Weekly Edition
Return to the Distributions page
| |||||||||||||
Linux Mint 12 released
The much-hyped Linux Mint 12 release has been announced. There is plenty of
new stuff in this release; see the "what's new"
page, the release
notes, and this LWN article for more
information. "The Linux Mint 12 desktop is a mix of old and
new. It's a brand new desktop but with traditional components. The new
technology in Gnome 3 is exciting but the components contributed by MGSE
make users feel at home. Linux Mint 12, like previous releases, and despite
the fact that it's based on Gnome 3, looks and behaves like a Mint
desktop."
(Log in to post comments)
Linux Mint 12 released Posted Nov 27, 2011 5:54 UTC (Sun) by b7j0c (subscriber, #27559) [Link]
it seems most of the interest in mint lies in the perception that it is the anti-ubuntu (insofar as unity)
i don't understand why you have to change your distro if you don't like unity. i'm typing this on ubuntu 11.10 with xmonad only...it was a trivial change to make
Linux Mint 12 released Posted Nov 27, 2011 9:45 UTC (Sun) by Rehdon (guest, #45440) [Link]
Why should I choose Ubuntu if I don't like Unity is the question: why should I choose a distro that breaks completely with its recent past, leaving only a namesake "classic" mode that's going to leave us soon? why should I have my working workflow dictated by people who admittedly don't care about current users? why should I use an environment as configurable as a kiosk? And those questions apply to Gnome 3.x too.
The interest in Linux Mint IMNSHO lies in the fact that they customize the desktop environment so that you can experiment with new stuff *if you want*, or have a more conservative experience *if you want*. They leave users choice, which both the Gnome and the Canonical devs don't do, because they know better than you anyway; and no, installing Ubuntu just to scrap Unity is not a 'choice' to me: it's all about the environment and sane defaults, I have reversed minor changes that I disliked (like the window buttons position) but after a certain number of changes it just doesn't make sense to continue using something which clearly doesn't suit your needs.
BTW I'm going to try Linux Mint and, if it works for me, this is the ONLY way for Gnome Shell to be installed on my boxes; otherwise it'll be XFCE. So if I were a Gnome dev I'd be grateful to the LM guys.
Rehdon
Linux Mint 12 released Posted Nov 27, 2011 17:53 UTC (Sun) by dmcguicken (guest, #57851) [Link]
> Why should I choose Ubuntu if I don't like Unity
The community? Package availability? The other stuff they list on their 'why use Ubuntu' page?
Also: you haven't really answered his question - you say you don't want to install Ubuntu and scrap Unity, yet you say later that XFCE is an option if Mint doesn't work out for you... XFCE on what? And why not just use it on a distro that you already use and know, since that same distro offers you an XFCE package? Is it purely a protest vote?
Also: 'sane defaults'... for whom?
Linux Mint 12 released Posted Nov 27, 2011 18:14 UTC (Sun) by Rehdon (guest, #45440) [Link]
Woah, lots of questions! Let's see:
- last that I checked Linux Mint is a Ubuntu derivative, so I'm not exactly giving up on package availability, community etc.
The problem for me is not Ubuntu, it's Unity (and/or Gnome Shell): if I can have a decent environment using an Ubuntu derivative then all the better.
Rehdon
Linux Mint 12 released Posted Dec 4, 2011 19:38 UTC (Sun) by gvy (guest, #11981) [Link]
> The community?
The fish rots from the head as they say.
Linux Mint 12 released Posted Nov 27, 2011 20:06 UTC (Sun) by Frej (subscriber, #4165) [Link]
So GNOME devs do not allow choice, but have created a platform that makes it possible for a small bunch of volunteers to create a gnome 2.x lookalike with very little work, and still keeping gnome 2.x on on the side.
Nobody is forcing anything on you. The guys behind linxmint shows how, and it's quite impressive. It does seem to be hard to grasp these days.
But luckily for you, linuxmint did the hard work. They even posted a very nice, balanced and informed blog post about MGSE earlier, with no complaining whatsoever. Try it, it makes you feel warm inside.
Linux Mint 12 released Posted Nov 27, 2011 23:26 UTC (Sun) by robert_s (subscriber, #42402) [Link]
"But luckily for you, linuxmint did the hard work."
I'd argue that debian did the hard work.
Linux Mint 12 released Posted Nov 28, 2011 0:06 UTC (Mon) by Frej (subscriber, #4165) [Link]
Either way works for me :)
Linux Mint 12 released Posted Nov 28, 2011 2:38 UTC (Mon) by atai (subscriber, #10977) [Link]
The original post was referring to the customization of GNOME 3 to make it look like GNOME 2. Doubt Debian can take credit for that
Linux Mint 12 released Posted Nov 27, 2011 10:31 UTC (Sun) by oldtomas (guest, #72579) [Link] "i don't understand why you have to change your distro if you don't like unity" Heh. Welcome to Diversity (i couldn't resist the somewhat cheap pun). I think that's one of the strengths of Free Software. Remember: whereas Unity might have no place for Diversity, Diversity surely has a place for Unity. To each her own and all that.
Linux Mint 12 released Posted Nov 28, 2011 8:51 UTC (Mon) by drag (subscriber, #31333) [Link]
The whole 'lets make a new distro' has always irritated me. I have had no problem at the login screen choosing 'KDE' vs 'Gnome' vs 'XFCE'.
It would require much less effort if Mint was basing their stuff on Ubuntu to just let Ubuntu users install it easily. Don't like Unity? Just download the mint-repo deb file, install the package by double clicking on it and then go 'apt-get install mint-desktop'.
