The opposite of Feature Creep

It’s time for a rant. Those sensitive to ranting should avert their eyes and go read something else today. But for those of you who enjoy such things, read on.

I have brought this up here and there over the past year on The Linux Critic, but I think it’s time I actually just dedicated a full discussion to it.

There’s a disturbing trend that I’ve been running into everywhere for a while now, and I feel that it’s worth a rant. I’m talking about the tendency of developers committing what I consider to be the cardinal sin of software:

Thou shalt not release a new version that has fewer features than the previous version.

This is the kind of thing that spins me up to no end, and I’m seeing it everywhere. It’s time more attention got brought to this problem, because it’s really running rampant.

I really, really don’t understand

Whenever I encounter this phenomenon, it always seems to be accompanied by statements like “We’re trying to simplify the user experience” and “This project is too bloated, so we decided to streamline things”.

Inevitably, the end result is something that is frustrating in its lack of features that I have grown to find useful, and is almost never “simpler” or “streamlined”. In fact, the opposite is true in most cases.

The thing is, when you remove features, you then force users to find other ways — which always end up being more complicated — to do what they need to do with whatever application/interface they’re using.

File managers

Take, for example, the one that has bit me most recently, PCManFM.

Now, I won’t lie, I’m a picky guy. Any of you who read this blog regularly would know that. Very picky, especially about my file managers. Worse still, I was spoiled by years of using Konqueror as my primary file manager, so I know how powerful a full-featured, well-designed file manager can truly be.

Needless to say, I ditched KDE when the KDE team decided to drop acid and go in a baffling set of directions that didn’t coincide with what I considered to be a user-friendly desktop experience, and along with it the now broken and hobbled Konqueror file manager, which sent me on a hunt for a replacement in my now non-KDE universe.

Thunar wasn’t terrible, but two things about it drove me nuts, namely its lack of configurability (for example, no way to shut off the delete confirmation dialog, which is always a FAIL in my book) and no tabs.

Yes, I’m hung up on tabs in my file managers, but again, after years of using Konqueror in KDE 3, I found I really miss them when I don’t have them.

Likewise, I also found that XFE was promising, and more configurable, but it too was sans tabs, and some of its default behavior was frustrating. For example, if you’re in a folder that has a lot of files in it, and you’re in detailed list view, right clicking in that folder is impossible without actually highlighting a file, even if you’re nowhere near the file name column… the entire row, all the way across will select the file, there’s no white space. The result of this is an inability to, say, right click and select “paste” to paste stuff into that folder (because when you right click anywhere in there you’re always clicking on a file, so the “paste” option isn’t available). This necessitates going up one level and clicking on the folder and pasting INTO it, which is counter-intuitive to me and I find myself constantly stumbling over it.

I dislike Nautilus because of a long list of arguments I really don’t want to go into here, so it only made sense that I gave that a miss.

So I ended up with PCManFM. It was tabbed, fast, it was (moderately) configurable, intuitive, and my only real complaint about it was a lack of a way to turn off the delete confirmation. While this is normally a FAIL for me, PCManFM was close enough on everything else that I decided I could live with that annoyance, hoping that a subsequent version might include a checkbox or something with which one could disable this dumb behavior.

All is well, until the new version

I found that I liked PCManFM despite the “are you sure?” B.S. every time I told it to delete something. It was a relatively new project, so I was willing to forgive this, and a lack of other options that I like to see in a file manager. But for the most part, it was pretty good. Heck, in PCManFM 0.5.2 there was even an option added in the Preferences to “always show the tab bar”, which I definitely liked.

Then, recently, while I was evaluating a new distro, I got a chance to use PCManFM 0.9.5, the newest version of the file manager.

I was actually kind of excited about this. I was expecting a big leap forward, lots more options, more flexibility, a better overall file manager.

I was wrong. In fact, the opposite was true.

Gone was the tree view, which rendered the entire left pane a useless waste of space. The Preferences window actually had even fewer options, including the elimination of the “always show the tab bar” option among others.


I initially thought there was something wrong with this version that had been packaged with this distro, and I did a manual install of the newest version of PCManFM on my own… nope. Same problem. The new version was a big leap backwards in usability, and this sent me on a brand new hunt for a new file manager.

I really don’t get this, people. Why on earth would anyone look at a promising project like PCManFM and say, “Okay, for the new version, let’s make this HARDER TO USE!”?

