in E just feels wrong.
Finally, Enlightenment is the only desktop that will make you say,”Oh sh#t! Wow! Oh sh#t! Wow!” continuously for an hour. I like Enlightenment because it doesn’t try to be Windows, Mac, CDE, GNOME or KDE.
The eye candy and animations in E make OS X look 1920, I ain’t even kidding you. But like Eugenia said, it is far from usable(read: awkward to use for day to day purposes).
“Its too bad that despite 30 different WM’s for Linux, you still can’t share themes between them, so that a background walpaper, common theme elements and (maybe) keyboard shortcuts are used by whatever WM you choose.”
When last I checked there were at least a half dozen windows theming engines, wms, and generally other themable elements for windows, none of which use the same format.
the proliferation of WMs, and desktops, does not negatively effect Linux. Any more than the proliferation of WMs for Windows or MacOS.
I can’t wait to see E17 finally released. I’ve been waiting for it for about as long as its been in development, it will be interesting to see how E17 stacks up against gnome 2.8 after not having used E17 since Gnome 2 was first released.
All I have to say about Enlightenment is that it’s unusable.
I’ll agree that it is truly a different GUI approach – while some compare GNOME to the Mac UI and KDE to Windows, E is definitely something original. Now, originality is not necessarily a good thing – especially in UIs, where people have come to expect a certain behavior of “how things work”. That’s why accusations of Linux being a copycat re: Windows miss the point – reinventing the wheel rarely is a good idea…
On the other hand, I wouldn’t call E “unusable”, because evidently some people use it – it has a small niche market, and it doesn’t seem to mind that much. Is it a mass market GUI? Of course not. But it is definitely useable by its few followers…
Visual Usability is really thrown out of the window in favor of eye candy.
Especially “brushed metal” eye candy. I guess it’s okay if you’re into the sci-fi desktop, with transparent terminals and chrome widgets. The new one looks good, though. If that OSX-like pager is faster and/or less resource-hungry than what you can get with gdesklets or superkaramba, I might be tempted to test drive the next Enlightment for my root desktop. Presently, I use KDE for my normal user logins, and XFCE4 for my root login – I like a lighter GUI when I’m just doing admin stuff, mostly installing KDE versions! 😉
Now, if the target of these developers is to simply research the “coolness” of things, then E is a great, fun project.
I think it appeals to a rather limited audience, and undoubtly the coolness of the thing is an important factor in that appeal. There is a research aspect to it which can also be positive to UI design in general – who knows, E may introduce new UI ideas that are eventually picked up by other WMs and DEs. That’s the beauty of the *nix “ecosystem” – marginal project that find niche market and serve specialized purposes.
There is no “one true UI” anyway, simply because people are different and have different tastes. There are general principles but at some point some users’ personal preferences will become incompatible with those of other users.
Video games have shown that you can have consistent elements in UIs (for example, the ubiquitous “press Triangle to cancel” on PSX/PS2 games) while having completely different styles. Some are better than others, but the fact is that users adapt quite quickly to a variety of different UI styles and structures. Granted, games and OSes different, but still this is something to consider.
If their target is real life end users though, E –the way it looks and behaves today– is useless.
I think you’re being too harsh. If by “real life” you mean “mass market”, then you’re right to doubt its chances at gaining market share – but for its intended sub-class of users, and to some extent the rest of the WM and DE community, it serves a definite purpose, and therefore isn’t useless.
I’ll try for my root session it when it comes out, but I’m not about to switch from KDE for my regular users…
that evidence (e file manager) looksinterestingly simple to use, im guessing that as you browse the file system the address to where you currently are grows along the top and at any moment you can pop back down the tree. i wonder tho if there is a rightclick menu so that say i can spawn a new window based on a section lower in the tree to browse from for when im moveing stuff from one part of the tree to the other…
i find this interesting, but then i have a soft spot for customizeable desktops (im pondering replaceing explorer.exe with sharpe when release 15 goes official)…
make that sharpe relase 5, not 15
I am so happy about this. I have been watching this project for along time and congrats to the developers who have put this together. And Oh I find it very useable. It’s a desktop WM for Geeks not for Joe Windows. It has always been fast because they (the developers) have taken there time in optimizing the code and are very good at it.
…who comment on how ugly E17 looks?! There IS NO E17 yet! Rasterman has been working on E17 for about 5 years. That’s how old the latest cutting edge E16 is. When E came out it was practically a miracle for users who wanted both something nice to look at AND usability. It was completely customizable and extremely simple, quick and useful!
The stuff you are seeing pictures of today isn’t even an alpha window manager. It’s just the bits and pieces of what will make up the window manager (the backend of it) slapped together to show you something. And yah, it may look tacky or not to your taste, but it is by NO means the end all for this future product.
This guy is full of ideas and is an excellent coder. When the REAL E17 comes out you will see something worth looking at. For now, just pay attention to what the article was REALLY about and that is the behind-the-scenes libraries this guy has been slaving over for years that has the potential to blow your multimedia sox off!!!!
Sheesh.
Mike
When you’re done commenting on e16/17 could someone please give some insight into EFL?
How dows it stack up against gtk/qt?
Should KDE/GNOME consider converting to EFL?
Any intressting new design filsofies? I understand that openstep (cocoa, gnustep) should have the best design, how is EFL compared to that?