Log out, select 'mint', log back in.
Done. Takes about 30 seconds of effort + whatever time it takes to download and install the files and that's it. He can even setup the defaults to use the search engines so that he can get the money from the advertising kick-backs (which is perfectly fine. It's not something that is being hid)
But whatever he wants to do is his choice. If the Mint dev likes making his own ISO images with Ubuntu artwork changed to Mint, then thats fine. If people appreciate his efforts and can find value from it, then that is terrific. Can't argue with that at all.
But it will prevent me from trying it out because I have better things to do then setting up my PC after a new install every time something half-way interesting pops up.
My opinion is not something that is unique for Mint. I have the same opinion for Kubuntu, Xubuntu, Lubuntu, Ubuntu Studio, and all the various Fedora Spins. All the choice is lovely, but after a while making choices seems more like punishment.
I am happy that people want to do stuff like that and if they and others derive some enjoyment and value then that is fantastic. This is just my personal opinion and I don't think any less of people that like to do spins or install them. I just think that there is a easier way, that is all.
Linux Mint 12 released Posted Nov 28, 2011 9:02 UTC (Mon) by dlang (✭ supporter ✭, #313) [Link]
kubuntu, xubuntu, and lubuntu are exactly what you are asking for.
to convert ubunto to kubuntu, just apt-get install kubuntu desktop and then at your next boot opt to use KDE instead of unity
shipping separate installation .iso files is mearly a convenience thing to save time and a little space by not installing the desktop environment you aren't planning to use.
Linux Mint 12 released Posted Nov 28, 2011 9:51 UTC (Mon) by drag (subscriber, #31333) [Link]
Cool. I stand corrected then.
Linux Mint 12 released Posted Nov 28, 2011 9:17 UTC (Mon) by ekj (guest, #1524) [Link]
Sure, you can tweak individual details. But if there's a trend where large decisions are repeatedly implemented in a direction you don't like, it makes sense to look for alternatives.
I'm on Ubuntu for now, but Unity does not work for me. It's broken in 20 different ways, and too many of them are unfixable. (or fixable, but not in a discoverable way)
I need to move my mouse to the top bar to discover what will happen, only to have one of 4 things happen:
1) Nothing - in the case of a non-maximized window with menues on the window itself (i.e. Firefox)
2) Application-menu appearing. (i.e. Pidgin)
3) De-maximize and close-buttons appearing in the case of maximized windows with menues on the window itself (i.e. Eclipse)
4) A combination of 2 and 3 in the case of maximized windows with menues in the top bar.
4 Different things *may* happen, and you don't know which one until you've physically moved the mouse up there to see what'll happen (in case 1 though, when you do, nothing happens)
Oh, and you must remember to activate the right window first, offcourse, if you're in a different window.
Thus where before to get the Pidgin menu, I'd click Pidgins menu, I must now first click a Pidgin-window to activate the program, and *then* move the mouse up to the top of the screen to see what (if anything!) will happen. This does not work.
Linux Mint 12 released Posted Dec 5, 2011 23:14 UTC (Mon) by ssam (subscriber, #46587) [Link]
"i don't understand why you have to change your distro if you don't like unity."
because ubuntu have patched some apps to support their indicator system. For example tomboy which has a very useful notify area applet, but a very weedy unity indicator applet. So if you just switch desktops you get broken apps.
I have been trying gnome3 in fedora for about a month now, and am getting more and more frustrated. Things like only having worksspaces on 1 monitor (the is a gconf setting to change that to broken workspaces across 2 monitors). alt+tab being app based instead of window based (you can change this in the key bindings, but the window based one has no window list, and graphics glitches). i could go on.
xfce does a lot right, but falls down at hot plugging monitors (i need to use xrandr to get them in the right place, and it moves windows to odd places). next stop KDE i suppose.
Linux Mint 12 released Posted Dec 6, 2011 15:54 UTC (Tue) by mathstuf (subscriber, #69389) [Link]
> xfce does a lot right, but falls down at hot plugging monitors (i need to use xrandr to get them in the right place, and it moves windows to odd places). next stop KDE i suppose.
I run xrandr manually and find it much faster than using any GUI. XMonad helps by putting a workspace on each monitor (which I find to be strictly better than extending each workspace to wrap every monitor), so I know that the workspace-after-current will be put on the projector or monitor or whatever.
Personally, I find that anything *other* than xrandr is a nightmare to use. I have a goal of 10 seconds between standing behind a podium and starting the presentation and any GUI just falls flat with this. Presenters asking "does anyone have a DisplayPort converter" or fiddling with getting projectors to work (fighting between mirroring and extending, detection, and resolutions) are painful to watch.
Linux Mint 12 released Posted Nov 27, 2011 21:02 UTC (Sun) by xxiao (subscriber, #9631) [Link]
tried it within a vbox for a few hours, looks well polished, like it.
Linux Mint 12 released Posted Nov 28, 2011 5:31 UTC (Mon) by jiu (subscriber, #57673) [Link]
I haven't tried it but that's something the GNOME devs should be watching. Why not provide their version of the Mint Gnome Shell extension by default in Gnome?
Linux Mint 12 released Posted Nov 28, 2011 8:29 UTC (Mon) by ovitters (subscriber, #27950) [Link]
By default, no. GNOME 3 is not finished yet and just because someone makes changes doesn't mean we have to provide _exactly_ that experience by default. That various people like it I don't doubt, but that is why extensions can be used.
