Frank, I'm talking about Qt, not QT. Qt is a toolkit from Trolltech, and is mostly used by KDE and a lot of other commercial apps, Google Earth being just one of them.
4.0 is the next version that will be cross-platform, Linux, Windows and OS X.
Kickaha:
Quote:
No amount of power or functionality can overcome a truly broken UI, but a good UI can make even an otherwise mediocre app reasonable to use.
That is true, but an app with a good UI and no significant functionality is also a mediocre app. I think that the UI should be well implemented, simple to figure out and not too cluttered. There, we are in complete agreement. But an app is much more than just a good UI... I think, as an apps purpose is first to provide functionality and only then provide that functionality in a simple, efficient, yet good-looking way.
That is true, but an app with a good UI and no significant functionality is also a mediocre app. I think that the UI should be well implemented, simple to figure out and not too cluttered. There, we are in complete agreement. But an app is much more than just a good UI... I think, as an apps purpose is first to provide functionality and only then provide that functionality in a simple, efficient, yet good-looking way.
Unfortunately, if you look at most apps, especially on *cough* other OSs, that last part gets left behind because hey, it *works* right? \
(Caution: rant upcoming, skip if you're bored.)
I think this is something that most programmers still don't get, and the l33t ones are the worst at it IME... all programming is communication. You are communicating your application's state and capabilities to the user, or the intent of your code to the compiler, or maybe other developers, or at the very least, to yourself in the future.
Unclear communication of intent means a poor product, bugs, and a maintenance nightmare in the making. If you always approach your code with "What is it I'm trying to state here?" instead of "What is this chunk supposed to do?" you'll find yourself creating elegant solutions that are clean, easy to use, and easy to maintain, almost effortlessly. It's a huge mental shift for most coders, unfortunately, and the academic and industrial training environments just push the 'get it to work and get it done' philosophy until it seems natural.
Sadly, we've got going on 40 years of experience that screams it doesn't work, but no one seems interested in changing. Of all the systems out there, the only one that even comes close to having a philosophy of clear communication is the Mac. It's far from perfect, and it's sad that it is the pinnacle, but there it is.
I've been toying with starting a thread on open-source development, but hesitated because I'm not a developer myself. So I'll just ramble here.
I've downloaded NeoOffice and I'll likely do the same for KOffice once it's out.
I've never contributed financially to either and contributing code isn't an option for me.
The problem, as far as I can tell, is that there's a lot of non-trivial programming gruntwork required to take open-source apps to the last mile. It's hard work and hard to get programmers to work on. Even NeoOffice's founders had to get outside work to pay the bills since donations simply aren't cutting it.
I think there are plenty of companies that would cut a check if they had a good idea of what their money was being used for, and when new software would be available.
Surely something like this could be developed, so Mac-based companies could contribute to a full MacOffice project, built in Xcode and with a real Aqua interface.
Just twenty Mac-based companies (legal offices, dental offices and many others) contributing $5000. each could pay for the bulk of the development.
The monies from the fund would be disbursed to participating programmers on completion of the project.
The problem, as far as I can tell, is that there's a lot of non-trivial programming gruntwork required to take open-source apps to the last mile. It's hard work and hard to get programmers to work on. Even NeoOffice's founders had to get outside work to pay the bills since donations simply aren't cutting it.
Isn't it true that you can take open-source software and sell it? Perhaps that's the answer - take something like OpenOffice, put a great Mac UI on it, and sell it for $25.
Unfortunately, if you look at most apps, especially on *cough* other OSs, that last part gets left behind because hey, it *works* right? \
Yes, that's also the sad truth, but there's one exception: GNOME. GNOME apps are usually simple, their UI is simple yet efficient and they pay a lot of attention to the published HIG. Of course, that's more true for GNOME apps themselves than third-party apps, but usually if you use GTK+ you can't stray too far away from the HIG anyway.
KDE, on the other hand, is a total mess when it comes to UI but not as bad as Windows, where every app is different, every developer abides by his own, made-up rules. Even Microsoft ignores their own rules, though Apple is not perfect on this either (Mail, GarageBand, etc.)
Isn't it true that you can take open-source software and sell it? Perhaps that's the answer - take something like OpenOffice, put a great Mac UI on it, and sell it for $25.
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
I like OpenOffice, but it has a lot of work to do when it comes to being a modular set of apps integrated together rather than a single behemoth that serves 17 different purposes. I use it, if only because I'm stuck with it because it's the only suite on the Mac that support OpenDocument so far.
I'll cherish the day when KOffice comes to the Mac.
P.S. Frank, thought you really meant it as opposed to joking. Sorry.
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
I like OpenOffice, but it has a lot of work to do when it comes to being a modular set of apps integrated together rather than a single behemoth that serves 17 different purposes. I use it, if only because I'm stuck with it because it's the only suite on the Mac that support OpenDocument so far.
I'll cherish the day when KOffice comes to the Mac.
