“What got me started on all this was a recent conversation I had with Ming Poon, the Xandros VP for software development, in which he explained why Xandros eschewed KDE’s generic UNIX file manager in favor of a home-grown Xandros File Manager [XFM], that’s tightly linked to the Linux OS. Now building a file manager has to be a huge task, and, in Ming’s case, it’s been his pet project ever since his days a manager of Corel Linux. Three questions come to mind: Why did they do it? Was it worth the effort? What’s the future of the XFM?” The article is focusing in the tweaked Xandros file manager.
Xandros is Managing Well, Thank You
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I’m trying Xandros out right now, and I have to say I am hugely impressed. It is a distribution that is suitable for just about anyone, from complete beginner, to a seasoned Linux user. (With that said, GNOME fans can skip it.)
The power of Debian is never far away. XFM is excellent. Networking is flawless. WINE is made easy. Hardware detection is as near perfection as any OS can manage.
I was initially skeptical of Xandros, because of its older packages ( KDE 2.2, GCC 2.95, … ) — however, they’ve really made the right choices. They’ve put stability and usability at the top of the priority list, without compromising functionality. In so many small aspects, usability tweaks and just a little extra love abound. After having tried almost every bleeding edge distro out there, I hardly miss a thing from them. Xandros really does just work.
Try it.
aside from the way xandros seems to be leveraging work out of the OSS community (using kparts, khtml and kioslaves) without contributing back, it looks really nice.
hopefully, the KDE developers will want to keep moving in this direction and have a really friendly file manager in linux.
somewhere in between xandros, lindows, knoppix and debian is where the future of linux should be…
They really should put a listing of things they have contributed on their site. I agree that this is a point over which people should be concerned.
I’ve also often thought it would be a good model if a company had a sort of contract with its users… A “we will GPL this sourcecode after X months, or your money back” type thing.
if a company had a sort of contract with its users… A “we will GPL this sourcecode after X months, or your money back”
it would be nice, but dont hold your breath. companies dont like to enter situations in which their holy IP might be at risk. a more viable plan might be to say, “if you dont maintain your application (ie: release every year, fix bugs within 3 months) then you must open the code.”
would you buy it, or wait until it was abandoned?
As a slackware user, where could I download a copy of xfm in source/tgz/etc..
-Jason
You can download the source here …
http://www.musikwissenschaft.uni-mainz.de/~ag/xfm/xfm.html
…but I bet this isn’t what you were talking about
Screenshot:
http://www.musikwissenschaft.uni-mainz.de/~ag/xfm/screen.gif
aside from the way xandros seems to be leveraging work out of the OSS community (using kparts, khtml and kioslaves) without contributing back, it looks really nice.
I think you’re being a bit harsh there. Xandros and before them when they were Corel have given back plenty of code. XFM is closed source yes – I think that’s reasonable.
For starters, they wanted to achieve something very specific, an Explorer clone basically. They also knew the KDE developers would never agree with them about certain (fundamental) things and so on, so they made their own. This is fair and imho reasonable.
I am especially pleased that they chose to use KIO rather than their own architecture – by keeping the interfaces open there is no risk of forking plugin development, ie if XFM was given away for free, even under the GPL Xandros would still be in control and so they’d have some control over linux by controlling the VFS plugins. This way, Xandros customers benefit from having a great file manager, KDE benefits from getting code back in other areas (and they’d not accept xfm anyway) and so we all win.
I have yet to try it, I wish they made a Live CD version for evaluation, but they seem to have it right.
On Windows XP any camera does not need drivers, automagically a removable harddiskn is mounted and I can manage the pictures on it.
Is this possible in XFM adn Konqueror. Seriosuly though, how come Konqueror doesen’t have such essential things liek the ones listed in XFM for example. This is pathetic. THey really need to work on Konqueror and as we speak it ahs 925 bugs.
to folks that make suggestions. I had a comment about it and made a suggestion, and he was quite rude to me like I insulted his intelegence. I hope he is not representative of all french.
At what point will everything but the kernel be closed-source? I mean, if you’re using OSS as one of the main selling features of the operating system, shouldn’t you follow your own advice?
Using free software and selling it _without_ giving anything back to the OSS community isn’t what OSS is about. At least companies like RedHat who are geared to make money contribute enormous amounts to OSS and GNU/Linux in perticular.
mean, if you’re using OSS as one of the main selling features of the operating system, shouldn’t you follow your own advice?
