Visual Basic is the worst, but yes gainful employment is great; it is even better when you don't have to wash dishes professionally.
They're perfectly aware I've never touched .NET before too. I'll be proactive but I'll ask if they're willing to train me too.
I don't think anything on my campus teaches VB. I know it's damn old.
VB.net isn't the same thing as the traditionally and rightfully hated VB6. VB.net is the same terrible syntax but with the nice .net api and features since it runs on .net.
I wonder what the employer is referring to when he says "we like .NET". It's not necessarily synonymous with "we like C#" but I went ahead and asked.
All my fancier data structures are in C and Java, and all my user interfaces are...well...websites.
.NET is, as much as anything, an exercise in marketing. Bosses saying "we like .NET" could just mean "hey, this kool-aid tastes pretty good."
From a web perspective, .NET provides a full Microsoft stack. Instead of Apache/Tomcat/Java/JSP, you'd have IIS/C#/F#/VB.NET/ASP.NET. It's a decent set of alternatives - I've found that Java/JSP ports over pretty easily to C#/ASP.NET.
He says it's VB.NET that they use, but that learning C# is good too.
In my extremely short ventures into the world of .NET, VB.NET looked a lot like a skin over C# written to verbose-ify the syntax. It seemed odd that anyone would use it, unless they were porting an application from VB to .NET ten years ago and feared anything that didn't remind them of VB6.
I could not think of a way of doing it without supplying 4 in G.
This is actually a really awesome mind-exercise.
To perform the computation without supplying 4 in G, you can grow {S(n)} by next immediate element before growing it with the remaining full range, but that adds an extra call to append each time through the loop. This isn't terribly bad because append is O(1) in python, but it definitely looks oddish. In reality since the sequence {S(n)} out-paces the {R(n)} sequence after n=2 (assuming n=1 for the start), I'd could argue that defining S(1)=2, and S(2)=4 as initial conditions is the more elegant way to define this programmatically.
None the less, I've adapted the solution anyway to provide an alternative (Note: Don't be concerned with the calls to len(), it is O(1), too):
F = [1]
G = [2]
for Fi in range(1,20):
F.append(F[Fi-1] + G[Fi-1])
G.append(F[Fi]+1)
G[len(G):] = range(G[len(G)-1]+1, F[Fi]+G[Fi])
print F
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
edited April 2012
I figured there had to be a way of doing it I just wasn't seeing it. The loop might have another line but it reinforces the idea that the sequences are generated inductively.
What I really like about your solution is that allows you to get rid of the extra index. I might rewrite it as
F = [1]
G = [2]
for i in range(1,20):
F.append(F[i-1] + G[i-1])
G.append(F[i]+1)
G.extend(range(G[-1]+1, F[i]+G[i]))
print F
Depending on if you find the idea of a -1 index intuitive.
Monkey Ball Warrior on
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
Depending on if you find the idea of a -1 index intuitive.
It doesn't make a lot of sense as an index, but as an offset.
Run you pigeons, it's Robert Frost!
0
jackalFuck Yes. That is an orderly anal warehouse.Registered Userregular
VB.Net is fine. The syntax isn't even that bad. The only real problem is everything is parentheses. While C# uses angle brackets for generics and parens for function calls, and brackets for indexes VB.Net uses parens for all of them. The lambda syntax is a bit verbose. "=" is both equality and assignment, but it's never ambiguous. I prefer C# to it, but being practically identical to C# makes it a lot better than a lot of other languages.
To answer the question as to why anyone would use it, it was much better with COM interop. C# has caught up in recent versions, and both languages are at parity, but it wasn't always like that. If you use C# there is no reason to not know VB.Net. All the language idioms are the same.
I really wish it didn't take so much work just to get a deliverable deadline out of someone. It seems like most people think "ASAP" is a good answer. If I just took "ASAP" as an acceptable answer every time and put their project beneath the 10+ other "ASAP"s in my list, it would take months to get to it. Obviously everyone wants their thing finished as fast as I can get to it, but depending on the urgency, that could be a day or it could be weeks or months away.
