Heck I earn my living currently with javascript on professional framework level but I still think the language is not cool. Namely the lack of namespaces and the lack of real inheritance are my biggest problems. And before someone comes you can write those with maps and prototype inheritance just think of this: a) Maps are slower than real namespaces and to get a better performance you have to add lookup indices yourself and even then you are not even close to the real thing.
b) Given that there is no real inheritance every framework who really wants to support this now rolls its own solution with its own syntax You wont get conflicts that way thanks to the unification on prototype level, but you cannot mix classes either, and to the worse documentation tools like jsdoc now have to be tailored towards every framework separately to get the meta information for namespaces and classes out of the code or what the runtime makes of it.
Sorry but I dont drink the coolaid here, I see it from a day to day problem point of view of having to deal with on industrial level. But it probably is cool to wank about the lack of this and to roll your own solution mentally on an academic point of view.
Since gnome shell nowadays is just nothing more than an extended window manager:-( Gone are the days of a desktop. I just wonder what they thought about sacrificing all the desktop space. I am glad that there are alternatives to Gnome. It was nice knowing you but Gnome 3 in many areas is a step in the wrong direction in some in the right.
Actually the japo rpgs never reached ultima 6 level gameplaywise. They were mostly stuck on Ultima 3 level and then added some story elements of Ultima 4. But I would not say Ultima really was that influental the games were designed like that in the 80s top down view was common random encounters were induced by the pen and paper rpgs and being on rails was a design element which was caused by the limitation of the hardware. Ultima always tried to break away from linearity and succeeded while the japanese mostly ramped up the graphics and cut scenes and did not change the core elements mostly inherited from pen and paper and computer limitations. The usual stick which worked until the death mentality of the japanes, I guess. Funny thing is when Final Fantasy 7 came out and people from the console crowd said it was better than Ultima, I gave the game a try, and felt like being catapulted back in time gameplaywise 10 years, never could find out why the consolers were praising the game so loudly. Now that I know more about the console scene I basically knew, they did simply not know better that there were other rpg designs. Not to say FF7 was a bad game, but it was not the style I preferred and was used to at that time (freeform heavy story driven sandbox)
Actually id rather have another decent Ultima on a tablet than yet another jrpg with random encounters dice roll battles and endless statistics and cutscenes.
Actually the problem with japanese development is mostly that once a formula works for them they stick with it in every detail until the steam has run out, exception to the rule is Nintendo. Their luck was mostly that console gamers until recently were very reliable on not being too much angry about constant rehashes that might have changed by a huge influx of former pc gamers to the modern consoles and a so called mass market. The entire game design of a japo rpg is basically stuck in the mid 80s when the western crpgs split to some degree with ultima 4 5 and 6 away from the pen and paper dice roll random encounter scheme. The japanese stuck with it and never changed it just added more story visuals and being on rails while western crpgs quickly removed the rails wherevery possible (Bioware being the exception) You can see this to some degree in many japanese gaming series and genres. From time to time they try a completely different formula which 90% fails and then they stick back to their copy cat road and 10% it works so they follow that road then.
Well the last tomb raider was quite good and a good break from the 20th rehash of the same the same goes for the next tomb raider, there is a new deus ex in the line as well. Ah yes I would love to see another Thief title. But for me both series are not really the same without Warren Spector and Doug Church at the helm.
First of all you still can sell ebooks over the web and there are enough ebook readers which allow direct drag and drop into the app. Apple allows also public epubs to be sideloaded into ibooks by synching. Also you can always sell a PDF. There is one thing however, if you want to sell your books over your own app nail them down with drm etc... you cannot do it unless you sell over apple. Fair choice you still can sell drmed books the kindle way by doing the billing over an external website.
Well if he uses massive subselects then he either uses a recursive call where postgres has special optimizations (very likely he uses several big non relational non transactional tables which mysql was built for initially), or he uses them over multiple tables where you also can use joins.
