I too didn't expect there to be a connection between the Swift language and the IBM partnership. I hadn't heard the 20% to 60% numbers before. When did Cook say that? If that is their goal, then that will indeed blow the lid off of sales of iDevices. In fact a 60% penetration of Exxon alone would be worth breaking out the century-old champaign.
That hole out behind Microsoft's Redmond offices needs to be expanded to hold all the Surface Pro 3s. When filled and covered, I understand there will a break dance exhibit done on the mound... that is after Ballmer does his obligatory monkey dance.
IBM has an army of 10,000's of developers/contractors around the globe, and a way to productize a custom developed app (they have another army of 10,000 tech writers and marketing folk). IBM sees enterprise problems, can develop the solution on the clients dime (client funded R&D), generalize it with a low infusion of product development $$, and push it out the door to every other enterprise client.
Given IBM's Analytics/Big Data bent, iOS integration into their analytics engines on big @ss IBM Virtual Data Center homed DB2 data farms is a big market. And in the Enterprise, keeping up with the Joneses, is par for the course.
As for Exxon Penetration... champagne? likely not. Assuming that half their employees use laptops, and half of those can switch, that's only 20K iPad sales. that's about 4 hours of sales. But if Exxon makes that move, then others will see less risk in it.
What you really want is something like the US Postal Service developing a custom app that makes an iPhone the primary mobile computing tool for a letter carrier (338,000 in the US alone).
Nurses/healthcare. 2.7Million nurses involved in healthcare. Assume that many more in allied-health (techs, transport, etc). IBM comes up with an iOS app for an iPad to talk to an IBM based medical records system... All those laptops on carts just fall out the window (literally, they take up more space in the charging bay).
The real benefit is that once IBM does it, the cost to deliver it to other business drops, it can scale down to small business (the real benefit of the cloud... I don't need a VM mainframe anymore, just an iPad, and a $100 month internet connect), and then it also drops for consumers (ditto).
When 'programming' gets to the point where one can create their own assets and then verbally or textually describe to Siri or Watson what you want done with those assets (in other words, the A.I. performs the coding for you), then I'd do nothing but 'program' all day.
No thank you, I have a hard enough time trying to get Siri to dial the correct Pizza service. Piazza Domination Hoes this is Elga, how may I punish you today, yeah, do you guys sell anchovies. I would like to see more IDE's offer better interfaces that incorporate a more visual approach to programming, elements you can simply drag and drop, easily selectable sources like database, RSS, other XML files, delimited files, selectable workable algorithms, etc. Out of the three main mobile OS's, iOS, Android and WM8.1 and the IDE's that are offered for them, I think I prefer programming with MS's tools, if you go here http://appstudio.windows.com, you can in 4 easy steps create a very usable app. Pretty cool little interface, I made an app in less then an hour that shows all of my subscriptions from AppleInsider, click on one, takes me to the last page, ready to post. It's a silly thing I know, but I have added notification support now that WM8.1 just got a Notification Center. Though I know how to write a Windows mobile app from scratch using Visual Studio, using a simple GUI that starts you with a working model is not only a perfect way to learn but makes it faster to produce simple tools, a lot of fun too. I also absolutely dig the virtual machine, it's so great that you can visualize your creation every step of the way.
Objective-C is extremely verbose. While typing may be only a small part of developing software, the more of it I have to do the more chance I have to make a typo. Also readability is essential for maintaining software. Objective-C can be UGLY. Reading extremely verbose code also leads to problems.
Swift aligns Apple with modern readability and syntax conventions of languages like Ruby, Groovy, Python and others. Yes, the vast majority of the real work is still done in libraries, but making those calls easier to parse (for the human) is very valuable.
While I'm primarily a Java developer, I've started using Groovy wherever I can because it cuts out a lot of verbose boilerplate that Java requires. I can get to the problem rather than writing lots of housekeeping code to keep Java happy.
I think Swift will make it easier for coders in other modern languages to jump in and out of Swift than Objective-C does.
I too didn't expect there to be a connection between the Swift language and the IBM partnership. I hadn't heard the 20% to 60% numbers before. When did Cook say that? If that is their goal, then that will indeed blow the lid off of sales of iDevices. In fact a 60% penetration of Exxon alone would be worth breaking out the century-old champaign.
That hole out behind Microsoft's Redmond offices needs to be expanded to hold all the Surface Pro 3s. When filled and covered, I understand there will a break dance exhibit done on the mound... that is after Ballmer does his obligatory monkey dance.