I really wish it didn't take so much work just to get a deliverable deadline out of someone. It seems like most people think "ASAP" is a good answer. If I just took "ASAP" as an acceptable answer every time and put their project beneath the 10+ other "ASAP"s in my list, it would take months to get to it. Obviously everyone wants their thing finished as fast as I can get to it, but depending on the urgency, that could be a day or it could be weeks or months away.
Figuring out schedules and deciding on priorities? That's pretty much exactly what your manager should be doing. Pass the buck. :P
0
thatassemblyguyJanitor of Technical Debt.Registered Userregular
I figured there had to be a way of doing it I just wasn't seeing it. The loop might have another line but it reinforces the idea that the sequences are generated inductively.
What I really like about your solution is that allows you to get rid of the extra index. I might rewrite it as
F = [1]
G = [2]
for i in range(1,20):
F.append(F[i-1] + G[i-1])
G.append(F[i]+1)
G.extend(range(G[-1]+1, F[i]+G[i]))
print F
Depending on if you find the idea of a -1 index intuitive.
Oh man! extend! yesssss, that is the more pythonic way of doing that statement, for sure. Personally, I'm ambivalent to the -1 notation; one one hand it is less characters to write, but on the other hand, like Seabass mentioned, it doesn't make sense as an index. Oh, well, I like the final flow and structure you have for that solution.
So my instructor gave us an extra "for fun" lab yesterday. It is supposed to ask for a number of assignments, then loop through and have you enter a students score and the max possible score. It then needs to figure out which three assignments to drop t o provide the largest improvement to the students grade and output the new grade percentage.
Figuring it out is fairly easy, since all points are weighted equally you just drop the three assignments that have the highest differential between max score and students score. I solved it at first with array lists, since that's what we've been studying and when you have a brand new hammer....... She told me I nuked it and that it could be done with just variables and three nested for loops. I went back and did it with a for loop and if/else chain, but I can't for the life of me see how you would do it with three for loops. Any pointers?
So my instructor gave us an extra "for fun" lab yesterday. It is supposed to ask for a number of assignments, then loop through and have you enter a students score and the max possible score. It then needs to figure out which three assignments to drop t o provide the largest improvement to the students grade and output the new grade percentage.
Figuring it out is fairly easy, since all points are weighted equally you just drop the three assignments that have the highest differential between max score and students score. I solved it at first with array lists, since that's what we've been studying and when you have a brand new hammer....... She told me I nuked it and that it could be done with just variables and three nested for loops. I went back and did it with a for loop and if/else chain, but I can't for the life of me see how you would do it with three for loops. Any pointers?
Remember that you don't have to iterate through a nested loop every time it's encountered. You can use the loop conditional to check both if every index in the array has been iterated through or if a variable has been set, for example.
Does anyone here have experience with the simple directmedia layer library (pref. version 1.2). I'm wondering what happens when too many keys are pressed at once on a keyboard. Specifically I'm curious if it is possible for a key to become unpressed without a corresponding SDL_KEYUP event.
I'm real excited that I just found out about this thread. As a professional programmer with not much experience (degree, internship + a little over a year of full time work), I can definitely pick up some pointers from some more experienced people. I learned C++, migrated to C# for my job, and I'm currently self teaching myself Python, which I have found to be quite fascinating. I assume that there's a lot of Python people in here, so I'll probably be sharing some of my growing pains into the language.
I have a programmer friend who has about six months more of professional experience than I do, and skill wise, his company has advanced him quite farther than my company has for me. We're a small company (our programming team consists of three people) and we do things sort of quick and dirty, which is nice, because it gives me a wide range of responsibility, while essentially being trusted to be able to make things work. He does web dev, and I do desktop applications, but he's using a crazy amount of the more "advanced" aspects of C#, like singletons and what not. The difference in skill set just really bothers me for some reason, but seeing as I don't have exposure to any other company, I can't tell what I should know, skill wise, at my experience level and age; should I be worried about that?
GnomeTankWhat the what?Portland, OregonRegistered Userregular
VB.net, while certainly not good, is a fuck ton better than VB6, and is a full .NET language. The syntax is hokey as fuck, but it still has lambdas, anonymous types, generics, full OO, etc. It's certainly not as optimal as C#, but it's a full .NET language.
HalibutPassion FishSwimming in obscurity.Registered Userregular
I do professional Java and .NET development. I prefer Java. but my experience with both is obviously anecdotal, so take this post with a grain of salt.
