An Interview with Jeff Waugh

An anonymous reader writes "LinuxWorld has published a nice interview with Jeff Waugh, one of the core members of the GNOME community. In the interview Waugh talks about the upcoming GNOME 2.6, his views on software patents and on the involvement of the big vendors in the GNOME development process. Waugh is the current chair of the GNOME release team."

Comments

Waugh

[Anonymous Coward]

"Well well well, what is he good for?" [Only the aussies here will get this. Nothing on *this* (Jeff) Waugh)]

Software developer talking about patents

[ObviousGuy]

What a sad state of affairs that this is one of the main topics that the GPL community has to discuss.

More than the progress of the GNU project, more than software engineering breakthroughs, more than new ideas in user interface design, software patents seems to have eclipsed all that.

I used to be excited about computers and sharing ideas, but when the community dedicated to sharing has become a one note wonder, I find myself dulled by such harping on technicalities rather than technologies.

Re:Software developer talking about patents

[Anonymous Coward]

When the "one note" seriously threatens the ability of the community to continue sharing, it has a tendency to cast a dark shadow over everything.

Re:Software developer talking about patents

[gaijin99]

I used to be excited about computers and sharing ideas, but when the community dedicated to sharing has become a one note wonder, I find myself dulled by such harping on technicalities rather than technologies.

Ahh, another slashdot poster who doesn't bother reading the article. Had you read the article you would have noticed that out of seven questions, one was about software patents. And it was a damn relivant question, too. Other questions focused on interface design, coder community design, etc.

Software patents are important, that's why people talk about them. If software patents are granted universally it won't do much good to talk about the other things, software engineering breakthroughs, etc, because it will be *ILLEGAL* for us to make any such breakthroughs. But, and again I really do have to recommend reading articles before posting like this, the article was hardly an example of FOSS becoming a "one note wonder".

This has to be asked...

[heironymouscoward]

What is the roadmap for convergence of Gnome and KDE? It is good to have choice, but sad to see a fragmentation at the application level. Apart from the different programming languages used in the two, is there any fundamental reason why a common API cannot be defined or added?

Right now it seems that the only solution for applications that want to be totally portable is to bypass KDE and Gnome entirely and use their own libraries (Mozilla, OOorg) and/or X.

Even being able to run Gnome and KDE side-by-side in the same sessions would be a good thing.

Re:This has to be asked...

[nzkoz]

A fundamental API like you're describing would be the 'lowest common denominator' between the two systems. No KIOSlaves for KDE and no Nautilus integration & panel applets on GNOME. Kinda like AWT from old java, and we all know how much that sucked.

A much saner approach is to ensure that the basic stuff is compatible. Window manager hints, preferences etc. Let application authors write with their preferred toolkit, but ensure it doesn't affect users.

Almost all linux users have both toolkits installed anyway. Yes, I realise some KDE users won't have gnome (Gentoo hackophiles etc.) however if they want to use CoolGnomeApp1.0 they'll just install some librarys and they're away.

Re:This has to be asked...

[unoengborg]

I certainly agree that fragmentation is a bad thing when it reduces choise. Even Microsoft is starting to use the lack of integration against free desktops in their get the facts campaign. It would be nice if we could prove them wrong.

Even if Gnome/Gtk and KDE/Qt are very different toolkits, that should not prevent users from having a good user experience even if toolkits are mixed. The tookit choise should be a developer only issue.This is possible in windows and on MacOS so why not on free desktops.

E.g. why must each browser have their own bookmark file format and bookmark file lokation? Both Gnome and KDE use a folder as Trash, why not use the same location for that folder by default?

Both Gnome and KDE have a postit applet for small desktop notes, why not use the same file format and file location.

Why not make it possible to do drag&drop between nautilus and konqueror. After all there is a XDND standard that both KDE and Gnome tries to follow in other applications. And if we drag a file from konqueror to the Trash in Nautilus we should get the expected behavior.

Browsers and some other applications have icons that have similar functions in both Gnome and KDE. E.g. Back and Forward icons for browsers. Why not let the icons have the same name in both Gnome and KDE.

