You know something, Coward, you're so right! What was I thinking? Desktop Metro apps are sweeping the marketplace. Across the globe, enterprises are chucking their vast investments in Web Apps and building desktop Metro Apps using.NET. Thank you for bringing me to my senses, Coward. Yes, I think you're right. I need to rush right out and buy up a bunch of books on.NET and get started right away, learning the language of the future,.NET and the CLR. Thank you! You're my salvation. You've saved me from a career steeped in the massive Java stack and the occasional RubyOnRails. All of those are going in the trash--Coward. You're so right, Coward! I've been such a fool to see the market dominance of Java as being meaningful. The future is Metro, Metro, Metro and.NET is along for the ride. Yay! Thank you for helping me to see the light!
Funny how your remarks are self contradictory. If you're writing in HTML and JavaScript, where do they run? Generally, that means an application server. But if you're writing standalone programs well yes they don't need an app server. But very few of us actually write stand alone apps that don't need some app server to run in.
Secondly, the point of bringing up Metro is the effect it will have on.NET, which means making it obsolete. In case you haven't noticed, nobody but MS is pushing.NET. So, if your primary champion gets on another horse, the future of.NET is not bright. Yes, everybody says they use the "best tool for the job" and that's why there are so many Scala jobs out there. That's why so many companies are clamoring to write SmallTalk apps. I worked for a big company in NYC that used Flex until we discovered it sucked and wasn't up to the task and then we switched our front end to Ruby on Rails which was awesome. So, I in no way am implying that Java is the only game in town. Instead, my point is that.NET is not a language with a bright future ahead--and that's why I would not recommend the used book store guy at the head of this article keep around a lot of old.NET books. Point made. Time for more Anonymous Cowards without the courage to write under their real handles to swarm.
Funny, if Java is such a terrible mess then why oh why is it kicking.NET's ass in the marketplace? Why are companies--including even Microsoft--dumping the entire CLR like last week's cottage cheese? Java is not perfect but it gets the job done and all across the country it is the language of choice. It's fast enough and secure enough unless you code like an idiot and are still 4 versions behind.
Finally, yet another meaningless AC, what application server do you stand up in a meeting and propose you use in a MS Shoppe, if not ISS? I've seen it happen before. A project takes on one tiny slice of the Borg from Redmond and before you know it, they have been forced to deploy the whole Redmond stack. It's comical. And I notice that you have not addressed the elephant in the room--Metro. Your own company--MS--has pitched you.NETers over the side. That had to smart.
Every single item you listed is crippleware.
On the Java side, for example, you get production class versions of everything.
Express = Crippled and you know it.
Ultimately, if a developer uses your crippleware to build an app, her company is just going to have to shell out big bucks to MS to buy expensive tool sets. I notice you did not list IIS on your list of CrippleWare. But I'm sure there's a crippled version of ISS floating around somewhere too.
Okay, here's why. As soon as you're using.NET and ASP.NET, you're informed that you have to use IIS. Oh, and IIS really only works well with SQL Server. And then and then...
Dude, nobody gets hired anymore for knowing "basic programming". Also, in most enterprises that I have worked in, mobile apps are novelties. Nobody really bets the farm on a mobile app unless that's their entire business. Enterprises have bigger fish to fry than making another Angry Birds.
In NYC, having.NET on your resume was considered a black mark--because it meant you were a button pusher who didn't understand what was actually happening. It meant you were akin to a script kiddy: Why we don't hire.NET coders
Dude,
If you read my first post, I prefaced it with "My opinion".
Second, I've never had a job that expected anyone to know both Java and.NET. Java and.NET are such massive languages--with all the ancillary technologies you must know to be competent--that I wouldn't expect anyone to know both. But knock yourself out.
Third, I have never been asked to spend a dime on anything I was learning. Learning "stuff" does not seem that fruitful to me.