From my perspective, .NET beats Java in some areas (language features and standalone GUI development). But, for what I do for a living, it falls behind Java in more important ones: web development, open source libraries, and build/support/deployment tools.
ASP.NET is awful for anything that's not a toy website. It's a web framework that puts a wrapper around HTTP/HTML in order to present it as something it's not. As soon as you try to do anything advanced, you have major issues. The problem is that Microsoft's chief design decision seems to be to insulate you as much as possible from your platform (HTTP/HTML). And in order to do this, you need their development environment because it generates and keeps track of an ungodly amount of glue code. Just give me request/response please, and then make it easy for me to layer my own features from there (binding to form data, validation, templating, authentication, authorization, compression). Maybe their MVC thing is better, I don't know, I haven't tried it yet.
The libraries thing is fairly self-explanatory. Java's been around longer. It's more open than Microsoft's platform. It has large, open-source-devoted organizations behind it creating libraries for anything you could ever want. And not only is there a library for everything, there's probably 2-3. Pick the one that fits your needs. If you prefer a web framework like ASP.NET that keeps you from knowing about HTTP, you can have that. Or you can choose from literally 10 other mature web frameworks that might suit your style/needs.
The build/deployment tools thing is related. How many different application servers can you deploy an ASP.NET app on? 1 (maybe 2 if there's a Mono thing?). In Java, there's at least 4-5 major ones, and they're all mature, and all compatible with your application, because the Java Servlet spec is an open standard.
Performing automated builds in .NET is frustrating. MSBuild is like Ant, but somehow even more verbose. Your only other option is to install Visual Studio and run devenv.exe. There's no Maven/Ivy equivalent to do dependency management. .project files keep track of individual source file names instead of source directories, and are automatically generated/regenerated when you make a change in Visual Studio, so SCM tools unnecessarily warn about conflicts. Microsoft built everything around using their IDE, so doing anything outside of it is painful. On top of that, Visual Studio costs money, and is missing basic features like refactoring and quick navigation to class/method definitions. If you want a full-featured IDE, you have to buy something extra like Resharper.
Finally, my experience with Java enterprise ecosystems versus .NET enterprise ecosystems has been night and day. It's almost like the .NET side hasn't ever heard of the word "enterprise". Just a hodgepodge of stuff someone threw onto a Windows Server. But, I'm willing to admit that this experience is probably specific to the clients I do work for.
There's fundamental design differences between the 2 platforms. .NET's is to get you up and running quickly as a single developer (or very small team). Java's is to leave it mostly to you, to do it how you want, which comes at the expense of making it difficult to get started as someone new to the platform. .NET is a Microsoft technology with Microsoft tools and resources. Java is a Sun/Oracle technology, with professional and community tools and resources.
0
GnomeTankWhat the what?Portland, OregonRegistered Userregular
edited April 2012
ASP.NET MVC is actually a great framework. If you are still doing classic ASP.NET dev, that's a shame. There may be more Java libraries, but you'd be hard pressed to find a type of library that the .NET platform doesn't have. A lot of which are ports/retools of Java libraries...and you could just use Nant, which is literally Ant for .NET. You can deploy ASP.NET on IIS and Apache, and probably Light with a bit of work, for Windows, OSX and Linux. I've written and deployed several MVC websites to Linux servers hosted on Apache using Mono.
Visual Studio 2010 does not lack those features you list, it has had basic refactoring and quick navigate built in since 2008. There are certainly addins like ReSharper to enhance them, but they are built in. I can't argue with the fact that Visual Studio costs money, but Microsoft has released the capable Express versions for years that are certainly on par with IDE's like NetBeans, though I'll concede Eclipse is very good.
The fact that you think .NET is just some hobbyist/single developer platform and Java is some ultra slick professional platform is telling. Basically, I can sum up your entire post as "I don't really know or understand the .NET eco-system, thus I prefer Java because it's what I know. I was not even aware of basic, available, and popular .NET solutions to issues I posted."...which is 100% fine, but it's not even kind of sort of evidence of one platform being better than the other.
