I don't believe in RAD. By the time anyone has developed a toolkit that can easily do something, it's not cool enough to do it anymore:)
RAD frameworks also tend to get bloated trying to do everything for everybody in my experience. At least when it comes to CRUD applications. I find myself more productive and have an easier time developing a site like that to do what I want by just using a good persistance (DAO, JDO, Hibernate, etc) framework or methodology that makes it easy to communicate with the DB.
Other stuff that i find useful are tools that build the classes you need to interact with various webservices.
Most sites using java don't serve JSP pages directly. They do it through some sort of MVC framework where servlets call the JSP behind the scenes. I've written sites where the url a user sees is a.html but it's actually just a call to a servlet that parses the url to server up the appropriate content and it uses a jsp for the presentation layer.
I'm pretty sure Blogger.com is written in java or at least uses java heavily.
This site I'm pretty sure uses Java because they used to use it in the past. The vendor they used is out of business and I would assume they switched to a different appserver but stuck with java. I could be wrong. An example url on their site is: http://www.bhphotovideo.com/c/browse/Computers-Solutions/ci/9581/N/4294542559
"/c" is the controller servlet. The rest of the url provides information like url parameters that tell the controller what to display.
Struts is a common Java web framework and by default urls for struts applications end in ".do"
I agree with what yous say. Just adding information.
Don't bother learning PHP it'll just slow you down learning the semantics of a new language and if you're used to dealing with Java PHP will feel like a joke. Java gets a bad wrap but I've found it to be faster than PHP in my tests.
I build most of my sites in Java using my own MVC framework. I've done some sites in PHP and have had to modify other PHP sites as well as looked into other languages. I still like Java the best and you can find a library to do almost anything you want. The only reason I'd pick something like PHP these days is if I don't want to build a site myself and want to use something prebuilt like wordpress or Joomla.
The only downside is that your servlet container (ie tomcat) is persistent and will take up a bit of memory. Not a huge amount but it makes it difficult to find cheap webhosting because providers can't throw thousands of websites on a server like they can with plain HTML and PHP. Try and find a good cheap VPS it's more secure and you won't have to worry about your site getting defaced because some other idiot didn't update their PHP software. That's happened to me.
Don't go with new frameworks. Go with popular ones that have been around for a while. I've been bitten in the ass when I built a website for a client and the framework I used was no longer around.
Spring is a good choice. I like to use NetBeans as my IDE. I've found it to work the best for me.
Almost all the sites I've built are written in Java. Stick with Java. I've written sites in PHP and I've also had to work on updates to some PHP sites. If you're already familiar with Java dealing with PHP will feel like a joke. PHP is great when you don't want to write your own software since there are so many publicly available stuff out there in PHP. Don't worry, you won't find a lack of Java libraries that will do anything you want to do.
Don't bother trying to learn a new language because you'll just slow yourself down trying to learn the semantics of the language instead of the details of the new libraries you'll be using. I know java gets a bad wrap in terms of performance but I've always found that Java kicks PHP's ass in terms of performance in the tests I've done.
The main issue with java is that when you're using a servlet container like Tomcat, the process runs constantly and takes up memory. It's not that much but it's hard to find Java hosting because the memory issues makes it hard for a webhost to put thousands of websites on the same server.
Your best bet is going to be to find a cheap VPN when you get started but check the big webhosts to. I remember LunarPages used to offer JSP support in the past.
There are a bunch of different frameworks. Stick to ones that are popular because you'd rather have some limited functionality now rather than an unsupported framework in the future. Which has happened to me.
I believe right now that's Spring but Struts is still pretty popular too.
I've found NetBeans to be a great IDE and it supports Spring.
It seems that Google+ will be connected with many systems and a Google+ account will be a very important if you put content online that you want to promote.
Want to get more clicks from google searches? Set up your site so it connects to your Google+ account so your picture can be included in the search pages.
I'm not saying if it's good or bad but it's different.
