An API is just an interface - there's nothing inherently open or closed about it. The code behind the API might be either open source or closed source, and in this case we appear to be talking about closed source.
Not only is open vs closed a meaningless adverb to apply to an API (the closest concept might be public vs private - e.g. hidden Winodows API's), but the trend the article is attempting to discuss is the growth of cloud based platforms - i.e services made available via web services (SOAP,.NET, etc).. it's got nothing to do with open or closed source.
Maybe the ultimate example of a "platformization" is Amazon - whose entire company is based on their own publicaly available (at a cost) web services, and this is where they make a lot of their money - from other companies building their web-based stores and services on top of the amazon platforms/services.
RISC isn't an instruction set - it's a design strategy.
RISC = reduced instruction set computing CISC = complex instruction set computing
The idea of RISC (have a small highly regular/orthogal instruction set) goes back to the early days of computing when chip design and compiler design wasn't what it is today. The idea was that a small simple instruction would correspond to a simpler chip design that could be clocked faster than a CISC design while at the same time being easier to compile optimized code.
Nowadays advances in chip design and compiler code generation/optimization have essentially undone these benefits of RISC, but the remaining benefits are that RISC chips have small die sizes hence low power requirements, high production yields and low cost, and these are the real reasons ARM is so successful, not the fact that the instruction set is "better".
Matching a search with useful information is kind of what google does best. For voice recognition, they've been doing voice-search on Android for a long time, plus their now defunct goog-411 and that's a lot of voice recognition experience.
Siri/Majel is really just a UI layer on top of those two things.
I have to disagree.
Certainly it's the
goal
of Google to decipher intent from search queries and deliver matching content, but I've yet to see any examples where what Google currently does really goes beyond basic keyword matching. Do you have any examples where Google is interpreting what you mean as opposed to what you actually typed (dumb keyword matching)?
The real value of Siri is in it's AI - it's ability to determine what you want from what you say (including prior context), and then of course act upon that via it's interfaces to real world systems and applications. This AI part is what Google will have a hard time replicating any time soon.
1 D "bisect = two equal halves. wow - tough" 2 H (1.5x)^3/x^3 = 1.5^3 3 C (got 4 sides, and it not a rectangle...) 4 628 cm^2 - they provide the formula for you! 5 56.3' (atan(12/8)\
Really, it's *pathetic* if a school board member can't solve "problems" like these.
Well, Qt is relatively hardware dependent - both due to assmbler used for low level stuff and needing OpenGL/VG for acceleration. They've done a lot of work in making Qt more portable, but not surprisingly it still takes some work to get it up and running and optimized on a new platform. Don't forget too that the Pi is ARM based.
Not sure... you can drag (copy) applications from the menu to the top menu bar as well as the desktop, but if you click on the menu bar there's no context menu to add custom lauchers.
Additionally, the menu bar functionality seems very limited. The icons there are (by default) tiny and can't be placed or spaced out. The only form of rearragement that seems to work is dragging icons to the end of the existing icons, which makes for rather laborous rearranging.
IMO it's a bit of a limited menu bar to launch mint 12 with, but hopefully it'll improve with time to bring back something closer to GNOME 2 in functionality.
OK, well that kind of makes sense of why the important apps are hided away, but not of the awful placement of the "Other" menu. I assume the menu can be reconfigured if I really want to, but I'm a recent Ubuntu to mint convert and havn't bothered to look into it yet.
I tried the Mint Xfce rolling edition briefly, but there seems to be an annoying bug where the window manager dies (or can accidently be killed during normal use) leaving you with unmovable borderless windows... You can recover by lauching a new window manager (so I've read) but it's a PITA so I switched back to the GNOME edition for stability.
I'm not so sure, and I wouldn't really characterize this as a move towards thin clients even if things are becoming more cloud-centric.
This is really about the computer market maturing and computers becoming consumer devices and converging on what the average consumer wants which is to consume (media and app content) and be entertained. The only folks who really want computers vs computer based appliances are hard code geeks and we're a tiny minority.