Well thats the idea, if you buy oracle you buy also into their consulting and/or training hours. Oracle can perform really well and can host loads of data if you have a person who knows his way around, if not and you are short on time forget about it. Oracle is something along the lines of we want banks big businesses etc... as customers and those do not care about expensive trainings and consulting hours, they never cared about the small time developers.
Yes Hjelfsberg is one of the best language architects/compiler builders worldwide. Turbo Pascal in its infancy was the fastest compiler for any high level language back then and with Delphi he had shown what to do to build a first class development tool (he and his team of course). There was a reason Microsoft hired that guy, you could not get anyone better for building up world class developers tools. Just look at the Microsoft tools before him and now.
Yes, thats one point apple definitely has overlooked they need to push a standardized gaming api for such controllers so that the game makers can hook into. There already are bluetooth gamepads for the iphone but none of them has standardized apis so they have to rely on the mercy of the gamemakers to support them.
Yes thats really weird, the cutscenes of the games usually are better movies than the movies based on video games. I guess this is due to the fact that the game creators care about the stories they want to tell while Hollywood only cares about cashing in on a name and thinks they can shove everything down the throat to their supposed 12 year old target audience. Hollywood has not gotten the clue that the majority of the people who play those games are in between 20 and 35. I have not even remotely cared about the PoP movie once I saw that Bruckheimer was directing it, they probably butchered anything Mechner had written for the movie to death before releasing it. It also was kind of interesting to see how bad the Tomb Raider movies had been, there was a chance to make another indiana jones style movie they blew it big time and those were some of the better adaptions.
Sheesh when was the last time you did some word editing on a phone... the office integration is moot, good file viewers with some editing capabilities are available for all platforms, as for outlook not even windows phone 7 has outlook integration anymore if you need it there are sync programs for every platform. The only thing where blackberry really still has an edge to some degree is the exchange connectivity and that keeps them afloat businesswise.
When I walked into an electronics store 18 months ago and saw a bunch of kiddies standing around the ipad playing games, while the NDS section was full but abandoned. Add to that that as a parent I rather pay 99c for a game then 45$ for a similar shoddy casual game.
I dont have a problem to pay full price but not for casual games nor for the 15th rehash of Mario 2d.
The funny thing is i never really have had a huge problem with dojo. Mastering Dojo was heavens sent for me to explain some non obvious constructs outside of that, the code + a little bit of brain and dojo campus got me around. As for the 20MB that is mostly like you would see a jquery distro with all sources and a load of plugins all unit tests included documentation etc.... You usually isolate your files and end up with a 50-300Kbyte single include source thats it. I have worked with several projects with dojo. Usually it was like that.
The dojo core was very good and always worked, not a single problem. Dijit widgets, there it dependend heavily on how far you would use them, the core widgets usually worked. Data bindings were somewhat problematic in a way that they are hard to use. And some widgets were a little bit buggy, but most of them more on the dojox side. The biggest project I was involved with just used the core and rolled its own widgets.
So it is mostly comparable to jquery, where you have a good core and depending on your luck good or bad extensions. The main difference is the dojo core covers way more ground.
PS: Jquery, I have worked with jquery as well extensively. You can use it nicely from the core, but as soon as you hit the widget area it becomes an utter mess, some widgets are really good but some are utter garbage and you often end up with fixing a myriad of bugs to get the thing going. Trying a pure functional approach and having a non existent real widget api does not help either. The code often ends up in a mess and things which could be isolated with inheritance are exposed fully which makes often widget code convoluted. As for prototype, nice try in language design, but a library which alters the core language underneath it just for the sake of it is possible is a no go, I have seen to many problems caused by this ghetto blaster of a library.
It really depends on your needs, what you do, but jquery for instance has its merits if i need to decorate a site quickly and if I can find a plugin which allows me to fix a problem quickly. I however would never program a bigger fully dhtml/ajax site with it, which does not do any page refreshes but provides you a huge rich client ui.