What benefit does this have to the user? I find that the newest version of PCManFM is almost useless to me. For what it’s worth, I hate Nautilus, and that’s what I actually ended up replacing it with.

Only to find that the newest version of Nautilus removed the location bar entirely from it, leaving only the stupid, clunky, awkward, and frustrating breadcrumb navigation at the top. I discovered that I had to actually install gconf-editor and go digging through that gigantic mess in order to find a way to bring the location bar back (come on, guys, seriously… did it ever occur to you that users MIGHT want to be able to type a path in their file manager so that they can navigate to something that you’re not showing in your borked, oversimplified default view?).

But even with all this (and the other long list of things I hate about Nautilus), it’s still more usable than the new PCManFM is. And that’s sad.

It’s everywhere

My file manager rant is only one example. This tendency to remove functionality and features from software is all over the place. Hardly any app I use anymore doesn’t suffer from it. Heck, I just read yesterday that Firefox 4 will have the menus hidden by default, requiring a hack that is likely beyond the average user to bring them back.

How on earth is this “simpler” and “easier to use”? I find this baffling. Yes, I’m not an “average user”, but generally speaking, for basic usability functions, if I have to go into Firefox’s “about:config” in order to fix their broken interface, it’s a severe FAIL. This doesn’t simplify anything, this complicates it.

It drives me nuts to find basic things like this missing in new versions of applications. It’s like being a concert pianist and every six months or so your piano is replaced with a new one that has fewer and fewer keys on it.

Eventually all we’ll be able to play is “Hot Cross Buns”. Won’t that be great? It’ll be SO MUCH SIMPLER and STREAMLINED!

I just don’t get it, folks.

So here’s my plea, in case any developers who are jumping on this feature assassination bandwagon happen to be reading this.

Please stop. New versions should be better for the user, not worse. For programs in particular that are lacking functionality (like PCManFM), there should be MORE configurability and flexibility, not less.

Repeat after me: Removing functionality doesn’t simplify the user experience.

Here ends my rant on this subject. Take it how you will, people. It had to be said.

— Trent

35 thoughts on “The opposite of Feature Creep

  1. Thank you!!Thank you!!Thank you!! Gnome, led by the chronically moronic Havoc Pennington, launched this awful trend of removing functionality and making life harder for end users in the name of “simplicity.” Gnome has never recovered and is still afflicted with the disease. KDE4 caught it, plus a severe case of eye candy at the expense of usability.

    Remind me again why Linux is “better”….? Oh know, because we can all learn to code, decipher the gawdawful sources written by nice simple developers, and fix junk that should never have seen the light of day. Cool!

  2. I agree wholeheartedly with you. “Simplifying” the product results in significant headache. I really hope that hiding the menu to give me a relatively small bit of real estate back is not one of the many “features” that Firefox developers are working on. If it weren’t for (what I perceive to be) a lack of features in Kazehakase (mainly, the ability for me to use a large assortment of keyboard shortcuts), I’d be using it over Iceweasel. If some of these silly changes come to be, chances are that I’ll be making the jump full-time (and like you, learning to live with a product that I’m not 100% happy with). I tend to set up a minimal, base Debian system and customize it to my liking anyway so …not grabbing one browser over the other won’t be any hair off of my back.

  3. I generally agree with what you say about feature removal, but I also wanted to defend KDE a little bit. I have no involvement with KDE other than a general user. The KDE desktop was a bit of a different animal in that they had a chicken and egg problem. The KDE3 series had become unmaintainable due to some of the legacy code and structure. So they rewrote it from scratch. But they needed more testers and developers to jump on board to make the base system solid. Then they could go back and add functionality. Of course, it’s difficult to get a critical mass without releasing something, so they released 4.0, which really should have been 4.0 alpha and we all know the fallout that occured. They’ve since done a good job of restoring features that they planned all along, but couldn’t add until the base was solid. So I applaud them mostly for getting the base overhauled for future improvements while working to restore features people liked from KDE3.

    But in general, you’re right — if you remove features for “simplicity” you’re probably shooting yourself in the foot as a project.

    • That’s a fair defense of KDE, in my opinion. There have been some decisions made in KDE4 (which I’ve read about on Aaron Seigo’s own blog) that fall squarely into the category about which I’m ranting, however.