Perhaps all such config options and data that is common to both Gnome and KDE could be held in a separate folder named e.g. .freedesktop.org.

Speaking of GTK 2.4

[$calar]

Wasn't it supposed to be released on the 2nd?

http://gtk.org/plan/2.4/

New file selector, yum.

Re:Speaking of GTK 2.4

[anarxia]

From the same page:
"A timescale of 8-9 months seems reasonable; that is, a final 2.4.0 in late August or early September. As always, we're a bit late, so our current target date is early in 2004."

I'd love to Gnome out!

[Basehart]

After a few lackluster attempts at installing Gnome on my OSX box I have to say that a nice easy step by step instruction would be most helpful.

For many users, all the untarring, compiling and whatnot is a major headache -- akin to grasping the concept of depth of field in photography for me. Once I finally got it, it was super easy, but getting it in the first place was a big struggle.

I guess there's something about the whole process that I either just don't get, or maybe I think it's a lot harder than it really is.

So anyone know an easy way to get Gnome on an OSX box?

Re:I'd love to Gnome out!

[This is outrageous!]

So anyone know an easy way to get Gnome on an OSX box?

Google is your friend, the first six hits (after which I stopped checking) all send you to the right place.

Re:I'd love to Gnome out!

[caseih]

yup. fink install gnome-bundle. or fink-install [gnome-package-name] to install individual components. I did this and it took about a day of compiling off and on, but I have a full gnome desktop that I run on top of the normal OS X desktop (turn off nautilus, use quartz-wm). On the left side I have the normal OS X panel, and on the bottom I have a small gnome panel. The best of both worlds.

See the fink home page for more information. But really, it's not hard at all.

Michael

Here is the roadmap

[Srin Tuar]

Here is the unification roadmap:

KDE: ----------X
GNOME: ------------------->

</biased_gnome_user>

But, seriously, it doesnt make sense to talk about unifying them, as they are built around fundamentally different toolkits. ( Qt uses a modified subset of C++, GTK+ uses C as a base but has a nice C++ wrapper)

So they cant really be unified, though they can be made quite compatible.

I'm personally biased towards GNOME, because as a C++ programmer I love the stl, and thus hate Qt and the moc. But that doesnt mean I really think that KDE will die off: Free code is, after all, immortal.

Re:Here is the roadmap

[be-fan]

Qt and the STL are not at odds. They are two different approaches for two different problems. The STL is a general-purpose container library. Its not at all object-oriented, and aggressively static.

Meanwhile, Qt is specialized for GUI programming, and moc and the Qt container library fit that very well. Both allow for much more dynamic code, and in my experience, GUIs are extremely well-suited to dynamism in the language. After all, two of the best GUI languages ever (Smalltalk and Objective C) were of the dynamic/object-oriented variety.

I'm a C++ coder too, and also love the STL. However, I've spent a bit of time doing Qt programming, and really do agree that a more dynamic approach is better suited for GUIs.

Re:Here is the roadmap

[claes]

There is more to convergence than just toolkits. It is unfortunate that every comparison between Gnome and KDE always involves comparing toolkits and the differences between them. A desktop is so much more than just the stuff that builds the GUI. Actually, I think we should stop talking about desktop at this point, since the "desktop problem" is more or less solved. I would say both Gnome and KDE have accomplished what was envisioned at the point when the KDE project was started. I say there is a need for a new vision now, something that goes beyond just the simple toolkits on the desktop.

Convergence can take place in a number of areas. The configuration problem needs to be dealt with. Ideally, all programs should have a common configuration mechanism. Apache, Samba, mail servers, X, drivers etc should be easier to configure. There is a need for a common approach to these problems. This is a major problem to solve, since it needs cooperation and a common vision between all developers, not just the desktop developers.

Better hardware handling. There is work in progress here, and it is more important than most other things going on in KDE or Gnome.