I installed the rpms, but I’m not sure how to start the actual Window Manager. I did get Entrance working (looks beautiful – certainly one step above gdm or kdm) but when selecting Enlightment as the session, it started TWM instead…
where I can run Enlightenment on a four-LCD layout running full-power graphics and video apps, with a two-LCD layout running database front-ends, spreadsheets and office stuff on KDE, another two-LCD GNOME setup for programming and suchlike, and XFCE running OpenOffice on another coupla LCDs.
That’s my ideal setup. With three grphics cards of course.
Any hope of that within the current millenium?
Is not a Window Manager, people.
Read the article summary again.
The Enlightenment Foundation Libraries are a collection of libraries that make up the foundation of the future enlightenment-0.17 window manager…
Note: future enlightenment-0.17
It doesn’t exist!
Finally, someone who actually bothered to READ what the news was about instead of launching off into brainless troll mode. The sheer cluelessness exhibited by many of the posters above is downright embarassing.
The EFL is a group of GUI libraries. Its chief components are:
Evas – a very fast, highly optimized, optionally hardware accelerated canvas library
Ecore – A convenience library that provides an optimized event loop, fd handling, timers, data structures, an easy-to-use Xlib wrapper API (ecore_x), a framebuffer wrapper API (ecore_fb), socket communications and IPC, and even a simplified xlib/evas wrapper API (create a window with an evas in it in about 5 lines of code).
Edje – Based on Evas, a graphical design and layout library, quite similar to Shockwave/Flash conceptually. Edje uses Embryo, which is a small scripting language that runs in an 8K virtual machine.
Edb – An implementation of the Berkeley DB API designed for easy storage and manipulation of app configurations.
Esmart – A collection of Evas Smart Objects that find common use in various applications, from GUI layout and text entry to pseudo-transparency and thumbnailing.
Emotion – This library takes advantage of Evas’s YUV converter abilities to make Evas video objects a reality — and they can be treated just like any other object (moved around, resized, colorized or even given alpha transparency values). It currently supports Xine for loading video objects, but its modular design allows for additional loader modules to be added in the future.
EWL – A widget library based on Evas, Ecore and Edje, still in early stages of development.
That is a brief synopsis, and there’s more to the EFL than what I’ve listed above. The app Engage is a nice example of what the EFL is capable of — it achieves perfectly smooth icon scaling and great usability without high CPU consumption, unlike any other attempts to clone the OSX docker that have been made previously. Please go to the web site and find out more about these libraries and what they do, rather than just blurting out whatever first impression you got from using E six years ago or complaining about fonts you see in the screenshots which have absolutely nothing to do with the EFL.
What’s nice is that the design seems to be made generally from the ground up, attempting to throw away some of the current prevailing assumptions about how a subsystem should act.
I certainly hope the design is done well…the API power means everything for this.
I absolutely appreciate that this system is generally system independent (doesn’t HAVE to run on xorg).
I think Rasterman’s approach for this was “build it, and they will come”. I hope it’s able to unify some currently disparate efforts for desktops in the Linux/BSD world.
Hi,
Kudos to Tuishimi and xcomputerman for pointing out the gross stupidity of some of the the posters here – do they even bother reading over what they’ve read before stupidly jabbing away at the submit button? How many of you mindless posters have even tried it (the libraries, that is), instead of stupidly spouting what other people have said?
This article isn’t about the E17 WM (non-existent), but about the underlying libraries which will power it..
bye,
Victor
(Off-topic rant on E16)
Eugenia: What the? Simply because a WM happens to look ‘cool’ (is that even quantifiable?), you ignorantly write it off as another ‘unusable’ WM.
Raster and co evidently put enormous effort into making E polished, from the well-written docs (Did you bother to read the docs, before saying, hmmm, I can’t use this?), to the *fast* code (I assume it’s well-written and clean, but, as I am not a proficient coder, I’m not going to make any blanket statements on something I don’t know about – unlike some other people.)
My apologies if my rant has offended anyone, but it would be nice if people bothered to explore a topic properly (isn’t that what Linux’s about) before making silly, blanket statments about things they’ve never tried. Sure, E16 may not have been to your liking, but to write it off as unusable because you don’t like configuring settings (which is not neccesarily a bad thing) is simply stupid (which is a bad thing.)
And no, not all themes are dark and gothic – see http://enlightenment.org/pages/img/efl_shots/rephorm-desktop.png
Usability is on a per user basis. I find enlightenment completely usable and have used it as my window manager for a couple of years now. I pulled away from it momentarily to try the new kde and gnome, and to try out all the other alternative window managers, but found all the others desktops to be annoying in comparison to enlightenment. Enlightenment is low weight, fast, extremely configurable, and just stays the hell out of my way, like it should. I can’t wait for e17 to get here and upgrade the enlightenment experience even more though, and this is a sign that it is coming, and by the looks of things, it’s worth the wait.
My fault.
I forgot to mention that e16 has NOTHING to do with e17.
New environment has been written FROM SCRATCH.
This looks fantastic. e17 looks like it will have definitely been worth the wait!
Much respect to Rasterman, xcomputerman and the rest of the Enlightenment team.
My god, I emerged entrance and the core library today, the are awesome, I hope that other DE teams will adopt those libraries into their projects because those are a great step forward for desktops
… when reading “EFL” in the headline, was an overview about the library , its advantages, some insight about its technical merits and differences from the other major toolkits
for example, i’d have liked to hear, whether or not the different approach Evas/Ecore takes toward window/widget rendering would allow for lower latencies and better responsiveness than what is often complained about the traditional X server…
//
The DR17 is now in early re-write, however it is nothing more than a code framework at this point, even if you did find the code it would only show you a dialog window. Nothing more. No menus, no controls, no anything. At this point, emphisis is still on developing the EFL, since that is the real power center of DR17.