      My KDE4 experiences (with 4.0, 4.1, 4.2, and 4.3) have frustrated me so much that I’ve vowed not to touch it again until they release KDE 4.5. So I’ll re-evaluate it then to see if they’ve brought forward enough features and functionality to make it viable again in my view of things. But based on some of what Seigo has discussed on his blog and on the KDE forums, there are some things that I think are just never going to be fixed… because they’ve been made “simplified” by their design, not by “we plan to implement this later”.

      Still, good comment, thanks for posting!

  4. great post. this can’t be said enough or shouted out from the rooftops enough. Seems the Linux Desktop projects have decided that the only way to be popular is to dumb things down to the level of Windows. Its aggravating and it won’t work.


  5. I agree. Too many developers are intent on turning Linux into the “OS for Retards” tm.

  6. This reduction in features also illustrates the failure of programmer-designed user interfaces. These guys (and I’m a programmer, too) will open a terminal without hesitation to do the things that their own GUI creations won’t. They do not understand why users don’t think the same way and just do what they need from the bash shell.

    To improve and actually compete with the Windows and OS X GUI’s (just the GUI’s here), the KDE and GNOME teams really need to be controlled and directed by marketing flacks and despicable GUI testers telling the programmers what to develop. Alas, this is contrary to the nature of volunteer Open Source software so we’re stuck with the WILI (“well I like it”) approach.

    I’d also really like to see how those user interfaces would evolve if you prevented the developers from using a terminal or a command line for _anything_. That includes those one-liner widgets for quick commands. Force them to do _everything_ from the GUI, and see how it evolves then.

    • These guys (and I’m a programmer, too) will open a terminal without hesitation to do the things that their own GUI creations won’t. They do not understand why users don’t think the same way and just do what they need from the bash shell.

      Oh most certainly. Heck, I’ve repeatedly pointed out that in GNOME, the lack of a “disable touchpad completely” checkbox complicates things. If you have to manually write a a shell script to do something that should be basic included functionality, you aren’t simplifying things for anyone.


      So does having to write shell scripts to manage mouse functionality.

      I like your idea about removing the terminal from developer workstations. That should be done for a minimum of a week when they get done designing new interfaces. Force them to use it and make them aware of its graphical limitations before letting it out the door.

  7. They act like choices are a bad thing.
    If that’s the case, shouldn’t we all be using Windows?

  8. What the Linux environment currently lacks is some Gentooism. No, not the distro. But it would be cool to have a central patch system. Whenever my package manager downloads a package, that I do not entirely like, there should occour an automatic rebuild.

    There are couple of more apps that have configuration settings COMPILED IN. So, if you need a certain feature you HAVE to recompile it. We need an automatism for this. Distributions and upstream don’t always provide what we users want. (I left Debian because of their DFSG packages.)

  9. This is exactly why I am stuck with an ancient Gutsy install running KDE3… I haven’t been able to find any filemanager better than konqueror for now , and – as you say – that’s only the beginning of my problems…

  10. In the case of both KDE 4 and PCManFM, these programs were complete rewrites of the previous version. Sometimes as a programmer there is just no way forward without starting over. It’s reality.

    When you rewrite, you take a hard look at what you had before, and you ask if it was all really necessary. Why add back every last little niggling feature just because “it was in the old version”? Not everyone thinks or works the same; just because it seems patently obvious to you that a feature should be included, doesn’t mean it was to the development team.

    On top of that, it may be that they wanted to get a rudimentary version out the door even if every last feature hadn’t been put in yet. KDE 4.0? Remember, in the FOSS world the idea is “release early, release often”.

    I think if you took the time to make a reasonable case for a feature, it would probably get re-included. And if that’s too much bother, just wait around a few releases; if it’s a worthwhile feature, someone else will suggest it sooner or later.

  11. First, let’s get this out of the way: when you write things like “[they] decided to drop acid and go in a baffling set of directions that didn’t coincide with what I considered to be…” you don’t lay the groundwork for a very constructive conversation. It’s very disappointing, especially if you’d like others to operate under the assumption that you’d like to be treated with common courtesy. *sigh*

    So …

    Konqueror was not rewritten, nor was much of KDE libs. Most of the changes to the libraries were additions. We kept a huge amount of work from KDE 3. Some parts were rewritten, but not as much as it may sound like from the comments here.