The entire extension infrastructure is there to make changes like this possible. If the extension is self-contained (didn't require any code changes), then Linux Mint could put it on extensions.gnome.org once available. Currently that site is password protected (closed testinggroup due to security implications). We're planning to open the site up on Dec 1st.
After that site is up, stuff like MGSE, those Frippery things or (hopefully) Wanda would be just a few clicks away for those who care.
Linux Mint 12 released Posted Nov 28, 2011 9:51 UTC (Mon) by drag (subscriber, #31333) [Link]
I had no idea that frippery was actually a word before your post. That makes me happy.
Linux Mint 12 released Posted Nov 28, 2011 11:42 UTC (Mon) by richo123 (guest, #24309) [Link]
I am glad to see gnome responding to user demand ;-)
Linux Mint 12 released Posted Nov 28, 2011 19:07 UTC (Mon) by jiu (subscriber, #57673) [Link]
sounds good! thanks for the heads up
Linux Mint 12 released Posted Nov 29, 2011 4:12 UTC (Tue) by bojan (subscriber, #14302) [Link]
> The entire extension infrastructure is there to make changes like this possible.
Unfortunately. The real problem is that Gnome Shell cannot be customised properly. Yeah, I dared to mention the dirty word. :-)
Linux Mint 12 released Posted Nov 29, 2011 9:43 UTC (Tue) by ovitters (subscriber, #27950) [Link]
I don't get why you say that extensions do not provide customization. Care to clarify your logic / reasoning? Maybe compare to Firefox, because I don't get it.
Linux Mint 12 released Posted Nov 29, 2011 14:18 UTC (Tue) by nye (guest, #51576) [Link]
>I don't get why you say that extensions do not provide customization.
Because the ability to write extensions is a skill available to an almost immeasurably small fraction of the potential userbase; it's only somewhat better than just saying 'the code is open source; modify it if you want changes'.
The effect is that the final product is good for two extremes of people: those who never change any defaults and are willing to use whatever UI paradigm is the flavour of the month because they either don't know or don't care about any options they may have, and those who are able and willing to write and maintain their own extensions - these are roughly speaking the same people who were happy to hand-write their XF86Config, maybe they've been running the same Slackware installation since 1994, etc.
In between those extremes is a class of users which I'd argue represent the majority, who want to have some control over the interface they spend eight hours a day using, without having to return to the batteries-not-included days of hacking something together manually.
Linux Mint 12 released Posted Nov 29, 2011 14:30 UTC (Tue) by bojan (subscriber, #14302) [Link]
+1
Example: to remove accessibility icon, one need an extension. Seriously?
Gnome Shell right now is like a sound card with jumpers. When did you last see one of those and had to open the box to change the IRQ?
Linux Mint 12 released Posted Nov 29, 2011 15:23 UTC (Tue) by ovitters (subscriber, #27950) [Link]
Your example is possible. I fail to see the logic.
Linux Mint 12 released Posted Nov 29, 2011 21:31 UTC (Tue) by bojan (subscriber, #14302) [Link]
I'm not surprised.
You also failed to see that requiring several GUI actions to change workspaces and loss off visibility is worse than the ability to see where you are and change workspaces with one click.
Linux Mint 12 released Posted Nov 29, 2011 15:26 UTC (Tue) by ovitters (subscriber, #27950) [Link]
So you're saying Firefox only has a few extensions? I don't think that is true. Suggest to take a look at http://addons.mozilla.org/.
Linux Mint 12 released Posted Nov 29, 2011 21:53 UTC (Tue) by bojan (subscriber, #14302) [Link]
I'm not saying one thing about Firefox.
What I'm saying is that Gnome Shell doesn't allow for the simplest of customisations without extensions. Especially if one compares Gnome Panel to it, it is quite obvious that many things that could be done cannot be done any more. These are usability regressions.
The shell itself needs to gain the ability to have things set the user wants them, without extensions.
PS. If you do want to talk about FF, please do go to preferences of it first and have a look at how many things _can_ be customised there, without loading _any_ extensions. In fact, I have FF just the way I want it, but I do not have one extension installed.
Linux Mint 12 released Posted Nov 29, 2011 22:45 UTC (Tue) by ovitters (subscriber, #27950) [Link]
So extensions provide a lot of customization. But you don't call that customization so you can complain about it? Extension infrastructure is provided by default btw so the user can change, without digging into code or anything.
I brought up Firefox for a reason. Mozilla Seamonkey has loads of preferences. Firefox removed loads and loads, suggested extensions instead.
GNOME 3 can be changed, by System Settings, extensions, then for the "power user": GNOME tweak tool, dconf-editor. Again similar to Firefox.
But oh well, seems glass has to be half empty and nothing is acceptable. Suggest MATE.
Linux Mint 12 released Posted Nov 29, 2011 23:01 UTC (Tue) by bojan (subscriber, #14302) [Link]
Please be serious.
Say you load MGSE or Frippery. It will give you a workspace switcher somewhere. How do you move it to another place on the panel without writing your own extension? That's right, you don't.
I gave you the accessibility icon example on purpose. It is trivial in the old panel to say you don't want something or that you want it somewhere else. It is impossible in the shell.
The fact that someone had to write an extension to remove an icon (I believe there is another one for bluetooth icon) is a the failure I'm talking about.
Linux Mint 12 released Posted Nov 29, 2011 22:47 UTC (Tue) by drag (subscriber, #31333) [Link]
> Because the ability to write extensions is a skill available to an almost immeasurably small fraction of the potential userbase; it's only somewhat better than just saying 'the code is open source; modify it if you want changes'.
No it's not the same at all.