//
this is what the enlightenment.org site says about DR17, judging the framework from the preliminary app built upon it is … pointless (IMHO) as would be judging GDI + from the look of the default theme in windows XP… it could be liked or not, but it’s not what’s to be judged
anyway, if one looks at the EFL component diagram at http://enlightenment.org/pages/img/efl-diagram.png it’s evident these dont form a simple “layer” but are pretty much intertwined among themselves… coherence in this apparent complexity couldnt be, without clear design choices, an dthis is one thing rasterman and the E team should be credited for
The EFL looks very promising and innovative! I am sure this will bring a small revolution to the linux desktop. AFAIK you can change the complete appareance and behaviour of your application without touching a single line of code (look at the file manager screenshots) – you can do far more advanced stuff than what the current winamp or glade can do (look at evidence). Applications become features packaged with their behaviour and look. The possibilities are endless!
Could someone explain me how this works on a technical level?
I only worry about consistency: can they make the application work and look consistent without sacrifying flexibilty? This seems quite hard to archive.
OK guys and gals enough of the namecalling. I think what Eugenia means (and shoot me down Eugenia if I am wrong) is that from the perspective of usability,ie. from the vantage point of trained UI people, what is being shown in these prototype images is a long, long way from forming a self-consistent, cohesive desktop. And if you meant this Eugenia-you are right-the release of the EFL makes it now possible for the E17 DE to finally pick up some real speed and become a really great desktop environment.
E has never been about producing a merely suitable desktop for the day to day usage-E has always striven to give the Linux desktop something which it was lacking-an incredibly fast, powerful, configurable desktop with an empahsis on the very latest, cutting edge graphic capabilities of modern computers. Rasterman and co seem more akin to graphic artists than usability experts. I totally agree with Rayiner Hashem-the technology which E represents is the real issue here. Rasterman is a merciless perfectionist-E17 has a long history now-and Rasterman has scrapped virtually everything in E17 multilple times in an attempt to produce a technology which is simply far more advaned and capable that what exists elsewhere in the Linux GUI world.
E17, which is still a way aways, is already capable of doing almost all of the things which cairo, libfixes and libdamges (xorg extensions) promise to do- evas, the canvas technology of E, is optionally opengl accelrated -which means blindingly fast and complete with true alpha compositing.
Rasterman has harshly criticized the RENDER extension and shown code which he has written which is much, much, much faster. Rasterman belongs to an elite group of ultra graphics hackers-his mastery of all things X places him in a league with only very few rivals. Beyond the ground breaking graphics capabilities which E represents and always has E17 is focused on creating an extremely clean, extremely powerful C library for the development of desktop applications.
The original E provided much of the technology which made GNOME what it is today-E was the original WM for GNOME 1.x and E’s imlib brought one of the best graphics libraries to the GNOME desktop and of course esd. It remains to seen what pieces of this new technology will be picked up by other projects. The EFL is a really impressive comprehensive set of libraries which could well form the basis for the most visually stunning desktop of tomorrow.
For those worried about “appearing ugly”- E17 is about enbaling new graphics technology which will likely trump much of what is already being offered- the images posted here range from very new to work that is already over three years old. Applications for E can embed scripting capabilities via Embryo, an 8k virtual machine(!)- E has it’s own ipc mechanism via Ecore- and E also works in directFB and Qtopia….man this stuff looks amazing…
————————-
I see that you are one of the developers behind E so here are a few questions:
1) Is there any dialogue or cooperation between the folks involved in E and those behind the new developments at xorg-ie. cairo, glitz, and the new X extensions by Keith Packard ? Will Evas also take advantage of cairo in the future(ie. display postscript/PDF, scalable SVG etc. ?)
2) Will the accelerated graphics offered by Evas also be available to other applications running on the future E17 desktop ? If so could you give us some insight into how that mioght work.
3) I see that Ecore has it’s own ipc mechanism-will work be done to allow this to interface with D-BUS ?
4) Is the E team committed to the ongoing developments at freedesktop.org -ie. will E be using the new MIME stuff, desktop notification etc.
5) Do you think the latest EFL release is the “final” basis for E17 development or are large parts of it likely to be rewritten again ?
6) Are we looking at a timetable of <6 months to see E17 in action or more like 1-2 years ?
I had E17 from cvs up and running last year but was constantly fighting with ebits … I would *love* to see it in action again and am really heartened by your projects latest releases. I even went thru Rastermans “lesson’s” using the EFL to create a xine-based dvd player….Keep up the good work
I find E’s design much like Blender and other 3D apps – It takes some getting used to, but once you’ve got it, it’s much better than everything else. The first time I tried E I gave up after 5 minutes because I couldn’t figure it out; Then a few months later I tried again and something clicked, and I haven’t looked back since
With E17’s libraries and demo apps themselves: As with the other guy, it’s the only thing to have actually made my jaw drop (I didn’t even think “jaw dropping” was real until it happened). As with flash, it’s easy to make something unusable, but there are some very talented people working there, and their work is some of the most addictively usable stuff out there (once you get over the “OMG! E isn’t like Windows!” thing).
3) I see that Ecore has it’s own ipc mechanism-will work be done to allow this to interface with D-BUS?
I had proof-of-concept support for DBUS .12 in evidence. While that is obviously outdated now, I think it shows that we follow DBUS developments with some interest. I’m pretty positive that we will eventually add current DBUS the right way, using eCore.