Lastly, where do you get off saying I'm fricking dictating to you--AC? Learn whatever the hell you learn. But I really can't understand why you would want to start learning.NET now since Metro is shit-canning it for HTML5 and Javascript. But it's your life. Enjoy your obsolescence.
Yeah, but the entire Microsoft stack is expensive. It's not the basis for learning. If you need to learn stuff, you should go Java where every damn last thing is free. Eclipse, app servers, everything.
And frankly--I just cannot take seriously anything said by someone who comments as AC.... Ballmer, you're retiring.Give it up. Move along. These are not the droids you're looking for.
This is my opinion.
Java--anything that doesn't say Java2 keep.
Spring -- anything
Application servers--keep anything.
Anything Windows--pitch. Anybody buying used books won't be able to afford Visual Studio.
Anything A+ -- pitch. Don't encourage that dead end.
Anything Networking--pitch, another dead end.
Anything design related--keep.
How amusing to have an Anonymous Coward telling me, a person who has used his real name for every single post, that I'm hiding behind Slashdot. Furthermore, on request I immediately posted the actual bug report I filed with Mozilla, thereby giving a second confirmation of my identity.
So, Anonymous Coward, I'm not sure what part of your hypocrisy I should focus on: that you are too scared to give your real name, that you challenged someone who was obviously speaking of a real issue that I have proven was a real issue, or that you can't deliver an intellectual argument without resorting to whimpering. I will leave it as an exercise for you to decide which of those factors are the best description of your opinions and unwillingness to come out of the shadows. But I do know that I will never cut any slack to you, an Anonymous Coward. Thank you for a laugh.
Not in the slightest. The blame is given to the application that is consuming the memory. That would be the browser in this case or the application running on an OS in your second example. This bug has manifested itself in both Windows and OS X, in my case. (And the code base is shared mostly for both so that rules out it being an OS problem.
To answer your essential question--yes--without question an application should be written to be bulletproof. There should be no use case possible that causes your application to eat up unlimited memory. That is a defect, pure and simple.
I defy you to make your argument in any boardroom across the United States. I can see you making a presentation to a board of executives, explaining how your application did this or that horrid thing because--it was the web page's fault. Do you know what they call engineers who make that argument? Unemployed.
As you can see from the bug report that I personally participated in creating, including providing before and after memory dumps, it's clear I am invested in getting this Firefox memory leak fixed. It is plaguing users across the globe. Read the complaints on Mozilla's forum.
As for your comment that I'm making a mountain out of a mole hill--Firefox has been documented to consume 3GB of RAM in about 10 minutes. Other users who had left it open and idle over night have seen the same result.
Finally, I am expressing my point of view forcefully because you are trying to poo-poo my concern. This is a serious show-stopping defect in the application Firefox and it needs to get fixed unless Firefox wants to die a slow painful death.
How is it that any site is able to make an application manifest a memory leak?
If you have any experience as a software developer, you know applications should be bullet proof. They should not have a vulnerability sitting around, waiting for some site to hit the correct use case to manifest it. A bug is a bug and the particular site that causes that pre-existing bug to manifest itself is not relevant. There should be no use case that causes Firefox--a browser used by billions of people world wide--to throw a memory leak this bad.
Furthermore, if you went to Mozilla's site, you will see that I have only the best documented example of this bug. There are many other co-reporters of the same problem.
But in my long experience as a developer, whenever the author of some code starts to blame the victim, I know they have a bug that they do not want to acknowledge or fix. This is a memory leak and nothing at all can change that other than finding and fixing the bug. Do you think they would have accepted this as a bug--and taken the memory-usage maps of before and after memory consumption--unless it was a legitimate bug? No. This is real and the memory maps prove it.
Yes, I have had a currently open bug with FF21.0--that got worse with 22.0. I have been fully co-operative and helpful for months as they work to resolve it.
I uploaded memory dumps of before and after. In the after state it was taking 2GB of RAM after TWO minutes.