I'm imagining an internet where trade groups disallow trade-group TLDs to any business that doesn't pay them dues. Not subscribed to the RIAA? No.singer,.artist,.music,.whatever for you. Not in the BSA? No.software...
Yes throw in two industry groups that everyone on here hates to make this sound like a bad idea.
The interenet already has restricted TLDs so this concept isn't new. When you visit a.gov site you know you're dealing with a government website. Starting in 2001.edu domains also were restricted to accredited schools.
Who controls access to these TLDs is important but it makes sense to have it be some sort of industry group that can fairly manage it..singer,.artist,.music,.software aren't being discussed and the reason for.bank and.insurance are very different. Millions, maybe billions of dollars are lost annually to phishing schemes in this sector and having restricted TLDs can help reduce that number.
Considering the #1 most repeated rule for general PC users is "if it doesn't contain just your bank's website and end in.com, it's fake," I'm not sure they're doing themselves any favors. A year from now if I can an e-mail from MyBanksName.bank, even I'd assume it's a fake and not click it.
The flaw is that anyone can get a.com domain name so I could go and by a domain name like Citi-Bank-Secure-For-Realz.com and some people would fall for it. I mean it does say it's for realz!
If the.bank and.insurance TLDs were restricted by an industry group that assured only legitimate banks and insurers could purchase domains in that space, I can see how it would increase trust as long as consumers are made aware of the new changes. If it does happen there would be a big push to raise awareness through direct contact with their customers and the media.
I agree IE isn't that hard to support anymore. I build websites for companies in various sectors including corporate clients. IE is still the dominant browser in some segments but not as much as it used to be.
For sites where the users tend to be more tech savy, and browse mainly from home IE isn't that popular. Maybe around 30-40% of these users are using IE.
For the business sites where visitors are generally browsing the site from work for business related reasons IE is still very strong. Around 60% and the most popular version is one version of the latest because of how software on corporate desktops is managed. It used to be a lot worse. 2-3 years ago IE was at around 90% for these client sites.
IE is still an important browser to support but if you limit support for certain areas, such as the admin area the site in TFA does it for, then you can be more restrictive on browser support because you have a limited set of users accessing those pages.
If you build websites for people that will be accessing the sites primarily from work, ignoring IE is shooting yourself in the foot.
no one is arguing against that point. So why do you keep raising it?
Not sure why it's such a sticking point so let me be clearer. Obviously when I refer to "Google" I'm talking broadly about the organization, considering a company has no free will and can not choose to do anything. Specifically I'm talking about the part of the organization that worked on Android. In this group I'm also including any contractors Google or it's subsidiaries might have hired. Instead of spending time trying to track down all the people that were involved with the code inside and outside of Google I just use the word "Google" to lump them all together. Because as I said, in the end it's their responsibility.
And I didn't infer anything from that line you're referring to. I was asking a question hoping to get an answer.
9 lines of what everyone thinks is trivial code would be faster to type than it would be to look up to copy and paste.
And the 9 lines is less than 2% of the code that was found to be infringing.
What I am inferring is that if Google didn't know that any of the code was infringing, that they knew that they told their contractor to develop a clean implementation, that information was likely communicated to Sun (I seem to recall some sort of conversation taking place but don't have a link) then any implied license Google thinks they had because of Sun's affirmative acts shouldn't be a factor because they were based on the belief by both parties that the code was not infringing.
Ironically doing a Google search for "Adnix and Preachnix" and clicking on the first page gets you a page full of ads, a popup and a warning message when you try and browse away.:)
Local broadcasts are free. Well sort of. You can still get over the air local channels most places but maybe 80-90%+ of people receive their local channels through cable tv, satellite, fios, whatever.
These companies can't just tap into the local airwaves and rebroadcast these channels. They have to pay for it and ever year or so there's a major issue with the contract of some channel holding out for more money.
The traditional broadcasting model is dead. The new one is get paid by advertisers, get paid again by distributors.
