.NET Framework

So, i'm currently studying programming in the .NET language, basically C#. I have to admit i like this thing, it gives you a lot of tools for developing, and the fact that i can program in different languages, including VB, without changing the program result (as in speed and whatnot), is quite good.

I know that usually you guys hate microsoft to death, but i think this time they did something acually useful. Does anyone have any opinions or experience with the thing?

please, NO FUCKING TROLLS, if that's even possible.

No one wants to admit it but .NET is pretty good.  People try to make excuses like it won't run on their neckbeard powered OS, but really people in the real world use windows whether you like it or not.

>>2
>People try to make excuses like it won't run on their neckbeard powered OS

oh yeah, linux is good when you're a young fool and have libeal software ideas, then you exit college and find work only on Microsoft stuff.

Usually, if you need to work, you'll find only Windows-related stuff to do. If you find it for linux, if you're that lucky, you'll get paid less for it than what you get paid for the same thing in Windows environment

>>1
Job issues aside, you should judge a language/environment/framework by what it offers.

C#/.NET are fairly good, or so I thought a few years ago. It's a Java clone done right. It's easy to use, somewhat of a relief if you've only coded in C and other low-level languages. The relief comes from not having to worry about memory management, a large library of code for doing a large body of common tasks you'll want to do, some slightly better possiblities for abstraction, not low-level, but let's you get your hands dirty when you need to.

That was my opinion a few years ago, but after a while, and writing a few medium sized projects in it, at least one of which was quite complex. That project showed me that the language had fundamental limitations, and to overcome them you would have to abuse standard design patterns and even do some nasty hacks. A few months later, I embarked onto reading my SICP and learning other new languages with much more powerful abstraction facilities than that of C# and other ALGOL derivates. I no longer think the same about programming as I did before, and feel much more free in expressing my ideas. I suggest you OP, try to take such a journey, it will be an eyeopener. As for C#, it will still be useful for getting a job.

>>4
that was inspiring as hell.

What language do you develop in? i've heard the best things about C#, C++ and such, but i don't know.
Care to give a little reccommending?

>>4
It's a Java clone done better, certainly, but I wouldn't go so far as to say that it's done right.

Whatever merits .NET may have, though, what's a deal-breaker for me is Microsoft's stated policy of not fixing bugs that cause their implementation to deviate from the published standard for the sake of backward compatibility, combined with the fact that their Community Promise only extends to other implementations insofar as they do follow the standard, so they can't emulate their bugs. That breaks even the tiny pretense at cross-platformness that Mono gave them.

That, and the fact that neither I nor anyone I know uses Windows. The fantasies of sheltered Microsoft weenies aside, Windows was never as ubiquitous as people believed, and its market share is dropping quickly.

>>5
Try looking into Scheme, Common Lisp, Haskell, O'Caml, Erlang, Smalltalk and others. Reading SICP first would be a good idea.

If you really want to stay in the MS universe and and don't want to install external toolchains or learn *nix'ish tools, you could try F#, it's Microsoft's O'Caml clone, and will let you use the .NET libraries too.

>>5
i've heard the best things about C#, C++ and such
You are surrounded by morons. You need better sources. You're lucky you found us.

"You are surrounded by morons. You need better sources. You're lucky you found us."

how is it that i start to feel incomprehensible skeptiscism (whatever, i'm a grammar nazi) upon your statement?

>>7
i'll look forward to it, thanks

>>6
you have some good points there, but i still stick to windows, for obvious reasons (AKA: it's the only thing that runs almost everything nowadays)

it's the only thing that runs almost everything nowadays
ihbt

>>10
i wasn't trolling

where do you find the logical fallacy in my post? So i can correct it, understand and stop being a moron, if that's the case?

>>11
If you were open to the suggestion, you wouldn't have to ask me to correct you; you would simply have corrected yourself through independent, self-interested research.

>>10
He was probably referring to you stating that everything runs on Windows which is false. Windows has a large market share in the desktop and enterprise market, but there's plenty of other markets where it's more rarely used, such as the server market which is mostly dominated by Unix-like OSes.

The OS of your choice won't really change anything as you can run implementations of the mentioned languages on both Windows and Unix-like OSes alike, but whichever OS you use, you will still have to learn new tools, some of which were primarily developed on *nix and then ported to Windows, so you should be aware of that.

>>12
i would but i'm too god damn tired to even get up the chair, i'll sleep here in front of the computer.

today was a bad day, don't ask. Later, i'll get some beer, and let the alcohol give me some sleep.

>>13
oh that
well i know that, but i'm implying the market part.
I worked with linux too and i know how the things work with that thing.

>>15
Windows only runs on the desktop and workstation. There's a far larger world of computing than just desktop machines.

>>16
Um, laptops, cell phones/PDAs, embedded devices, and servers too.  I can't think of much else other then microprocessors, and those don't need an OS.

>>17
ever used a winmob phone?? i was stuck with one work gave me fo a year and change. shit sucks, man, i mean really sucks.

a slimmed down desktop o/s for a phone? lolwut?!?!

talk about cramming 10 pounds of shit into a 5 pound bag

>>17
WinCE supports four architectures, the mainstream Windowses support one and a half. Your sheltered ignorance does not imply ubiquity on the part of Windows.