Actually, I think probably the main draw to Xandros would be the inclusion Crossover Office. Is this not also closed source, or does Codeweavers contribute back?
actualy…..that IS what OSS is about. I can just build everything from stock OSS crap and not do anything to it then sell it for profit and I do not give anything back.
that is my right.
I can also make a special program that makes the desktop a better experience for my users and keep it closed. I am not tweaking any OSS code. I am offering a feature for my customers that is not available anywhere else. that is also my right.
I mean if I was like read hat who hacks the stock OSS stuff, then yes, I am leagaly bound to give back what I ship. but if RH made a special program to only work with RH, that would be there right to do so.
Xandros gives back the stuff that is OSS that they hack, but they have all the right to keep there home grown stuff closed.
I can just build everything from stock OSS crap and not do anything to it then sell it for profit and I do not give anything back. that is my right.
youre absolutely right and there is nothing wrong with that. the annoying thing is working really hard towards a goal (say, a linux file manager that sucks less) and youre nearly there and then someone comes out with exactly what you wanted, using components which you built and its closed.
i’m not saying that its wrong, or should be, but it is very annoying to those of us who now have to duplicate what Xandros has done.
THey really need to work on Konqueror and as we speak it ahs 925 bugs.
yeah, konq needs a lot of work, but 925 bugs isnt a big deal. most OSS projects have hundreds of outstanding bugs and mozilla frequently has thousands. just because marketing says windows has 0 bugs doesnt mean it is so. if you could see microsoft’s internal bug-tracker you’d probably be horrified
if you are almost done with makeing the less crappy FM, then just finish the one you are building and release it under the GPL.
Darius wrote:
>Actually, I think probably the main draw to Xandros would >be the inclusion Crossover Office. Is this not also closed >source, or does Codeweavers contribute back?
As far as I am aware, only parts of CodeWeavers Crossover code are ‘closed’. I guess these bits are probably the sheen that make their commercial products so slick and therefore sellable. Which seems fair enough to me. 🙂
It should be noted that CodeWeavers make a very significant contribution to The Wine Project (open source project to reimplement (reverse engineer?) the Win32 API under Linux upon which Codeweavers software is based).
Konqueror supports digital cameras just fine.
Go to the control center. Set up your digital camera (Type, port (usb/serial).
Open a Konqueror window. Type Camera: into the location bar. Enjoy.
youre absolutely right and there is nothing wrong with that. the annoying thing is working really hard towards a goal (say, a linux file manager that sucks less) and youre nearly there and then someone comes out with exactly what you wanted, using components which you built and its closed.
Isn’t the point of a component based architecture the ability for the components to be reused by any number of applications? Why is it so wrong for XFM to be using the components of KDE? Wouldn’t it be worse if XFM wrote their own version of everything and eschewed the ability link in very tightly to KDE?
i’m not saying that its wrong, or should be, but it is very annoying to those of us who now have to duplicate what Xandros has done.
I’m confused as to why it is so improper for the Xandros team to give their distro a competitive edge.
yeah, konq needs a lot of work, but 925 bugs isnt a big deal.
Just fixing bugs isn’t going to make Konq better. It, like most OSS projects is designed by programmers for programmers. The results of usibility testing/feedback just aren’t there (and I say that for most all OSS projects). The end-user experience isn’t nearly as polished as it could (should) be.
I had no idea that Xandros had XFM until I read this article. I had written it off as “just another Linux distro that offers the exact same thing as every other Linux distro”. Now I realize it has something that sets it apart from the crowd. It has made me interested.
Most Linux distro producers need to understand that the “same-old same-old” perception is going to start hurting them. I applaud distro manufacturers that are trying to offer things to make Linux easier for end users.
The part of the article about XFM listing all of your storage devices, with no mounting required, is something that should really be burned into the minds of all distro makers. I didn’t have to manually mount things on my Amiga 10 years ago. That I have to do it on an OS in the year 2002 is inexcuseable. If Linux makers want a bigger desktop share then they need to stop trying to code the latest and greatest knock-off of Windows applications, take 10 steps back, and concentrate 100% on the end-user experience of the OS. It’s refreshing to see that some distro makers are finally going down that path, regardless of whether the end results are OSS or closed.
When I read the first couple of posts, I realised all these people want to do is get a hold of a product that is closed source.
Remember they only used some features of KDE, such as Kpart. That does not mean they should release any code that is their own. If they made some changes to KDE specific code, fair enough they should release that, not XFM.
