If it's all about speed and lightness they might want to drop firefox for chrome. The difference is very noticeable on an Atom 1.6gighz, and in the upcoming ARM powered devices it will be too.
The screenshot ( http://www.markshuttleworth.com/wp-content/uploads/2010/05/L... ) does use Chromium. What I find remarkable about this "light" version of Ubuntu, though, is that it seems to use up more screen real estate for panels than normal Ubuntu! What the hell?
The left panel has "this should auto-hide" written all over it.
As a Ubuntu Netbook Remix user, though, I'm looking at the window titlebar and wondering, what happened to the titlebar-integrates-into-top-panel stuff from UNR? Shouldn't that remain a part of the "conserving precious vertical display space" approach?
Does it have anything to do with betting on the browser they can probably influence more in the future rather than one they essentially have no control of?
I am not sure what you mean. How do they have more control on Firefox and not Chromium? Both are open source, both can be forked and customized in to whatever they want it to be. One is exceptionally faster than the other. Since they are concentrating on speed, shouldn't Chrome/Chromium be the obvious choice?
Firefox already includes a lot of customization points that allow entities like Ubuntu to package a just-right version of Firefox without needing to go whole-hog and fork the entire system (remember Iceweasal?).
I've seen a few tests that claim Firefox beats Chrome on overall page load (and e.g. memory usage). Do you have anything that shows Chrome to be exceptionally faster in the general case, or is it just reducing Javascript execution on the average page from 1 millisecond to 0.5 milliseconds?
I'm not aware of benchmarks either way, but Chrome certainly feels a lot faster. Pages snap up almost instantly and events fire immediately. Firefox, by contrast, comes across as comparatively sluggish.
It pains me to write this, because I've been a happy and devoted Firefox user since the early days of Mozilla, but I am gradually shifting over to using Chrome for straight browsing - though I still rely on Firefox and the unparalleled Firebux addon for troubleshooting web applications.
Some of these instant-on linuxes actually run on a separate ARM chip from the main OS to further differentiate it as the low-capability/low-power/long-battery alternative to the full blown Windows OS, as well as the speed of access element.
Latitude ON runs MontaVista Linux on an ARM-based subprocessor. This so-called MontaVista Montabello Mobile Internet Device Solution provides a customizable, Linux-based Mobile Internet Device (MID) platform the laptop is able to boot almost instantly and view Email, document reader, calendar, contacts and access the Internet
... Dell claims that up to 19 hours of battery life on standard lithium ion batteries can be achieved with this system.
I think they always have done at a certain level, but of course good developers do not make good "joe average" users, and the open source movement has really always been about programmers, not graphic designers/UX/IA people. That's always where companies have added the real value when they build off open source imo. While I admire what people like KDE and Gnome do in terms of effort, the results they return have always been questionable.
When 90% of your end users are old people who rarely use a computer you have to ask them about there needs. You will collect the 'votes' and may discover you have to drop some functionalities. And maybe a slider UI will be better than a number ticker/spinner control.
I watch them make bad decisions all the time; they want to put a box of 60 categories here in the left nav of the page; they want to have a drop-down menu for every department in the hospital. "Asking" ten users how a user-interface should work, and you'll get ten incompatible and stupid answers.
The trick to using users to get good UI is to have a programmer watch them try a prototype (or using paper prototypes). Of course, the programmer has to be tied to a chair and have his mouth duct-tape'd closed so he doesn't try and help, but the fact is when he sees the "inputs", he'll optimize the layout correctly to suit them. Programmers make fine user-interface designers when they're properly motivated.
You shouldn't ask users how the UI should be. You have to listen to there needs.
Then you have to translate this 'democratic' information into a UI design.
When a user wants a drop-down for every department in the hospital you could translate this into "I want to select a department as easy and quickly as possible".