Documentation and help systems. Every program should deliver documentation in a way so that can be integrated in a common help system. It should contain relevant metadata, be easily translatable, viewable in different environments. The information about available programs in the system today is scattered: there is information in the package management database, in the man pages, in the doc directory, in the menu hierarchy, but it is loosly coupled and it is not easy to find the documenation given the .desktop entry in the menu directory. I believe the free software community should define a metadata format like the one that freshmeat uses. Every tarball should include descriptions in a common format, and it should be usable in a number of contexts. There is a need for a distributed web of metadata. Today it seems the metadata is centralized in the package repositories and on freshmeat. There is a gap between description of packages and descriptions of the programs they install. Every available application needs good descriptions. Not just "Mozilla" "Web browser". "Konqueror" "Web browser".

Desktop Apps

[petabyte]

In terms of the technology, we've basically got all of the desktop applications solved. Between OpenOffice.org, GNOME, Mozilla and a number of other projects, the stack of stuff people generally use on the desktop is pretty much there.

Which really makes me wish that GNUCash was in that group. I do everything (word processing, email, spreadsheets, gaming) on Linux inside Gnome except for managing my finances. I keep a windows box with Quicken around for that. GNUCash could replace that for me but probably not before GNUCash-2 which is supposed to be GTK2. I heard they were short on developers and that was stalling progress on that. I guess personal finance doesn't have much of a place on a business desktop and gets less attention. I've been playing around with SQL-Ledger but thats a bit overkill for my needs.

That aside I love Gnome and am looking forward to 2.6 and Epiphany 1.2. :)

Re:Desktop Apps

[Anonymous Coward]

I do everything (word processing, email, spreadsheets, gaming) on Linux inside Gnome except for managing my finances

Let me guess, all the Tuxracer you can handle?

Re:Desktop Apps

[mcrbids]

I keep a windows box with Quicken around for that.

Dunno about you, but some time ago, I put a version of Quicken5 for DOS on a Linux system in FreeDos using DosEMU. I can now access this instantaneously via SSH from any computer in the world on the Internet, with a very high degree of security.

Nice. Very, very NICE.

KDE and Gnome *do* run side-by-side

[jonathanbearak]

I don't understand how people keep saying that KDE and Gnome don't work together. They're different environments, but all they're parts are pretty darn interchangeable. A while ago, for the heck of it, I replaced gnome-panel in Session prefs with kicker. Worked perfectly. After reading your post, I called kwin --replace to switch from metacity to kde's wm.

And OO.org ... that's for running across OS's, not KDE/Gnome. Besides, Native Widget Framework is due for the next major release AFAIK.

Mozilla ... it uses gtk+ or gtk2, many of which would consider to be (sort of) Gnome. XUL is not a KDE/Gnome issue. Like OO.o, it's another platform issue.

Gnome and KDE don't need to converge. At this point, they're aiming at different markets. KDE is uber-customizable. Gnome is focusing on KISS usability issues. The important backend stuff is already being taken care of via freedesktop.org.

Re:KDE and Gnome *do* run side-by-side

[theantix]

Bah.

The problem you are complaining about is a problem with Xandros's distribution, not linux. If gnome doesn't run properly in Xandros that just means that they haven't bothered to properly package it. Many distributions have the ability to install GNOME and KDE on the same installation, and applications written for one DE have always run just fine in the other one in every distributions I have tested.

Re:KDE and Gnome *do* run side-by-side

[Namaseit]

The problem is their *frameworks* are completely different. Because of this if you wanted to bring the two together you would have to scrap one DE's entire framework as "melting" them together would take so many man hours I don't even want to think about it. The closest your going to get is esthetic integration. KDE uses a lot of things that integrating into GNOME would be more work then it's worth. Kpart's(embedding applications in other applications aka konqueror), Kioslaves(doing ftp:// in kate and being able to remotely modify files directly), and many other things that make up KDE. I don't know much about GNOME's framework to tell you the truth.
There will never be true integration. Accept it.

Re:KDE and Gnome *do* run side-by-side

[True Grit]

1. Uber-customizability and uber-simplicity are two mutually exclusive things

You know, call me crazy, but I just don't believe this. Think of the apps out there that have an "advanced interface" button. With it turned off, the app adopts the KISS principle, with a streamlined interface with only basic options, but if the user selects the advanced mode, the interface expands or even alters itself to support the needs of the power user. So the problem with KDE is that they don't adopt a KISS interface as standard and hide the complex stuff by default, while GNOME doesn't have the complex stuff at all.