I think you really hit the nail on the head here. OpenOffice is a monster. NeoOffice takes a minute to load. A minute, even if all I want is a lousy, simple, free spreadsheet app!
Granted, they've done a great deal to cut down the load times in OpenOffice 2.0, with a NeoOffice 2.0 coming out before the end of the year.
Still, the idea of KOffice intruiges me. I want to try it out (Intel iMac), even if I have to run it through X11, but DarwinPorts tells me that:
{My Computer}:/Users/{me} {me}$ sudo port install koffice
koffice is not supported on this platform yet
Bummer! Hopefully more apps will leverage Qt and Java to be truly cross-platform and run natively on Mac. NetBeans is a Java app (Java IDE) that runs natively on Mac.
Bummer! Hopefully more apps will leverage Qt and Java to be truly cross-platform and run natively on Mac. NetBeans is a Java app (Java IDE) that runs natively on Mac.
What annoys me about Java apps is they are so slow. Is it just the Apple Java VM? I've used Java apps on pocket computers and they actually feel more responsive. Qt on the other hand is pretty fast because it's pretty much just a GUI toolkit like Tcl/Tk, though I read it does more than that.
[B]I think you really hit the nail on the head here. OpenOffice is a monster. NeoOffice takes a minute to load. A minute, even if all I want is a lousy, simple, free spreadsheet app!
Granted, they've done a great deal to cut down the load times in OpenOffice 2.0, with a NeoOffice 2.0 coming out before the end of the year.
Still, the idea of KOffice intruiges me. I want to try it out (Intel iMac), even if I have to run it through X11, but
Running through X11 is a pain. I just don't do it unless I have to: say, if I have lots of things to do on the GIMP or Inkscape, it's worth booting into X11, otherwise not. So it's great news that NeoOffice 2.0 is out, in an alpha release which works very well indeed. It costs as an early adopters release at the moment, but will be free in a month, for those who want to try. It's a full-featured office suite [although I've heards negs re the database program]. It runs native in os x. It looks less Windows-ish than ever before. And you can option Crystal icons, which gives it a real KDE look, FWIW! This is PPC only now, Intel in a few months.
I was wondering why Google Earth looked like it was designed by a retard.
Okay I understand that it's easier to port if they leave it in X11 or GtK or Qt or whatever the crap that stuff is. But that's just doing it half way. Be honest. X11 apps look retarded compared to System 1. Let alone MacOS X. And THEN on top of that they emulate MS Office UI (should have an option to turn on intelligent interface as was suggested.)
I'm sorry, it's laziness. Don't be open source apologists. Half of the time, open source is the product of laziness. No one feels like doing the UI after they did all the coding perhaps. Sort of like when I finish a huge GIS map and then I just can't bring myself to put another few minutes in to making it look neater in Layout view. Or the same could apply to writing a report. OK that was just a guess. But if they can't be bothered to port it to Aqua, most Mac users will not even bother trying it. And if they do, they will laugh after looking at it for a few seconds then trash it. Like I do.
If they were smart they'd get an Aqua port out of an open source office suite before MS catches up with a Universal Office, which I bet will take a while, at least until mid to late next year. But they won't, because most open source users are hippies who think X11 or the latest Linux window manager is the best interface ever. Get with the program.
And another thing, why are there 3 office suites? Maybe 2...but 3? Sounds like a lot of reinventing the wheel. Instead of 3 shitty ones how bout one good one? I'm sure it's just three groups that want to "code differently" or something, but in the end, hey, they are all office suites. And they could have three times more people working on one.
There are no "X11 apps". There are GTK+ and Qt apps that use the X11 server instead of Aqua. They do that because using Aqua would most likely require a re-write of a lot of their code. XFree86 (used to be) a free (but piss-poor) implementation of a Window System, and Xorg is another (better) implementation of it. Apple uses XFree86 because Apple is too busy to offer the latest X11 version with all its enhancements. And they don't have to, but since they tout it as a feature, they should do a better job keeping up with the changes (but that's another matter).
X11 is like Aqua, in it that it provides a framework for developers to create graphical apps. A lot of people choose to write apps that work with X11 because a lot of people choose to have their apps work just fine between all Unices that have at least some version of X11 installed. Others, choose to use Aqua and limit themselves to OS X (which is a decision that depends on the apps, company/developer, and the stated goal of the app). X11 apps do not look "out-of-place" or "bad" when they are run among their own kind - other, X11-based apps, of which there are many. Take a look at Xgl and what Novell is doing with it.
BTW, a "window manager" is not a "best interface" thing by any stretch of the imagination. A window manager is something like Metacity, or Kwin, or lately Compiz. But because you have no clue what you're talking about, and mix things as if it were a Russian salad, you think what you're saying makes sense. It doesn't. You should educate yourself on the issue(s) before calling other people, and especially their work, lazy and hippies. The rejection of anything that isn't "Aqua" is, in fact, hippiness.