I don't normally comment on grammar, but you've made this mistake in both of your posts: "there needs" should be "their needs". "There" indicates a place, "their" is a possessive pronoun. It's common to also confuse those two homophones with "they're", a contraction of "they" and "are". Getting those straight makes one's writing more intelligible (and more native-sounding if it's being written by a non-native user of English).
Otherwise, I agree with what you're saying about letting users "design" the UI without actually asking them to design a UI.
I think you should take it one step further and along with listen to their needs, put them in front of a computer, and WATCH what they're doing. Find out what their quirks are, how they expect the software to work and build on that.
Ignorance is not stupidity. I agree that asking users to make UI decisions isn't the best approach, but calling users dumb because they can't use your software is dumb.
It's instructive to compare this with Windows 7 (often comes installed on modern netbooks). Recipe: 1. move default taskbar to the left and 2. use Chrome. The result is pretty good, I think:
It's funny that they are talking about maximizing the vertical pixels while letting title bar, menu bar, url bar, bookmark bar, tab bar, horizontal scroll bar, and status bar eating up 1/3 of vertical pixels of the primary web browser window on the home page photo at <http://www.canonical.com/products/unity>.
Seriously, Open Source marketing should be improved in this level of detail. Yes netbooks have shallow screens, then why wasting 145 precious vertical pixels to show non-content controls on a web browser window only 428 pixels high?
Some naive ways to improve the impression:
* Make the browser window taller, preferably taking all available vertical space. Present more web content when you really want to improve the web experience. (Right now I can only see the big heading of NYT, not really interesting.)
* Bundle one of those Hide Menu plugin to kill menu bar. How often do you really need it? And when you do need it, just click one button on the url bar to summon it.
* Choose a page that is slightly narrower or widen the window a little bit (you have the horizontal pixels to waste!) so horizontal scroll does not show up.
* Hide the status bar. Most users don't look at it anyway.
* If there is only one tab, don't show the tab bar.
Or better yet, replacing Firefox with Chrome solves it once and for all.
On my netbook with Ubuntu I wanted to move the top gnome panel to the right side so I'd have more vertical space; apparently this is impossible to do cleanly.
We have wide screens now, and a lack of vertical space. The solution seems obvious to me; don't take up vertical space with desktop UI and move it to the side instead.
I love the gnome-shell, but I'd love it more if I could get those top pixels back. On a netbook, every little bit counts.
>On my netbook with Ubuntu I wanted to move the top gnome panel to the right side so I'd have more vertical space; apparently this is impossible to do cleanly.
I tried this with the past four versions of Ubuntu (8.10, 9.04, 9.10 and 10.04) on my Aspire One. In the first three cases I went back to a top or bottom panel because the side panel was ugly and unusable. With 10.04, a right-side panel is actually decent for the first time. It's not great - e.g. the shutdown applet is still too wide and doesn't flip vertically - but it's decent.
"It’s about how fast you have a running system that is responsive to the needs of the user." - This is the key.
Make it look nice with excellent fonts and i am sold. I know this is not designed for a workstation - but for standard browsing, checking emails and general usage , this could be it.
Dock on the left, title bar and sys tray at the top, window controls on the left side... this vaguely reminds me of some other OS, and I don't think that's at all a bad thing. People like OS X's UI. It makes sense to go with what works.
If they wanted to save more vertical screen real estate, Ubuntu could take another hint from Apple and enable Global Menu Bar by default ( http://code.google.com/p/gnome2-globalmenu/ )
They are doing this. "Third, we will make the top panel smarter. We’ve already talked about adopting a single global menu, which would be rendered by the panel in this case."
My OS X dock was always on the left, but the default positon was at the bottom. Later design changes actually re-inforced this decision as I believe the 3-D effect didn't work vertically.
Does anybody that has used both have an idea of how this compares to the Ubuntu Notebook Remix default UI?
I've been using it on a small acer netbook with a 3G card in it and it has worked pretty well in that configuration for me, I'm curious how they compare but I'm loathe to install stuff on it without some idea of what I'm getting in to because it is a machine I rely on in emergencies (on the road, server down).