HalibutPassion FishSwimming in obscurity.Registered Userregular
I totally get it. I haven't done any new .NET development in a year and a half (it's mostly maintenance stuff now), so I'm willing to concede that things have probably changed. We were on Visual Studio 2008 at the time, and the only refactoring support it had was for renaming methods/classes. You could do the same thing with search/replace. Navigating to method/class definitions shouldn't take more than 1 key/mouseclick combination. All the guys in my office who only do .NET stuff were surprised we were even asking for this stuff and basically just told us to get Resharper.
Nant was something we looked into, but keeping Visual Studio, and the build script working together wasn't fun. We switched to MSBuild because it seemed to be the standard way to do automated build stuff. We tried using NUnit for testing, but setting it up as an external tool in Visual Studio was a pain, as was setting it up in MSBuild, so we resorted to the built in testing tool (which tied us even more to Visual Studio). The verbose compilation unit model in .project files is just bizarre, and is what causes most of the problems. It's like there's no standard way to organize code/resources so it just enumerates everything explicitly. Lack of something like Maven is a real turn-off.
We quickly came to the understanding that even if there was a N-version of a Java library, it wasn't considered the "right" or "standard" way to do it in .NET, and as a result it normally had a small, only somewhat-active community around it. ASP.NET is awful, even compared to something like Struts. I'll take your word for it that MVC is better.
All of that is fine. I'm willing to admit that .NET stuff is different, and I'm also willing to admit that most of what I think is "better" is purely subjective. Some of the differences stem from the fact that I know how to do something in Java, and can't seem to find an equivalent or better way to do it in .NET. That's probably mostly on me, but partially on .NET and the .NET community/ecosystem.
And, I'm also willing to admit that a lot of the stuff that turns me off about .NET is purely based on initial bad experiences with it. It's probably similar to a lot of peoples complaints about JSP.
0
GnomeTankWhat the what?Portland, OregonRegistered Userregular
edited April 2012
I think Java gets a lot of crap around here, but I actually don't think it's a terrible platform. My issues with Java tend to be the lack of movement in the language (the committee moves glacially slow). Microsoft is very forward looking, with things like their Patterns & Practices group, that builds and releases working implementations of a lot patterns and practices (shockingly) that become defacto .NET "canon" if you will. A lot of these things end up rolled in to the base framework in someway. In fact, Microsoft has really embraced the open source thing with their bleeding edge stuff. A lot of what translates in to the .NET framework in someway, came from one of Microsoft's open source efforts (the WPFToolkit is one of the biggest examples, most of that was included in .NET 4.0). MSP&P was what spawned the original ASP.NET MVC, and has spawned professional grade frameworks I use every day at work (Unity for IoC/dependency injection and Prism for composite WPF applications).
I'll be the first to admit Microsoft hasn't gotten everything right, but overall, .NET is absolutely one of their successes of the last decade. They've grown a vibrant platform, and are pushing the edges with their open source work. They've embraced the open source Mono project to provide a complete and functional implementation of the .NET framework on other platforms, including running ASP.NET stuff on webservers other than IIS. I get the anomosity between the Java and .NET worlds, since .NET and C# were unabashedly born from the ashes of the Sun/Microsoft thermonuclear explosion...but in most senses, they are pretty comparable platforms now, with a lot of the difference being personal preference. With .NET, you're going to get a faster moving platform, that embraces new methodologies and patterns quickly, and adds new language and platform features relatively quickly. The speed of this change can be off-putting to some people. They feel like Microsoft is always changing the platform and they can't get their feet under them. Java is certainly an older and more stable platform, but it moves glacially slow and retains some interesting idiosyncrasies that are like archaeological markers for the 3GL -> 4GL transition era that Java represents.
Re: post warping stuff, I'm currently seeing my posts show up around 1-2 minutes ahead of the actual time, so it's definitely a server timestamp issue with whatever server I'm contacting to post
I think Java gets a lot of crap around here, but I actually don't think it's a terrible platform. My issues with Java tend to be the lack of movement in the language (the committee moves glacially slow). Microsoft is very forward looking, with things like their Patterns & Practices group, that builds and releases working implementations of a lot patterns and practices (shockingly) that become defacto .NET "canon" if you will. A lot of these things end up rolled in to the base framework in someway. In fact, Microsoft has really embraced the open source thing with their bleeding edge stuff. A lot of what translates in to the .NET framework in someway, came from one of Microsoft's open source efforts (the WPFToolkit is one of the biggest examples, most of that was included in .NET 4.0). MSP&P was what spawned the original ASP.NET MVC, and has spawned professional grade frameworks I use every day at work (Unity for IoC/dependency injection and Prism for composite WPF applications).