I'm saddened to see you defend X11. It's a terrible system that should have been replaced long ago. Better alternatives such as DirectFB have existed for a long time. The networking support of X11 is a nice feature, but face it, it's for a small niche, and for the rest, it needlessly complicates things.
X11 lags far behind Quartz* in terms of features. And by 'far', I mean 'years'. Yes, extensions exist for pretty much anything, but guess what? The reason they're extensions is mainly because they're simply not deemed good enough to become a standard feature at this point. You can only secondarily judge a piece of software by how much better it becomes through extensions; primarily, what matters is that the defaults are right. And that's where Windows's GUI and OS X's Aqua GUI are far, far ahead of anything the open source / Unix community have put out.
And finally, regardless of whether a Qt app is compiled to run through X11, or to run 'natively' on Aqua, it's still not a true Mac app. I expect more from Google Earth's developers. There's a very good reason Skype uses Cocoa on the Mac while using Qt on Linux and Windows: it's that users simply won't put up with mediocrity.
*) Since you're being so pedantic about distinguishing between window managers, desktops, frameworks and whatnot, let's do the same here, shall we? Aqua is a desktop and a window manager, not a GUI system.
Aqua is the name Apple uses to compare, contrast and tout Quartz with X11 as complementary technologies. It's not about being pedantic, it's about being able to rationally disuss something with the prior knowledge of what it is, rather than some flavor-of-the-month dissing term Mac zealots came up with (since when do Linux people claim that a window manager designed by them has the best user interface? Since that's the claim going around...)
I'm not defending X11 per se (I already called it piss-poor, at least some of its implementations) - I'm just trying to let him know that X11 doesn't make apps look bad automagically, and that it's job is much different than being a UI Builder. Or even a toolkit. People don't seem to know that X11 apps are usually created with either GTK+ or Qt. They think that X11 = Interface Builder therefore, MacIntel OO.org = bad so OO.org = X11 and X11 = bad. It's a lot more complicated than that, as I'm sure you know.
X11 (XFree86, and to some extent Xorg) are not the best Window Systems out there, but nobody claimed they were. That's why there's so much work being done by companies and other developers to replace X11 with something a lot more modern, like Xgl (which I linked to) or AiXgl. Xgl is out there, and AiXgl is expected to be in FC6. They are, from what I've read, seen and tried, on par with or sometimes even better than anything Apple has to offer now* and Microsoft will offer with Vista. This is not me being a Linux zealot, this is from what I've experienced so far.
Of course, the whole Qt/Mac app is a little bit overrated. Yes, I too like Mac apps (whatever that means? Apps that use Carbon/Cocoa?) but some apps have existed for a long time as something else and re-writing those apps to Cocoa is a lot more effort, time and money than some companies are willing to spend on such a small marketshare. It's the sad truth. Google Earth wasn't born yesterday - it has existed for years now, and I believe, as a Windows-only app. I'd much rather see that app on the Mac as a Qt app then not have it at all.
It's the opportunity cost of having one thing while not having the other. It could be better, you could have both, but I've never seen an Apple app respect the UI of the platform it has existed on (QT, iTunes look anything but a standard Windows apps. Some Microsoft apps break that rule, but at least their Mac apps look like Mac apps.) It's hard to expect others to re-write their apps according to Apple HIG when Apple itself doesn't respect others' HIG.
*except Quartz 2D Extreme, which for all we know, is never going to be released.
Firefox is better than Safari. UI is but one part of the app, and it's almost always the least important part in determining an apps "goodness".
Safari is a browser with a piece of crap javascript implementation, without any significant plug-in interface, slower than almost any browser out there at rendering pages (sans OmniWeb) and the browser with the most retarded implementation of stop/reload buttons.
Firefox, on the other hand, is none of these. Yes, it does not have a native UI, but I'll sacrifice that any day to achieve functionality. After all, what good do looks bring when I can't login to GMail?
X11 lags far behind Quartz* in terms of features. And by 'far', I mean 'years'.
Decades.
Look, my point is this. Is there a single "X" app with a good interface? Probably not. Constraints due to X11, the stuff that runs on it like window managers (yes I know X is a window server, but that's...not relevant.) And, laziness.
Running stuff on OS X in X11 = laziness. So it needs a rewrite. That's my point. Aqua is so far ahead of every other interface in existence. Particularly all the *NIX GUIs/Windows Managers/Whatever.
I did take a look at NeoOffice today and I can not wait to grab that 2.0 Alpha later in May. That's three-fourths of the way there, doing it Firefox style and at least "pretending" to be Aqua the best it can. It's ugly but vastly better. I have tried to use stuff in X11 before like GIS and it is...dumb. If they all teamed up surely they could get a good office app in Aqua. And perhaps there will be an impetus. I bet Mac marketshare will jump through the roof this year and next due to the whole Intel, maybe by 5 to 10%, that's huge. And all these Switchers might just be Joe Sixpacks who might be interested in an open source office. But only if it's Aqua most likely. Because X11 is decades behind Aqua.