Have I understood you people correctly? If not I apologise.
http://www.zdnet.com/anchordesk/story/story_4466.html
How quickly we forget.
>>
with no mounting required, is something that should really be burned into the minds of all distro makers. I didn’t have to manually mount things on my Amiga 10 years ago. That I have to do it on an OS in the year 2002 is inexcuseable.
>>
What OS are you running? I have been running Linux (mostly mandrake) as my primary desktop for about 5 years and I haven’t HAD to mount or unmount anything manually for a long time now….
Mandrake (usually) even mounts your windows and other fat partitions in /mnt for you during installation so you don’t have to do it later. Something my recent Red Hat 8.0 install didn’t even do for me.
I am pretty sure I read some where a while ago that the work that corel did with KDE did actually end up getting back into the official KDE release (like the most recent applications used section on the kmenu…
Most of the features in the ConsultingTimes article can already be done with konq:
——–
Another XFM exclusive: simply right-click on a local folder to enable Windows and NFS file sharing. No other Unix/Linux file manager can currently do this.
——–
Bullshit…mandrake does this, lycoris does this (i think), and ELX does this!
———
XFM can also rip CDs into wav, ogg or mp3 formats.
———
Konq. can do this too…has been able to for a long time now
———
thumbnail preview for images, text, HTML and PDF files
tooltip display of file details
browser-type navigation
an integrated web browser
URL auto-completion
bookmarks
configurable toolbars
built-in search tools
———
konq. does all this too…
———
They would much prefer to embed a media player into the preview panel, the way Windows has it, complete with stop and play buttons.
———
I believe konq. in 3.1 does this as well
———
XFM has a “split view” feature which lets you see different locations without opening the second window.
———
again, konq. does this as well
———
Xandros is mum, but considering their flexible plug-in architecture, it’s not hard to come up with a list of intriguing possibilities.
———
what? “their” flexible plug-in architecture? this is what they have been saying about konq. since it was released with the 2.x series…about how it is sooo easy to add functionality because of the same thing…
XFM looks to me like it is just konq. refined exactly for Linux (as opposed to generic UNIX) and displays some stuff differently then most distro’s (like the storage thing)
I like the look of Xandros, at the moment I run Mandrake 9, but I have lots of Windows programs still, and wine won’t even install on Mandrake.
Although I’m happy with my distro at the moment, I must admit Xandros is tempting me, and I may well buy it sohuld I find $100 lying around somewhere. Almost all the reviews are positive, so they must be doing something right.
Lindows has been in the news lots lately, but most of it’s just been hype, not that much positive feedback. If it was a choice between the two, I’d pick Xandros, it wins hands down.
Why is it so wrong for XFM to be using the components of KDE? Wouldn’t it be worse if XFM wrote their own version of everything and eschewed the ability link in very tightly to KDE?
it would be exactly the same, whether it was completely closed or only partially closed, because its not the closed-ness that is bothering me.
I’m confused as to why it is so improper for the Xandros team to give their distro a competitive edge.
its not improper or wrong, just highly annoying for hundreds of us developers to be pipped to the post. its like devoting years to the next new cancer medicine, which you’ll publish, and suddenly some company comes along and beats you to the patent.
if you are almost done with makeing the less crappy FM, then just finish the one you are building and release it under the GPL.
well, thats the problem, we’ve been ‘nearly finished’ for a long time now. we could have done what xandros did, but chose to go a different route. hey, i know thats life, but its a bitter pill to swallow.
I don’t know if I need to say anything more than this. I think they would have been better served to have just sold a modified KDE because all they really changed was the look. Are there license limits? Most (but not all, and less these days) Linux Distro’s allow one to install on as many computers as possible. How many systems can you install with this?
I’m not totally against this, but Xandros is not what I was hoping for.
One purchase equals:
– As many PCs at home as you wish
– Plus one install at work.
i.e. if you purchase it, you can use it at your work computer and at home on the one license.
Another XFM exclusive: simply right-click on a local folder to enable Windows and NFS file sharing. No other Unix/Linux file manager can currently do this.
——–
>Bullshit…mandrake does this, lycoris does this (i think),
>and ELX does this!
Folder sharing that is there in KDE 3.1 is very primitive, it does not have ACL support, i.e. everything you share is world readable. Xandros has proper permissions support, so you can give fine-grained access to local or domain users.
the Xandros way is more advvanced and more intuitive…it is like saying that Linux has DVD support that is comprable to windows or Mac…it is not even close even though it is there.