4) Is the E team committed to the ongoing developments at freedesktop.org -ie. will E be using the new MIME stuff, desktop notification etc.
Speaking for myself, evidence supports XDND, XDS, clipboards, and thumbnails as per the standard. MIME detection is supported using libmagic (the library of the “file” command) and/or one of the fd.o implementations.
fd.o are working on more MIME fu that determines what app handles what file-types (like mailcap, basically); once that is ready, I intend to support that as well. While current DBUS might not work with evidence, ev still listens for eCore-IPC (for its shell “evsh”, the GUI configuration tool “examine” or whatever), and optionally KDE’s DCOP.
5) Do you think the latest EFL release is the “final” basis for E17 development or are large parts of it likely to be rewritten again ?
EFL is pretty good this time around, thankyou. : )
PS Speaking of usability, it’s not intuitively obvious how threaded comments/replies work on this site? Thanks in advance for, um, enlightening me!
DR16 is better than you think, you just have to find the right theme.
Check out:
http://www.flea.org/starenli/index.html
Its got neat (and usable) sliding buttons for all the window operations, and you can change the border style, so the gimp will have its title on the left side, and mozilla can have the border on the top.
Turn sliding on, and it’s pretty neat, especially when you change workspaces. One thing that is somewhat annoying is the menu, but apparently you can use it quite well with GNOME or KDE panel(if you turn on the right menu option). If you want to use E’s menu with Debian(and it’s huge menus of apps), you have to turn off the automatic sliding of the menus, I forget which option it is.
On another note, does anyone notice that it’s really hard to browse themes on themes.org now? I haven’t looked at it for a few years.
For one thing, that’s not exactly true anymore — kwo updated e16, fixing some bugs, adding transparency for menus and the like, adding NetWM support, basically bringing things up to speed.
More importantly, though, e16 still does what I need; it gets the job done, and it’s so configurable that it’s not even that important what the job is (clean or eyecandy, keyboard- or mouse-driven, …) — if you take the trouble of configuring it, chances are it’ll be good to you.
On the other hand, some younger WMs failed me on things as mundane as setting a window to be “always on bottom” (“always on top” was supported).
e is about empowering the user. It’s about choice. If your choice is to build a usability nightmare, e won’t stand in your way. If your choice is to build a usability dream, likewise.
While I acknowledge that choice may be scary to some, and that it usually entails extra work in the short run, the lack thereof can be just as bad or worse.
Usability may have some near-absolutes (Fitt’s Law etc.), but usefulness requires a purpose, appeal a taste. I have a feeling that mixing up the three won’t help the discussion.
That considered, when I’m evaluating new applications, I’m a fan of “nasty” themes. Take for instance the evidence’s over-the-edge “fishtank” theme — it’s pseudo-transparent, it has loud colours, it has bloody bubbles rising up. It’s the equivalent of the dreaded blink-tag. Would I want to use it for longer than five minutes? Hell, no. But it tells me that the application won’t stand in my way, no matter how deranged my theme-idea is by the programmer’s standards. It’s empowering. It’s about my choice, not the coder’s. And it’s not even bloat, since the whole thing is modular — what I don’t need, I don’t load.
And that is probably the reason “e users have a certain taste”, as someone put it — if you want something middle of the road, in handling as well as in appearance, whatever came with your distribution will do. The reason why e has a good share of outlandish themes might be that these wouldn’t have been possible with any other Window Manager.
That said, if you want more “staid” themes, of course that’s possible with e, too. Consider the almost aristocratic elegance of the Winter theme that comes with the current e16. It’s stylish, but not in your face. It’s intuitive. And, relevant to those working with images, it’s almost devoid of colour.
e. Something for everyone. : )
Methinks Eugenia is forgetting the fact that E is extremely configurable…
Ok, some might be complaining that configuring E is difficult compared to GNOME and KDE…but, that’s just out of question I think….without their control panels, GNOME and KDE is as hard as E to configure…oh, no, it will be much harder to configure, since GNOME and KDE has no built-in configuration dialogs inside….
And of course, E’s configuration dialogs are not that difficult after reading docs and manuals…so, give it some try before saying some trolling stuff like “E is unusable w/ default setting!!!”
@Eugenia: Did you tried to configure E?
Wow… people read “enlightenment” and instantly their brains fall out! why is it they cant read the rest of the sentence or paragraph?
Never mind…
As for EFL and future things… Xrender/Cairo. OK. Cairo is built on Xrender or glitz. glitz implements xrender via opengl (as best it can). while this is cool – it has limits, and couldnt sanely be used too well in conjunction with xfixes+xdamag for a x compositing manager – limiting it – but then again evas suffers from the same issue, so it boils down to xrender. cairo just sits on top of xrender and anything else. xrender is the core of the issue. last time i checked (a few weeks back) my xserver and its drivers on every machine i have (2 nvidia, 1 radeo, 1 intel i865) still blew chunks in terms of speed. imlib2 was still running rings around it. i dont run the latest fd.o/x.org servers – i run whatever debian unstable currently has as i dont have time to rtrack every xserver version, cvs tree, fork and release. buit the point is – once xrender is properly accelerated and doing what it SHOULD do (and that is run rings around imlib2 instead as the gfx hardware these days CAN do so much bertter if used properly), it is a simple matter of adding an xrender engine to evas alongside software_x11, fb, directfb, buffer, qtopia and opengl. i dont’ see there being any value in this work until xrender is properly accelerated and properly accelerated servers AND drivers are wide-spread enough to make it worht it. renderbench exists as a tool for driver writers and developers on xfree86 and x.org to test their work and see if they have made it good. the day the software fallbacks for xrender stop being 1/30th – 1/50th of the speed of imlib2 is the day i will look seriously at xrender again. it’s really more a fmormality as the mechanisms and features are in-place already, abstracted and working cpu/client-side at a reasonable speed.