I'll be the first to admit Microsoft hasn't gotten everything right, but overall, .NET is absolutely one of their successes of the last decade. They've grown a vibrant platform, and are pushing the edges with their open source work. They've embraced the open source Mono project to provide a complete and functional implementation of the .NET framework on other platforms, including running ASP.NET stuff on webservers other than IIS. I get the anomosity between the Java and .NET worlds, since .NET and C# were unabashedly born from the ashes of the Sun/Microsoft thermonuclear explosion...but in most senses, they are pretty comparable platforms now, with a lot of the difference being personal preference. With .NET, you're going to get a faster moving platform, that embraces new methodologies and patterns quickly, and adds new language and platform features relatively quickly. The speed of this change can be off-putting to some people. They feel like Microsoft is always changing the platform and they can't get their feet under them. Java is certainly an older and more stable platform, but it moves glacially slow and retains some interesting idiosyncrasies that are like archaeological markers for the 3GL -> 4GL transition era that Java represents.
I think I agree with pretty much everything you said. Java the language is old, and can sometimes be painful to program in (especially compared to lot of more modern languages). But, it's tried and true, and the JVM and the rest of the Java ecosystem are really great. There's lots of libraries and ways of designing applications that can make Java (the language) fun and rewarding.
I definitely agree regarding pace of language evolution. I think I would actually prefer something in between Java and .NET (which is the only thing I really like about the Oracle takeover). Sometimes it seems like Microsoft moves faster than it thinks (I don't have anything to back this up, it's just a feeling). You can look at parts of C# and wonder if anyone asks "should we really do this?" Look at the number of reserved keywords in C# versus other languages, and you'll see a bit of what I'm talking about: http://stackoverflow.com/questions/4980766/reserved-keywords-count-by-programming-language. Obviously there's not a 1:1 correlation between language size/complexity and keyword count, but they're definitely related.
Java feels like it has the opposite problem. Like it's in a cycle of thinking to the point of inaction. For one thing, it's older, and strives to maintain binary backwards compatibility with previous compiled code. The first versions of Java hang over its head to this day. One of its key features is also one of its most frustrating: it doesn't trust developers to understand how to do complicated things. As a result, it's pretty easy to read and debug, but it makes writing code feel clunky and verbose at times. If Java had lambdas and reified types, I'd be pretty happy with the language (and then I'd ask for more ). Fortunately, lambdas are being planned for Java 8 (reified types, probably never).
The thing I really like about both Java (the VM), and the CLR is that they are producing so many new and interesting languages. Sure, you've got ports of existing languages like Python or Ruby, but you've also got totally unique things like Scala and F#.
Anyway, I was mainly trying to give a counterpoint to the ".NET is better than Java" arguments that I saw. In the end, most of these debates are anecdotal or based on personal preference, so it's good to know about other people's experiences.
Well, obviously they are... I'm presuming that as soon as it detects something like one of the SQL keywords there, it just drops the entire message.
Either that or the form is completely susceptible to injection attacks and they know it but don't know how to stop it other than asking people not to use those words.
Does anyone here have experience with the simple directmedia layer library (pref. version 1.2). I'm wondering what happens when too many keys are pressed at once on a keyboard. Specifically I'm curious if it is possible for a key to become unpressed without a corresponding SDL_KEYUP event.
@seabass I don't think so, I think it queues them up in order of how you facerolled them, assuming your buffer doesn't get overflowed. But I'd imagine the buffer is larger than 101+ keycodes.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Now I want to write a message that contains all of those words.
Brief attempt.
"Dear Congressman Peters, please ALTER the upcoming spending bill to DELETE the passage that DECLAREs that SELECT parents should not be able to DROP their insurance coverage. Thanks." 5/7 isn't bad.
i am so utterly tired of the request/response model and the baggage that comes with it
my ideal scenario is one where I can make programs that run entirely offline, or with only a single big web call to get some seed data or something, or do something like websocket programming, where the flow of code (to me at least) is more intuitive
i am so utterly tired of the request/response model and the baggage that comes with it
my ideal scenario is one where I can make programs that run entirely offline, or with only a single big web call to get some seed data or something, or do something like websocket programming, where the flow of code (to me at least) is more intuitive
Why don't you?
