Visual Studio a) will never run on Unix, and b) is not required to do .NET development.
I know people who do their .NET work in Sublime Text with a nant build chain. It's not my preferred method, but it works great for them. Hell, I know a guy that uses ASP.NET to host an Angular app, and to unify his build system, he has Grunt pick up the heavy lifting of running his MSBuild files whenever he does the pre-deploy steps for his JavaScript.
If you think you need Visual Studio to do .NET development, it's because you fundamentally don't understand the technologies at play and just wanted to say something provocative and witty.
The only reason I bring up Visual Studio is because most people don't even see the world out side of that IDE. "If it's not MS and it's not possible inside of Visual Studio it doesn't exist." That's how a whole lot of .NET developers, and .NET houses work -- and so that leads to my comment on VS.
> If you think you need Visual Studio to do .NET development, it's because you fundamentally don't understand the technologies at play and just wanted to say something provocative and witty.
On the off chance you're not just being knee-jerk and defensive because Something You Like is under fire--people think that VS is necessary because life is way too short to spend building something half as good as the other options in twice the time. I mean, could write Scala in TextMate, too, if I had the jones for it. But I could also not, and use IntelliJ, because I value having tools that are able to improve my productivity. (And decrease my frustration; by sticking with the JVM I have the benefit of not have to rely on Mono, whose GC, from experience, I do not trust in long-running processes; after spending a year owning a trivial Mono ASP.NET app that presented no end of troubles, I am very happy not needing to have a watchdog ready to kick it over when it undergoes its daily spaz-out.)
Xamarin Studio is not an acceptable product, either. I've been trying to use it for a couple weeks now and about ready to pitch a computer out a window--the basic functions of editing code are magnificently broken and have been for a long time. Random phantom line breaks in the editor that aren't in the code file, bad auto-indenting, phantom error highlighting, no partial-build error solving--and Visual Studio doesn't work on OS X. (To say nothing of the "you have to restart because XS forgot how to talk to xbuild" bugs...) That leaves the shitty bodges you well-actually'd your way into or not doing .NET at all. The latter makes a lot more sense for a lot more people.
I really enjoy C# and F# and I'd love to use .NET more if the frustrations in using it on not-Windows were not of such a magnitude. I am critical because it should be better and isn't. (Worse, Xamarin removed most reasons to contribute to the Mono ecosystem a while ago by making so much of the useful bits commercial, which is understandable--gotta eat--but also unfortunate.)
>"I mean, could write Scala in TextMate, too, if I had the jones for it. But I could also not, and use IntelliJ, because I value having tools that are able to improve my productivity."
I would argue that it would depend on the person behind the keyboard. I love VS but I happen to be working all day on vi at work and here I know colleges that are as fast or even faster than a lot of people relying on Intellisense. So my guess is that YMMV
I don't understand the thinking behind phrases like "too little, too late" about a company whose tooling will likely span the entire careers of a number of generations of developers.
I say that a lot and I feel like .NET is loosing relevance. Embracing open source and making Linux a first-class citizen in the eco system would be the only way for .NET to increase adoption or stay technologically relevant, and unfortunately its happening too slow.
Sure you can, but the technology of the future is not being built or run on windows stacks. You find it everywhere in the enterprise, yet almost every big new tech company that has come up in the last 10 years all rely primarily on open source technology.
Xamarin / Mono and Unity run pretty much everywhere. Unity even runs on WebGL, none of this is tied to windows stacks.
Also, Microsoft has a great startup programme with Bizspark, we built a company on it a few years ago that eventually ended up as part of a $700m acquisition
Yes. Most users of .NET is in the enterprise (as opposed to tech and startups) and Java has positioned itself as the go-to platform for big data analytics and processing.
I think this comes down to clinging to an antiquated product model. A lot of the technologies that are seeing rapid adoption are open source and driven by transparent communities. In some cases companies back them (e.g. cloudera, datastax, 10gen, typesafe).
And where exactly do you think most devs work? As far as startups go, and we may be usual, but we are a healthcare analytics company using Sql Server for medium? data and processing, with a pretty light .Net middleware/front-end.