It's only a recent thing that you can pack enough technology into a small portable package to make a really compelling "internet appliance", but now that it's happened, it's hard to see what would make people want to go back to clunky difficult to maintain desk-bound computers.
Does anyone know why the default menus are so oddly organized - such as the catch-all "Other" sub-menu being in the middle of the menu, and containing important stuff like the Update Manager and Synaptic Package Manager?
Is this menu organization something Mint is inheriting from GNOME 3? In Mint 11 the system stuff was in some System menu where you more expect to find it.
I was expecting the menu to be cleaned up during the Mint 12 beta, but it's still there know in what appears to be the release version.
Well, the desktop in general, Windows included, is rapidly becoming inconsequential other than for business use. The non-business computer market is rapidly moving to smartphones, tablets and laptops - all smaller screen devices where a traditional screen-real-estate-hungry user interface isn't the best option. This is the market that Ubuntu is obviously targeting with Unity, and Android and Windows also appear to be moving in the same direction - Windows 8 and Ice Cream Sandwich UIs both are geared towards small-screen appliance-type use.
But, that said, there's always going to be a demand for a more traditional general purpose compute devices, for development work if nothing else, and for that use Linux always has been a great option, and only getting better with age, even if the path it's taking is a little uncertain. RIP Ubuntu. Long live Linux Mint!
There's no need to make an copy of class (not instance) specific information per instance, but that doesn't change the fact that the constructor is the cleanest place to do it, otherwise you'd need to explicity initialize each type of class before use.
The overhead of in-constructor global initialization is minimal - just test/set a static member initialized flag.
All you need for a meta object system/etc is an appropriate QObject base class to replace the moc & Q_OBJECT preprocessor kluges. Wanting to avoid standard C++ features like RTTI and dynamic_cast in favor of Qt-specific hacks is a horrible case of no-invented-here syndrome. Just stick to the standard language facilities, please.
One obvious way to cleanly implement introspection without preprocessor hackery would be to have each object's constructor register it's method in an appropriate way with the proposed QObject base class.
When Qt was first implemented it was *perhaps* excusable, given the state of template, STL, etc support in target compilers, to use preprocessor hackery, but for many years now that's been an invalid excuse. Sure it would be considerable work to change Qt into a truly native C++ library and ditch moc, etc, but there's no valid *technical* reason why it couldn't be done.
If you're in Arizona then Grand Canyon is a must (just unbelievably huge), but another really interesting place to visit is Petrified Forest National Park - full of fossilized (petrified) trees laying as they fell.
Try Linux Mint Debian Xfce (rolling update) edition. Just tried it myself this weekend (looking to get away from Ubuntu because of Unity) and first impressions are very good.
Xfce is basically like Windows XP - just a bottom menu/task bar, and the rest of the screen free to put icons on.
About the only other really low latency service (and by this I mean service where 20ms becomes significant) is supercomputing
Well, there's program trading. There was a story a couple of days ago of a new $300M transatlantic cable being laid whose sole purpose is to reduce transit times (latency) by 6ms from the current 60ms. The consumers for this are hedge funds/etc doing program trading - the article said that a large hedge fund might make $100M/yr extra from a 1ms data advantage.
I checked Google news maybe a minute after I felt it (in NJ) and was impressed to see they had it already. Not sure hos fast these things travel hence how long after it hit in VA.
I'm in Pearl River, NY - right beside northern NJ - and we felt it on the 6th floor of our building. I was sitting at my desk and first thought someone was behind me pushing my chair as a joke.
Once you've represented the lawn area as a tessellation of (slightly overlapping) lawnmower-sized patches, then isn't this just the traveling salesman problem - visit all patches with the least distance traveled?
This is a classic NP problem... if the problem size (N) is too large to fully evaluate (in this case 6 acres = 29,000 square yards, tractor area = 1 square yard, so N = 29,000 which is rather large for this type of problem), then heuristics are you're friend.
The optimal solution, which would only apply for a circular lawn, is obviously a spiral pattern. For an irregular shape lawn one obvious heuristic would be to decompose the lawn area into a set of various sized circular blobs and do each of these in an expanding spiral pattern, then onto the next.