Extensible software is vastly more powerful and flexible then non-extensible software. This is one of the major reasons why Microsoft Office is so successful. Same thing with Emacs. And Firefox and a whole host of software.
A very good approach to software design is to attempt to provide what does 90% of what is required and allows people to extend it for whatever else they want. With a proper extension frame work you can keep the interface simple and clean. And you can reduce the development effort required to support it. Users that want different functionality can have it while users that don't want to see it or deal with it do not. It is about as close to 'having your cake and eating it, too' as you can get.
Your basic logic also does not hold any water:
On one hand you are saying that Gnome is missing significantly important functionality. On the other hand you are saying that extensions require effort that is only slightly less then just hacking on the source code.
Don't you see the fault in that line of reasoning?
If there is functionality that is commonly desired then in short order somebody is going to develop way through a extension or other method to enable it. Just like how people complained that they couldn't set fonts in Gnome, but it was already there through the gnome-tweak application since nearly the beginning. So if there is some item or configuration option that people really want then, if it's possible, there is going to be a extension to enable it.
Users then can use that extension to get the required functionality without knowing how to write a single line of javascript easily. The Mint extensions are a perfect example of this.
If it turns out that features are in high demand that people have unfortunately a hard time with on extensions then future versions of Gnome can correct this by incorporating the desired functionality.
Linux Mint 12 released Posted Nov 29, 2011 23:03 UTC (Tue) by bojan (subscriber, #14302) [Link]
There is nothing wrong with having the ability of extensions. What is wrong is that the most trivial UI customisations (e.g. removal of an icon, placement of UI elements etc.) require someone to write one.
Care to explain? Posted Nov 30, 2011 7:04 UTC (Wed) by khim (subscriber, #9252) [Link] What is wrong is that the most trivial UI customisations (e.g. removal of an icon, placement of UI elements etc.) require someone to write one. Why is it so wrong? You already said: I'm not saying one thing about Firefox. Ok. But... the GUI to rearrange icons and other UI elements was one the things which were lost in Mozilla to Firefox transition. Can you explain why this change is Ok for Firefox (AFAICS it never regained this ability: you still need to write an extension to disable search box, for example), but if GNOME does the exact same thing (not even similar one - the exact one, down to the minute details) then it's "sky is falling, end the world is near" moment?
It's possible with Firefox Posted Nov 30, 2011 8:40 UTC (Wed) by spaetz (subscriber, #32870) [Link]
> But... the GUI to rearrange icons and other UI elements was one the things which were lost in Mozilla to Firefox transition.
I'm staying out of the flames here, just one correction. Right-click on e.g. the navigation toolbar and select customize. You can easily click and move and drop and put on any icon/button to your hearts content without any add-on.
Sorry for stepping in :-)
Care to explain? Posted Nov 30, 2011 13:41 UTC (Wed) by bojan (subscriber, #14302) [Link]
Never claimed sky was falling. Just that making a GUI where it takes a programmer to remove an icon is nothing short of ridiculous. Ditto exuses about users being distracted by workspace switcher and taskbar (and not by the presence of Activities, clock, your own name etc.). Ditto changes where GUI now requires multiple actions for what one click was enough.
Factually, these are usability regressions.
Care to explain? Posted Nov 30, 2011 14:03 UTC (Wed) by daniels (subscriber, #16193) [Link] Factually, these are usability regressions. Factually, you're wrong. (See how tedious this line of argument is?)
Care to explain? Posted Nov 30, 2011 14:08 UTC (Wed) by nye (guest, #51576) [Link]
>Factually, you're wrong. (See how tedious this line of argument is?)
Just so we're clear, is it your position that all else being equal, a GUI in which only a programmer can remove a button is as usable or more usable than the same GUI in which an unwanted button can be removed by an end-user?
Care to explain? Posted Nov 30, 2011 14:21 UTC (Wed) by daniels (subscriber, #16193) [Link]
Well, my point was mainly that your line of argument is massively unhelpful to everyone. To answer your question though - making a blanket declaration is impossible. Do you think Chromium would be more usable if there was exposed UI for removing the back button?
Care to explain? Posted Nov 30, 2011 15:05 UTC (Wed) by nye (guest, #51576) [Link]
>Well, my point was mainly that your line of argument is massively unhelpful to everyone
(Note that I'm not the person to whom you were originally responding)
>To answer your question though - making a blanket declaration is impossible.
Well the example given by the OP was an 'Accessibility' button, and his claim was that the inability to remove it constitutes a regression in usability. I can't envisage any argument that would oppose this but can see clear reasoning in its favour so have come to the conclusion that it is objectively true.
Perhaps some contrary reasoning will come up and I will conclude that there is some room for subjectivity, but currently it seems that the statement is simply factually correct, and don't understand why that would be considered 'massively unhelpful to everyone'.
>Do you think Chromium would be more usable if there was exposed UI for removing the back button?
Interesting question. I follow your point that not all UI components can be considered equal and so the same conclusion can not necessarily be made for every element. In this case though I would say 'yes', and my reasoning is as follows:
There exist people who would like that option (eg http://superuser.com/questions/90007/hide-toolbar-buttons...). Having it available (as it is in Safari, Opera, Firefox at least) would come at nobody's expense so even a small improvement for usability for a small number of people would come as a net positive.
Care to explain? Posted Nov 30, 2011 15:45 UTC (Wed) by daniels (subscriber, #16193) [Link] Having it available (as it is in Safari, Opera, Firefox at least) would come at nobody's expense Again, this is an assertion which many people have disagreed with over the years, and continue to vehemently disagree with. If that's your point of view, that's fine and you're more than entitled to it. But presenting it as an obvious fact isn't helpful in any way, and just makes arguments really tedious because both sides are presenting diametrically opposed opinions as fact.