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Going to take the Java certification exam again on Saturday. I'm taking Friday off so I can study/cram all day. I'm so stressed right now... I cannot wait until this is over with.
0
GnomeTankWhat the what?Portland, OregonRegistered Userregular
Posts
They're perfectly aware I've never touched .NET before too. I'll be proactive but I'll ask if they're willing to train me too.
I don't think anything on my campus teaches VB. I know it's damn old.
Results:
I could not think of a way of doing it without supplying 4 in G.
In my extremely short ventures into the world of .NET, VB.NET looked a lot like a skin over C# written to verbose-ify the syntax. It seemed odd that anyone would use it, unless they were porting an application from VB to .NET ten years ago and feared anything that didn't remind them of VB6.
This is actually a really awesome mind-exercise.
To perform the computation without supplying 4 in G, you can grow {S(n)} by next immediate element before growing it with the remaining full range, but that adds an extra call to append each time through the loop. This isn't terribly bad because append is O(1) in python, but it definitely looks oddish. In reality since the sequence {S(n)} out-paces the {R(n)} sequence after n=2 (assuming n=1 for the start), I'd could argue that defining S(1)=2, and S(2)=4 as initial conditions is the more elegant way to define this programmatically.
None the less, I've adapted the solution anyway to provide an alternative (Note: Don't be concerned with the calls to len(), it is O(1), too):
F = [1] G = [2] for Fi in range(1,20): F.append(F[Fi-1] + G[Fi-1]) G.append(F[Fi]+1) G[len(G):] = range(G[len(G)-1]+1, F[Fi]+G[Fi]) print FWith no change in the result:
What I really like about your solution is that allows you to get rid of the extra index. I might rewrite it as
F = [1] G = [2] for i in range(1,20): F.append(F[i-1] + G[i-1]) G.append(F[i]+1) G.extend(range(G[-1]+1, F[i]+G[i])) print FDepending on if you find the idea of a -1 index intuitive.It doesn't make a lot of sense as an index, but as an offset.
To answer the question as to why anyone would use it, it was much better with COM interop. C# has caught up in recent versions, and both languages are at parity, but it wasn't always like that. If you use C# there is no reason to not know VB.Net. All the language idioms are the same.
Figuring out schedules and deciding on priorities? That's pretty much exactly what your manager should be doing. Pass the buck. :P
Oh man! extend! yesssss, that is the more pythonic way of doing that statement, for sure. Personally, I'm ambivalent to the -1 notation; one one hand it is less characters to write, but on the other hand, like Seabass mentioned, it doesn't make sense as an index. Oh, well, I like the final flow and structure you have for that solution.
Figuring it out is fairly easy, since all points are weighted equally you just drop the three assignments that have the highest differential between max score and students score. I solved it at first with array lists, since that's what we've been studying and when you have a brand new hammer....... She told me I nuked it and that it could be done with just variables and three nested for loops. I went back and did it with a for loop and if/else chain, but I can't for the life of me see how you would do it with three for loops. Any pointers?
Remember that you don't have to iterate through a nested loop every time it's encountered. You can use the loop conditional to check both if every index in the array has been iterated through or if a variable has been set, for example.
I have a programmer friend who has about six months more of professional experience than I do, and skill wise, his company has advanced him quite farther than my company has for me. We're a small company (our programming team consists of three people) and we do things sort of quick and dirty, which is nice, because it gives me a wide range of responsibility, while essentially being trusted to be able to make things work. He does web dev, and I do desktop applications, but he's using a crazy amount of the more "advanced" aspects of C#, like singletons and what not. The difference in skill set just really bothers me for some reason, but seeing as I don't have exposure to any other company, I can't tell what I should know, skill wise, at my experience level and age; should I be worried about that?
Press X To Podcast
Press X To Watch
From my perspective, .NET beats Java in some areas (language features and standalone GUI development). But, for what I do for a living, it falls behind Java in more important ones: web development, open source libraries, and build/support/deployment tools.