Though I can imagine dish caving just like tivo did.
Because it doesn't matter whether it was a contractor or not. Google is ultimately responsible for abiding by the law with the products they release whether they were written by employees, obtained through an acquisition or created for them by a third party at their request.
I'm not talking about rangeCheck specificially but the entirety of code that the judge ruled was decompiled and copied from Java.
That leaves the question of whether the contractor copied other parts of the Java that are not as easy to identify or if they decompiled and copied then rewrote to make it look like it wasn't copied. Maybe they didn't do that with this function because it was trivial. Who knows. But to me this just doesn't look good.
I don't see Oracle as the bad guy here. They had their (Sun's) IP stolen and are right to seek damages.
I don't see Google as a bad guy here either. Their intention was not to violate Sun's IP.
Things got screwed up and now need to be resolved. In a perfect world Sun would have successfully negotiated a deal when Google initially approached them. That didn't happen and now here we are.
This suit wasn't to stop Google from using the car any longer, but to seek billions in damages for misuse of the car during a time that Google thought they owned it (due to deception by Noser)
Uhm... that's my point. Noser took short cuts to deliver on their contract to Google. Those shortcuts where violated Sun's IP. Violating Sun's IP helped them get android out quicker and provided financial benefits.
Just because Oracle can only legally prove that there was a limited amount of IP theft going on, doesn't mean that it wasn't more widespread. That's what it makes me believe anyway.
Because car analogies are so much fun. If the car was a limo and Google was a limo service, Just because Google upgraded their car after finding out it was stolen doesn't mean Sun/Oracle isn't entitled to loses.
I made the mistake of lumping in this method with the other decompiled code and you seem to be making similar mistakes.
The 9 lines of code were copied. They were copied by a Google employee Josh Boch. Josh actually wrote the code that is in the JDK while working for google but on his own time. He assigned copyright for that code to Sun.
Google had expressly directed them not to do.
The reason Google expressly directed them not to is because they wanted a clean room implementation that would not be encumbered by Sun's IP.
They didn't get what they paid for. It may be the contractor's fault that Sun's IP was violated but Google did reap the benefits of that violation. If it was the contractor's fault that the Sun's IP was violated, Sun now Oracle still has a valid claim for damages. Oracle may want to seek remedy against their contractor for failing to deliver as promised.
The most interesting point is if Bloch did work on the code for OpenJDK and also worked on Andriod code, doesn't that throw the whole "cleanroom implementation" out the window?
"Judge William Alsup ruled that evidence presented during the trial had shown that Google infringed on Oracle’s copyrights by decompiling eight Java files and copying them in their entirety for use with Android."
As someone who really likes Java and Google I don't like this lawsuit one bit.
Sun created Java and mostly gave it away for free except for the mobile part which they were licensing to create revenue. Sun and Google couldn't come to terms regarding licensing and Google decided to just make it themselves.
Sun didn't seem to care, or didn't have the money/will to fight it. Then comes Oracle.
Please resolve this amicably soon. I don't like it when mommy and daddy fight.
He gave up a lot of his wealth, and even potential wealth to spend more time doing things he thought were more rewarding in other ways. So when he asks people to give up their share of the pie it's not a do as I say not as I do thing.
Slashdot Mirror
I agree. Still too soon to be publishing stories with sensational headlines about aircraft crashing into things in NY.
I'm not trying to be funny or sarcastic.
I don't believe in RAD. By the time anyone has developed a toolkit that can easily do something, it's not cool enough to do it anymore :)
RAD frameworks also tend to get bloated trying to do everything for everybody in my experience. At least when it comes to CRUD applications. I find myself more productive and have an easier time developing a site like that to do what I want by just using a good persistance (DAO, JDO, Hibernate, etc) framework or methodology that makes it easy to communicate with the DB.
Other stuff that i find useful are tools that build the classes you need to interact with various webservices.