Care to explain? Posted Nov 30, 2011 16:34 UTC (Wed) by nye (guest, #51576) [Link]
>>Having it available (as it is in Safari, Opera, Firefox at least) would come at nobody's expense
>Again, this is an assertion which many people have disagreed with over the years, and continue to vehemently disagree with.
I honestly cannot conceive of how this could possibly be spun as a matter of opinion. I believe it to be an objective fact, no more worth arguing about than the question of whether 1!=0.
You say 'again', but in fact nobody here has yet made the claim that the feature in question would come at somebody's expense, so this is the first I've heard that you don't consider this point to be a matter of fact. (Indeed, it hadn't occurred to me that you would disagree on that particular point; I simply don't see any opportunity for disagreement so your response came as a total surprise.)
Clearly I've failed to understand what you're trying to say; I think I'm now more confused than ever. Unfortunately you haven't made any attempt at making a coherent point and actively avoided my attempt to understand your position better.
Please understand that when I see person A making what I believe to be a purely factual point, and then see person B claiming that the given assertion is actually a matter of opinion, without even *trying* to make some argument as to why the given statement might be open to some subjectivity, it is hard to avoid the conclusion that person B is engaged in pointless obstructionism.
Allow me to illustrate this with an analogy: it is a fact that the process of evolution is known to have taken place. There are those who disagree; they are objectively wrong. Periodically those people attempt to interrupt serious work by proclaiming it to be only an opinion, but their claim does not make it so. It is simply an act of derailment.
Care to explain? Posted Nov 30, 2011 16:44 UTC (Wed) by daniels (subscriber, #16193) [Link]
Yeah, this is exactly what I mean. It's extremely tedious and 100% pointless - I'll leave you to it.
Care to explain? Posted Nov 30, 2011 18:23 UTC (Wed) by nye (guest, #51576) [Link]
>Yeah, this is exactly what I mean. It's extremely tedious and 100% pointless - I'll leave you to it
Okay, I don't understand what you want. Have I said anything in this thread to indicate that I'm not sincere in my desire to understand you?
So far you have been persistently rude and dismissive while refusing to make any actual points, and I can't see a reason why.
And waht about another situation? Posted Nov 30, 2011 17:56 UTC (Wed) by khim (subscriber, #9252) [Link] Please understand that when I see person A making what I believe to be a purely factual point, and then see person B claiming that the given assertion is actually a matter of opinion, without even *trying* to make some argument as to why the given statement might be open to some subjectivity, it is hard to avoid the conclusion that person B is engaged in pointless obstructionism. Right. This is not what happened here. You've started with incorrect assertion and then used it to present your subjective opinion as a purely factual point... Forgive me, but I'm not impressed. You have said: Having it available (as it is in Safari, Opera, Firefox at least) would come at nobody's expense so even a small improvement for usability for a small number of people would come as a net positive. First part (having it available ... would come at nobody's expense) is factually incorrect so you can not use it as base for your "objective" conclusion. Options have price. Always. Someone must write the code to support it, someone else should test it, this increases burden on the support personnel (because someone may toggle option and then forget about it, or do that accidentally and have no idea how to go back, etc) and makes options dialog less comprehensible, etc. It even affects seemingly unrelated people. For example if you decide to remove back/forward buttons then you assume that web sites will not intercept Alt-Left and Alt-Right combinations. In short: your facts are not facts, but just your beliefs. It may be possible to argue if the ability to disable Back/Forward buttons (or just the ability to disable Forward button while keeping Back button intact: Firefox had this ability few years back, but now it does not) is a net positive or not, but to present this as an objective fact and claiming it's it's like the question of whether 1!=0 is just wrong.
And waht about another situation? Posted Nov 30, 2011 18:08 UTC (Wed) by nye (guest, #51576) [Link]
>Options have price. Always. Someone must ...
>... >In short: your facts are not facts
I believe there has been a small misunderstanding here. My apologies; I should have been more precise. When I say "at nobody's expense" I meant to say "at the expense of no users", or "at no expense in terms of usability".
I though this was implicit when I pointed out:
Does that clarification change your assessment?
And waht about another situation? Posted Nov 30, 2011 18:30 UTC (Wed) by mjg59 (subscriber, #23239) [Link]
The cost of implementing it. The cost of ensuring that it interacts correctly with all other options. The cost of documenting it. The cost of there now being more documentation that the user has to dig through when looking for something unrelated. The cost of user confusion when they accidentally hide their back button. The cost of having to turn it back on when you find a website that manages to eat your back key.
And waht about another situation? Posted Nov 30, 2011 20:25 UTC (Wed) by nye (guest, #51576) [Link]
(Thanks for making a serious and helpful response; I think you win the thread.)
Would it be fair to say that, regarding usability alone (because that was specifically the point being made), your position is that there is a potential cost to allowing a user to remove a button which essentially boils down to 'some user might do it accidentally'[0]?
If so, I think I'm finally starting to understand the problem here - I would probably never have considered that as a potential usability cost since it seems so small that I've unconsciously dismissed it.
I get the impression people have been trying to make more of what I'm saying than I mean to put into it. When I asked "is it your position that all else being equal, a GUI in which only a programmer can remove a button is as usable or more usable than the same GUI in which an unwanted button can be removed by an end-user?" I couldn't think of any plausible reason why the answer might be yes, so daniels' evasion of the question seemed to be implicitly saying either "yes, but I don't have any real reason" or "no, but I just want to argue with you".