as for dialogue? well apart from the odd swaps of ideas/words/comments, not a lot. we’ve done our part – we wrote a benchmarker to encoruage driver authors to make xrender better so it can be finally useful to us. beyond that i don’t see what else there is to say there.
as for svg – we see not need for it, but it could be added.
as for other open standards. dbus has been discussed – but we’re on a “look an wait” position there – we can add debus suport fairly easily. all the infrastructure is there and is demonstrated by things like ecore_con and ecore_ipc that servers our “dbus-like” needs for now. exactly how we interface to dbus is yet to be decided. do we provide a ecore_ipc<->dbus gateway server, or have thigns slowly just use a hypthetical ecore_dbus module? don’t know, but time will tell. at this stage we dont need to make such a decision yet, so it’s open.
as for e17 “fional” – there are tarballs available (rasterman.com/files.html) shoing the current framework i have for e17. this is final, but its thin. the menus work but they need to be laid out properly by edje and for that it needs some more features. designign these feaytures means designing them to work not just for mensu, but work 5+ yuears from now for 10000 other things too, and thus this takes immense amounts of time, but the payoff will come in the long-term.
as for timetable – hard to say. defnintely not < 6 months, but since this is done in spare time, as we have spare time in between day-jobs (that consume 50+ hours a week) and actually having a social life, families, children, partners, friends etc, we can’t guarantee time to work on it. when u have no guaranteed time available, that incidentally is often at night after a long day at work working on other code – so after 10 hours of that, straight you’re probably too burnt out to work on E sutff, there isnt a lot of time left. we do what we can as we can. it’s probably a testament to the solidity and felxibility of what IS there that we have as much as we do. it’s a testament to all the people who have contributed and worked so hard, and keep doing so, knowing there is a light at the end of the tunnel, but that it’s just far off.
(PS – i am too tired to bother checking this for typos)
When I first installed Linux on my computer, I decided to also install Enlightenment just to check it out because a friend HIGHLY recommended it. When I got it, I wasn’t exactly impressed…
The default skin was ugly and I don’t remember any of the other options being much better, it took me a LONG time to get used to the button controls, and most importantly I discovered I could resize things to be larger but not smaller again. THAT was a real pain; I’d click and drag and get things bigger, then click and drag and do nothing. I’m sure there was some key combo but I was too annoyed. I guess Enlightenment probably isn’t for me. Maybe with a better skin or knowing what I do know about Linux, I might like it better.
Weird….why does Rasterman have to clear things up on this site everytime osnews runs an article on e? People simply don’t really bother to read up on things anymore.
As for Eugenia’s initial comment, install Enlightenment and reread Eugenia’s comment. Sit back. Reread. Notice anything?
…configure configure configure….
1) Is there any dialogue or cooperation between the folks involved in E and those behind the new developments at xorg-ie. cairo, glitz, and the new X extensions by Keith Packard ? Will Evas also take advantage of cairo in the future(ie. display postscript/PDF, scalable SVG etc. ?)
We haven’t had any formal interaction with the folks at freedesktop.org yet … not a lot of what we’ve been doing so far intersects with their work (although some does). However, we do consider freedesktop.org important especially with respect to standardization and make use of their published standards for reference often. For example, Ecore fully supports the EWMH (extended window manager hints), work is going on to make full use of UTF8 across the board, and support for the Xdnd spec is almost complete. As for other specific projects such as the new X extensions, those are marked up for future support. Evas’s modular design allows for additional backends to be created in the future, so it is possible that when XRender finally stops being as slow as molasses in winter, we’ll have an evas backend for it and with ARGB visuals be able to create full alpha transparency capable apps for X. Display/PDF is interesting; there is no plan to support it right at the moment (not enough time, not enough developers) but if it becomes standardized across Linux desktops it will likely become a priority too.
2) Will the accelerated graphics offered by Evas also be available to other applications running on the future E17 desktop ? If so could you give us some insight into how that mioght work.
Any app is free to take advantage of the EFL suite of libraries. They are extremely easy to code for and make things a lot better. With the new Evoak canvas server in CVS, for example, one could technically rewrite the gdesklets suite to make use of a desktop canvas and Edje to make desktop applets that are far more beautiful and graphically complex than what you have today. I recommend you read the “Systems” page on the web site for more information.
3) I see that Ecore has it’s own ipc mechanism-will work be done to allow this to interface with D-BUS ?
We already have drawn up plans for DBUS support (I actually have this on my TODO list); evidence already has some preliminary DBUS support which will likely make its way into Ecore in the future. Ecore’s IPC mechanism is much more convenient, however, we do plan to provide an interface between it and DBUS so that E apps & applets can have easy access to system messages.
4) Is the E team committed to the ongoing developments at freedesktop.org -ie. will E be using the new MIME stuff, desktop notification etc.
See (1) above.
5) Do you think the latest EFL release is the “final” basis for E17 development or are large parts of it likely to be rewritten again ?
This is certainly going to be the foundation for E17. There are no plans to rewrite anything — one of the main reasons for the last big rewrite was to introduce cleanness and structure into the API, and we now have this consistently across the board. You can be sure this will be the foundation of E17.