IBM has an army of 10,000's of developers/contractors around the globe, and a way to productize a custom developed app (they have another army of 10,000 tech writers and marketing folk). IBM sees enterprise problems, can develop the solution on the clients dime (client funded R&D), generalize it with a low infusion of product development $$, and push it out the door to every other enterprise client.
Given IBM's Analytics/Big Data bent, iOS integration into their analytics engines on big @ss IBM Virtual Data Center homed DB2 data farms is a big market. And in the Enterprise, keeping up with the Joneses, is par for the course.
As for Exxon Penetration... champagne? likely not. Assuming that half their employees use laptops, and half of those can switch, that's only 20K iPad sales. that's about 4 hours of sales. But if Exxon makes that move, then others will see less risk in it.
What you really want is something like the US Postal Service developing a custom app that makes an iPhone the primary mobile computing tool for a letter carrier (338,000 in the US alone).
Nurses/healthcare. 2.7Million nurses involved in healthcare. Assume that many more in allied-health (techs, transport, etc). IBM comes up with an iOS app for an iPad to talk to an IBM based medical records system... All those laptops on carts just fall out the window (literally, they take up more space in the charging bay).
The real benefit is that once IBM does it, the cost to deliver it to other business drops, it can scale down to small business (the real benefit of the cloud... I don't need a VM mainframe anymore, just an iPad, and a $100 month internet connect), and then it also drops for consumers (ditto).
Your comments, and my monthly retirement check from IBM, are making me feel good about Big Blue again
Swift is indeed a big improvement wrt Objective-C. After a learning curve, it will reduce the development time for app developers. But the main drawback remains: like its predecessor it is an Apple only language.
As a developer of smartphone apps I have to listen to the market demands and make my apps available for both iOS and Android platforms. Porting an app from one platform to another typically requires 50% additional effort and cost. So currently I try to stick, whenever possible, to Cordova/Phonegap which makes it possible to develop cross platform smartphone apps.
Swift is indeed a big improvement wrt Objective-C. After a learning curve, it will reduce the development time for app developers. But the main drawback remains: like its predecessor it is an Apple ony language.
As a developer of smartphone apps I have listen to the market demands and make my apps available for both iOS and Android platforms. Porting an app from one platform to another typically requires 50% additional effort and cost. So currently I try to stick, whenever possible, to Cordova/Phonegap which makes it possible to develop cross platform smartphone apps.
What I'd really like to see happen is for Apple to embrace the open source side with Swift. Microsoft doesn't like the open source community much at all and there's some uncertainty over C#:
If Swift became the main language for iOS and Android, they could drive C# and maybe even Java out. It would also allow apps like Unity and maybe other SDKs to use the language directly. This allows 3rd party SDK developers to directly call Apple's OS APIs and possibly avoid cross-compiling through XCode. One problem there might come from people not liking Apple directing the language's features but it wouldn't be much different from Oracle controlling Java. Once the language reaches an acceptable level and is open source, there shouldn't be significant changes that disrupt workflows.
Beyond that, they can push for Swift support in browsers and web servers because if Google takes it on, they can support it in Chrome and that can become the language for Chrome OS too. Google tried to get a new language adopted called Go but they never gave an incentive to use it. If they had tied it to Android and Chrome OS development the way Apple will do with iOS and OS X, it would have gained much better adoption.
The platform-specific APIs would still hamper portability but having a common language helps a lot and it shouldn't harm Apple in any way to do this. They'd just be expanding their pool of developers.
As a Java Enterprise programmer, why would I want to use a programming language that isn't cross platform compatible, isn't that the whole point. No, I see web driven apps replacing Java not Swift, maybe a few will utilize it for specialized apps that require their apps to be ran locally.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
As a Java Enterprise programmer, why would I want to use a programming language that isn't cross platform compatible, isn't that the whole point. No, I see web driven apps replacing Java not Swift, maybe a few will utilize it for specialized apps that require their apps to be ran locally.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question :???:
As a Java Enterprise programmer, why would I want to use a programming language that isn't cross platform compatible, isn't that the whole point. No, I see web driven apps replacing Java not Swift, maybe a few will utilize it for specialized apps that require their apps to be ran locally.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question :???:
Relic raised some good issues. First, cross platform compatibility. Apple hasn't said anything about opening Swift up to other platforms, and it may not happen soon, if at all. Swift is still beta and Apple has warned that Swift will undergo a lot of changes over the next two years. I'm even wondering if it will remain beta during that whole time span. If so, it wouldn't surprise me if Apple doesn't open it up to other platforms until that phase is complete. Secondly, I thought web-driven apps had been declared a closed issue. I must have missed a memo. As far as "few will utilize it for specialized apps that require their apps to be ran locally." My understanding is that IBM is all over Swift as a programming language. We will see what the next two years may bring.