You do make a good point in rather having an app, say Google Earth, as nonAqua vs. not at all. I just hope that maybe the open source community will start to embrace Aqua more and let go of "their" stuff like X11 WMs.
Lastly IMHO the Safari Reload button is brilliant! It's I think a huge belwhether in a small package: the computer thinking for us. OK, we can't stop a page that is loaded, or reload a page that is loading, so the buttons are mutually exclusive, and one button becomes two dynamically. A sign of things to come, our Mac thinking for us. Sweet.
I did take a look at NeoOffice today and I can not wait to grab that 2.0 Alpha later in May.
Oh my. People are actually selling Alpha software now?
I understand that NeoOffice needs a helping hand, but how on Earth can selling 'early access' to ALPHA software be a viable business solution?
That makes zero sense. Early access to Alpha software is not a benefit.
The computing world has gone from fully developed software for sale, to less developed software (with later bug fixes) for a lesser price (Aperture), to free access to early 'beta' versions (lightroom), to now actually paying for alpha software.
The distinction between the window server, the desktop environment, the window manager, etc. is a problematic one because users could never understand that, and an unnecessary one because the choice between different ones is unlikely to be of any advantage.
The window server could easily be a one-size-fits-all thing: Windows GDI lacks 3d acceleration, Quartz lacks networking and X11 lacks common sense. X11 could easily become obsolete on the Mac by adding networking to Quartz. X11, given various extensions, has all features a user could ever want, only they're all implemented poorly and impractically. It doesn't even have a universal interface for copy and paste, for drag and drop, etc. Zealots will be quick to jump up and say that this is the way it's supposed to be, that it adds flexibility, and that choice is good. Well, good for them, but X11 is a huge part of the reason desktop Linux still royally sucks. Choices are no use when they all fight against each other; few good choices outweigh many bad choices.
The desktop environment, too, is an unnecessary choice. The correct answer to a Linux distro installer's question of "do you want to install GNOME, or KDE?" should be: "I just want to fucking use my computer". KDE has shown time and again that they are unable to understand the nature of human-computer interaction; thankfully, they are slowly disappearing from most distributions. It is unfortunate, however, that projects like Kubuntu still exist, and it is even worse that someone like Torvalds doesn't realize that an engineer could never understand this matter, and has chosen to speak up about it.
The window manager is an even more quixotic matter.
Quote:
it's about being able to rationally disuss something with the prior knowledge of what it is, rather than some flavor-of-the-month dissing term Mac zealots came up with (since when do Linux people claim that a window manager designed by them has the best user interface? Since that's the claim going around...)
I have no idea what you're trying to say here.
Quote:
I'm just trying to let him know that X11 doesn't make apps look bad automagically,
No, but see, Quartz/Aqua and Windows's UI make apps look somewhat good automatigally. That's the difference.
Quote:
and that it's job is much different than being a UI Builder. Or even a toolkit.
But see, that's not the point. There is no reason such a distinction should exist to begin with. An application shouldn't be built for a toolkit; it should be built for an OS.
Quote:
People don't seem to know that X11 apps are usually created with either GTK+ or Qt.
Maybe because people shouldn't need to know such useless information?
Quote:
They think that X11 = Interface Builder therefore, MacIntel OO.org = bad so OO.org = X11 and X11 = bad. It's a lot more complicated than that, as I'm sure you know.
Yes, and this very complicatedness is part of the problem.
Quote:
X11 (XFree86, and to some extent Xorg) are not the best Window Systems out there, but nobody claimed they were. That's why there's so much work being done by companies and other developers to replace X11 with something a lot more modern, like Xgl (which I linked to) or AiXgl. Xgl is out there, and AiXgl is expected to be in FC6.
Um, those are still X11 implementations. They're still just extensions or reimplementations or refactorings of the same old tired crufty system. They're not innovation. They're a lackluster attempt at saving and old system for no good reason.
Quote:
Yes, I too like Mac apps (whatever that means? Apps that use Carbon/Cocoa?)
I can write a Mac app with WxWidgets. (It'll internally use Carbon anyway, but that's another matter.) I can put a lot of effort intno making it look good, and behave well, and make my users happy, and all that.
Or I can use Carbon or Cocoa and get most of that for free. Automagically. I get proxy icons (hi X11!). I get drag and drop (hi X11!). I get a unified copy and paste behaviour (hi X11!). I get a unified look of toolbars, of widgets, of a zillion things. Automagically.
Quote:
I'd much rather see that app on the Mac as a Qt app then not have it at all.
Ah, but that's the big problem, isn't it? The developers will immediately turn into jackass mode and respond "well, if you don't like our app the way it is, we'll just pull it entirely; Windows is a larger market anyway blah blah blah".
It's not that a well-designed Google Earth couldn't be done. It's that Google isn't fucking trying.
Comments
Originally posted by Frank777
You can build an Office suite with QuickTime 4.0?