6) Are we looking at a timetable of <6 months to see E17 in action or more like 1-2 years ?
It will be ready when it is ready.
Seriously though — I cannot give you a timeframe because there isn’t one. In all possibility we could see a usable version of E17 arrive in CVS in a few months. Just stay tuned.
But in the mean time, more and more apps continue to be written that take advantage of the EFL and showcase its abilities, and you can check those out.
and here why :
1° incredebly faster than any other wm
2° the pager is really needed (you can drag apps within desktiop into the pager)
3° I set more easyly than in gnome the keyboard control :
alt+arrow : move desktop
alt+f4 : kill windows
4° ctrl+alt+sup : launch gnome-system-monitor
5° all windows maker doklet are loaded as if windows maker was running (I mean no windows decoration, etc….)
the stuf that drives me crazy anyway is that
1) nautilus 2.4 (or 2.6) hide the drag bar (it’s exactly like the scrolling between scren in amiga os and plus if you click on it you can see every apps running and switch to them even if they’re not on the same desktop
2) if you set the desktop wallpaper into enlighhtenment nautilus just don’t care (I know it’s gnome fault)
should I add that without enlightenment the concept of theming desktop would be still in air (and on winamp, to be fair)
my best desktop ever (ximian gnome 1.4 with enlightenment)
some pics (using shiny blue as theme for enlightnement)
http://djame.seddah.free.fr/osnews1.jpg
http://djame.seddah.free.fr/osnews3.jpg
Djamé
ps :
I have posted them here already, it was to cry about the lack of consistency in my gtk/gnome apps
look at the 3 different file selector in gnome 2.4
http://djame.seddah.free.fr/osnews2.jpg
and look at ximian gnome 1.4 one : really cool, isn’t it ?
http://djame.seddah.free.fr/osnews4.jpg
good holydays all
No, there actually wasn’t a menu that let you open e.g “/usr/local” in a new window if the path was shown as /usr/local/bin/whatever”. There was a “middle-click on backdrop” menu that offered those directories, however.
That said, I’ve been wondering forever what to do with middle and right clicks on the path (left clicks go to that directory, of course). So, I have added “open new window with that dir” for right clicks on the path (now in a cvs near you : ), thanks for the idea. Also, you can now drop items on the path, so instead of opening a second window, you could just drag the files in question on the “local” part of “/usr/local/bin/whatever” to copy or move them there.
I really, realy appreciate your feedback. I know it can be discouraging when reading some of the posts here-but a whole lot more people are interested in what you folks have been working on and where things are going than is evidenced by the some of the silly and thoughtless posts on this forum. I just wish I knew enough to be able to help out in the coding…at somepoint I would love to find a way to contribute, perhaps in documentation. I have infinite patience as regards E-even if it does take a couple of years-for some reason E just triggers the little child in me that was so fasinated by the world of computer graphics back when home computers first came out.
Rasterman,
From what I have read the work being done on libdamages and libfixes will make it possible to make render much, much better than it is now. I won’t pretend to know all the details-but I have gotten this impression from multiple postings on the xorg mailing list. I still am frustrated that xorg’s render support on nvidia chips is much, much better than that which comes from the propietary nvidia drivers themselves(on the order %200-300 faster). In the absence of more complete hardware information concerning the GPU’s-which is what hinders advancement more than anything else(IMHO)-alternate approaches via different X extensions(composite/fixes/damages) can change the problem that render is trying to solve-at least this my take on it. At any rate I am grateful for your hardwork on E over the years and fully appreciate how difficult it is when juggling family, work and coding E in the time left over. Please keep up the great work.
Azundris,
Thanks for your reply. I recently(yesterday) installed evidence again to see the progress being made. I love your approach of redundant dependencies falling back in the absence of this or that library. I did see, on your web site, that you mentioned dbus and I was really impressed that you also will support gnome-vfs and kioslaves. Thanks for the great work.
xcomputerman,
Thanks for your response. I see from your answer regarding my 2nd question that the answer to whether or not other applications will be able to take advantage of the benefits offered by Evas(acclerated graphics/transparency etc.) is dependent upon their being linked against the EFL. The reason I asked is that with the current xorg work going on with cairo, fixes, and damages to the extent that the canvas being used by applications makes use of cairo a lot of the rendering work can be handed off to glitz and directly to the 3d accelerated hardware support of the GPU. (at least in my primitive understanding)-so work is now going on to rewrite gnomecanvas and pango to take advantage of cairo and supposedly this is also planned for QT4.0. I guess what I am trying to get at is whether an opengl accelarated desktop will be able to offer these advantages to the whole range of applications written using different technologies(mozilla, OpenOffice, things which are based on wx, fltk, motif etc.) -or will only those applications which specifically use these new technologies, ie. evas and cairo, benefit and seamlessly fit into the new desktop(transparency would be rather silly if only 1 out of 3 applications actually use it).
I guess I am not entirely clear about the logical seperation in the design of these things-at least not sufficient to understand what is dependent upon what and how they interrelate. Again, please keep up the excellent work…..
I reread your response about render-about 6 times to see if I am capable of “grok”ing it. It seems to me that you are saying that evas sits on top of the extensions much like cairo does and that one is forced to either try to do all the compositing and acceleration stuff via the extensions(render, damages, fixes, composite) or to simply try to reimpliment the same kind of functionality in opengl-which of course was not really designed for 2d anyway and the opengl support of the GPU’s is logically distinct from the acclerated font drawing hardware in modern GPU’s which is *supposedly* used by render.