When 'programming' gets to the point where one can create their own assets and then verbally or textually describe to Siri or Watson what you want done with those assets (in other words, the A.I. performs the coding for you), then I'd do nothing but 'program' all day.
I can see it now. Only two years hence as the human race reckons the illusion of time.
I gesture to my iDevice which summons a hologram of Siri from her virtual bath. (that girl spends most of her time in soapy water).
"Siri," I say, "Let Watson know I need an app written. There are three specifications. First, Watson needs to search his Big Data and find a human need that has not been realized yet and write an app to fulfill that need. Second: It must be a need strong enough to trigger the auto-buy on enough iDevices that it brings me $10,000 a month. No, make that per week. Finally have it on the Apple Global app store before tomorrow. I need the cash for this weekend."
"Oh, Siri, have the lout, Ballmer, bring my limo around to the front. I feel an invigorating drive through the country to cool back down after this little mental workout."
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question
Those Java programmers weren't very good.
Web applications are defiantly the present and future, I think I mentioned this before but when I was healthier and working I was employed by the largest bank in Switzerland. I was the managing director of internal programming, though we utilized compiled applications for specialized solutions such as our our trading platform the bulk of our applications were web driven. It's exactly the same situation in almost every company, in every field. Even the applications that ran specifically on a tablet, such as the iPad were for the most part web driven. Yes, I'm sure company's will utilize all those wonderful new apps that IBM will produce but mark my words any internal custom solutions will most likely be web driven and not compiled using Swift. No company in there right mind is going to send any of their programming staff to school to learn Swift just for the sole purposes of making iPad apps, not when they could easily make a web app in less time, less money, utilize the talent they currently have and especially use those apps across their entire IT infrastructure. There will of course be those exceptions, specialized apps are always needed but for the most part they will be purchased from third party vendors.
Swift is a clearly a very clever thing but it will not be shaking up the Enterprise world anytime soon as most here believe. I've seen many different programming solutions and languages come and go, those that stuck and were utilize the most were the ones that could be used multiple hardware and OS environments, i.e. Java, Python and web applications. it's just the way it is, things like running naively, faster, intuitiveness of the UI, whatever, just isn't a match for portability, speed of releases, budget constraints, especially when the output is identical.
Those who will benefit from Swift the most are company's or individuals specifically selling iPad apps.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Then honestly your Java programming team is sub-par if they can't get their applications working the way they're supposed too. I've heard this many times in forums but when it comes right down to it, writing and supporting multiple versions of one app to run on different platforms is no match for writing it once, supporting just that one build and then using it on the multiple platforms. It's fine to think that Apple's solution is superior and more company's should be using it but what are the real benefits, speed, not really important, reliability, sure but your delusional if you think a company wouldn't expect a programmer to make sure their Java version isn't also reliable, look and feel, not important, it's all about the output in the end. For the most part Enterprise apps are data entry, data retrieval, reports and calculations, things that can be accomplished with a web app and there isn't much that can't be done with HTML5, PHP/Python/Ruby and a Oracle/MySQL/PostgreSQL database.
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question :???:
Those Java programmers weren't very good.
Web applications are defiantly the present and future, I think I mentioned this before but when I was healthier and working I was employed by the largest bank in Switzerland. I was the managing director of internal programming, though we utilized compiled applications for specialized solutions such as our our trading platform the bulk of our applications were web driven. It's exactly the same situation in almost every company, in every field. Even the applications that ran specifically on a tablet, such as the iPad were for the most part web driven. Yes, I'm sure company's will utilize all those wonderful new apps that IBM will produce but mark my words any internal custom solutions will most likely be web driven and not compiled using Swift. No company in there right mind is going to send any of their programming staff to school to learn Swift just for the sole purposes of making iPad apps, not when they could easily make a web app in less time, less money, utilize the talent they currently have and especially use those apps across their entire IT infrastructure. There will of course be those exceptions, specialized apps are always needed but for the most part they will be purchased from third party vendors.