    When it comes to Konqueror, there was one very big part that was rewritten: the file views. This had exactly zero to do with any desire to dumb things down. It had everything to do with Qt4 providing a model/view framework that would allow Konqueror (and other KDE file managers) to be more feature rich and more performant.

    The existing file views in Konqueror were not efficient and couldn’t sustain more features (they hardly dealt with that was already implemented).

    Unfortunately, there aren’t many people who like to work on file managers, and so the guy who did a lot of the work in Konqueror in KDE 2 and 3, David Faure, was mostly on his own on this .. until we brought Dolphin in and attracted some more KDE libs developers.

    Still, it’s taken time to get the file views back up to parity with what was in KDE 3. Which they are now. I have the same arbitrary depth split views, network transparency, context menus, etc. in addition to some new features and tricks thrown in.

    It would have been great to have had the model/view based file views complete for 4.0, but manpower is often scarce. Thankfully Dolphin brought some more hands to the table and sped up that work, but it still took a while.

    It had, however, nothing to do with any “dumb it down” conspiracy. 😉

    • Aaron,
      I’m not Trent (the original poster), but I think it’s fair to say that where KDE 4 is concerned, usability was half of the equation. The other half was that when asking “we were able to do this in KDE 3, how do you do it in KDE 4?” we would get rather rude responses saying essentially, “why do you hate change? Get with the times!!” And, as Trent mentioned, the straight answers he did get were of the, “we’re removing that functionality” type. To be fair, Konqueror was one of the strongest reasons to use KDE.

      I don’t really want to get into a flame war or a KDE bash-fest; that’s not really the point. Trent has documented his KDE saga fairly well, I think, and I think it’s also fair to say he’s not written KDE off completely; after all, there are still several features of KDE for which he has yet to find an acceptable substitute (also well-documented.)

      The bigger point is that software development needs to be a two-way street. The mention of KDE developers “dropping acid” is really the culmination of a lot of frustration and trying to deal with having his environment of choice pulled out from under him, then having his frustrations thrown back in his face for daring not to like it “out of the box.” That’s not necessarily on the developers so much as on the larger community, but it’s an issue nonetheless.

      Community is one of the strengths of the FOSS universe, and that’s really one of the reasons Trent started this blog. Personally, I’m glad that you stopped by and said what you did, because it shows that KDE really isn’t a “my way or the highway” sort of deal, and it shows that the community matters. I have no delusions that any desktop environment development team is going to give me precisely what I want; if I want it that badly, I should program it myself. But at the same time, every environment appeals to a certain audience and a certain type of user, and I think it’s critical that the development team continue to identify and remain in touch with that audience, and that the audience contribute actively regarding non-programming functions (documenting, providing informal help, providing feedback and bug reports and such.)

      • “And, as Trent mentioned, the straight answers he did get were of the, “we’re removing that functionality” type.”

        When it came to Konqueror, I’d be surprised.

        “To be fair, Konqueror was one of the strongest reasons to use KDE.”

        For some that was, and is, absolutely true. For others … it wasn’t. In fact, for many, it was a reason for them _not_ to use KDE. I know, hard to imagine (and I don’t say that facetiously, I mean it!), but that’s what we discovered.

        When you mention that software development is a two-way street, for us as developers it’s more like a metropolis of streets with a lot of variation out there.

        You’re correct as well IMO that no single piece of software will cater to everyone well, but we do try to cast our net pretty broad with the KDE Workspace and related software. Many of our users do not seem to fully appreciate how that means that their “two way street” is just one of many, and we end up getting flak (undeservedly) for that.

        “really the culmination of a lot of frustration and trying to deal with having his environment of choice pulled out from under him, then having his frustrations thrown back in his face for daring not to like it “out of the box.””

        That’s an over simplification of the events that serves nobody well. There were some features/ideas/issues that we certainly did say “they are going to be different now” (folderview was one such decision, for instance; one which I personally got insane amounts of negative feedback on before anyone got to really try it out, and not just routine negative feedback but really, really out of line stuff), but many issues we simply said “we haven’t gotten to that yet” or “yep, that’s a bug”. Now some people use the issues where we said “no” and held the line on as a broad sweeping generality to cover any issue they ran across. That isn’t how it happened, that isn’t how we’ve dealt with things and it doesn’t reflect the KDE software we’re shipping today.