I had no idea that the two are mutually exclusive-ie. I thought using opengl would work great with the extensions… Am I even in the ball park with my primitive comprehension here ? I am just curious-I don’t need an introduction to X architecture 101-in case my question is itself insulting…. At any rate it sounds like evas and cairo are closer to each other than I had first understood-closer in the sense that they are both libraries sitting atop the extensions, whereas I was under the impression that evas could sit atop cairo-which now seems like nonsense. What is somewhat disappointing is that from the sounds of your response glitz and the opengl-backend of evas are supposed to implement all of the compositing and acceleration stuff in opengl so either one uses opengl for this stuff, stuff for which opengl was not concieved, or one tries and gets hardware accleration via the X extensions. So if I am anywhere near understanding this the only real question is how effective the two approaches are in utilizing the GPU power-opengl support is only really good on a small handful of different cards(nvidia, ati and perhpas matrox)-and the X extensions are supported by most cards(???) but not particularly good-and only will get better with a) more and better documentation from the manufacturers or b) better native extension support in the propietary drivers. I am rambling here- don’t wish to waste your time…
RE: xrender, acceleration etc.
evas ALREADY can use opengl to accelerate – glitz under cairo is not needed. we’re already there
cairo and glitz are technically playing catch-up.
as for 200-300% speedups. thats still abysmal, when imlib2 is 30-50 TIMES faster in most operations. thats 3000-5000% faster . sure with a 200-300% speedup now its 1000-2000% faster – its still 10-20 TIMES faster. until its SOFTAWRE rendering is approaching the speed imlib2 has (ie its hlaf the speed of imlib2 or better) we can talk. as for hardware accel- any hardware aqccelerated renderong on a modern card SHOULD be 10-30 TIMES faster than imlib2. that is the kind of speedup evas gets by going from software engine to opengl. it gets somewhere like a 10-30 TIMES speedup. (it has a benchmarking test program to show this)
as i said. i haven’t run x.org’s server as i just dont have time to keep chanign xservers. i have code to write and the abstraction layers support all the features and concepts needed. all it needs is for the drivers to finally catch up and work as they should for us to spend the time to put in that layer to hook it together. cairo buys us nothing. xrender COULD – if it were to be actually an acceleration mechanims, NOT a deceleration mechanism. again – renderbench. run that.
btw – thanks for the support
it’s really appreciated 
OK – I should explain evas a bit i guess.
evas is a graphical object abstraction. an image/icon is an object. a string of text is an object. you deal with is at a higher level and create/destroy, move, resize, show, hide etc. these objects. evas then figures out how it most efficiently can do the least work in redrawing the canvas in which these objects exist when they change. the redrawing of thisis handled by a rendering engine. evas has svereal rendering engines available. software_x11 is one. it uses the cpu entirely to do all the graphics then copies the rendered pixels to the screen as fast as it can. this works on every exserver since the dawn of time (X11R6). it’s fast and scales with the speed of your memory bus/cpu. this is also the core of the enigne for the framebuffer enigne that can do the same thing in the linux framebuffer – no x needed. it can also render to a allocated chunk of memory so you can take the pixels and write them to disk or whatever. it can also use the directfb engine to do the same rendering tasks, or the qtopia engine, or opengl. xrender could be another engine – so could cairo. but i don’t see the need for it until xrender is sufficiently fast to warrant the effort.
and yes you are right. opengl IS fast – but it has its limitations. it does not work well for “small windows” ie lots of small ones. it is very much designed and intended for big windows or full-screen, with few of them, doing 3d. you CAN use it for 2d and it DOES work but it does present problems that xrender makes easier to deal with, thus making xrender a preferred pipeline to render via, BUT since its speed issues hobble it, that takes it out of the race for now.
i can tell you exactly why xrender has its issues – and its not because of lacking docs for chipsets. that just stops xrender from being 10-30 times faster. iut deosnt make it 30-50 timres slower. this slowdown is caused by naieve code that does all the rendering with the CPU within xrender but doing pixel modifications ACROSS the AGP or PCI bus. this means a read/modify/write cycle for EACH pixel is amazingly slow – this covers MOST of the slowdown alone. we can go into detail and i’ve missed out some here – but the SOFTWARE FALLBACK code in xrender needs to be MUCH smarter about how it does things when the GPU cannot help.
Because this is enabling technology, I can’t see how this is anything *but* usable. Sure, as azundris said, there’s nothing keeping the turbo-733t from turning their desktops into usability nightmares. But by the same token, look at reporm’s desktop screenshot (posted earlier). It’s light, consistent, easy on the eyes and looks painfully simple to use for even the greenest newbie. The framework has been layed. Friendlier themes will follow. Right now, I think it’s safe to say that this is very, very exciting technology and an underground leap into the future. I for one plan on reinstalling e17 as soon as I get back home. I want to start theming so that all the usability naysayers out there can finally come to understand the flexibility of e. After all, I’m a user too. A simple UI benefits us all, hacker or grandma.
I don’t understand how anyone can see E as unusable. To me its the only one that _is_ usable. Its very aesthetically pleasing.
It also cannot be compared to GNOME/KDE since these are DEs and not WMs. If you were to give E a fair comparison against GNOME, then that would be Metacity. IMHO Metacity has zero features. It offers nothing more than the bare minumum you need to call something a WM, with a pager slapped on for good measure. I personally love GNOME as a DE, but Metacity completly kills it for me.
As for those making the bloat comments: KDE = bloat. Sure there is lots of eye candy, which I think is nice, except that when you are constantly dealing with screen artifacts, and slow resposivness, it defeats the whole purpose.
