netbooks:origins success: small is cute focus shift challenges of first models rule changes attempt to restore old rules: xp what we've seen on a netbook: - ad hoc system: eee and linpus something linux with an atempt of a completely different interface, closed and 0 talk with upstreams -> fail - the bogstandard appeal: xp -> success in the west: an habit is stronger than real benefits - It seems a game over but we shouldn't have to get fooled on what we can find in a random shopping mall in North America and Europe. Other countries starting from a blank sheet can simply skip the legacy and this will funningly put them in an advantage position compared to Europe and North America. - moblin: interesting approach, not totally unlike ours: an harware accelerated canvas as the workspace with a mix of existing and ad hoc GNOME applictions on top, we have the advantage of an already good library of components to put in this said canvas - ChromeOS: they're right 90% of the use of a netbook is probably web browsing - the remaining 10% is sometimes vital - a good share of what is done with the browser is a suboptimal way What we can learn from this: - things like CromeOS express a change in how computers are used, but they don't totally get the point: - in the 80's and 90's computers were used as workhorses, mainly to produce content, with the main exception as gaming, but i both cases isolated each other. - Now we are seeing a shift in how computers are used: while the typical workplace computer is still used mainly for content creation and administrative task, computers at home are using mainly to consume content, this is mainly due to the pervasiveness of the Internet. - This has been mistaken for a transition from rich client aplications to a kind of an universal thin client that is the web browser, and here we have ChromeOS. -This is what is actually happening but is just the effect, not what is really happening. -What we really need is a way to efficiently pull the content we need and visualize it in the best way possible - using a browser, optimized to visualize "pages" to watch a video really isn't that great - Example: a reason for the success of the iPhone is the myriad of applications that let to use some web service in a more convenient way than the broser - So, the internet is not the interwebs Plasma Netbook Shell - In Plasma we never made assumptions on what the formfactor would have been - The first thing that was implemented with it was a desktop shell, because this is what was needed at the moment, but we carefully tried to avoid every reference to the "Desktop" in the plasma library - As soon they surfaced it immediately was obvious that the scale of new mobile devices, from really smart phones such as the n900 to netbooks, even to stuff that still isn't surfaced in the market were the best occasion for the widespread of Linux and hopefully of KDE/Qt technologies - The first logical step is to go in the direction of the less "alien" device compared to what uses to run KDE: the netbook. - The development of Plasma Netbook Shell is done with those firm targets in mind. - As much code reuse as possible, few specific code, only where is needed - think how its components could scale up (desktop) or scale down (smaller mobile devices) Components - Search and Launch In today Desktp computers we see that the usual applications menus really are no longer up to their task: the amount of possible applications and options is simply too much to be possible to display it to the user. - In the desktop this remains an open problem - In the netbook we start from a blank canvas, so how can we do? - Menus gives too much interface to the user, this is really not necessary - The user opens a menu because he want to perform a certain task, the easiest way is to just tell the computer what he wants to do: a query interface is what works best for the internet - in this way we go behyond "launching applications" but we can do every sort of query (eg wikipedia runner) - mention to openbossa research on queryes, so we made KRunner the most prominent thing instead of being a semi hidden feature (icons for defult queries are revealing themselves to be somewhat counterproductive) - Newspaper, or Widgets Pages - different types of content: - content that requires full attention (e.g. reading an article, watching a movie) - content that requires only a very quick glimpse (e.g weather, microblogging) - content that from a quick overview of it is possible to decide if it's important enough to give full attention (e.g. news feeds, emails) - The newspaper activity is built to offer a very efficient and qick way to access the latter two types of content. - Sometimes content to be valuable must be absolutely immediate and arrive in the least time possible - It will be possible to receive quick updates as soon as the device is turned on, being a significative shortcut compared to the burden of opening a web browser. An activity like posting a microblog entry needs to be immediate, if you think about it a second time you maybe won't post at all, and the time taken to load a web page is simply too much, even if the browser is always on, like in ChromeOS. It will be pretty efficient in presenting you a quick overview of activities that could be important enough to give to it your full attention: for instance a quick overview of your pending appontments, unread mails and news feeds, with an easy and immediate way to launch the full application designed to manage the task in the best possible way, like KMail, akregator or yes, even the web browser sometimes :) Your average Newspaper widget and what means for the future - they all have a similar look and behaviour, we can see a direction of standardization - their main part is a scrolling item view, that diplays the info of what you're iterested in, microblog entrie, news, emails.. - the item view will have scrollbars only when needed, in the newspaper they tend to become long enough to avoid scrollbars-in-scrollbars - most of the area is occupied by content, there is as less chrome as possible - the available real estate is really tiny in most of the possible Plasma shells, in the destop there must be room for many widgets, in the netbook the screen is little and there must be room as well for different widgets in the screen as well. - in a tiny mobile device the widget will become probably full screen, and here the screen will be really tiny (even if screen resolution is rising, inches of glass can't grow) - this kind of widgets are a perfect candidate for fullscreen smartphone applications: we must work on the touchscreen friendlyness (is already possible do do flick scrolling in the ScrollWidget component used by most of those widgets) In the future we will see plasma aplications colonizing many different and eterogeneous kind of devices, the key difference compared to the past is that it won't be necessary anymore having different software stacks for them and the work given to make it better with handheld devices will directly benefit all the other targets, like the desktop and the netbook.