Most sites using java don't serve JSP pages directly. They do it through some sort of MVC framework where servlets call the JSP behind the scenes. I've written sites where the url a user sees is a .html but it's actually just a call to a servlet that parses the url to server up the appropriate content and it uses a jsp for the presentation layer.
I'm pretty sure Blogger.com is written in java or at least uses java heavily.
This site I'm pretty sure uses Java because they used to use it in the past. The vendor they used is out of business and I would assume they switched to a different appserver but stuck with java. I could be wrong. An example url on their site is:
http://www.bhphotovideo.com/c/browse/Computers-Solutions/ci/9581/N/4294542559
"/c" is the controller servlet. The rest of the url provides information like url parameters that tell the controller what to display.
Struts is a common Java web framework and by default urls for struts applications end in ".do"
I agree with what yous say. Just adding information.
oops.. I thought the first post didn't go through.
Don't bother learning PHP it'll just slow you down learning the semantics of a new language and if you're used to dealing with Java PHP will feel like a joke. Java gets a bad wrap but I've found it to be faster than PHP in my tests.
I build most of my sites in Java using my own MVC framework. I've done some sites in PHP and have had to modify other PHP sites as well as looked into other languages. I still like Java the best and you can find a library to do almost anything you want. The only reason I'd pick something like PHP these days is if I don't want to build a site myself and want to use something prebuilt like wordpress or Joomla.
The only downside is that your servlet container (ie tomcat) is persistent and will take up a bit of memory. Not a huge amount but it makes it difficult to find cheap webhosting because providers can't throw thousands of websites on a server like they can with plain HTML and PHP. Try and find a good cheap VPS it's more secure and you won't have to worry about your site getting defaced because some other idiot didn't update their PHP software. That's happened to me.
Don't go with new frameworks. Go with popular ones that have been around for a while. I've been bitten in the ass when I built a website for a client and the framework I used was no longer around.
Spring is a good choice. I like to use NetBeans as my IDE. I've found it to work the best for me.
Almost all the sites I've built are written in Java. Stick with Java. I've written sites in PHP and I've also had to work on updates to some PHP sites. If you're already familiar with Java dealing with PHP will feel like a joke. PHP is great when you don't want to write your own software since there are so many publicly available stuff out there in PHP. Don't worry, you won't find a lack of Java libraries that will do anything you want to do.
Don't bother trying to learn a new language because you'll just slow yourself down trying to learn the semantics of the language instead of the details of the new libraries you'll be using. I know java gets a bad wrap in terms of performance but I've always found that Java kicks PHP's ass in terms of performance in the tests I've done.
The main issue with java is that when you're using a servlet container like Tomcat, the process runs constantly and takes up memory. It's not that much but it's hard to find Java hosting because the memory issues makes it hard for a webhost to put thousands of websites on the same server.
Your best bet is going to be to find a cheap VPN when you get started but check the big webhosts to. I remember LunarPages used to offer JSP support in the past.
There are a bunch of different frameworks. Stick to ones that are popular because you'd rather have some limited functionality now rather than an unsupported framework in the future. Which has happened to me.
I believe right now that's Spring but Struts is still pretty popular too.
I've found NetBeans to be a great IDE and it supports Spring.
It seems that Google+ will be connected with many systems and a Google+ account will be a very important if you put content online that you want to promote.
Want to get more clicks from google searches? Set up your site so it connects to your Google+ account so your picture can be included in the search pages.
I'm not saying if it's good or bad but it's different.
I'm imagining an internet where trade groups disallow trade-group TLDs to any business that doesn't pay them dues. Not subscribed to the RIAA? No .singer, .artist, .music, .whatever for you. Not in the BSA? No .software...
Yes throw in two industry groups that everyone on here hates to make this sound like a bad idea.
The interenet already has restricted TLDs so this concept isn't new. When you visit a .gov site you know you're dealing with a government website. Starting in 2001 .edu domains also were restricted to accredited schools.