If the two camps really wanted to, they could work together, I don't buy the argument that a feature from KDE couldn't be ported to GNOME or vice versa, but the reason they still won't work together on a common interface is because of the problems that were present from the very beginning of both DEs:

Qt is not free on all platforms

GNOME=C but KDE=C++
Even though Qt is now free on *nix, that isn't enough for the diehard Free Software folks, while the latter reason is still the real fundamental stumbling block, because GNOME people refuse to work with anything other than C, and the KDE people refuse to use any language they feel is inferior to C++. Having "wrappers" for other languages is a smoke screen, the language issue for the core of the DE is still what separates the 2 groups.

Theoretically, if the 2 groups could agree on a common language for the core, the rest of the integration could happen with little technical difficulty. Having a default KISS interface that can morph into a power user's interface in the same DE is not some technically impossible challenge, its just a matter of agreeing on an underlying standard framework and then adopting a coherent policy on the cosmetic and asthetic issues. The real reasons they won't integrate their work actually goes back to the very old reasons that still separate them.

Re:KDE and Gnome *do* run side-by-side

[True Grit]

1. HIG research shows that such interfaces do not work. People overestimate their capabilities and try to use the advanced mode when they shouldn't, or need to use it the minute they need even one thing not in the default interface.

It seems to me, people overestimating their own ability isn't the fault of the interface. I agree its a dilemma, but not as bad as not offering the advanced interface at all. If GNOME takes the KISS principle too far they are just going to drive the power users to another alternative. Perhaps that has already happened.

1. The KDE code and GNOME code are completely different.

Yes, but not in the same way as saying COBOL code and Lisp code are completely different. GNOME is C written in a OO style, while C++ is, of course, very similar to C but with OO features. You can port between C and C++ a lot easier than most other combinations, indeed, unlike COBOL/Lisp, C++ compilers today will compile most ANSI C code without complaint. While a lot of the GUI code on both sides is very different because they use different GUI libraries, a lot of the foundation code in both systems isn't GUI dependent and could migrate a lot easier I believe than you think.

1. Microsoft Office, Photoshop, etc, are very powerful, but have very complex interfaces. iMovie, iPhoto, etc, are very simple, but limited in power.

Ultimately that is always the tradeoff you are going to have. Complex apps with complex features require a complex interface *to* those features. I was speaking less in terms of applications, and more in terms of just the DE's interface. To solve this problem, you don't need different DE's, I think, you just need different apps ranging from the dead simple to the complex. The environment remains the same, but the complexity is moved to the applications, with the whole point being that we end up with a common basic DE, but with a wide variation of file managers, editors, and other apps to solve the needs of different users.

Does a DE really have to have just one official file manager for it? I don't think so, decisions have to made about what the defaults are, but beyond that choice is essential to satisfy a large user base, and the whole argument is about trying to come up with one DE that could satisfy most folks, and provide Linux with a GUI standard that it currently doesn't have. If the KDE and GNOME people could agree on a common core, they can still go their own ways on the apps and utilities to satisfy different users and different goals, while at least saving all of us from having to keep 2 completely different and massive library suites on our system to use the different apps.

I just don't buy the argument that you have to have an entirely different system, from the foundation libraries providing I/O and GUI widgets right up to the file manager and other utilities, just to satisfy the differing goals of KISS and customizability. The 2 groups *could* find a lot of common ground, *if* they really wanted to, and really *tried*.

Re:KDE and Gnome *do* run side-by-side

[be-fan]

GNOME is C written in a OO style, while C++ is, of course, very similar to C but with OO features.
--------
Its not just a matter of language. Its a matter of the fundemental structure of the code.

a lot of the foundation code in both systems isn't GUI dependent and could migrate a lot easier I believe than you think.
---------
I am reasonably familier with the KDE architecture and parts of the KDE code, and I can tell you that nearly everything is fundementally tied to Qt. Even non-GUI stuff like kio and dcop are dependent on Qt. Unless GNOME is willing to adopt a Qt dependency (never going to happen) that code cannot be integrated. In the other direction, a lot of GNOME code has dependencies on glib and GTK+. There is no way in hell that KDE will adopt a GTK+ dependency, but it is porting some of the glib-dependent parts over. However, that is not terribly useful, because GNOME really has few non-GUI things that are worth porting.