And I and the other watchers of the bug I opened at Mozilla will dispute your contention that Chrome uses more memory. Simply not true! Chrome with its process-per-tab manages memory much better than FF does.
As I said, this bug has been validated by DEVs at Mozilla and they have admitted my use case exposes a valid problem that they themselves have been able to replicate.
Next troll?
I won't be downloading any new versions of Firefox--nor will I enable automatic updates--until they fix the danged memory leaks that have been present since they began their whirlwind upgrade cycle with FF 4.0. Chrome is a handy replacement for what used to be a reliable friend--Firefox.
Not a valid example. I can and do have several technical books open at the same time, side by side, with the ability to refer to one right after the other. For example, right at this moment I have "Definitive Guide to CentOS" open and right next to that "Spring Security" and next to that "MySQL" (5th Ed). Each of them is open to a specific page.
And you know that having to go find the other bookmark really follows the "out of sight, out of mind" principle. Your example fails. Any one of those books I could take to the bathtub to read--something you would never do with a Kindle--but something books handle easily. (If they get wet along the bottom edge, no biggie.)
I don't play any games so I can't speak to that issue.
I do have two iMacs with three screens for the exact reason I specified. One will have Eclipse open, the second may have a stackoverflow page and a third email or a movie. Three screens.
Slashdot Mirror
I did two regular VHS tapes for about $30. Good quality. They were tapes I made in Russia when I lived there, hence: priceless.
Get a COSTCO membership and they digitize videos. I had it done with some precious ones and they did a great job.
I have a site that is free and has three semesters of Java lectures called: http://sites.google.com/site/f...
I have a Java site: http://sites.google.com/site/f... They're pretty self explanatory. Three semesters worth.
Fisher Price is the only maker of a phone you should give to a 4-year old.
You know something, Coward, you're so right! What was I thinking? Desktop Metro apps are sweeping the marketplace. Across the globe, enterprises are chucking their vast investments in Web Apps and building desktop Metro Apps using .NET. Thank you for bringing me to my senses, Coward. Yes, I think you're right. I need to rush right out and buy up a bunch of books on .NET and get started right away, learning the language of the future, .NET and the CLR. Thank you! You're my salvation. You've saved me from a career steeped in the massive Java stack and the occasional RubyOnRails. All of those are going in the trash--Coward. You're so right, Coward! I've been such a fool to see the market dominance of Java as being meaningful. The future is Metro, Metro, Metro and .NET is along for the ride. Yay! Thank you for helping me to see the light!
Funny how your remarks are self contradictory. If you're writing in HTML and JavaScript, where do they run? Generally, that means an application server. But if you're writing standalone programs well yes they don't need an app server. But very few of us actually write stand alone apps that don't need some app server to run in. .NET, which means making it obsolete. In case you haven't noticed, nobody but MS is pushing .NET. So, if your primary champion gets on another horse, the future of .NET is not bright. Yes, everybody says they use the "best tool for the job" and that's why there are so many Scala jobs out there. That's why so many companies are clamoring to write SmallTalk apps. I worked for a big company in NYC that used Flex until we discovered it sucked and wasn't up to the task and then we switched our front end to Ruby on Rails which was awesome. So, I in no way am implying that Java is the only game in town. Instead, my point is that .NET is not a language with a bright future ahead--and that's why I would not recommend the used book store guy at the head of this article keep around a lot of old .NET books. Point made. Time for more Anonymous Cowards without the courage to write under their real handles to swarm.
Secondly, the point of bringing up Metro is the effect it will have on
Funny, if Java is such a terrible mess then why oh why is it kicking .NET's ass in the marketplace? Why are companies--including even Microsoft--dumping the entire CLR like last week's cottage cheese? Java is not perfect but it gets the job done and all across the country it is the language of choice. It's fast enough and secure enough unless you code like an idiot and are still 4 versions behind.
.NETers over the side. That had to smart.