A similar heuristic would be to start by spiraling inwards around the entire lawn, and "recurse" into smaller areas when they (via having narrow "neck" entrances) are about to be cut off from the main spiral - specifically when the neck has been reduced to two tractor widths wide (one path in, one path out). In fact, this may well be the optimum strategy, particularly as it takes advantage of the specific problem topology rather than being a generic traveling salesman heuristic.
Please send cash to SpinyNorman c/o Slashdot if this makes you money!
The changes appear due to Larry Page (one of the founders) having this year taken over from Eric Schmidt as CEO. Obviously he'd not have done it if he intended to run the company the same way, and he's made a number of moves in the direction of more focus.
A decent attention span is obviously a prerequisit for most jobs, but it's no substitute for experience.
Someone with raw talent has the potential to be great, but it's only experience that will make them so (assuming their raw talent includes the ability to learn from experience).
Remember - the best 40yr old programmers have raw talent plus 20years of experience. The best 20yr old progammers *only* have the raw talent.
First off, 60-70% isn't crippled - you'd only notice the difference for an app that was stretching the graphics card to the max (i.e. games), and we're talking Linux here... not exacly famous as a high-perf gaming platform regardless of what drives you're running.
Secondly, for a complex piece of software like this you don't just beat on the competition at first attempt - it's an ongoing process of learning and experimentation. 60-70% is totallty usable today, and tomorrow it'll be faster...
Finally, I don't share your faith that the proprietary driver is optimally designed and coded and is therefore squeezing max ("100%") performance out of the hardware. Who knows how efficient it is, especially since AMD's doesn't have any incentive to keep improving older drivers - there going to always be moving on to the next one.
Slashdot Mirror
An API is just an interface - there's nothing inherently open or closed about it. The code behind the API might be either open source or closed source, and in this case we appear to be talking about closed source.
Not only is open vs closed a meaningless adverb to apply to an API (the closest concept might be public vs private - e.g. hidden Winodows API's), but the trend the article is attempting to discuss is the growth of cloud based platforms - i.e services made available via web services (SOAP, .NET, etc).. it's got nothing to do with open or closed source.
Maybe the ultimate example of a "platformization" is Amazon - whose entire company is based on their own publicaly available (at a cost) web services, and this is where they make a lot of their money - from other companies building their web-based stores and services on top of the amazon platforms/services.
RISC isn't an instruction set - it's a design strategy.
RISC = reduced instruction set computing
CISC = complex instruction set computing
The idea of RISC (have a small highly regular/orthogal instruction set) goes back to the early days of computing when chip design and compiler design wasn't what it is today. The idea was that a small simple instruction would correspond to a simpler chip design that could be clocked faster than a CISC design while at the same time being easier to compile optimized code.
Nowadays advances in chip design and compiler code generation/optimization have essentially undone these benefits of RISC, but the remaining benefits are that RISC chips have small die sizes hence low power requirements, high production yields and low cost, and these are the real reasons ARM is so successful, not the fact that the instruction set is "better".
More than $10K at current prices.
The base model has 160mi range then you pay $10K for 70mi more (230mi range), or $20K for 140mi more (300mi).
So... at $10K per 70mi range of battery even the base 160mi range battery is $20K, and the 300mi one is $40K!
That's a damn expensive battery!
Matching a search with useful information is kind of what google does best. For voice recognition, they've been doing voice-search on Android for a long time, plus their now defunct goog-411 and that's a lot of voice recognition experience.
Siri/Majel is really just a UI layer on top of those two things.
I have to disagree.
Certainly it's the
of Google to decipher intent from search queries and deliver matching content, but I've yet to see any examples where what Google currently does really goes beyond basic keyword matching. Do you have any examples where Google is interpreting what you mean as opposed to what you actually typed (dumb keyword matching)?
The real value of Siri is in it's AI - it's ability to determine what you want from what you say (including prior context), and then of course act upon that via it's interfaces to real world systems and applications. This AI part is what Google will have a hard time replicating any time soon.
You'd have to me a moron not to get those.