Actually JS is both, it is rather messy because everything is basically a function and an object or we could name it both is the same. What you dont get is inheritance namespacing etc.. out of the box all you get is access to the prototype function to build such structural patterns upon. The problem by not delivering both simply is, that you have to build your own solutions upon it which has several disadvantages. Namespacing must be built with maps which means speed hit and automatic documentation tools will fail. The same then goes for classes inheritance etc... documentation tools automatically will fail if not custom tailored to the syntax of the framework.
Actually I am working with dojo as we speak, we dont use too much of it just the core the query part and a handful of widgets, and it works out quite well. The core itself can be reduced to 50k of code if you pack everything in with all the dependencies you have about 200k of code. What you talk about the 21 MB is the source all its tests all the documentation and whatever comes with it. As soon as you isolate what you are using you are down to something smaller.
You can compare a full dojo distribution with a full jquery distribution with every plugin on earth written for it. The huge difference is outside of the jquery core you have a huge mess of plugins with different quality, while dojo tries to enforce a certain quality standard by using extensive unit testing on its core and dijit widget. As for the documentation, it could be better, but given the extensiveness it is ok. Not like YUI standards but given all the resources out there good enough to use it. For me the code ultimately is the last reference the books usually are a jump starter.
The jquery core itself is just about 1% of the extensiveness of the dojo core, mainly the query part and a little bit of xhr and events, and that part is well documented, everything else is a myriad of extensions. Some of them well documented some of them utter garbage not even working not a single line of documentation. I would not even closely compare dojo and jquery. Jquery does not even cover the ground the dojo core covers and both have about the same bootup times. Things become slow in dojo as soon as you dont package and use a lot of digit widgets. Which means dojo issues load of xhr requests to fetch additional code in. This can be fixed by packaging everything.
Slashdot Mirror
Heck I earn my living currently with javascript on professional framework level but I still think the language is not cool.
Namely the lack of namespaces and the lack of real inheritance are my biggest problems.
And before someone comes you can write those with maps and prototype inheritance just think of this:
a) Maps are slower than real namespaces and to get a better performance you have to add lookup indices yourself and even then you are not even close to the real thing.
b) Given that there is no real inheritance every framework who really wants to support this now rolls its own solution with its own syntax
You wont get conflicts that way thanks to the unification on prototype level, but you cannot mix classes either, and to the worse documentation tools like jsdoc now have to be tailored towards every framework separately to get the meta information for namespaces and classes out of the code or what the runtime makes of it.
Sorry but I dont drink the coolaid here, I see it from a day to day problem point of view of having to deal with on industrial level.
But it probably is cool to wank about the lack of this and to roll your own solution mentally on an academic point of view.
Sun not really had been pursuing the Workstation business actively the last 10 years before the Oracle merger. It was all about servers.
Since gnome shell nowadays is just nothing more than an extended window manager :-(
Gone are the days of a desktop. I just wonder what they thought about sacrificing all the desktop space.
I am glad that there are alternatives to Gnome. It was nice knowing you but Gnome 3 in many areas is a step in the wrong direction
in some in the right.
Actually the japo rpgs never reached ultima 6 level gameplaywise. They were mostly stuck on Ultima 3 level and then added some story elements of Ultima 4. But I would not say Ultima really was that influental the games were designed like that in the 80s top down view was common random encounters were induced by the pen and paper rpgs and being on rails was a design element which was caused by the limitation of the hardware. Ultima always tried to break away from linearity and succeeded while the japanese mostly ramped up the graphics and cut scenes and did not change the core elements mostly inherited from pen and paper and computer limitations. The usual stick which worked until the death mentality of the japanes, I guess.
Funny thing is when Final Fantasy 7 came out and people from the console crowd said it was better than Ultima, I gave the game a try, and felt like being catapulted back in time gameplaywise 10 years, never could find out why the consolers were praising the game so loudly. Now that I know more about the console scene I basically knew, they did simply not know better that there were other rpg designs.