Who controls access to these TLDs is important but it makes sense to have it be some sort of industry group that can fairly manage it. .singer, .artist, .music, .software aren't being discussed and the reason for .bank and .insurance are very different. Millions, maybe billions of dollars are lost annually to phishing schemes in this sector and having restricted TLDs can help reduce that number.
Considering the #1 most repeated rule for general PC users is "if it doesn't contain just your bank's website and end in .com, it's fake," I'm not sure they're doing themselves any favors. A year from now if I can an e-mail from MyBanksName.bank, even I'd assume it's a fake and not click it.
The flaw is that anyone can get a .com domain name so I could go and by a domain name like Citi-Bank-Secure-For-Realz.com and some people would fall for it. I mean it does say it's for realz!
If the .bank and .insurance TLDs were restricted by an industry group that assured only legitimate banks and insurers could purchase domains in that space, I can see how it would increase trust as long as consumers are made aware of the new changes. If it does happen there would be a big push to raise awareness through direct contact with their customers and the media.
I think this makes a lot of sense.
All those people that got lasik eye surgery will now be kicking themselves.
Very true.
If it wasn't for email I wouldn't even know about all these new social networks that are constantly springing up.
I agree IE isn't that hard to support anymore. I build websites for companies in various sectors including corporate clients. IE is still the dominant browser in some segments but not as much as it used to be.
For sites where the users tend to be more tech savy, and browse mainly from home IE isn't that popular. Maybe around 30-40% of these users are using IE.
For the business sites where visitors are generally browsing the site from work for business related reasons IE is still very strong. Around 60% and the most popular version is one version of the latest because of how software on corporate desktops is managed. It used to be a lot worse. 2-3 years ago IE was at around 90% for these client sites.
IE is still an important browser to support but if you limit support for certain areas, such as the admin area the site in TFA does it for, then you can be more restrictive on browser support because you have a limited set of users accessing those pages.
If you build websites for people that will be accessing the sites primarily from work, ignoring IE is shooting yourself in the foot.
no one is arguing against that point. So why do you keep raising it?
Not sure why it's such a sticking point so let me be clearer. Obviously when I refer to "Google" I'm talking broadly about the organization, considering a company has no free will and can not choose to do anything. Specifically I'm talking about the part of the organization that worked on Android. In this group I'm also including any contractors Google or it's subsidiaries might have hired. Instead of spending time trying to track down all the people that were involved with the code inside and outside of Google I just use the word "Google" to lump them all together. Because as I said, in the end it's their responsibility.
And I didn't infer anything from that line you're referring to. I was asking a question hoping to get an answer.
9 lines of what everyone thinks is trivial code would be faster to type than it would be to look up to copy and paste.
And the 9 lines is less than 2% of the code that was found to be infringing.
What I am inferring is that if Google didn't know that any of the code was infringing, that they knew that they told their contractor to develop a clean implementation, that information was likely communicated to Sun (I seem to recall some sort of conversation taking place but don't have a link) then any implied license Google thinks they had because of Sun's affirmative acts shouldn't be a factor because they were based on the belief by both parties that the code was not infringing.
Ironically doing a Google search for "Adnix and Preachnix" and clicking on the first page gets you a page full of ads, a popup and a warning message when you try and browse away. :)
Local broadcasts are free. Well sort of. You can still get over the air local channels most places but maybe 80-90%+ of people receive their local channels through cable tv, satellite, fios, whatever.
These companies can't just tap into the local airwaves and rebroadcast these channels. They have to pay for it and ever year or so there's a major issue with the contract of some channel holding out for more money.
The traditional broadcasting model is dead. The new one is get paid by advertisers, get paid again by distributors.
Though I can imagine dish caving just like tivo did.
I take it assembly and microprocessors weren't part of your education?
Because it doesn't matter whether it was a contractor or not. Google is ultimately responsible for abiding by the law with the products they release whether they were written by employees, obtained through an acquisition or created for them by a third party at their request.