Frank, I'm talking about Qt, not QT. Qt is a toolkit from Trolltech, and is mostly used by KDE and a lot of other commercial apps, Google Earth being just one of them.
4.0 is the next version that will be cross-platform, Linux, Windows and OS X.
Kickaha:
No amount of power or functionality can overcome a truly broken UI, but a good UI can make even an otherwise mediocre app reasonable to use.
That is true, but an app with a good UI and no significant functionality is also a mediocre app. I think that the UI should be well implemented, simple to figure out and not too cluttered. There, we are in complete agreement. But an app is much more than just a good UI... I think, as an apps purpose is first to provide functionality and only then provide that functionality in a simple, efficient, yet good-looking way.
Originally posted by Gene Clean
That is true, but an app with a good UI and no significant functionality is also a mediocre app. I think that the UI should be well implemented, simple to figure out and not too cluttered. There, we are in complete agreement. But an app is much more than just a good UI... I think, as an apps purpose is first to provide functionality and only then provide that functionality in a simple, efficient, yet good-looking way.
Unfortunately, if you look at most apps, especially on *cough* other OSs, that last part gets left behind because hey, it *works* right?
(Caution: rant upcoming, skip if you're bored.)
I think this is something that most programmers still don't get, and the l33t ones are the worst at it IME... all programming is communication. You are communicating your application's state and capabilities to the user, or the intent of your code to the compiler, or maybe other developers, or at the very least, to yourself in the future.
Unclear communication of intent means a poor product, bugs, and a maintenance nightmare in the making. If you always approach your code with "What is it I'm trying to state here?" instead of "What is this chunk supposed to do?" you'll find yourself creating elegant solutions that are clean, easy to use, and easy to maintain, almost effortlessly. It's a huge mental shift for most coders, unfortunately, and the academic and industrial training environments just push the 'get it to work and get it done' philosophy until it seems natural.
Sadly, we've got going on 40 years of experience that screams it doesn't work, but no one seems interested in changing. Of all the systems out there, the only one that even comes close to having a philosophy of clear communication is the Mac. It's far from perfect, and it's sad that it is the pinnacle, but there it is.
Okay, rant off.
Originally posted by Gene Clean
Frank, I'm talking about Qt, not QT.
Hence the winking smiley.
----
I've been toying with starting a thread on open-source development, but hesitated because I'm not a developer myself. So I'll just ramble here.
I've downloaded NeoOffice and I'll likely do the same for KOffice once it's out.
I've never contributed financially to either and contributing code isn't an option for me.
The problem, as far as I can tell, is that there's a lot of non-trivial programming gruntwork required to take open-source apps to the last mile. It's hard work and hard to get programmers to work on. Even NeoOffice's founders had to get outside work to pay the bills since donations simply aren't cutting it.
I think there are plenty of companies that would cut a check if they had a good idea of what their money was being used for, and when new software would be available.
Surely something like this could be developed, so Mac-based companies could contribute to a full MacOffice project, built in Xcode and with a real Aqua interface.
Just twenty Mac-based companies (legal offices, dental offices and many others) contributing $5000. each could pay for the bulk of the development.
The monies from the fund would be disbursed to participating programmers on completion of the project.
Originally posted by Frank777
The problem, as far as I can tell, is that there's a lot of non-trivial programming gruntwork required to take open-source apps to the last mile. It's hard work and hard to get programmers to work on. Even NeoOffice's founders had to get outside work to pay the bills since donations simply aren't cutting it.
Isn't it true that you can take open-source software and sell it? Perhaps that's the answer - take something like OpenOffice, put a great Mac UI on it, and sell it for $25.
Originally posted by Kickaha
Unfortunately, if you look at most apps, especially on *cough* other OSs, that last part gets left behind because hey, it *works* right?
Yes, that's also the sad truth, but there's one exception: GNOME. GNOME apps are usually simple, their UI is simple yet efficient and they pay a lot of attention to the published HIG. Of course, that's more true for GNOME apps themselves than third-party apps, but usually if you use GTK+ you can't stray too far away from the HIG anyway.
KDE, on the other hand, is a total mess when it comes to UI but not as bad as Windows, where every app is different, every developer abides by his own, made-up rules. Even Microsoft ignores their own rules, though Apple is not perfect on this either (Mail, GarageBand, etc.)
Originally posted by BRussell
Isn't it true that you can take open-source software and sell it? Perhaps that's the answer - take something like OpenOffice, put a great Mac UI on it, and sell it for $25.
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
I like OpenOffice, but it has a lot of work to do when it comes to being a modular set of apps integrated together rather than a single behemoth that serves 17 different purposes. I use it, if only because I'm stuck with it because it's the only suite on the Mac that support OpenDocument so far.
I'll cherish the day when KOffice comes to the Mac.
P.S. Frank, thought you really meant it as opposed to joking. Sorry.
Originally posted by Gene Clean
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
Oh, OK, I wasn't aware it was different.