I now suppose that an accurate answer might have been "yes, because there can be a cost for people who remove the button accidentally, and that cost might outweigh the benefit to the people who do so intentionally".
Having established that a counterpoint actually exists that's a sensible starting point for a dialogue about trade-offs, but now I'm far too exhausted.
[0] I know, the digging-through-documentation point is also usability-related, but I really think that's a red herring. I guess we could argue the toss there but I'm sure it would be unproductive.
And waht about another situation? Posted Dec 1, 2011 7:39 UTC (Thu) by daniels (subscriber, #16193) [Link]
Havoc Pennington elaborated this quite well here, many years ago: http://ometer.com/free-software-ui.html
And waht about another situation? Posted Dec 1, 2011 11:14 UTC (Thu) by paulj (subscriber, #341) [Link]
But those costs only go away if the option goes away. If, on the other hand, you replace a limited set of options with an extension language that opens up the innards of the programme for run-time customisation, and then the options are implemented in that extension language then you have:
a) still the cost of those options
b) the added costs of everyone potentially having a different run-time
If you had argued it *moves* the costs, e.g. from the core GNOME developers to others (hacker users), then fair enough. However, even then, many of the extensions for popular options will end up being maintained by distributors, who may even also be core GNOME developers - so for some code the move may be simply be between *roles* of the same person!
There's one other cost which has yet to be realised:
c) the security costs that come with a run-time that is trivially modifiable.
But that will happen, if gnome-shell becomes popular enough. I gather it's something the extensions.gnome.org project may already be grappling with.
In short, options do have costs, but to get rid of those costs you actually have to get rid of the options - not just shift the burden for them.
And waht about another situation? Posted Dec 1, 2011 11:45 UTC (Thu) by dlang (✭ supporter ✭, #313) [Link]
look at firefox extensions as an example of doing it wrong. they were one of the first ones to offer browser extensions, but what they did was to allow the extension to modify anything.
this makes the extensions significant security risks
but it also means that updates to firefox break extensions that are mucking around deep in the code.
look at all the complaints there are to mozilla about breaking extensions and think hard about if you want to risk being in a similar situation.
Still no cigar... Posted Nov 30, 2011 20:12 UTC (Wed) by khim (subscriber, #9252) [Link] When I say "at nobody's expense" I meant to say "at the expense of no users", or "at no expense in terms of usability". This is still not true. As I've said: I'm not even convinced that number of users who will benefit from the ability to disable certain button is larger then number of users who may accidentally turn the feature off and then be bitten by the fact later. Both sets are small, true, but they both are not empty and IMNSHO "problem" (with quotes) for the users who must tolerate additional button is much smaller then problem (without quotes this time) for the users who try to find the button - and can not. We may decide that we don't care about such dump users - but this, too, is conscious decision and is not like the question of whether 1!=0 P.S. BTW I want to apologize about my mistake: I completely forgot about the fact that it actually was Firefox which added the ability to easily add new button to toolbar, not the other way around. You may guess what I think about importance of said feature from the very fact that I forgot this piece of information :-)
Still no cigar... Posted Nov 30, 2011 20:52 UTC (Wed) by nye (guest, #51576) [Link]
>this, too, is conscious decision and is not like the question of whether 1!=0
Okay, I acknowledge that now that I'm aware of the existence of the second set.
I suppose you think I'm pretty stupid for having to have that spelled out, but it's a concept so alien to me that I'm afraid it just hadn't entered my head. Now that it's been pointed out, I see it should have been obvious.
For what it worth... Posted Nov 30, 2011 21:27 UTC (Wed) by khim (subscriber, #9252) [Link] Well, this concept was alien for me too till I've started helping my sister to fix the problems on her work (she worked in a small firm which had no dedicated IT staff). Most of the "problems" I've encountered were trivial - for me, that is. Examples: Note: these small problems led to significant loss of productivity - they coped somehow (for example when they lost the ability to reach documents from different computer they used e-mail to send these documents back and forth), but still... every time I've encountered their "serious problems" I was not sure if I should laugh or cry. Granted, few times they indeed had real serious problems well beyond their ability to solve. For example when someone moved printer attached via Centronix cable and destroyed both cable and connector on printer (I've solved it by going out and buying USB cable) - but most of the time "problems" were of the sort described above... First few times I was as baffled as you are - but over times I've understood that they are not dumb - they just perceive computer as some kind of magic and don't even expect to see any logic behind all these buttons and knobs. Till I encountered this first-hand I, too, was under impression that all such stories are just bogeymen invented by some people to justify locked down systems.
And waht about another situation? Posted Nov 30, 2011 21:36 UTC (Wed) by bojan (subscriber, #14302) [Link]
> In short: your facts are not facts, but just your beliefs.
Love this bit. Thanks.
It is my belief that if I want to change a workspace using GUI in Gnome Shell that I have to click on Activities, wait for the desktop to rearrange itself, go to the right edge, wait for workspaces to pop up, _find_ the one I actually want and then click it. Not fact, sorry. Just my belief. It is also my belief that I can do that with one click with workspace switcher.
I think I'm going to change my belief system now to The Worship of Gnome 3, in which case, all of the above beliefs will simply disappear. :-)
Oh, I forgot - I should have used keyboard shortcuts. That would surely change my belief about the GUI immediately. ;-)
Seriously for a second. Some really ridiculous stuff started popping up in these threads. Common sense, completely abandoned. Previous usage patterns familiar to millions, completely abandoned.
I think I'll bail.
You win.