I for one can’t wait for E17 to come out. I’m sure even in a “no one but the core devs should be messing with this” cvs pull it will be more usable than any other WM.
Enlightenment is, for me, the epitome of what a window manager should be, and works beautifully. I use it for work, for play, and just for the sheer hell of it. It performs well, it’s extremely usable (I barely have to touch the mouse), the epplets rock (cheers gilbertt), and the themes are beyond amazing. I’m in the process of moving over to e17, but for now, just take a look at this: http://www.madaxe.net/transfer/screenie.jpg .
Your work is appreciated and as far as I can tell the new EFL are amazing.
And thanks to all the people making it worthwhile to read this comments section and I certainly don’t mean Euginia by that.
A site owner trolling her own site can you believe it.
glad i could be of some help. it was just logical for me as they looked like buttons and i guessed that if one was clicked i would be taken back to that folder and that to me indicated that the next logical step would be for it to support drag and drop. and the rightclick to open folder in new window is perfect if you have two deep branches that split at some point and you only have one of them open, rightclick the point of split and dig into the tree in the new window. then its just a matter of dragging the stuff you want to move over from one window to the next
one other interesting way of doing it would be a kind of “snap to” system. drag the file to the point of split in the “text bar” and the window would snap to that location if you hover for a bit. then drag to any of the folders shown and it would snap into view (again when hoverd), continue down. when you release the mouse the items dragged would be pasted in. question is, should the content then revert to the source folder or stay in the destination folder?
this is just some crasy idea of mine, dont bother to put it in tho as i have a feel that rightclick to open new window is smoother
even more so when you can do it to any depth of the branch your on rather then just the folders shown in the window as some file managers do it…
(Warning: I may be highly confused…)
Since the EVAS canvas seems to do the compositing and rendering on top of a range of backends, achieving the same benefits that the DAMAGE extension may provide, would it be possible/beneficial to retarget existing toolkits such as GTK+ to render to an EVAS canvas?
Wouldn’t this be preferable to developing a stand-alone E widget library?
Python bindings… Give me those, and I’ll be floating on air.
All I want to say is, that E16 was one of the catchers, that hooked me onto linux some years ago. As I remember, It’s been over 4 years ago when I was amazed by its excellent themes, configuration possibilities, menu (containing backround thumbnails in it), desktop switcher that other WMs/DEs can only dream about… I’ve tried since that time a lot of WM’s but always got back to E.
Well, it’s been some years, and from time to time I was checking the E homepage, hungry for the new 0.17 release. Based on the stable E version 0.16, someone could really say that the project is dead, which is absolutely wrong. The 0.16 still gets new stuff in it and is being released.
Tried the new EFL & apps (entice, entrance, …) and the impression is that one I remember when I first saw the E years ago – whoooaa, that kicks ass !!!
As the EFL is now a bit stabilized, I believe that it will not take that much time to see the new commits in CVS of the e17/apps/e
And last words – want to thank rasterman & other E developers for their hard work, it’s realy realy worth it.
Once more, it seems as though a clarification is in order…
E17 does not, at the moment, exist. Except for a very thin codebase linked to by raster, there is no window manager. _Every_ single screenshot that you have seen is of Enlightenment DR16.
Now, as for E16’s usability. To start, E16 was written several (5+) years ago. Recently kwo has put in a lot of work updating it, but the overall framework from a usability point of view (configuration, etc) is still that of 5 years ago. So, its rather silly to compare the ease of configuring e16 to the ease of configuring Gnome 2.x. By lack of ‘usability’, I assume you mean the difficulty in doing things such as changing fonts, which are unfortunately set by the theme. I fail to see how normal everyday usability (running applications, moving windows around, min-/maximizing them, making windows stick, etc…) could be any easier. (Its the same as in any other wm, except with more flexibility — as Azu. mentioned, its difficult to find things like ‘keep on bottom’ or ‘maximize the height/width only’ in other wm’s).
Now, on to the ‘eye-candy’ issue. I for one have a complete distaste for brushed metal interfaces. (I don’t care if others want to use them, but I can’t take them for long). Thus, I created my own theme to fit my tastes and needs. Sure, not everyone can do this, but ‘winter’ is about as far from a ‘futuristic-brushed-metal-type’ interface as you can get. Sure, it still needs some work, but its getting there. (Time is lacking these days). [In DR16.7, winter will be the default theme.]
Now, the the last point. This article is not even about enlightenment-the-window-manager. It is about the EFL, which is an altogether different entity. I can’t understand how the same person who made the original post (Eugenia), can follow it up with a comment complaining about the window manager… Make the distinction between a cleaned up version of a 5 year old window manager that uses only the oldest module of the EFL (imlib2), and the rest of this suite of libraries.
ENlightenment does still impress me. I use it for 4 years now. And i do hope i shall see E17 in quite little time. I saw Entice in action, i tried Evidence. They blew my mind
I hope that whole E-team will make what it wants.
ps: i have few screenshots with E on mall.deviantart.com
i prefer minimalistic desktop… and i love brushed metal also
That sounds like, “once you add spring-loaded folders, make sure it works with the text-path, too.” I will — but we’ll need spring-loaded folders first. : )
i was just going nuts there, i realy was just going to say i was glad i could help but then my mind kicked into overload for some reason…
I got winter theme for some time now and i use it (as far as i remember i downloaded it from xcomputerman website when it was not quite finished:). And im modifying it a little. Just like this: http://www.deviantart.com/view/9456536/