Originally posted by Gene Clean
I would actually be more in favor of taking KOffice, a small, tightly coded, efficient set of apps, slapping an Aqua interface on it, than taking OpenOffice, a single behemoth and try to break it into little pieces first and then add the interace.
I like OpenOffice, but it has a lot of work to do when it comes to being a modular set of apps integrated together rather than a single behemoth that serves 17 different purposes. I use it, if only because I'm stuck with it because it's the only suite on the Mac that support OpenDocument so far.
I'll cherish the day when KOffice comes to the Mac.
I think you really hit the nail on the head here. OpenOffice is a monster. NeoOffice takes a minute to load. A minute, even if all I want is a lousy, simple, free spreadsheet app!
Granted, they've done a great deal to cut down the load times in OpenOffice 2.0, with a NeoOffice 2.0 coming out before the end of the year.
Still, the idea of KOffice intruiges me. I want to try it out (Intel iMac), even if I have to run it through X11, but DarwinPorts tells me that:
{My Computer}:/Users/{me} {me}$ sudo port install koffice
koffice is not supported on this platform yet
Bummer! Hopefully more apps will leverage Qt and Java to be truly cross-platform and run natively on Mac. NetBeans is a Java app (Java IDE) that runs natively on Mac.
Originally posted by JavaCowboy
Bummer! Hopefully more apps will leverage Qt and Java to be truly cross-platform and run natively on Mac. NetBeans is a Java app (Java IDE) that runs natively on Mac.
What annoys me about Java apps is they are so slow. Is it just the Apple Java VM? I've used Java apps on pocket computers and they actually feel more responsive. Qt on the other hand is pretty fast because it's pretty much just a GUI toolkit like Tcl/Tk, though I read it does more than that.
Originally posted by JavaCowboy
[B]I think you really hit the nail on the head here. OpenOffice is a monster. NeoOffice takes a minute to load. A minute, even if all I want is a lousy, simple, free spreadsheet app!
Granted, they've done a great deal to cut down the load times in OpenOffice 2.0, with a NeoOffice 2.0 coming out before the end of the year.
Still, the idea of KOffice intruiges me. I want to try it out (Intel iMac), even if I have to run it through X11, but
Running through X11 is a pain. I just don't do it unless I have to: say, if I have lots of things to do on the GIMP or Inkscape, it's worth booting into X11, otherwise not. So it's great news that NeoOffice 2.0 is out, in an alpha release which works very well indeed. It costs as an early adopters release at the moment, but will be free in a month, for those who want to try. It's a full-featured office suite [although I've heards negs re the database program]. It runs native in os x. It looks less Windows-ish than ever before. And you can option Crystal icons, which gives it a real KDE look, FWIW! This is PPC only now, Intel in a few months.
PM
Okay I understand that it's easier to port if they leave it in X11 or GtK or Qt or whatever the crap that stuff is. But that's just doing it half way. Be honest. X11 apps look retarded compared to System 1. Let alone MacOS X. And THEN on top of that they emulate MS Office UI (should have an option to turn on intelligent interface as was suggested.)
I'm sorry, it's laziness. Don't be open source apologists. Half of the time, open source is the product of laziness. No one feels like doing the UI after they did all the coding perhaps. Sort of like when I finish a huge GIS map and then I just can't bring myself to put another few minutes in to making it look neater in Layout view. Or the same could apply to writing a report. OK that was just a guess. But if they can't be bothered to port it to Aqua, most Mac users will not even bother trying it. And if they do, they will laugh after looking at it for a few seconds then trash it. Like I do.
If they were smart they'd get an Aqua port out of an open source office suite before MS catches up with a Universal Office, which I bet will take a while, at least until mid to late next year. But they won't, because most open source users are hippies who think X11 or the latest Linux window manager is the best interface ever. Get with the program.
And another thing, why are there 3 office suites? Maybe 2...but 3? Sounds like a lot of reinventing the wheel. Instead of 3 shitty ones how bout one good one? I'm sure it's just three groups that want to "code differently" or something, but in the end, hey, they are all office suites. And they could have three times more people working on one.
X11 is like Aqua, in it that it provides a framework for developers to create graphical apps. A lot of people choose to write apps that work with X11 because a lot of people choose to have their apps work just fine between all Unices that have at least some version of X11 installed. Others, choose to use Aqua and limit themselves to OS X (which is a decision that depends on the apps, company/developer, and the stated goal of the app). X11 apps do not look "out-of-place" or "bad" when they are run among their own kind - other, X11-based apps, of which there are many. Take a look at Xgl and what Novell is doing with it.
BTW, a "window manager" is not a "best interface" thing by any stretch of the imagination. A window manager is something like Metacity, or Kwin, or lately Compiz. But because you have no clue what you're talking about, and mix things as if it were a Russian salad, you think what you're saying makes sense. It doesn't. You should educate yourself on the issue(s) before calling other people, and especially their work, lazy and hippies. The rejection of anything that isn't "Aqua" is, in fact, hippiness.
Well, peace out.