ASP.NET is awful for anything that's not a toy website. It's a web framework that puts a wrapper around HTTP/HTML in order to present it as something it's not. As soon as you try to do anything advanced, you have major issues. The problem is that Microsoft's chief design decision seems to be to insulate you as much as possible from your platform (HTTP/HTML). And in order to do this, you need their development environment because it generates and keeps track of an ungodly amount of glue code. Just give me request/response please, and then make it easy for me to layer my own features from there (binding to form data, validation, templating, authentication, authorization, compression). Maybe their MVC thing is better, I don't know, I haven't tried it yet.
The libraries thing is fairly self-explanatory. Java's been around longer. It's more open than Microsoft's platform. It has large, open-source-devoted organizations behind it creating libraries for anything you could ever want. And not only is there a library for everything, there's probably 2-3. Pick the one that fits your needs. If you prefer a web framework like ASP.NET that keeps you from knowing about HTTP, you can have that. Or you can choose from literally 10 other mature web frameworks that might suit your style/needs.
The build/deployment tools thing is related. How many different application servers can you deploy an ASP.NET app on? 1 (maybe 2 if there's a Mono thing?). In Java, there's at least 4-5 major ones, and they're all mature, and all compatible with your application, because the Java Servlet spec is an open standard.
Performing automated builds in .NET is frustrating. MSBuild is like Ant, but somehow even more verbose. Your only other option is to install Visual Studio and run devenv.exe. There's no Maven/Ivy equivalent to do dependency management. .project files keep track of individual source file names instead of source directories, and are automatically generated/regenerated when you make a change in Visual Studio, so SCM tools unnecessarily warn about conflicts. Microsoft built everything around using their IDE, so doing anything outside of it is painful. On top of that, Visual Studio costs money, and is missing basic features like refactoring and quick navigation to class/method definitions. If you want a full-featured IDE, you have to buy something extra like Resharper.
Finally, my experience with Java enterprise ecosystems versus .NET enterprise ecosystems has been night and day. It's almost like the .NET side hasn't ever heard of the word "enterprise". Just a hodgepodge of stuff someone threw onto a Windows Server. But, I'm willing to admit that this experience is probably specific to the clients I do work for.
There's fundamental design differences between the 2 platforms. .NET's is to get you up and running quickly as a single developer (or very small team). Java's is to leave it mostly to you, to do it how you want, which comes at the expense of making it difficult to get started as someone new to the platform. .NET is a Microsoft technology with Microsoft tools and resources. Java is a Sun/Oracle technology, with professional and community tools and resources.
Visual Studio 2010 does not lack those features you list, it has had basic refactoring and quick navigate built in since 2008. There are certainly addins like ReSharper to enhance them, but they are built in. I can't argue with the fact that Visual Studio costs money, but Microsoft has released the capable Express versions for years that are certainly on par with IDE's like NetBeans, though I'll concede Eclipse is very good.
The fact that you think .NET is just some hobbyist/single developer platform and Java is some ultra slick professional platform is telling. Basically, I can sum up your entire post as "I don't really know or understand the .NET eco-system, thus I prefer Java because it's what I know. I was not even aware of basic, available, and popular .NET solutions to issues I posted."...which is 100% fine, but it's not even kind of sort of evidence of one platform being better than the other.
Nant was something we looked into, but keeping Visual Studio, and the build script working together wasn't fun. We switched to MSBuild because it seemed to be the standard way to do automated build stuff. We tried using NUnit for testing, but setting it up as an external tool in Visual Studio was a pain, as was setting it up in MSBuild, so we resorted to the built in testing tool (which tied us even more to Visual Studio). The verbose compilation unit model in .project files is just bizarre, and is what causes most of the problems. It's like there's no standard way to organize code/resources so it just enumerates everything explicitly. Lack of something like Maven is a real turn-off.
We quickly came to the understanding that even if there was a N-version of a Java library, it wasn't considered the "right" or "standard" way to do it in .NET, and as a result it normally had a small, only somewhat-active community around it. ASP.NET is awful, even compared to something like Struts. I'll take your word for it that MVC is better.