Not to say FF7 was a bad game, but it was not the style I preferred and was used to at that time (freeform heavy story driven sandbox)
Who really cares i never could get into their games. :-)
I am currently waiting for the unlock of The Witcher 2
Actually id rather have another decent Ultima on a tablet than yet another jrpg with random encounters dice roll battles and endless statistics and cutscenes.
Actually the problem with japanese development is mostly that once a formula works for them they stick with it in every detail until the steam has run out, exception to the rule is Nintendo. Their luck was mostly that console gamers until recently were very reliable on not being too much angry about constant rehashes that might have changed by a huge influx of former pc gamers to the modern consoles and a so called mass market.
The entire game design of a japo rpg is basically stuck in the mid 80s when the western crpgs split to some degree with ultima 4 5 and 6 away from the pen and paper dice roll random encounter scheme. The japanese stuck with it and never changed it just added more story visuals and being on rails while western crpgs quickly removed the rails wherevery possible (Bioware being the exception)
You can see this to some degree in many japanese gaming series and genres. From time to time they try a completely different formula which 90% fails and then they stick back to their copy cat road and 10% it works so they follow that road then.
Well the last tomb raider was quite good and a good break from the 20th rehash of the same the same goes for the next tomb raider, there is a new deus ex in the line as well. Ah yes I would love to see another Thief title. But for me both series are not really the same without Warren Spector and Doug Church at the helm.
First of all you still can sell ebooks over the web and there are enough ebook readers which allow direct drag and drop into the app.
Apple allows also public epubs to be sideloaded into ibooks by synching. Also you can always sell a PDF.
There is one thing however, if you want to sell your books over your own app nail them down with drm etc... you cannot do it unless you sell over apple.
Fair choice you still can sell drmed books the kindle way by doing the billing over an external website.
So it is a fair choice.
This is no proof at all, a one liner on the command line produces such a file.
Well if he uses massive subselects then he either uses a recursive call where postgres has special optimizations (very likely he uses several big non relational non transactional tables which mysql was built for initially), or he uses them over multiple tables where you also can use joins.
It was not Anyonmous, it was John Doe...
All I can say is, use PostgreSQL, if you need professional support, there are various options linked from the PostgreSQL homepage.
Well thats the idea, if you buy oracle you buy also into their consulting and/or training hours. Oracle can perform really well and can host loads of data if you have a person who knows his way around, if not and you are short on time forget about it.
Oracle is something along the lines of we want banks big businesses etc... as customers and those do not care about expensive trainings and consulting hours, they never cared about the small time developers.
Yes Hjelfsberg is one of the best language architects/compiler builders worldwide. Turbo Pascal in its infancy was the fastest compiler for any high level language back then and with Delphi he had shown what to do to build a first class development tool (he and his team of course).
There was a reason Microsoft hired that guy, you could not get anyone better for building up world class developers tools.
Just look at the Microsoft tools before him and now.
http://www.wired.com/gadgetlab/2011/01/stick-on-joypad-upgrades-ipad-gaming/
Yes, thats one point apple definitely has overlooked they need to push a standardized gaming api for such controllers so that the game makers can hook into. There already are bluetooth gamepads for the iphone but none of them has standardized apis so they have to rely on the mercy of the gamemakers to support them.
Yes thats really weird, the cutscenes of the games usually are better movies than the movies based on video games. I guess this is due to the fact that the game creators care about the stories they want to tell while Hollywood only cares about cashing in on a name and thinks they can shove everything down the throat to their supposed 12 year old target audience. Hollywood has not gotten the clue that the majority of the people who play those games are in between 20 and 35.
I have not even remotely cared about the PoP movie once I saw that Bruckheimer was directing it, they probably butchered anything Mechner had written for the movie to death before releasing it.
It also was kind of interesting to see how bad the Tomb Raider movies had been, there was a chance to make another indiana jones style movie they blew it big time and those were some of the better adaptions.
Sheesh when was the last time you did some word editing on a phone... the office integration is moot, good file viewers with some editing capabilities are available for all platforms, as for outlook not even windows phone 7 has outlook integration anymore if you need it there are sync programs for every platform. The only thing where blackberry really still has an edge to some degree is the exchange connectivity and that keeps them afloat businesswise.