Care to explain? Posted Nov 30, 2011 20:58 UTC (Wed) by bojan (subscriber, #14302) [Link]
Now you're trying to be a comedian.
It's like this: grandma now has to walk 5 km to the nearest shop, instead of 50 m.
Me: It's a regression.
You: You're wrong.
Sure, you win.
Care to explain? Posted Dec 1, 2011 1:26 UTC (Thu) by bronson (subscriber, #4806) [Link]
>> Factually, these are usability regressions.
> Factually, you're wrong. (See how tedious this line of argument is?)
A regression is when existing functionality has been removed / broken / bugged / etc. So, factually and objectively, these really do sound like usability regressions. If you disagree, you're going to have to do a little better than just "you're wrong."
I'm not saying regressions are evil or wrong or anything. They're often necessary for moving forward. In moderation of course. :)
Linux Mint 12 released Posted Nov 30, 2011 12:53 UTC (Wed) by nye (guest, #51576) [Link]
>Extensible software is vastly more powerful and flexible then non-extensible software.
I don't disagree.
>Extensible software is vastly more powerful and flexible then non-extensible software. This is one of the major reasons why Microsoft Office is so successful.
True, but one of the other reasons that it's so successful is that it provides, out of the box, more functionality than any competitor. Barring statistical anomalies, home users aren't writing their own macros, and in business use they're usually written for the user by someone in IT.
>A very good approach to software design is to attempt to provide what does 90% of what is required and allows people to extend it for whatever else they want
Required by whom? A common adage regarding MS Word is that nobody uses more than 10% of the features, but they all need to be there because nobody uses the *same* 10%.
Allow me to take your point ad absurdum:
Fundamentally it's not possible for a software developer to determine "what's required" by looking only at their own personal usage pattern, *unless* they are explicitly setting out to target themselves as the only userbase.
>With a proper extension frame work...you can reduce the development effort required to support it
Whilst this sounds like an appealing idea, in reality it generally proves to be false. If you have software which is sufficiently limited that most of its users start having several extensions installed then what you actually end up with is a combinatorial explosion of complexity as each possible combination of extensions is a potential source of problems.
In theory a well-designed extension framework as you say can solve that problem, but in practice such a framework either needs to be so restrictive that its utility is limited or ends up unable to solve all the attendant problems of the extension buffet.
The end result is that while extensibility is a good thing, *requiring* extensions to perform even simple tasks is generally 'too much' of a good thing.
Personally I stopped using Firefox largely because at the time I found it too frustrating to use without extensions, and even wrote a couple, but the effort to maintain them - and the dependency on other extension authors to keep their extensions working as Firefox was updated - eventually grew too much to bear, and I switched back to Opera, because it comes with batteries included.
>On one hand you are saying that Gnome is missing significantly important functionality. On the other hand you are saying that extensions require effort that is only slightly less then just hacking on the source code.
> Don't you see the fault in that line of reasoning?
Not at all. Perhaps I've missed your point (in which case I apologise, but would like clarification), but it seems you haven't even tried to point out the flaw you see.
I will add though that personally I'm not saying that Gnome is missing significant functionality (though others in this thread are); I don't really care about Gnome since I'm not using it. What I *do* care about is the attitude previously expressed that when large groups of users *do* complain about missing functionality, the appropriate response is 'if you really care about it, write your own extensions'.
That attitude is entirely appropriate when the software in question is written by a developer purely for him/herself, and also made available to the public as a bonus. It's *not* appropriate when the software sets a goal of being as widely used as possible by a large and varied userbase, and it's *even less* appropriate when that varied userbase is expected to include non-trivial numbers of non-technical users.
>If there is functionality that is commonly desired then in short order somebody is going to develop way through a extension or other method to enable it
Doubtless.
>If it turns out that features are in high demand that people have unfortunately a hard time with on extensions then future versions of Gnome can correct this by incorporating the desired functionality.
And doubtless they will do this too, but to me it sounds like what you're describing is software which is not yet feature-complete, ie. alpha stage at best. Software released for the general public should be past the stage where it's dependent upon that public to provide for themselves the features they need.
Linux Mint 12 released Posted Nov 28, 2011 8:22 UTC (Mon) by job (guest, #670) [Link]
LMDE sounds interesting, as a rolling release for the desktop. Does it break? Is it usable for my grandmother, the same way Ubuntu is(/was)?
LMDE stability Posted Nov 28, 2011 13:51 UTC (Mon) by Cato (subscriber, #7643) [Link]
I believe LMDE is not recommended yet for non-experts as it can break - see question 4 here: http://blog.linuxmint.com/?p=1818 - also see the Steve Rosenberg section here: http://blog.linuxmint.com/?p=1543
LMDE does now have monthly Update Packs which should reduce the risk of breakage - however, the Ubuntu-based Mint is still recommended for people who don't have the skills to fix Debian Testing breakages. Linux Mint 9 is the LTS release and should work well for long enough for the MGSE/MATE work started in Mint 12 to stabilise in Mint 13 onwards.
I think the only caveat to using the Ubuntu based Mint is if the Ubuntu base of the Mint release is flaky in some areas or with some hardware - e.g. Ubuntu 10.04 LTS is very broken on Intel GPUs. Hence I am switching a PC with Intel GPU using 10.04 not to Mint 9 (based on 10.04) but to Mint 11 (based on 11.04), which should be better in this case and also avoids the GNOME 3 / MATE stuff for now as it uses GNOME 2.3x.