Looking at that Ubuntu Light example image, they could still move the window title and buttons to bar at the top, next to the search field, to save more vertical space.
In the other hand, if they want to make it more usable for touch users, they'll need to enlarge the main toolbar and all those tiny indicators (battery, volume etc.) at the lost of some vertical space.
Yes, the top of that screenshot looks messy. If they are moving in this direction, they could also get rid of the top panel. Or at least redesign it.
I have been working in a similar desktop on my wide-screen laptop for nearly a year now. I feel much more productive. It took me some time and skills to configure it, but the tools are already available. I use Openbox+Tint2+Conky and some other lightweight utilities. Also Tree Style Tab extension for Firefox:
https://addons.mozilla.org/en-US/firefox/addon/5890
Since they are talking about getting users to the browser faster, they may as well tweak the browser to work better in this environment.
It is more natural when window/tab hierarchy flows from left to right, than from left and then from top.
Now that I've acclimated to Chrome on Windows, I don't think I could give up having the tabs on the very top of the screen. Me and my man Fitts love having the browser tabs at the top and the open applications at the bottom. It's really fast to switch tabs using the mouse that way.
To get the fonts I like (Mac OS X style) I turn hinting off completely. Ubuntu seems to be moving closer to this as the default as time goes by, probably a smart move as there will be less and less people used to old-skool Windows font rendering.
Yes, the released it in Mac OS X 10.2, one year after Windows XP had cleartype, but they didn't hide the feature away so that only experts could find it and so it was widely used.
But, sub-pixel rendering is just a diversion from the real differences between the two systems, one fits to the grid via hints, the other anti-aliases. I believe both iPhone and Windows Phone are now using this system, and without bothering with sub-pixel rendering.
Sorry I misread that. But I'm fairly certain that Apple's method positioned text to less than a pixel anyway, just as a side effect of their general method. It's the Microsoft school of type rendering that tries to grid fit and therefore is excited about having a finer grid of (sub-)pixels to fit to.
I can't Google up a definitive answer but this document comparing Adobe type rendering against Cleartype seems to suggest that what I say above is true.
Microsoft and Adobe: Sub-pixel Positioning and Kerning:
It's fairly obvious that Apple use the same basic system as Adobe. This renders text on a grid that is far higher resolution than the screen then resamples it down. Anything that doesn't round to a pixel is pretty much by definition sub-pixel, and the colour fringes you see when zoomed in clearly show that they're using the individual sub-pixels independantly. I'm not really sure how you could achieve this without what Microsoft call "sub-pixel positioning". It only appears to be a feature with a distinct name because they left it out in their first few versions of cleartype.
Old-skool referred to a) the fact that it's subjectively more pleasing to those who are used to aliased, bitmap fonts hence the big demand for "terminal" or programmer fonts in that style, even today and b) the fact the rising DPIs have been on the cards for a long time and are now arriving, making cleartype technology, and the aesthetic philosophy behind it, basically redundant.
I don't really believe it's more technically advanced, just a different style of rendering that worked with a different set of technological and user-defined constraints.
(For what it's worth, when I looked up what sub-pixel positioning was, it was listed next to Y-axis anti-aliasing with equal billing. Again this is just built into the Adobe and Apple approach from the start: http://msdn.microsoft.com/en-us/library/ms749295.aspx#y-dire...)
Do all of those render identically on Mac OS X, or are the colour fringes different for different i's? They render differently on Windows with DirectWrite. If they render differently then you know you support sub-pixel positioning. (I can't claim the inverse, of course.)
> b) the fact the rising DPIs have been on the cards for a long time and are now arriving, making cleartype technology, and the aesthetic philosophy behind it, basically redundant.
How is this so? As DPI goes to infinity, both ClearType and Quartz will look exactly like they would look in print.
If it's all about speed and lightness they might want to drop firefox for chrome. The difference is very noticeable on an Atom 1.6gighz, and in the upcoming ARM powered devices it will be too.