X11 lags far behind Quartz* in terms of features. And by 'far', I mean 'years'. Yes, extensions exist for pretty much anything, but guess what? The reason they're extensions is mainly because they're simply not deemed good enough to become a standard feature at this point. You can only secondarily judge a piece of software by how much better it becomes through extensions; primarily, what matters is that the defaults are right. And that's where Windows's GUI and OS X's Aqua GUI are far, far ahead of anything the open source / Unix community have put out.
And finally, regardless of whether a Qt app is compiled to run through X11, or to run 'natively' on Aqua, it's still not a true Mac app. I expect more from Google Earth's developers. There's a very good reason Skype uses Cocoa on the Mac while using Qt on Linux and Windows: it's that users simply won't put up with mediocrity.
*) Since you're being so pedantic about distinguishing between window managers, desktops, frameworks and whatnot, let's do the same here, shall we? Aqua is a desktop and a window manager, not a GUI system.
I'm not defending X11 per se (I already called it piss-poor, at least some of its implementations) - I'm just trying to let him know that X11 doesn't make apps look bad automagically, and that it's job is much different than being a UI Builder. Or even a toolkit. People don't seem to know that X11 apps are usually created with either GTK+ or Qt. They think that X11 = Interface Builder therefore, MacIntel OO.org = bad so OO.org = X11 and X11 = bad. It's a lot more complicated than that, as I'm sure you know.
X11 (XFree86, and to some extent Xorg) are not the best Window Systems out there, but nobody claimed they were. That's why there's so much work being done by companies and other developers to replace X11 with something a lot more modern, like Xgl (which I linked to) or AiXgl. Xgl is out there, and AiXgl is expected to be in FC6. They are, from what I've read, seen and tried, on par with or sometimes even better than anything Apple has to offer now* and Microsoft will offer with Vista. This is not me being a Linux zealot, this is from what I've experienced so far.
Of course, the whole Qt/Mac app is a little bit overrated. Yes, I too like Mac apps (whatever that means? Apps that use Carbon/Cocoa?) but some apps have existed for a long time as something else and re-writing those apps to Cocoa is a lot more effort, time and money than some companies are willing to spend on such a small marketshare. It's the sad truth. Google Earth wasn't born yesterday - it has existed for years now, and I believe, as a Windows-only app. I'd much rather see that app on the Mac as a Qt app then not have it at all.
It's the opportunity cost of having one thing while not having the other. It could be better, you could have both, but I've never seen an Apple app respect the UI of the platform it has existed on (QT, iTunes look anything but a standard Windows apps. Some Microsoft apps break that rule, but at least their Mac apps look like Mac apps.) It's hard to expect others to re-write their apps according to Apple HIG when Apple itself doesn't respect others' HIG.
*except Quartz 2D Extreme, which for all we know, is never going to be released.
Originally posted by Gene Clean
Firefox is better than Safari. UI is but one part of the app, and it's almost always the least important part in determining an apps "goodness".
Safari is a browser with a piece of crap javascript implementation, without any significant plug-in interface, slower than almost any browser out there at rendering pages (sans OmniWeb) and the browser with the most retarded implementation of stop/reload buttons.
Firefox, on the other hand, is none of these. Yes, it does not have a native UI, but I'll sacrifice that any day to achieve functionality. After all, what good do looks bring when I can't login to GMail?
Yeah, actually, Safari is 10 times faster at rendering CSS and almost twice as fast at most JavaScript commands (http://www.howtocreate.co.uk/browserSpeed.html). Nice try, though.
Originally posted by JavaCowboy
Does Pages have a spreadsheet? Hmmmm..... didn't think so. KOffice does.
It does, in fact.
http://www.apple.com/iwork/pages/features/tables.html
X11 lags far behind Quartz* in terms of features. And by 'far', I mean 'years'.
Decades.
Look, my point is this. Is there a single "X" app with a good interface? Probably not. Constraints due to X11, the stuff that runs on it like window managers (yes I know X is a window server, but that's...not relevant.) And, laziness.
Running stuff on OS X in X11 = laziness. So it needs a rewrite. That's my point. Aqua is so far ahead of every other interface in existence. Particularly all the *NIX GUIs/Windows Managers/Whatever.
I did take a look at NeoOffice today and I can not wait to grab that 2.0 Alpha later in May. That's three-fourths of the way there, doing it Firefox style and at least "pretending" to be Aqua the best it can. It's ugly but vastly better. I have tried to use stuff in X11 before like GIS and it is...dumb. If they all teamed up surely they could get a good office app in Aqua. And perhaps there will be an impetus. I bet Mac marketshare will jump through the roof this year and next due to the whole Intel, maybe by 5 to 10%, that's huge. And all these Switchers might just be Joe Sixpacks who might be interested in an open source office. But only if it's Aqua most likely. Because X11 is decades behind Aqua.
You do make a good point in rather having an app, say Google Earth, as nonAqua vs. not at all. I just hope that maybe the open source community will start to embrace Aqua more and let go of "their" stuff like X11 WMs.