When I walked into an electronics store 18 months ago and saw a bunch of kiddies standing around the ipad playing games, while the NDS section was full but abandoned.
Add to that that as a parent I rather pay 99c for a game then 45$ for a similar shoddy casual game.
I dont have a problem to pay full price but not for casual games nor for the 15th rehash of Mario 2d.
The funny thing is i never really have had a huge problem with dojo. Mastering Dojo was heavens sent for me to explain some non obvious constructs outside of that, the code + a little bit of brain and dojo campus got me around.
As for the 20MB that is mostly like you would see a jquery distro with all sources and a load of plugins all unit tests included documentation etc.... You usually isolate your files and end up with a 50-300Kbyte single include source thats it.
I have worked with several projects with dojo. Usually it was like that.
The dojo core was very good and always worked, not a single problem. Dijit widgets, there it dependend heavily on how far you would use them, the core widgets usually worked. Data bindings were somewhat problematic in a way that they are hard to use. And some widgets were a little bit buggy, but most of them more on the dojox side.
The biggest project I was involved with just used the core and rolled its own widgets.
So it is mostly comparable to jquery, where you have a good core and depending on your luck good or bad extensions.
The main difference is the dojo core covers way more ground.
PS: Jquery, I have worked with jquery as well extensively. You can use it nicely from the core, but as soon as you hit the widget area it becomes an utter mess, some widgets are really good but some are utter garbage and you often end up with fixing a myriad of bugs to get the thing going. Trying a pure functional approach and having a non existent real widget api does not help either. The code often ends up in a mess and things which could be isolated with inheritance are exposed fully which makes often widget code convoluted.
As for prototype, nice try in language design, but a library which alters the core language underneath it just for the sake of it is possible is a no go, I have seen to many problems caused by this ghetto blaster of a library.
It really depends on your needs, what you do, but jquery for instance has its merits if i need to decorate a site quickly and if I can find a plugin which allows me to fix a problem quickly. I however would never program a bigger fully dhtml/ajax site with it, which does not do any page refreshes but provides you a huge rich client ui.
Actually JS is both, it is rather messy because everything is basically a function and an object or we could name it both is the same.
What you dont get is inheritance namespacing etc.. out of the box all you get is access to the prototype function to build such structural patterns upon.
The problem by not delivering both simply is, that you have to build your own solutions upon it which has several disadvantages. Namespacing must be built with maps which means speed hit and automatic documentation tools will fail.
The same then goes for classes inheritance etc... documentation tools automatically will fail if not custom tailored to the syntax of the framework.
Actually I am working with dojo as we speak, we dont use too much of it just the core the query part and a handful of widgets, and it works out quite well.
The core itself can be reduced to 50k of code if you pack everything in with all the dependencies you have about 200k of code. What you talk about the 21 MB is the source all its tests all the documentation and whatever comes with it. As soon as you isolate what you are using you are down to something smaller.
You can compare a full dojo distribution with a full jquery distribution with every plugin on earth written for it.
The huge difference is outside of the jquery core you have a huge mess of plugins with different quality, while dojo tries to enforce a certain quality standard by using extensive unit testing on its core and dijit widget.
As for the documentation, it could be better, but given the extensiveness it is ok. Not like YUI standards but given all the resources out there good enough to use it. For me the code ultimately is the last reference the books usually are a jump starter.
The jquery core itself is just about 1% of the extensiveness of the dojo core, mainly the query part and a little bit of xhr and events, and that part is well documented, everything else is a myriad of extensions. Some of them well documented some of them utter garbage not even working not a single line of documentation.
I would not even closely compare dojo and jquery. Jquery does not even cover the ground the dojo core covers and both have about the same bootup times. Things become slow in dojo as soon as you dont package and use a lot of digit widgets. Which means dojo issues load of xhr requests to fetch additional code in. This can be fixed by packaging everything.