        It would be great if we could therefore get beyond that “you hurt my sensibilities, that’s my excuse for everything now” rubbish. It’s not January 2008 anymore.

        “I think it’s critical that the development team continue to identify and remain in touch with that audience, and that the audience contribute actively regarding non-programming function”

        Agreed on both points. I would add this: the audience should not expect to play director, anymore than people who drive on a bridge should expect to play bridge engineer. Let us do our job, even if it doesn’t line up with how you think a bridge should be built (so to speak). As for the audience contributing actively, that needs to be done with the same constructive mindset they would like in return. If we are all in this together, we need to treat each other like it. We’re slowly seeing that happen more again, but for years (predating KDE 4.0, and certainly not unique to KDE either) we have seen repeated and systemic failures in this regard.

        How do you think Trent-the-author-of-this-blog-entry would feel if I said something like, “You have obviously been dropping some sort of bad acid which has led you to believe you are the center of our userbase universe.”? Probably not overly well. Now, guess how often I get to read that kind of message pointed at the developers who put their blood, sweat and tears into things? Well .. this blog entry is an example of that.

        We can all do better. Let’s stop defending poor behaviour, including when it comes from users. Let’s really start improving how we treat each other and build up a community of strong, constructive communication. Who knows, we may even find ways to be positive! 😉

        • I’m pretty much in agreement with everything you’ve said, although there’s a difference between “you’ve hurt my sensibilities, so I’m taking my toys and leaving” and “I can’t make the software behave as desired, and the user community has shown no inclination to help me, so I’ll look elsewhere.”

          To be very clear, that is NOT a developer issue, and it’s not something I’m pinning on the KDE development team. In this case, the correct response to “how do I do XYZ?” should have been, “that’s currently not possible, right now we recommend ABC.”

          At any rate, if you poke around this blog enough, I hope you’ll find at least some evidence of what the community OUGHT to be: users helping users and providing honest, non-fanboyish feedback. Sometimes emotional reactions get in the way of things, but we do generally try to keep that somewhat in check.

          The very least result of this conversation, though, is that at some point in the near future, I’ll be giving KDE another closer look myself.


          I think I just committed myself to another blog post *sigh*

        • >>there’s a difference between “you’ve hurt my sensibilities, so I’m taking my toys and leaving” and “I can’t make the software behave as desired, and the user community has shown no inclination to help me, so I’ll look elsewhere.”<<

          The fact that Seigo **still** can't distinguish between the two tells me all I need to know about the state of KDE and why so many former users are leaving it for Gnome (!) of all things. When your users are disgusted enough with your removal of functionality to the point where they think Gnome has more features…!

          Personally I'm hoping these missteps will be enough to propel one of the smaller 'third-party' desktops into the dominant position for awhile, give LXDE, Esther, or XFCE a chance in the driver's seat for awhile and maybe these arrogant developers will suddenly see why they need users again.


          • Personally I’m hoping these missteps will be enough to propel one of the smaller ‘third-party’ desktops into the dominant position for awhile, give LXDE, Esther, or XFCE a chance in the driver’s seat for awhile and maybe these arrogant developers will suddenly see why they need users again.

            You and me both. Personally, I have high hopes for LXDE in particular, which is why Joe and I have both been writing so many posts here about it lately.

            Thanks for the comment!

            • Thanks for the rant (which I find myself in full agreement with despite having been a long time Gnome fan–this sickness is destroying that desktop too!)

              If it matters I’ve added the blog to Straw my feed reader, so I’ll be following along from now on.

              By the way, I assume you’re already aware but just in case you were not–did you know VectorLinux just released a live cd of their 6.0 release featuring the “classic” KDE desktop? Since VectorLinux IS Slackware or at least is based on it, this might buy you yet a little more time before you are forced to upgrade to a different desktop–KDE based or other desktop. That announcement can be found here:

              Hope this helps!


    • Wow. If I knew that you’d be reading this, I’d have been more diplomatic about it.

      To be honest, Aaron, when I wrote this post today, it was first thing this morning, and I was pissed off, hadn’t had my coffee yet, and frankly didn’t expect anybody would bother reading it. I just pounded out my rant this morning and hit “submit” without even proof reading it.

      No, the way I put my jab at KDE4 was not conducive to constructive discussion on the subject, and for that I will apologize. In my defense, I was ranting unconstructively — note a complete lack of any kind of positive suggestion content.