Freedesktop.org is the fulcrum point

[Ars-Fartsica]

A great deal of work is taking place to push commonalities up the food chain to Freedesktop.org. Mostly this is X related, but I suspect this will grow over time into a true interoperability effort.

Once we've reached a point where the projects are not duplicating effort needlessly, we can truly say vive la difference with no guilt over wasted efforts.

gnome 2.5

[prockcore]

I'm using Gnome 2.5 (Subscribed to the 2.5 channel in Red Carpet, automagically upgraded everything for me). I have to say that Nautilus in the 2.5/2.6 branch is amazing.

How amazing, you ask? It's as fast as gmc used to be. Although it is a little strange to switch back to the old OS9 style Spatial Finder style of file management.

Things are a little buggy, Nautilus crashes every once in a while, and Evolution sometimes doesn't quit correctly. But in general, the whole desktop is great. Gimp1.3 is super sweet, and finally supports re-editable Text layers (ala photoshop)

Re:gnome 2.5

[prockcore]

hah. Try konqueror 3.2.
Blows the pants off nautilus.

No, it really doesn't. The new Nautilus is perfect. It handles files, it's clean and simple. It's the least cluttered file manager I've ever used. Konq on the other hand, is as unweildy as Explorer.

There's even a thread by the konq devs talking about how nice the spatial nautilus is and how they want to do something similar.

Re:gnome 2.5

[prockcore]

I can't seem to find the link to the thread in google, but here is the blog entry that I think kicked the thread off:

http://www.kdedevelopers.org/node/view/218

GNOME is excellent

[Beavis!]

I've recently been introducing my staff at my day job to GNOME since we are moving away from OpenVMS to Unix. Since HP-UX will be coming with GNOME as a default in future releases, I figured it would be good to get the guys used to it by having them use it on a daily basis for basic work stuff. So far they have taken to it pretty well. The most amazing thing is that some of them actually find it EASIER and more FLEXIBLE than Windows. Thank you for a terrific project!

Re:GNOME is excellent

[bonch]

I think the thing that keeps me going back to Gnome is both its simplicity and its speed. KDE feels way too much like an overload of thrown-in features, although the 3.2 release really impressed me.

I often switch back and forth between the two as new releases come out--I will be using Gnome again when its new version is released.

Re:GNOME is excellent

[Beavis!]

Actually, I do the same thing myself. I find elements in each that I like, but GNOME seems to feel more polished and put together these days. I'm not saying that KDE doesn't, but... GNOME apps seem more flexible in terms of look and feel. I can't stress the importance of a really slick looking desktop when it comes to the usability of a computer from a non-techie perspective. It makes it much easier for me to introduce people to a new desktop when it looks more impressive than Windows XP.

The reverse?

[Gyan]

Waugh: The whole point of the patent system is that they're supposed to be obvious things. But there are a lot of things in computing that are unobvious to a point

Umm, isn't it the opposite? Only those insights and ideas which are "non-obvious".

Re:The reverse?

[jdub!]

Yeah, the interview was done in the middle of the linux.conf.au wireless area, with a tape recorder. It seems a fair few things were lost along the way. ;-)

Linus

[bonch]

Why do you think Linus Torvalds is so popular? He's so down-to-earth about these things and interested in the technology and not the technicalities. This SCO mess forced him into it, but even then he still spits out the choice quotes, like the infamous "crack" comment.

Untarring, compiling, etc.

[Animats]

Ah, yes. Download, decompress, untar, compile, discover missing package, download, decompress, untar, compile, discover missing package...

Re:Gnome or KDE

[extra the woos]

this wasn't flamebait (no one is gonna bother flaming the first part of his post lmao), however, it may have been a bit redundant as this question comes up EVERY TIME an article like this rolls around...now i'm not karma whoring here, BUT there is only one answer:

USE BOTH... Whichever one you like better (and you will, its never a toss up!) use!