Swift is a clearly a very clever thing but it will not be shaking up the Enterprise world anytime soon as most here believe. I've seen many different programming solutions and languages come and go, those that stuck and were utilize the most were the ones that could be used multiple hardware and OS environments, i.e. Java, Python and web applications. it's just the way it is, things like running naively, faster, intuitiveness of the UI, whatever, just isn't a match for portability, speed of releases, budget constraints, especially when the output is identical.
Those who will benefit from Swift the most are company's or individuals specifically selling iPad apps.
I suspect your enterprise experience is a lot more current than mine -- when I left IBM in 1980 there was no web.
I retired in 1989, and didn't touch a computer again until 1997 ... To get myself current, I taught myself JavaScript -- and implemented a web site with a shopping cart entirely in JavaScript.
Later Perl, then ColdFusion which was built atop Java.
I haven't done any for pay development since my wife Lucy died, suddenly, unexpectedly in 2001.
So, I'm a dilettante -- I piddle around and program for my own amazement -- and a few things for friends and family.
Much of what you say is true -- especially cross-platform, platform-independent points.
Here are some things I found out researching Swift.
Swift runs on a thing called an LLVM -- which stands for Low-Level Virtual Machine -- similar to the Java JIT ByteCode compiler, but designed to be closer to the iron. They soon realized that the LLVM could be used for a lot more than runtime execution -- things like writing tools and compilers.
Microsoft, Google, Sony and Apple * have LLVM development environments in various stages of testing and production.
So, it is possible that LLVM will realize the promise of Java for the desktop and web server-side programming.
* Apple's XCode has been transitioning to LLVM for several years and it is mostly complete.
Today at WWDC 2014, Apple announced the beta availability of a new programming language, Swift, which is set to ship with iOS 8 and OSX Yosemite later this year. Swift is a high-level programming language that will be familiar to JavaScript developers, but is compiled using LLVM to produce highly performant executable code for both OSX and iOS platforms.
Apple has heavily invested in LLVM technologies, which provide an abstract instruction set that can be translated for specific architectures. Clang replaced GCC as the compiler of choice for C and Objective-C programs, and both of those are translated with Clang into LLVM instructions, which are then optimised into an executable for the platform. The new programming language, Swift, produces LLVM bytecode in the same way and can co-exist with existing Objective-C applications and libraries.
In addition, swift also comes with a REPL environment for testing code. Normally used by interpreted languages such as JavaScript or Python, a REPL provides a Read-Evaluate-Print Loop that can be used to evaluate individual expressions and statements at the command line for easy debugging. Combined with powerful looping, string interpolation and printing/debugging options, it allows an interactive style of development and testing that is frequently missing for compiled languages such as C and Java.
Now, the designers of the LLVM have a goal for a mobile LLVM.
Apple and Google Together Again?
Considering the size of Google’s operation, this is a wonderfully impressive hack. But like Vikram Adve back at the University of Illinois, the company wants more out of LLVM. It wants to achieve something akin to that “write once, run anywhere” nirvana.
The search giant is also using LLVM to create a tool that can safely run any code inside a browser on practically any machine. Still under development, this tool is known as Portable Native Client, and you can think of it as a kind of uber browser plug-in. In many ways, it echoes the original goal laid down by Lattner and Adve: a programming paradigm somewhere between Java and machine code.
Adve questions whether Portable Native Client will find success — browsers already have JavaScript — but believes that LLVM may achieve a similar success on mobile phones. He envisions a world where you can use LLVM not only to create applications that can move from machine to machine, but that can move from across different types of processors on the same machine — from, say, CPU to graphics processor.
One final point: Apparently, Apple started writing Swift in 2010 using LLVM/CLANG -- As the language syntax and definition solidify, I suspect that Apple could make it open source ala WebKit ... Or other platforms can write their own implementations.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Then honestly your Java programming team is sub-par if they can't get their applications working the way they're supposed too. I've heard this many times in forums but when it comes right down to it, writing and supporting multiple versions of one app to run on different platforms is no match for writing it once, supporting just that one build and then using it on the multiple platforms. It's fine to think that Apple's solution is superior and more company's should be using it but what are the real benefits, speed, not really important, reliability, sure but your delusional if you think a company wouldn't expect a programmer to make sure their Java version isn't also reliable, look and feel, not important, it's all about the output in the end. For the most part Enterprise apps are data entry, data retrieval, reports and calculations, things that can be accomplished with a web app and there isn't much that can't be done with HTML5, PHP/Python/Ruby and a Oracle/MySQL/PostgreSQL database.