      I do have hope for KDE. I was put off by every 4.* release I’ve tried so far, but I’m hearing positive things — from credible sources (to me, anyway) that I may have been premature in my panning of your hard work.

      Last year I found myself in the position of having nothing positive to say whatsoever, which is what prompted my necessary step of just taking a big jump back, re-examining how I did things, and just walking away from KDE for my own good.

      It was a good decision. Change is, after all, hard. In taking a long walk in the non-KDE universe for a while, I’ve learned a lot, and I can’t say that that is ever a “bad thing”.

      That said, it didn’t come without its pain. I can appreciate the tact with which you worded your comment on this haphazard, off-the-cuff blog post of mine I made while bleary-eyed this morning, and I can say that I also appreciate the points you’ve made.

      Having followed the KDE4 saga from the beginning, I have to say, you’ve one thick skin, Mr. Seigo.

      So, let me offer an olive branch in my saying thank you for bothering to take the time to comment on someone so spiteful and angry and lacking in constructive criticism in what must seem to be just-another-KDE-rant at least in part.

      I do have substance to my argument (and Joe very eloquently put some of it in his defense of my position), and I know I speak for others in at least some of my sentiments.

      However, as Joe put it, this is after all, about community, something I’ve posted about recently myself, and I’d be remiss to dismiss such an obvious example of that by being an ass even on my own blog on the subject of change.

  12. As a further point of clarification: that work on the fileviews started _before_ Dolphin came on to the scene and was part of the plans for Konqueror from “day 0” of KDE 4 development.

  13. Thanks for the article. It is something which has been on my mind since I left GNOME 1.something for KDE (and haven’t looked back since, thank you.)

    Reverse feature creep (feature drain?) has unfortunately become the buzzword for Ubuntu (at least the GNOME version). In fact, GNOME and GNOME along, IMO, is to blame for it!!!

  14. In defense of Pcmanfm2:

    Pcmanfm has been rewritten from scratch because it was broken in many ways. Trent, i’m surprised you didn’t come across problems which concern basic functionality like mounting/unmounting external drives, file management dialogs (like the one you get wen you paste a file in a folder with a file with the same name), desktop management, etc. As far as “features” and functionality goes, I think it’s fair to assume these things are more “basic” than “tree view”, for example.

    Further, I don’t think Pcman (the main developer of the project) considers the first version of Pcmanfm2 a complete work yet. As far as I know, the LXDE team were concerned with having a working version of their file manager in time for this spring’s distribution releases, Lubuntu in particular. Pcmanfm2 has been out for less than a month! I’m sure the missing functionality will return sooner than you expect.

    • Further, I don’t think Pcman (the main developer of the project) considers the first version of Pcmanfm2 a complete work yet.

      Hey, if that’s the case, I’ll happily stand corrected, particularly if it means it’s going to be worth using once more work is done on it!

      Thing is, from the purely end-user perspective, you must understand what it looks like though. Without being privy to the goings-on behind the scenes, it PCManFM really just looks like the proverbial piano that keeps having keys taken away. Heck, by comparison to earlier versions, the latest version looks like a piano with no black keys at all. Pretty hard to play!

      I think in order to avoid that misconception, it may have been a better idea to get it developed a bit more completely before subjecting users to it as the “latest-and-greatest”, don’t you think? I mean, unless it’s clearly labeled as a “beta” or an “alpha” or something.

  15. Yeah, I hear you, but you also have to take into account that, at least on the offical LXDE blog, which I regularly follow, the new PcmanFM was never announced or advertised as “the latest and greatest”, let alone a complete product. It was always referred to as the first stable version of the new series, with the emphasis on “stable”, nothing less, nothing more.
    All I can say is that it is indeed “stable”, and basic functionality, like mounting drives, is much better implemented right now.

  16. Just wanted to pop in and say that I’ve been using KDE 4 fulltime since 4.3 and I haven’t really found ANY things I dislike! (obviously I don’t speak for everyone). I mostly use PCLinuxOS, with occasional tryouts of Linux Mint, and although I wasn’t using Linux during the 4.0 to 4.2 period, I’m just as happy with the new version as I ever was on 3.5. As far as removed functionality, I am not sure what you miss – I find that some was added (the easy access to a terminal from within a Konq folder being one of the best).
    Trent I do hope you’ll give KDE 4 another try, you needn’t wait until 4.5 😉

  17. This article makes some very good points. Many developers seem to think ‘less options’ == ‘easier to use’. But the two concepts are only tangentially related.