Re:Gnome or KDE

[be-fan]

KDE is whatever you want it to be. I've got mine set up like OS X --- menubar at top, panel at bottom, toolbars and menus simplified. Out of box, its pretty Windows-y, but I'd wager most KDE users don't use the desktop the way it comes out of box.

Re:Smart People

[Anonymous Coward]

gtk 2 renders slower due to internationalization added when gtk 2 was being worked on. Its much harder to render when you have to be able to render in everyone's native language.

Re:Gnome or KDE

[be-fan]

It depends on where your priorities are. Yeah, OS X is more polished and more pretty, but KDE is a whole lot more flexible and powerful.

Mono?

[Anonymous Coward]

Does anyone know about the legal status of Mono? I mean, if Gnome starts using it all over the place and M$ decides to shut it down (they have like a million lawyers, so they can probably do that), won't Gnome be, well, dead? Or at least in a very uncomfortable position?

Re:Waugh ? == Most hated !

[Anonymous Coward]

What ? They are still allowing that guy to be part of that project ?

Evolution Dataserver version 2.0

[axxackall]

Something magical is upcoming. I've tried to find anything about Evolution Dataserver version 2.0 mentioned in interview, and all I found so far were references to cvs. Looks like apart few developers accessing thisnew wombat no one else knows what it is, how it is designed and how it works.

Re:Evolution Dataserver version 2.0

[jerky42]

Yeah, I searched high and low for info on this too. Why is it such a big secret?

Re:Waugh ? == Most hated !

[Coverfire]

Can you support any of that with facts? Examples from IRC, or the GNOME mailing lists? I seriously doubt it.

Re:Smart People

[twener]

The only argument you gave says that GTK was ahead of Qt in internationalization. Anything confirming that it still is ahead?

Re:Gnome or KDE

[justsomebody]

As a guy who sold all of my Macs (had 4 and 1 powerbook (part of my job was Mac support), now I only have one G3 running yellowdog), I can say I'm happy.

And don't think that I haven't tried to use OSX. Common, I'm a sysadmin administering few companies and about 50 servers, I'm not unemployed, I don't have time to tweak my box to be usable for my work. Installing X11, installing Fink, compiling, searching for missing libraries, compiling, searching...

The common fact that Linux works nothing like OSX is main reason that I use it.

And believe me, Gnome is nothing like OSX. Main problem of OSX is when people like me start to use it. I always have 3x19" or 3x22" monitors, just to fit my windows. And just think about it, menu bar is on one monitor only, what a distance traveller this OSX mouse is (at least in my case). Also there is no support for separated screens, on Linux I just set Xinerama off and voila! Every monitor it's main menu and main panels with separated Window task list and it's own virtual screens.

Basically, my setup it's just as it would be as if I would use three different computers, but with one keyboard, one mouse and one storage.

But then again some people seem happy with OSX.

Re:LGPL

[donscarletti]

If you don't like the LGPL you are in a little trouble because all the KDE core libraries are licenced under the LGPL KDE Licencing policy as well. You will find that no KDE library can be anything tighter than LGPL and many of them are far looser

It is QT that is duel licenced under GPL and QPL but that is not part of the KDE project and they are connected very little except of one depending on the other.

Re:Not sure I'd agree

[be-fan]

Admittedly, it is all compile-time, but that doesnt mean its not dynamic
---------
If its compile-time, then its *not* dynamic. C++ has very limited support for any sort of dynamism, and the STL and libsigc++ continue that tradition. Like I said, I'm a fan of "modern" C++, but I have to admit that its static nature isn't great for GUIs.

The MOC isnt going to make C++ into an interpreted language or anything.
-----------
Since when do you need an interpreted language to have dynamism? Smalltalk and Lisp are among the most highly dynamic languages out there, yet have the most advanced native-code compilers available.

that the code size tends to be small, and the executables tend to be fast.
----------
In my experience, Qt is a lot faster than GTK+. GTK+ is really glacial for redraws (especially resizing) and certain things like GtkListView and Pango are quite slow. It doesn't really matter if the language binding is faster if the underlying toolkit is slower.