Yes, but the HTML 5 solution uses a lot of bandwidth -- sending down the wire:
verbose CSS
verbose Javascript
verbose XML or JSON (usually as web services)
verbose HTML
images
links
cookies
multiple connections
Paraphrasing US Senator, rtd Fritx Hollings:
"Day's a whole lotta' bandwidth consuming' goin' on out dare"
I suspect your enterprise experience is a lot more current than mine -- when I left IBM in 1980 there was no web.
I retired in 1989, and didn't touch a computer again until 1997 ... To get myself current, I taught myself JavaScript -- and implemented a web site with a shopping cart entirely in JavaScript.
Later Perl, then ColdFusion which was built atop Java.
I haven't done any for pay development since my wife Lucy died, suddenly, unexpectedly in 2001.
So, I'm a dilettante -- I piddle around and program for my own amazement -- and a few things for friends and family.
Much of what you say is true -- especially cross-platform, platform-independent points.
Here are some things I found out researching Swift.
Swift runs on a thing called an LLVM -- which stands for Low-Level Virtual Machine -- similar to the Java JIT ByteCode compiler, but designed to be closer to the iron. They soon realized that the LLVM could be used for a lot more than runtime execution -- things like writing tools and compilers.
Microsoft, Google, Sony and Apple * have LLVM development environments in various stages of testing and production.
So, it is possible that LLVM will realize the promise of Java for the desktop and web server-side programming.
* Apple's XCode has been transitioning to LLVM for several years and it is mostly complete.
One final point: Apparently, Apple started writing Swift in 2010 using LLVM/CLANG -- As the language syntax and definition solidify, I suspect that Apple could make it open source ala WebKit ... Or other platforms can write their own implementations.
Thanks for the info, I love reading your posts,always filled with lovely new stuff to research, I'm still going though it all but so far very impressive. I need to ask though, will Apple provide a Swift runtime for Windows and Linux because without it, well, it's kind of like figuring out cold fusion and only keeping it for yourself.
I suspect your enterprise experience is a lot more current than mine -- when I left IBM in 1980 there was no web.
I retired in 1989, and didn't touch a computer again until 1997 ... To get myself current, I taught myself JavaScript -- and implemented a web site with a shopping cart entirely in JavaScript.
Later Perl, then ColdFusion which was built atop Java.
I haven't done any for pay development since my wife Lucy died, suddenly, unexpectedly in 2001.
So, I'm a dilettante -- I piddle around and program for my own amazement -- and a few things for friends and family.
Much of what you say is true -- especially cross-platform, platform-independent points.
Here are some things I found out researching Swift.
Swift runs on a thing called an LLVM -- which stands for Low-Level Virtual Machine -- similar to the Java JIT ByteCode compiler, but designed to be closer to the iron. They soon realized that the LLVM could be used for a lot more than runtime execution -- things like writing tools and compilers.
Microsoft, Google, Sony and Apple * have LLVM development environments in various stages of testing and production.
So, it is possible that LLVM will realize the promise of Java for the desktop and web server-side programming.
One final point: Apparently, Apple started writing Swift in 2010 using LLVM/CLANG -- As the language syntax and definition solidify, I suspect that Apple could make it open source ala WebKit ... Or other platforms can write their own implementations.
Thanks for the info, I love reading your posts,always filled with lovely new stuff to research, I'm still going though it all but so far very impressive. I need to ask though, will Apple provide a Swift runtime for Windows and Linux because without it, well, it's kind of like figuring out cold fusion and only keeping it for yourself.
Comments
It pretty much guarantees that if an app compiles it will not crash at runtime.
That would be a major academic achievement and, if true, would have been the first thing out of their mouths at WWDC.
Actually, it was the third thing out of their mouths when describing Swift:
They didn't dwell on code safety other than showing some unsafe constructs in Obj-C that go away with Swift -- it's kind of hard to demonstrate.
I too didn't expect there to be a connection between the Swift language and the IBM partnership. I hadn't heard the 20% to 60% numbers before. When did Cook say that? If that is their goal, then that will indeed blow the lid off of sales of iDevices. In fact a 60% penetration of Exxon alone would be worth breaking out the century-old champaign.