    My question is: is there any graphical file managers that have an option to show the “.” and “..” directories? I would love to be able to have icons for those two. Even with a full page of files you could always right click on “.” and select paste. Or you could change the permissions on the current directory. Or right click on “..” and paste files into the parent directory without having to navigate out of the current directory. The possibility are endless!

  18. @RyanPatterson

    Krusader does have the functionality you asked about.

    In my opinion it is one of the best twin-panel file managers ever available.

    Very customizable, lots of options. Set it to work the way you want.

    I really hope the Krusader Krew keeps it up.

  19. I’m the developer of PCManFM and I accidentally found this post. Well, things are quite different from what you think.
    The original PCManFM is started in 2007 and is finished in 2009. PCManFM2 is a complete “rewrite” from scratch and it’s not a derived work. What you can see now are done in less than 1 year by very few developer. So it’s not possible to be feature-rich ATM. However, it now supports gvfs and seamless access to remote filesystems and trash can, which are all lacking in old series. Of course, this is only the beginning. With the rewrite, the source code and APIs are all carefully redesigned, so it’s much easier to extend and add new functionality to. Adding new features is just a matter of time and man power. In the old series, the internal structure is problematic and this make adding new features very difficult. In addition, with recent changes happened in Linux (HAL being replaced by udisks , the use of PolicyKit was enforced, and some specs are changed arbitarily by its owners), the old PCManFM no longer works and cannot compatible with those rapidly emerging new (actually gnome) standards. So, like it or not, if I don’t do this rewrite, the old pcmanfm will be broken soon due to incompatible changes happened in Linux, Gnome, and People said free software is all about choice, but actually, IMO all too often we have no choice at all.

    • PCMan thanks for the comment and the clarifications, I appreciate that a lot.

      I was hopeful that this was in fact the case… the original PCManFM kind of became my favorite file manager over the last year or so. Forgive me if I got a little worked up when — from my point of view — it looked like it just took what appeared to be a gigantic step backward!

      I’m glad this is still a work-in-progress (and not the final product), and I look forward to the continued development of PCManFM2. Thanks again for the comment, you didn’t have to respond to an off-the-cuff rant on a niche Linux blog. 🙂

      I’m actually a big fan of LXDE too and I’m pretty excited to think of it as a brand new step in the right direction when it comes to desktop environments.

  20. I agree completely with this post. This is one of my biggest pet peeves and started when I first used MS Word for the Mac in 1986. It was the first computer program I had ever used and I was very impressed — until I used WordPerfect for DOS. It was harder to learn but so much more powerful and logical.

    The latest iterations in this are the MacOS X interface and the MS Office “ribbon.”

    I call this the “Make everything easier until most things are completely impossible” school of software development.

  21. I hate this sort of thing too. You don’t need to remove features to make an application accessible to people.

    Hide the complexity by default, but leave it available for those that need it. A highly configurable application with sane defaults can cover both use cases well.

    If an application becomes bloated because of too many features, follow UNIX philosophy. Split the application into smaller pieces and make them work well together. Don’t completely remove the features, just separate them so individual components are fast and make sure they work together.

    I find GNOME to be the worst offender here, at least since version 2. Applications completely remove options and take choices away from the user in a very Windows-like fashion. The disease started with Metacity and infected the rest of the project, to the point that I dislike using most of the environment, except for a few apps.

  22. ctrl+l in nautilus opens the location bar. at first it may seem like an extra/unnecessary step, but once you take into account that it removes the step of navigating to the bar, it breaks even.

  23. I really hate linux’s file manager choices, I recently started using Linux and the one thing that bugged me was Xubuntu’s Thunar, although It is simple to use it lacked most day to day things, such as formatting devices, file/films info… so on. I toyed with the idea with using nautilus as it was built in as it is required for the desktop and a bunch of others, I really didn’t like what I found. So I started seaching how to mod either Thunar or Nautilus and found nautilus-elementary Which is a beautiful modified version And I recommend it to to everyone 🙂

    (Link: )

    And I also found how to Permanently make the Location bar Text (instead of Ctrl L) Place into the terminal

    gconftool-2 –type=Boolean –set /apps/nautilus/preferences/always_use_location_entry true

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s