First five:
1 D "bisect = two equal halves. wow - tough" ...)
2 H (1.5x)^3/x^3 = 1.5^3
3 C (got 4 sides, and it not a rectangle
4 628 cm^2 - they provide the formula for you!
5 56.3' (atan(12/8)\
Really, it's *pathetic* if a school board member can't solve "problems" like these.
Really? Where did you read this? Was there a massive price cut?
It's pretty surprising since the most recent TOTAL sales number I read was 10 million, so 0.7 million in 1 day is staggering if true.
Thanks very much for taking the time for the detailed reply - very useful!
Well, Qt is relatively hardware dependent - both due to assmbler used for low level stuff and needing OpenGL/VG for acceleration. They've done a lot of work in making Qt more portable, but not surprisingly it still takes some work to get it up and running and optimized on a new platform. Don't forget too that the Pi is ARM based.
Not sure... you can drag (copy) applications from the menu to the top menu bar as well as the desktop, but if you click on the menu bar there's no context menu to add custom lauchers.
Additionally, the menu bar functionality seems very limited. The icons there are (by default) tiny and can't be placed or spaced out. The only form of rearragement that seems to work is dragging icons to the end of the existing icons, which makes for rather laborous rearranging.
IMO it's a bit of a limited menu bar to launch mint 12 with, but hopefully it'll improve with time to bring back something closer to GNOME 2 in functionality.
OK, well that kind of makes sense of why the important apps are hided away, but not of the awful placement of the "Other" menu. I assume the menu can be reconfigured if I really want to, but I'm a recent Ubuntu to mint convert and havn't bothered to look into it yet.
I tried the Mint Xfce rolling edition briefly, but there seems to be an annoying bug where the window manager dies (or can accidently be killed during normal use) leaving you with unmovable borderless windows... You can recover by lauching a new window manager (so I've read) but it's a PITA so I switched back to the GNOME edition for stability.
I'm not so sure, and I wouldn't really characterize this as a move towards thin clients even if things are becoming more cloud-centric.
This is really about the computer market maturing and computers becoming consumer devices and converging on what the average consumer wants which is to consume (media and app content) and be entertained. The only folks who really want computers vs computer based appliances are hard code geeks and we're a tiny minority.
It's only a recent thing that you can pack enough technology into a small portable package to make a really compelling "internet appliance", but now that it's happened, it's hard to see what would make people want to go back to clunky difficult to maintain desk-bound computers.
Does anyone know why the default menus are so oddly organized - such as the catch-all "Other" sub-menu being in the middle of the menu, and containing important stuff like the Update Manager and Synaptic Package Manager?
Is this menu organization something Mint is inheriting from GNOME 3? In Mint 11 the system stuff was in some System menu where you more expect to find it.
I was expecting the menu to be cleaned up during the Mint 12 beta, but it's still there know in what appears to be the release version.
Well, the desktop in general, Windows included, is rapidly becoming inconsequential other than for business use. The non-business computer market is rapidly moving to smartphones, tablets and laptops - all smaller screen devices where a traditional screen-real-estate-hungry user interface isn't the best option. This is the market that Ubuntu is obviously targeting with Unity, and Android and Windows also appear to be moving in the same direction - Windows 8 and Ice Cream Sandwich UIs both are geared towards small-screen appliance-type use.
But, that said, there's always going to be a demand for a more traditional general purpose compute devices, for development work if nothing else, and for that use Linux always has been a great option, and only getting better with age, even if the path it's taking is a little uncertain. RIP Ubuntu. Long live Linux Mint!
Huh?
There's no need to make an copy of class (not instance) specific information per instance, but that doesn't change the fact that the constructor is the cleanest place to do it, otherwise you'd need to explicity initialize each type of class before use.
The overhead of in-constructor global initialization is minimal - just test/set a static member initialized flag.
All you need for a meta object system/etc is an appropriate QObject base class to replace the moc & Q_OBJECT preprocessor kluges. Wanting to avoid standard C++ features like RTTI and dynamic_cast in favor of Qt-specific hacks is a horrible case of no-invented-here syndrome. Just stick to the standard language facilities, please.