That hole out behind Microsoft's Redmond offices needs to be expanded to hold all the Surface Pro 3s. When filled and covered, I understand there will a break dance exhibit done on the mound... that is after Ballmer does his obligatory monkey dance.
IBM has an army of 10,000's of developers/contractors around the globe, and a way to productize a custom developed app (they have another army of 10,000 tech writers and marketing folk). IBM sees enterprise problems, can develop the solution on the clients dime (client funded R&D), generalize it with a low infusion of product development $$, and push it out the door to every other enterprise client.
Given IBM's Analytics/Big Data bent, iOS integration into their analytics engines on big @ss IBM Virtual Data Center homed DB2 data farms is a big market. And in the Enterprise, keeping up with the Joneses, is par for the course.
As for Exxon Penetration... champagne? likely not. Assuming that half their employees use laptops, and half of those can switch, that's only 20K iPad sales. that's about 4 hours of sales. But if Exxon makes that move, then others will see less risk in it.
What you really want is something like the US Postal Service developing a custom app that makes an iPhone the primary mobile computing tool for a letter carrier (338,000 in the US alone).
Nurses/healthcare. 2.7Million nurses involved in healthcare. Assume that many more in allied-health (techs, transport, etc). IBM comes up with an iOS app for an iPad to talk to an IBM based medical records system... All those laptops on carts just fall out the window (literally, they take up more space in the charging bay).
The real benefit is that once IBM does it, the cost to deliver it to other business drops, it can scale down to small business (the real benefit of the cloud... I don't need a VM mainframe anymore, just an iPad, and a $100 month internet connect), and then it also drops for consumers (ditto).
When 'programming' gets to the point where one can create their own assets and then verbally or textually describe to Siri or Watson what you want done with those assets (in other words, the A.I. performs the coding for you), then I'd do nothing but 'program' all day.
No thank you, I have a hard enough time trying to get Siri to dial the correct Pizza service. Piazza Domination Hoes this is Elga, how may I punish you today, yeah, do you guys sell anchovies. I would like to see more IDE's offer better interfaces that incorporate a more visual approach to programming, elements you can simply drag and drop, easily selectable sources like database, RSS, other XML files, delimited files, selectable workable algorithms, etc. Out of the three main mobile OS's, iOS, Android and WM8.1 and the IDE's that are offered for them, I think I prefer programming with MS's tools, if you go here http://appstudio.windows.com, you can in 4 easy steps create a very usable app. Pretty cool little interface, I made an app in less then an hour that shows all of my subscriptions from AppleInsider, click on one, takes me to the last page, ready to post. It's a silly thing I know, but I have added notification support now that WM8.1 just got a Notification Center. Though I know how to write a Windows mobile app from scratch using Visual Studio, using a simple GUI that starts you with a working model is not only a perfect way to learn but makes it faster to produce simple tools, a lot of fun too. I also absolutely dig the virtual machine, it's so great that you can visualize your creation every step of the way.
all web content is PHP.
These guys, these guys, these guys and all the .NET developers might disagree with that.
---
Objective-C is extremely verbose. While typing may be only a small part of developing software, the more of it I have to do the more chance I have to make a typo. Also readability is essential for maintaining software. Objective-C can be UGLY. Reading extremely verbose code also leads to problems.
Swift aligns Apple with modern readability and syntax conventions of languages like Ruby, Groovy, Python and others. Yes, the vast majority of the real work is still done in libraries, but making those calls easier to parse (for the human) is very valuable.
While I'm primarily a Java developer, I've started using Groovy wherever I can because it cuts out a lot of verbose boilerplate that Java requires. I can get to the problem rather than writing lots of housekeeping code to keep Java happy.
I think Swift will make it easier for coders in other modern languages to jump in and out of Swift than Objective-C does.
Your comments, and my monthly retirement check from IBM, are making me feel good about Big Blue again
Swift is indeed a big improvement wrt Objective-C. After a learning curve, it will reduce the development time for app developers. But the main drawback remains: like its predecessor it is an Apple only language.
As a developer of smartphone apps I have to listen to the market demands and make my apps available for both iOS and Android platforms. Porting an app from one platform to another typically requires 50% additional effort and cost. So currently I try to stick, whenever possible, to Cordova/Phonegap which makes it possible to develop cross platform smartphone apps.