I'm not talking about rangeCheck specificially but the entirety of code that the judge ruled was decompiled and copied from Java.
That leaves the question of whether the contractor copied other parts of the Java that are not as easy to identify or if they decompiled and copied then rewrote to make it look like it wasn't copied. Maybe they didn't do that with this function because it was trivial. Who knows. But to me this just doesn't look good.
I don't see Oracle as the bad guy here. They had their (Sun's) IP stolen and are right to seek damages.
I don't see Google as a bad guy here either. Their intention was not to violate Sun's IP.
Things got screwed up and now need to be resolved. In a perfect world Sun would have successfully negotiated a deal when Google initially approached them. That didn't happen and now here we are.
This suit wasn't to stop Google from using the car any longer, but to seek billions in damages for misuse of the car during a time that Google thought they owned it (due to deception by Noser)
Uhm... that's my point. Noser took short cuts to deliver on their contract to Google. Those shortcuts where violated Sun's IP. Violating Sun's IP helped them get android out quicker and provided financial benefits.
Just because Oracle can only legally prove that there was a limited amount of IP theft going on, doesn't mean that it wasn't more widespread. That's what it makes me believe anyway.
Because car analogies are so much fun. If the car was a limo and Google was a limo service, Just because Google upgraded their car after finding out it was stolen doesn't mean Sun/Oracle isn't entitled to loses.
when Oracle didn't actually own it, either.*
Source?
Seriously, what kind of code do you expect them to write after decompiling a function like rangeCheck?
If they wanted a clean room implementation they should be decompiling code at all.
I made the mistake of lumping in this method with the other decompiled code and you seem to be making similar mistakes.
The 9 lines of code were copied. They were copied by a Google employee Josh Boch. Josh actually wrote the code that is in the JDK while working for google but on his own time. He assigned copyright for that code to Sun.
Google had expressly directed them not to do.
The reason Google expressly directed them not to is because they wanted a clean room implementation that would not be encumbered by Sun's IP.
They didn't get what they paid for. It may be the contractor's fault that Sun's IP was violated but Google did reap the benefits of that violation. If it was the contractor's fault that the Sun's IP was violated, Sun now Oracle still has a valid claim for damages. Oracle may want to seek remedy against their contractor for failing to deliver as promised.
It doesn't really matter who decompiled the files. The point is that they were decompiled and copied.
If I hire you to go buy me a car and you do so I pay you for it. If it turns out you stole the car, should I get to keep it?
That also leaves the question of whether that contractor, or other contractors had violated Sun/Oracle copyrights in less obvious ways.
I am not even convinced that Google copied those 9 lines.
Then you might want to try reading at least the text of the link I posted if not the linked court documents themselves.
And maybe this 9 lines of code don't seem like a big deal? What about the other hundreds of lines that were decompiled and copied? There's no story about that? This filing has some more information. http://docs.justia.com/cases/federal/district-courts/california/candce/3:2010cv03561/231846/1157/
The most interesting point is if Bloch did work on the code for OpenJDK and also worked on Andriod code, doesn't that throw the whole "cleanroom implementation" out the window?
What does it matter how easy the code was to write? And if it was so easy why did google need to copy it?
I tried to submit this over the weekend but it wasn't selected. http://www.wired.com/wiredenterprise/2012/05/google-oracle-decompile/
As someone who really likes Java and Google I don't like this lawsuit one bit.
Sun created Java and mostly gave it away for free except for the mobile part which they were licensing to create revenue. Sun and Google couldn't come to terms regarding licensing and Google decided to just make it themselves.
Sun didn't seem to care, or didn't have the money/will to fight it. Then comes Oracle.
Please resolve this amicably soon. I don't like it when mommy and daddy fight.
Found some interesting quotes from Wozniak related to apple and jobs.
He gave up a lot of his wealth, and even potential wealth to spend more time doing things he thought were more rewarding in other ways. So when he asks people to give up their share of the pie it's not a do as I say not as I do thing.