Finally, yet another meaningless AC, what application server do you stand up in a meeting and propose you use in a MS Shoppe, if not ISS? I've seen it happen before. A project takes on one tiny slice of the Borg from Redmond and before you know it, they have been forced to deploy the whole Redmond stack. It's comical. And I notice that you have not addressed the elephant in the room--Metro. Your own company--MS--has pitched you
Every single item you listed is crippleware.
On the Java side, for example, you get production class versions of everything.
Express = Crippled and you know it.
Ultimately, if a developer uses your crippleware to build an app, her company is just going to have to shell out big bucks to MS to buy expensive tool sets. I notice you did not list IIS on your list of CrippleWare. But I'm sure there's a crippled version of ISS floating around somewhere too.
Okay, here's why. As soon as you're using .NET and ASP.NET, you're informed that you have to use IIS. Oh, and IIS really only works well with SQL Server. And then and then...
Dude, nobody gets hired anymore for knowing "basic programming". Also, in most enterprises that I have worked in, mobile apps are novelties. Nobody really bets the farm on a mobile app unless that's their entire business. Enterprises have bigger fish to fry than making another Angry Birds. .NET on your resume was considered a black mark--because it meant you were a button pusher who didn't understand what was actually happening. It meant you were akin to a script kiddy: Why we don't hire .NET coders
In NYC, having
Dude, If you read my first post, I prefaced it with "My opinion". .NET. Java and .NET are such massive languages--with all the ancillary technologies you must know to be competent--that I wouldn't expect anyone to know both. But knock yourself out.
.NET now since Metro is shit-canning it for HTML5 and Javascript. But it's your life. Enjoy your obsolescence.
Second, I've never had a job that expected anyone to know both Java and
Third, I have never been asked to spend a dime on anything I was learning. Learning "stuff" does not seem that fruitful to me.
Lastly, where do you get off saying I'm fricking dictating to you--AC? Learn whatever the hell you learn. But I really can't understand why you would want to start learning
Well, sounds like crippleware to me. On the Java side free gets you the best stuff--not crippleware.
Yeah, but the entire Microsoft stack is expensive. It's not the basis for learning. If you need to learn stuff, you should go Java where every damn last thing is free. Eclipse, app servers, everything.
And frankly--I just cannot take seriously anything said by someone who comments as AC.... Ballmer, you're retiring.Give it up. Move along. These are not the droids you're looking for.
This is my opinion.
Java--anything that doesn't say Java2 keep.
Spring -- anything
Application servers--keep anything.
Anything Windows--pitch. Anybody buying used books won't be able to afford Visual Studio.
Anything A+ -- pitch. Don't encourage that dead end.
Anything Networking--pitch, another dead end.
Anything design related--keep.
How amusing to have an Anonymous Coward telling me, a person who has used his real name for every single post, that I'm hiding behind Slashdot. Furthermore, on request I immediately posted the actual bug report I filed with Mozilla, thereby giving a second confirmation of my identity.
So, Anonymous Coward, I'm not sure what part of your hypocrisy I should focus on: that you are too scared to give your real name, that you challenged someone who was obviously speaking of a real issue that I have proven was a real issue, or that you can't deliver an intellectual argument without resorting to whimpering. I will leave it as an exercise for you to decide which of those factors are the best description of your opinions and unwillingness to come out of the shadows. But I do know that I will never cut any slack to you, an Anonymous Coward. Thank you for a laugh.
I am not the only person suffering with this bug.
And it only makes sense for the owners of programs to fix them.
Finally, son, this is slashdot. If you can't handle the heat, run off to Reddit.
Not in the slightest. The blame is given to the application that is consuming the memory. That would be the browser in this case or the application running on an OS in your second example.
This bug has manifested itself in both Windows and OS X, in my case. (And the code base is shared mostly for both so that rules out it being an OS problem.