What I'd really like to see happen is for Apple to embrace the open source side with Swift. Microsoft doesn't like the open source community much at all and there's some uncertainty over C#:
http://en.wikipedia.org/wiki/C_Sharp_(programming_language)
http://www.fsf.org/news/dont-depend-on-mono
If Swift became the main language for iOS and Android, they could drive C# and maybe even Java out. It would also allow apps like Unity and maybe other SDKs to use the language directly. This allows 3rd party SDK developers to directly call Apple's OS APIs and possibly avoid cross-compiling through XCode. One problem there might come from people not liking Apple directing the language's features but it wouldn't be much different from Oracle controlling Java. Once the language reaches an acceptable level and is open source, there shouldn't be significant changes that disrupt workflows.
Beyond that, they can push for Swift support in browsers and web servers because if Google takes it on, they can support it in Chrome and that can become the language for Chrome OS too. Google tried to get a new language adopted called Go but they never gave an incentive to use it. If they had tied it to Android and Chrome OS development the way Apple will do with iOS and OS X, it would have gained much better adoption.
The platform-specific APIs would still hamper portability but having a common language helps a lot and it shouldn't harm Apple in any way to do this. They'd just be expanding their pool of developers.
In any good diverse ecosystem there are important niches for bottom-feeders and sucker fish.
Don't forget the Babel fish - Dude
And the wrecks split into millions of fragments.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question :???:
Relic raised some good issues. First, cross platform compatibility. Apple hasn't said anything about opening Swift up to other platforms, and it may not happen soon, if at all. Swift is still beta and Apple has warned that Swift will undergo a lot of changes over the next two years. I'm even wondering if it will remain beta during that whole time span. If so, it wouldn't surprise me if Apple doesn't open it up to other platforms until that phase is complete. Secondly, I thought web-driven apps had been declared a closed issue. I must have missed a memo. As far as "few will utilize it for specialized apps that require their apps to be ran locally." My understanding is that IBM is all over Swift as a programming language. We will see what the next two years may bring.
I can see it now. Only two years hence as the human race reckons the illusion of time.
I gesture to my iDevice which summons a hologram of Siri from her virtual bath. (that girl spends most of her time in soapy water).
"Siri," I say, "Let Watson know I need an app written. There are three specifications. First, Watson needs to search his Big Data and find a human need that has not been realized yet and write an app to fulfill that need. Second: It must be a need strong enough to trigger the auto-buy on enough iDevices that it brings me $10,000 a month. No, make that per week. Finally have it on the Apple Global app store before tomorrow. I need the cash for this weekend."
"Oh, Siri, have the lout, Ballmer, bring my limo around to the front. I feel an invigorating drive through the country to cool back down after this little mental workout."
Someone said ... "Java: Write once -- debug everywhere!"
And, if web-driven apps are the answer -- what was the question
Those Java programmers weren't very good.
Web applications are defiantly the present and future, I think I mentioned this before but when I was healthier and working I was employed by the largest bank in Switzerland. I was the managing director of internal programming, though we utilized compiled applications for specialized solutions such as our our trading platform the bulk of our applications were web driven. It's exactly the same situation in almost every company, in every field. Even the applications that ran specifically on a tablet, such as the iPad were for the most part web driven. Yes, I'm sure company's will utilize all those wonderful new apps that IBM will produce but mark my words any internal custom solutions will most likely be web driven and not compiled using Swift. No company in there right mind is going to send any of their programming staff to school to learn Swift just for the sole purposes of making iPad apps, not when they could easily make a web app in less time, less money, utilize the talent they currently have and especially use those apps across their entire IT infrastructure. There will of course be those exceptions, specialized apps are always needed but for the most part they will be purchased from third party vendors.
Swift is a clearly a very clever thing but it will not be shaking up the Enterprise world anytime soon as most here believe. I've seen many different programming solutions and languages come and go, those that stuck and were utilize the most were the ones that could be used multiple hardware and OS environments, i.e. Java, Python and web applications. it's just the way it is, things like running naively, faster, intuitiveness of the UI, whatever, just isn't a match for portability, speed of releases, budget constraints, especially when the output is identical.
Those who will benefit from Swift the most are company's or individuals specifically selling iPad apps.
using java and thinking you're writing cross-platform apps is a mistake. a java app on multiple platforms just means you have an app that looks poor and behaves unusually on multiple platforms. native is superior.