Lastly IMHO the Safari Reload button is brilliant! It's I think a huge belwhether in a small package: the computer thinking for us. OK, we can't stop a page that is loaded, or reload a page that is loading, so the buttons are mutually exclusive, and one button becomes two dynamically. A sign of things to come, our Mac thinking for us. Sweet.
Originally posted by Aquatic
I did take a look at NeoOffice today and I can not wait to grab that 2.0 Alpha later in May.
Oh my. People are actually selling Alpha software now?
I understand that NeoOffice needs a helping hand, but how on Earth can selling 'early access' to ALPHA software be a viable business solution?
That makes zero sense. Early access to Alpha software is not a benefit.
The computing world has gone from fully developed software for sale, to less developed software (with later bug fixes) for a lesser price (Aperture), to free access to early 'beta' versions (lightroom), to now actually paying for alpha software.
This is progress?
Originally posted by Gene Clean
It's not about being pedantic,
The distinction between the window server, the desktop environment, the window manager, etc. is a problematic one because users could never understand that, and an unnecessary one because the choice between different ones is unlikely to be of any advantage.
The window server could easily be a one-size-fits-all thing: Windows GDI lacks 3d acceleration, Quartz lacks networking and X11 lacks common sense. X11 could easily become obsolete on the Mac by adding networking to Quartz. X11, given various extensions, has all features a user could ever want, only they're all implemented poorly and impractically. It doesn't even have a universal interface for copy and paste, for drag and drop, etc. Zealots will be quick to jump up and say that this is the way it's supposed to be, that it adds flexibility, and that choice is good. Well, good for them, but X11 is a huge part of the reason desktop Linux still royally sucks. Choices are no use when they all fight against each other; few good choices outweigh many bad choices.
The desktop environment, too, is an unnecessary choice. The correct answer to a Linux distro installer's question of "do you want to install GNOME, or KDE?" should be: "I just want to fucking use my computer". KDE has shown time and again that they are unable to understand the nature of human-computer interaction; thankfully, they are slowly disappearing from most distributions. It is unfortunate, however, that projects like Kubuntu still exist, and it is even worse that someone like Torvalds doesn't realize that an engineer could never understand this matter, and has chosen to speak up about it.
The window manager is an even more quixotic matter.
it's about being able to rationally disuss something with the prior knowledge of what it is, rather than some flavor-of-the-month dissing term Mac zealots came up with (since when do Linux people claim that a window manager designed by them has the best user interface? Since that's the claim going around...)
I have no idea what you're trying to say here.
I'm just trying to let him know that X11 doesn't make apps look bad automagically,
No, but see, Quartz/Aqua and Windows's UI make apps look somewhat good automatigally. That's the difference.
and that it's job is much different than being a UI Builder. Or even a toolkit.
But see, that's not the point. There is no reason such a distinction should exist to begin with. An application shouldn't be built for a toolkit; it should be built for an OS.
People don't seem to know that X11 apps are usually created with either GTK+ or Qt.
Maybe because people shouldn't need to know such useless information?
They think that X11 = Interface Builder therefore, MacIntel OO.org = bad so OO.org = X11 and X11 = bad. It's a lot more complicated than that, as I'm sure you know.
Yes, and this very complicatedness is part of the problem.
X11 (XFree86, and to some extent Xorg) are not the best Window Systems out there, but nobody claimed they were. That's why there's so much work being done by companies and other developers to replace X11 with something a lot more modern, like Xgl (which I linked to) or AiXgl. Xgl is out there, and AiXgl is expected to be in FC6.
Um, those are still X11 implementations. They're still just extensions or reimplementations or refactorings of the same old tired crufty system. They're not innovation. They're a lackluster attempt at saving and old system for no good reason.
Yes, I too like Mac apps (whatever that means? Apps that use Carbon/Cocoa?)
I can write a Mac app with WxWidgets. (It'll internally use Carbon anyway, but that's another matter.) I can put a lot of effort intno making it look good, and behave well, and make my users happy, and all that.
Or I can use Carbon or Cocoa and get most of that for free. Automagically. I get proxy icons (hi X11!). I get drag and drop (hi X11!). I get a unified copy and paste behaviour (hi X11!). I get a unified look of toolbars, of widgets, of a zillion things. Automagically.
I'd much rather see that app on the Mac as a Qt app then not have it at all.
Ah, but that's the big problem, isn't it? The developers will immediately turn into jackass mode and respond "well, if you don't like our app the way it is, we'll just pull it entirely; Windows is a larger market anyway blah blah blah".
It's not that a well-designed Google Earth couldn't be done. It's that Google isn't fucking trying.
Originally posted by Frank777
Oh my. People are actually selling Alpha software now?
Selling?
I understand that NeoOffice needs a helping hand, but how on Earth can selling 'early access' to ALPHA software be a viable business solution?
What gives you the idea that NeoOffice is a business?