All of that is fine. I'm willing to admit that .NET stuff is different, and I'm also willing to admit that most of what I think is "better" is purely subjective. Some of the differences stem from the fact that I know how to do something in Java, and can't seem to find an equivalent or better way to do it in .NET. That's probably mostly on me, but partially on .NET and the .NET community/ecosystem.
And, I'm also willing to admit that a lot of the stuff that turns me off about .NET is purely based on initial bad experiences with it. It's probably similar to a lot of peoples complaints about JSP.
I'll be the first to admit Microsoft hasn't gotten everything right, but overall, .NET is absolutely one of their successes of the last decade. They've grown a vibrant platform, and are pushing the edges with their open source work. They've embraced the open source Mono project to provide a complete and functional implementation of the .NET framework on other platforms, including running ASP.NET stuff on webservers other than IIS. I get the anomosity between the Java and .NET worlds, since .NET and C# were unabashedly born from the ashes of the Sun/Microsoft thermonuclear explosion...but in most senses, they are pretty comparable platforms now, with a lot of the difference being personal preference. With .NET, you're going to get a faster moving platform, that embraces new methodologies and patterns quickly, and adds new language and platform features relatively quickly. The speed of this change can be off-putting to some people. They feel like Microsoft is always changing the platform and they can't get their feet under them. Java is certainly an older and more stable platform, but it moves glacially slow and retains some interesting idiosyncrasies that are like archaeological markers for the 3GL -> 4GL transition era that Java represents.
I think I agree with pretty much everything you said. Java the language is old, and can sometimes be painful to program in (especially compared to lot of more modern languages). But, it's tried and true, and the JVM and the rest of the Java ecosystem are really great. There's lots of libraries and ways of designing applications that can make Java (the language) fun and rewarding.
I definitely agree regarding pace of language evolution. I think I would actually prefer something in between Java and .NET (which is the only thing I really like about the Oracle takeover). Sometimes it seems like Microsoft moves faster than it thinks (I don't have anything to back this up, it's just a feeling). You can look at parts of C# and wonder if anyone asks "should we really do this?" Look at the number of reserved keywords in C# versus other languages, and you'll see a bit of what I'm talking about: http://stackoverflow.com/questions/4980766/reserved-keywords-count-by-programming-language. Obviously there's not a 1:1 correlation between language size/complexity and keyword count, but they're definitely related.
Java feels like it has the opposite problem. Like it's in a cycle of thinking to the point of inaction. For one thing, it's older, and strives to maintain binary backwards compatibility with previous compiled code. The first versions of Java hang over its head to this day. One of its key features is also one of its most frustrating: it doesn't trust developers to understand how to do complicated things. As a result, it's pretty easy to read and debug, but it makes writing code feel clunky and verbose at times. If Java had lambdas and reified types, I'd be pretty happy with the language (and then I'd ask for more
The thing I really like about both Java (the VM), and the CLR is that they are producing so many new and interesting languages. Sure, you've got ports of existing languages like Python or Ruby, but you've also got totally unique things like Scala and F#.
Anyway, I was mainly trying to give a counterpoint to the ".NET is better than Java" arguments that I saw. In the end, most of these debates are anecdotal or based on personal preference, so it's good to know about other people's experiences.
Look, if you want to be completely safe and can't figure it out, just blacklist it!
I don't think congressional technology is that advanced. We're on the cutting edge of progress just by having databases, here.
Either that or the form is completely susceptible to injection attacks and they know it but don't know how to stop it other than asking people not to use those words.
Both are pretty amazing options.
hahahaha, this is exactly what I wanted to wake up to. Thank you
edit: I'd know that if I bothered to check the source on the image. Duh.
@seabass I don't think so, I think it queues them up in order of how you facerolled them, assuming your buffer doesn't get overflowed. But I'd imagine the buffer is larger than 101+ keycodes.
Brief attempt.
"Dear Congressman Peters, please ALTER the upcoming spending bill to DELETE the passage that DECLAREs that SELECT parents should not be able to DROP their insurance coverage. Thanks." 5/7 isn't bad.
i am so utterly tired of the request/response model and the baggage that comes with it
my ideal scenario is one where I can make programs that run entirely offline, or with only a single big web call to get some seed data or something, or do something like websocket programming, where the flow of code (to me at least) is more intuitive
Why don't you?
They pay me to do .NET too, except I get to those offline apps you talked about :P