Yes this might be true right now, but take a look at a lot of the tech startups are cropping up - humor yourself and look up a list of the tech ipos the last two years. They are overwhelmingly based on open source technology and they are also displacing or disrupting your traditional 'enterprise' businesses.
Even though C# and .NET has a huge developer following, .NET is losing ground to OSS within the top traffic driving sites. 10 years ago, people would build their MVP in .NET because it was the best technology around, but nowadays I see more people building new businesses and placing their strategic technology bets elsewhere (Java/Python/Ruby or Node).
A lot of these developers don't want to be married to the Windows eco-system, for a variety of reasons. Microsoft cannot continue to thrive on the sole basis of it being the only game in town anymore. So all in all what I'm trying to say is I don't think the long-term prospect is looking great.
When company X that the hivemind dislikes makes it to the news, they are often criticized with not open sourcing their product Y. When they finally open source it, they are met with "Too late!". Sometimes there is no winning with people who have already made up their mind.
>Too little
Really? Its MS, their core products and the open source we are talking about. Heck they even open sourced it under a liberal license.
Too little? Biggest change for ASP.NET from release of ASP.NET!
"Too much of pain" was the first (and last) thing I said when trying to configure ruby on ubuntu. Or tried to download some app and run it.
I guess all those thousands of Rails apps running on Ubuntu had some kind of magic or secret knowledge to get that working. Going open source and posting some libs on Github is 'Huge'? That's kind of funny considering how much is out there, top notch, and did it way before MS did, and yet, it wasn't "huge" for them.
I think Hudo's point is that experience with technologies is subjective. I loathe Windows after having been comfortable in terminal, and I loathe Eclipse after having been comfortable in VS.
I've gotten apps working on .net and ubuntu -- configuration was a drag in both, but it works, and isn't the entire basis for loving or hating a platform. So what I'm saying is I agree with your comment.
The catch here (and point) is why should you have to switch from VS at all? If it's a real source code editor shouldn't it handle Java, Clojure, Haskell, Python, Ruby, JavaScript, etc all just fine? Some of the oldest and free editors do and do it well. I realize MS has an agenda (perhaps) or they've de-prioritized any of these features into oblivion -- but who are they serving, I certainly didn't feel it was me even after giving them a lot of $ for VS.
Other editors certainly 'do it', but they're so embarassingly bad that a huge amount of non-VS devs resort to using text editors instead of dealing with the clunky IDEs.
There's definitely some places where this may be true, but there's a lot of locations and industries where .NET is firmly planted and not going anywhere anytime soon. For instance, Houston and the oil industry here (not just the big companies like Exxon, but the massive ecosystem surrounding petroleum). It may not be about winning new industries, but gaining the benefits of open source where they already are.
I believe one of the big changes in ASP.NET vNext is the ability to deploy outside of IIS.
Other types of projects, notably WebAPI, have been able to easily deploy outside of IIS for years. I actually just converted two WebAPI projects to self-hosting. It took about a day, which is not bad considering that includes time to learn what I was doing and reworking an existing deployment process.
Nope, I'm not a fan of Php either actually. Mostly I rock the Java, Node, Clojure, Scala, *nix (mac), C++11 on Ubuntu and Mac with JS+Angular. I guess that's the stack mate.
This entire version of ASP.NET is specifically about modularizing the stack and breaking it's dependence on IIS so it will work well with Mono + Apache and Nginx. That's quite literally why this version is being written, as aside from MVC, ASP.NET as an IIS interface technology has been feature complete for years and wasn't in line for any major changes.
As a person who likes C# and MVC but doesn't like Windows, I've hoped for this for a long time. The fact that Microsoft is going to include mono in their tests is awesome. It's like a genie granted me a wish.
I know people who do their .NET work in Sublime Text with a nant build chain. It's not my preferred method, but it works great for them. Hell, I know a guy that uses ASP.NET to host an Angular app, and to unify his build system, he has Grunt pick up the heavy lifting of running his MSBuild files whenever he does the pre-deploy steps for his JavaScript.
If you think you need Visual Studio to do .NET development, it's because you fundamentally don't understand the technologies at play and just wanted to say something provocative and witty.