To answer your essential question--yes--without question an application should be written to be bulletproof. There should be no use case possible that causes your application to eat up unlimited memory. That is a defect, pure and simple.
I defy you to make your argument in any boardroom across the United States. I can see you making a presentation to a board of executives, explaining how your application did this or that horrid thing because--it was the web page's fault. Do you know what they call engineers who make that argument? Unemployed.
As you can see from the bug report that I personally participated in creating, including providing before and after memory dumps, it's clear I am invested in getting this Firefox memory leak fixed. It is plaguing users across the globe. Read the complaints on Mozilla's forum.
As for your comment that I'm making a mountain out of a mole hill--Firefox has been documented to consume 3GB of RAM in about 10 minutes. Other users who had left it open and idle over night have seen the same result.
Finally, I am expressing my point of view forcefully because you are trying to poo-poo my concern. This is a serious show-stopping defect in the application Firefox and it needs to get fixed unless Firefox wants to die a slow painful death.
How is it that any site is able to make an application manifest a memory leak?
If you have any experience as a software developer, you know applications should be bullet proof. They should not have a vulnerability sitting around, waiting for some site to hit the correct use case to manifest it. A bug is a bug and the particular site that causes that pre-existing bug to manifest itself is not relevant. There should be no use case that causes Firefox--a browser used by billions of people world wide--to throw a memory leak this bad.
Furthermore, if you went to Mozilla's site, you will see that I have only the best documented example of this bug. There are many other co-reporters of the same problem.
But in my long experience as a developer, whenever the author of some code starts to blame the victim, I know they have a bug that they do not want to acknowledge or fix. This is a memory leak and nothing at all can change that other than finding and fixing the bug. Do you think they would have accepted this as a bug--and taken the memory-usage maps of before and after memory consumption--unless it was a legitimate bug? No. This is real and the memory maps prove it.
Here is the bug: https://bugzilla.mozilla.org/show_bug.cgi?id=896016
https://bugzilla.mozilla.org/show_bug.cgi?id=896016
Whether Chrome temporarily uses more RAM is not the point. I have never seen Chrome get into a runaway 2-3GB memory leak like so frequently happens to FF https://bugzilla.mozilla.org/show_bug.cgi?id=896016
Yes, I have had a currently open bug with FF21.0--that got worse with 22.0. I have been fully co-operative and helpful for months as they work to resolve it. I uploaded memory dumps of before and after. In the after state it was taking 2GB of RAM after TWO minutes. And I and the other watchers of the bug I opened at Mozilla will dispute your contention that Chrome uses more memory. Simply not true! Chrome with its process-per-tab manages memory much better than FF does. As I said, this bug has been validated by DEVs at Mozilla and they have admitted my use case exposes a valid problem that they themselves have been able to replicate. Next troll?
Thank you for some solace and good ideas, friend. Yes, Chrome is creepy too. I will explore your suggestions. Thank you and wishing you up mods!
I won't be downloading any new versions of Firefox--nor will I enable automatic updates--until they fix the danged memory leaks that have been present since they began their whirlwind upgrade cycle with FF 4.0. Chrome is a handy replacement for what used to be a reliable friend--Firefox.
Not the same. You ignored my explanations of why above--coward.
Not a valid example. I can and do have several technical books open at the same time, side by side, with the ability to refer to one right after the other. For example, right at this moment I have "Definitive Guide to CentOS" open and right next to that "Spring Security" and next to that "MySQL" (5th Ed). Each of them is open to a specific page. And you know that having to go find the other bookmark really follows the "out of sight, out of mind" principle. Your example fails. Any one of those books I could take to the bathtub to read--something you would never do with a Kindle--but something books handle easily. (If they get wet along the bottom edge, no biggie.) I don't play any games so I can't speak to that issue. I do have two iMacs with three screens for the exact reason I specified. One will have Eclipse open, the second may have a stackoverflow page and a third email or a movie. Three screens.