>>5
You know, you're going to get trolled a lot. Just thought I'd let you know.

>>4
This is actually a brilliant post, although I'd disagree with the final note and suggest that C#, like all tools, has its place.

His underlying point is indisputably correct, though. You should learn as much as you're able and find new ways to think about problems and solutions, which radically different paradigms force you to do. That is the true path to satori.

>>6
That's a fair argument, though I don't think it precludes .NET languages from being useful tools.

As much as I hate Java and all its shortcomings, even that has its uses.

We're your friends.  We're not like
            the others.

>>17
I can't think of much else other then microprocessors, and those don't need an OS.
IHBT, on two accounts.

Bump; the front page really needs less him.

Fucking framework disturb me.
I learned C# and VB .net, sure this is easy but shitty too...

Too many framework class loaded for few things.

>>22
Are you implying microprocs are used for something other than blinking LEDs?

>>25
Quite the opposite, firstly, every computer you have ever used is a microprocessor and secondly, implying that a microprocessor doesn't need an OS is quite frankly bullshit.

>>26
s/doesn't\ need/doesn't\ usually\ need/

>>19
You think I'm saying that vista would work on an embedded device?  Of course I meant windows CE.  CE is still a Windows OS, just made for embedded devices, the same way there are linux distros for embedded devices.

As of the current stable release, the official ports are:[86]
i386 – x86 architecture designed for Intel/AMD 32-bit PCs. Also compatible but not recommended on Intel/AMD 64-bit single/multi core PCs[87]
amd64 – x86-64 architecture designed for Intel/AMD 64-bit single/multi core PCs
alpha – DEC Alpha architecture
sparc – Sun SPARC architecture on sun4u/v systems
arm, armel – ARM architecture on Risc PC and various embedded systems (both little-endian – old ABI and EABI)
powerpc – PowerPC architecture
hppa – HP PA-RISC architecture
ia64 – Intel Itanium (IA-64) architecture
mips, mipsel – MIPS architecture (big-endian and little-endian)
s390 – IBM ESA/390 architecture and z/Architecture

The m68k port was the second official port in Debian, and has been part of five stable Debian releases. Due to its failure to meet the release criteria, it has been dropped before the release of etch. Still, it continues to be available as part of the unstable distribution:
m68k – Motorola 68k architecture on Amiga, Atari, Macintosh, and various embedded VME systems

Ongoing efforts include ports to Hitachi SuperH (sh4), Renesas M32R (m32r), and Atmel AVR32 (avr32) architectures, big-endian ARM port (armeb), and 64-bit-only PowerPC port (ppc64).

>>28
i've run windows 7 in qemu on a handheld device. it was slow as fuck, but it worked.

OP here

you guys are fucking awesome

you guys are fucking awesome
This is sarcasm, right?

>>4
That project showed me that the language had fundamental limitations
Elaborate please?

>>32
No

>>34
;_;

sage mah balls

bump

>>36
GET OUT

>>33
The language forced one to think in terms of its limited object system, and didn't even have true first class types (at that time, things have improved somewhat in newer .NET versions). It also lacked at that time (and still lacks some of them): first class functions, closures, anonymous functions, a powerful macro system or other kinds of ways to extend syntax, ability to chose scoping rules, and had some other small kludges. You can work around most of these problems as the language is turing complete, but it will result in useless boilerplate code to work around language issues. Why should the programmer fight against the language when he can just use a language which offers him much better possibilities? When one codes in a more powerful language, he can think in better terms about the problems he wants to solve. The language influences ones ideas greatly. A programmer who has only worked in assembler, won't dare think about higher level aproaches to solving his problem, and instead of chosing the natural declarative solution, he will chose an imperative solution which just painfully describes each little step that needs to be taken to reach the goal he wants. Neither solution is inferior without defining any metrics to judge the quality of it(clarity, execution speed, programmer time required to write it, scalability, maintanability and others). As a side note, programmer time is a lot more expensive today than it was 30-40 years ago.

>>37
u mad

>>38
Ah. But why would you need first class types? And you do know that C# supports lambda expressions?

Though I guess that I never came across many limitations simply because I never really tried making a complex system in C#.

>>40
Yes, I'm aware they added support for lambda expressions, however they were not around the time I wrote that project (.NET 1.1).

As for the lack of first-class types, I did find some workarounds at that time. Useful practical things anybody could do if they were implemented include being able to switch on types, do pattern matching or instantiate objects dynamically. These things are possible through ugly workarounds, but if the language would have been more flexible, the user would either have the feature, or he would have the option to add it himself.

>>6
The fact that neither I nor anyone I know uses Windows. The fantasies of sheltered Microsoft weenies aside
neither I nor anyone I know uses Windows. The fantasies of sheltered Microsoft weenies aside
neither I nor anyone I know uses Windows
sheltered

Truly, 0/10.

46 Name: Anonymous 2011-02-04 17:36

Don't change these.

Name:		Email:
Entire Thread Thread List

All trademarks and copyrights on this site are owned by their respective parties. All comments and posts are the responsibility of their own posters.
- Tinychan + 2ch Mode -