LMDE stability Posted Dec 5, 2011 18:01 UTC (Mon) by jospoortvliet (subscriber, #33164) [Link]
And, just to add to this - Debian Testing is currently in stabilization mode. As it's meant to be the testing stage for the next release, which is due to come out, it is frozen. So right now, I'd say it's actually quite stable. Soon, it will become unstable when new packages rush in.
'Real' rolling release distro's like Gentoo and Arch are better in this regard as they don't stop for releases. The same is true for openSUSE's Tumbleweed rolling release repository, which also doesn't require you to install anything special other than add that repo.
LMDE stability Posted Dec 11, 2011 0:00 UTC (Sun) by pixelpapst (guest, #55301) [Link]
Debian is not frozen. Where did you get that idea ?
Linux Mint 12 released Posted Nov 28, 2011 16:57 UTC (Mon) by dashesy (subscriber, #74652) [Link] The new default search engine is Duck Duck Go. It doesn't show different results depending on who's making the search, it doesn't track or record user information, it provides you with optmized results and it's built on and contributes to Open Source.Very interesting, after Google 'instant' frenzy I tried a couple of different search engines to find to my dismay that none is good enough to search for things like some weird compiler error codes (me being lazy)!. After suffering Google "instant", search result URL obfuscation and redirection, left-hand-side bar, localization, eyeglass popups, re-ordering search phrases!, ... I hope an open source search engine can be a sane and elegant substitution for searching. I will change my homepage to "Duck Duck Go", it was greater if they had the "suggestion" mechanism which for me is unobtrusive helper and spell checker (me being lazy). I think I am spoiled by Google's past when they cared more about the search quality, rather than looking at their search website to promote their other services. I will try Mint because of the beautiful screenshots presented here, but I like Fedora so much I cannot change it, so I will stick to XFCE for now.
Linux Mint 12 released Posted Nov 30, 2011 10:20 UTC (Wed) by Cato (subscriber, #7643) [Link]
To provide Linux Mint with some revenue from using DuckDuckGo, the URL is: http://duckduckgo.com/?t=lm (can also use https://duckduckgo.com/?t=lm)
Linux Mint 12 released Posted Nov 28, 2011 20:23 UTC (Mon) by tdwebste (guest, #18154) [Link]
Mint is popular for the same reason Ubuntu gained popularity. It builds on a great distributions and tunes the user experience to so that users can just do their work and don't need to individually fiddle their user environment.
What i don't like about Mint, it is not easy to upgrade from one mint release to the next.
Linux Mint 12 released Posted Nov 30, 2011 10:24 UTC (Wed) by Cato (subscriber, #7643) [Link]
I prefer Mint's upgrade philosophy (backup your home directory and list of installed packages, and do a fresh install, using Mint Backup) to Ubuntu's (do an in-place upgrade using "apt-get dist-upgrade" or equivalent).
The few times that I've done an Ubuntu in-place upgrade, I've spent many hours fixing subtle problems with the configurations carried over (e.g. Xkb problem meaning keyboard didn't work properly). I suspect it would take far less time to just re-apply the configurations from the original setup, particularly as I use the excellent etckeeper to manage the /etc configurations in Mercurial, so it's quite easy to see which changes are relevant.
Linux Mint 12 released Posted Nov 30, 2011 22:51 UTC (Wed) by tdwebste (guest, #18154) [Link]
What about all your config files in /etc or do you just use your computer for browser the web and editing docs?
I use a separate home partition, which makes upgrades like mint similar.
I user etckeeper. It helps, but it is definitely more work than debian style upgrading. I wish debian upgrades config file upgrades had some intelligence to know when a release results in an incompatible config file changes. Otherwise just do a 3way of the original new and patched config file.
With mint backup home directory method, I need to do the same thing using my etckeeper repository of etc config files. I am modifying my etc keeper repository to include the /var/lib/dpkg as sub module.
I rarely change the gnome settings, but when I do I have had mixed luck with upgrades.
---------
If you have heavily customized your systems how can you LIKE Mint's method of backup your home and wipe EVERYTHING!!!
Linux Mint 12 released Posted Dec 1, 2011 23:16 UTC (Thu) by Cato (subscriber, #7643) [Link]
I use etckeeper, as mentioned, and have done this sort of upgrade between Ubuntu releases in some cases. I try to only upgrade infrequently, but I prefer to incrementally migrate config files, which makes it easy to see if something has just broken - with the Debian/Ubuntu dist-upgrade model, all of your packages are upgraded, and in some cases the config breaks but you may have no idea which package is causing. Xkb was an example here going from 8.04 LTS to 10.04 LTS.
Linux Mint 12 released Posted Dec 7, 2011 0:19 UTC (Wed) by jospoortvliet (subscriber, #33164) [Link]
It gets even better if you use btrfs and it's snapshot functionality to do upgrades... Esp if snapshots are integrated in your package manager like openSUSE's zypper, providing rollback etc :D
Linux Mint 12 released Posted Dec 7, 2011 10:23 UTC (Wed) by Cato (subscriber, #7643) [Link]
Good idea, I know that NexentaOS does something similar with APT dist-upgrades using ZFS snapshots.
However, btrfs really isn't ready for prime time (still no fsck although I know it's being worked on) so I'd suggest a less bleeding-edge model.
Although LVM snapshots are rather slow and can be buggy (see the snapshots section of http://serverfault.com/questions/279571/lvm-dangers-and-c... ) they can be used with care in a similar way during an upgrade. You need to allocate a large chunk of the original FS size to the snapshot particularly for a major upgrade - 50 to 100% is probably advisable. LVM snapshots are not as nice as btrfs snapshots of course, so I look forward to btrfs becoming fully baked.
|
|||||||||||||