One obvious way to cleanly implement introspection without preprocessor hackery would be to have each object's constructor register it's method in an appropriate way with the proposed QObject base class.
When Qt was first implemented it was *perhaps* excusable, given the state of template, STL, etc support in target compilers, to use preprocessor hackery, but for many years now that's been an invalid excuse. Sure it would be considerable work to change Qt into a truly native C++ library and ditch moc, etc, but there's no valid *technical* reason why it couldn't be done.
If you're in Arizona then Grand Canyon is a must (just unbelievably huge), but another really interesting place to visit is Petrified Forest National Park - full of fossilized (petrified) trees laying as they fell.
http://www.petrified.forest.national-park.com/
Try Linux Mint Debian Xfce (rolling update) edition. Just tried it myself this weekend (looking to get away from Ubuntu because of Unity) and first impressions are very good.
Xfce is basically like Windows XP - just a bottom menu/task bar, and the rest of the screen free to put icons on.
About the only other really low latency service (and by this I mean service where 20ms becomes significant) is supercomputing
Well, there's program trading. There was a story a couple of days ago of a new $300M transatlantic cable being laid whose sole purpose is to reduce transit times (latency) by 6ms from the current 60ms. The consumers for this are hedge funds/etc doing program trading - the article said that a large hedge fund might make $100M/yr extra from a 1ms data advantage.
Hopefully the US economy will recover now that the geeks can get back to work and stop browsing /.
I checked Google news maybe a minute after I felt it (in NJ) and was impressed to see they had it already. Not sure hos fast these things travel hence how long after it hit in VA.
I'm in Pearl River, NY - right beside northern NJ - and we felt it on the 6th floor of our building. I was sitting at my desk and first thought someone was behind me pushing my chair as a joke.
Once you've represented the lawn area as a tessellation of (slightly overlapping) lawnmower-sized patches, then isn't this just the traveling salesman problem - visit all patches with the least distance traveled?
This is a classic NP problem... if the problem size (N) is too large to fully evaluate (in this case 6 acres = 29,000 square yards, tractor area = 1 square yard, so N = 29,000 which is rather large for this type of problem), then heuristics are you're friend.
The optimal solution, which would only apply for a circular lawn, is obviously a spiral pattern. For an irregular shape lawn one obvious heuristic would be to decompose the lawn area into a set of various sized circular blobs and do each of these in an expanding spiral pattern, then onto the next.
A similar heuristic would be to start by spiraling inwards around the entire lawn, and "recurse" into smaller areas when they (via having narrow "neck" entrances) are about to be cut off from the main spiral - specifically when the neck has been reduced to two tractor widths wide (one path in, one path out). In fact, this may well be the optimum strategy, particularly as it takes advantage of the specific problem topology rather than being a generic traveling salesman heuristic.
Please send cash to SpinyNorman c/o Slashdot if this makes you money!
The changes appear due to Larry Page (one of the founders) having this year taken over from Eric Schmidt as CEO. Obviously he'd not have done it if he intended to run the company the same way, and he's made a number of moves in the direction of more focus.
Huh?
A decent attention span is obviously a prerequisit for most jobs, but it's no substitute for experience.
Someone with raw talent has the potential to be great, but it's only experience that will make them so (assuming their raw talent includes the ability to learn from experience).
Remember - the best 40yr old programmers have raw talent plus 20years of experience. The best 20yr old progammers *only* have the raw talent.
First off, 60-70% isn't crippled - you'd only notice the difference for an app that was stretching the graphics card to the max (i.e. games), and we're talking Linux here... not exacly famous as a high-perf gaming platform regardless of what drives you're running.
Secondly, for a complex piece of software like this you don't just beat on the competition at first attempt - it's an ongoing process of learning and experimentation. 60-70% is totallty usable today, and tomorrow it'll be faster...
Finally, I don't share your faith that the proprietary driver is optimally designed and coded and is therefore squeezing max ("100%") performance out of the hardware. Who knows how efficient it is, especially since AMD's doesn't have any incentive to keep improving older drivers - there going to always be moving on to the next one.