Then honestly your Java programming team is sub-par if they can't get their applications working the way they're supposed too. I've heard this many times in forums but when it comes right down to it, writing and supporting multiple versions of one app to run on different platforms is no match for writing it once, supporting just that one build and then using it on the multiple platforms. It's fine to think that Apple's solution is superior and more company's should be using it but what are the real benefits, speed, not really important, reliability, sure but your delusional if you think a company wouldn't expect a programmer to make sure their Java version isn't also reliable, look and feel, not important, it's all about the output in the end. For the most part Enterprise apps are data entry, data retrieval, reports and calculations, things that can be accomplished with a web app and there isn't much that can't be done with HTML5, PHP/Python/Ruby and a Oracle/MySQL/PostgreSQL database.
I suspect your enterprise experience is a lot more current than mine -- when I left IBM in 1980 there was no web.
I retired in 1989, and didn't touch a computer again until 1997 ... To get myself current, I taught myself JavaScript -- and implemented a web site with a shopping cart entirely in JavaScript.
Later Perl, then ColdFusion which was built atop Java.
I haven't done any for pay development since my wife Lucy died, suddenly, unexpectedly in 2001.
So, I'm a dilettante -- I piddle around and program for my own amazement -- and a few things for friends and family.
Much of what you say is true -- especially cross-platform, platform-independent points.
Here are some things I found out researching Swift.
Swift runs on a thing called an LLVM -- which stands for Low-Level Virtual Machine -- similar to the Java JIT ByteCode compiler, but designed to be closer to the iron. They soon realized that the LLVM could be used for a lot more than runtime execution -- things like writing tools and compilers.
Microsoft, Google, Sony and Apple * have LLVM development environments in various stages of testing and production.
So, it is possible that LLVM will realize the promise of Java for the desktop and web server-side programming.
* Apple's XCode has been transitioning to LLVM for several years and it is mostly complete.
http://www.infoq.com/news/2014/06/apple-swift
Now, the designers of the LLVM have a goal for a mobile LLVM.
http://www.wired.com/2013/07/apple-google-llvm/all/
It's an interesting read!
One final point: Apparently, Apple started writing Swift in 2010 using LLVM/CLANG -- As the language syntax and definition solidify, I suspect that Apple could make it open source ala WebKit ... Or other platforms can write their own implementations.
Yes, but the HTML 5 solution uses a lot of bandwidth -- sending down the wire:
Paraphrasing US Senator, rtd Fritx Hollings:
"Day's a whole lotta' bandwidth consuming' goin' on out dare"
Now, off to have some raclettes
I suspect your enterprise experience is a lot more current than mine -- when I left IBM in 1980 there was no web.
I retired in 1989, and didn't touch a computer again until 1997 ... To get myself current, I taught myself JavaScript -- and implemented a web site with a shopping cart entirely in JavaScript.
Later Perl, then ColdFusion which was built atop Java.
I haven't done any for pay development since my wife Lucy died, suddenly, unexpectedly in 2001.
So, I'm a dilettante -- I piddle around and program for my own amazement -- and a few things for friends and family.
Much of what you say is true -- especially cross-platform, platform-independent points.
Here are some things I found out researching Swift.
Swift runs on a thing called an LLVM -- which stands for Low-Level Virtual Machine -- similar to the Java JIT ByteCode compiler, but designed to be closer to the iron. They soon realized that the LLVM could be used for a lot more than runtime execution -- things like writing tools and compilers.
Microsoft, Google, Sony and Apple * have LLVM development environments in various stages of testing and production.
So, it is possible that LLVM will realize the promise of Java for the desktop and web server-side programming.
* Apple's XCode has been transitioning to LLVM for several years and it is mostly complete.
http://www.infoq.com/news/2014/06/apple-swift
Now, the designers of the LLVM have a goal for a mobile LLVM.
http://www.wired.com/2013/07/apple-google-llvm/all/
It's an interesting read!
One final point: Apparently, Apple started writing Swift in 2010 using LLVM/CLANG -- As the language syntax and definition solidify, I suspect that Apple could make it open source ala WebKit ... Or other platforms can write their own implementations.
Thanks for the info, I love reading your posts,always filled with lovely new stuff to research, I'm still going though it all but so far very impressive. I need to ask though, will Apple provide a Swift runtime for Windows and Linux because without it, well, it's kind of like figuring out cold fusion and only keeping it for yourself.
I suspect they will -- and for Android, too!