Live data from Hacker News

Announcing .NET Core 1.0

blogs.msdn.microsoft.com

121–130 of 327 posts

Re: Announcing .NET Core 1.0

#121
post #82
post #39

Earlier quoted context omitted.

Way too many compatibility issues unaddressed by that PR to have a hope of getting anything changed there. Also it looks like what you (?) are trying to do could easily be solved with the Path.* and Directory.* path manipulation functions. No need to put a million apps at risk when you can wrap your call to Uri() with a three line helper.

3line helper? facepalm ...

How many lines would be acceptable? Your frustration is inscrutable.

Re: Announcing .NET Core 1.0

#122
post #28

I'd LOVE to see some Linux web/micro benchmarks... particularly against Java, Go, Node, Python.

Scott Hunter of MS did an interview not so long ago re .NET, Kestrel (the new libuv HTTP server they build), and their march to build up as a top contender for Tech Empower.

http://dotnetrocks.com/?show=1291

They have a team who does it all for fun and has an impressive testing environment build out. Fascinating to listen to given today's new. I listened a few weeks ago myself.

I have friends at work who liked C# and moved on to frontend. The fact they are so less interested in .NET now is amazing to me as I finally want them to teach me to use ... on Linux!? I never thought I would say that. Haha.

Re: Announcing .NET Core 1.0

#123

Earlier quoted context omitted.

If you need COM interop, you don't need cross platform compatibility. You can just use the full .Net framework.

I remember working with COM on MacOS 8... And it supported remote objects, so, it'd be, in theory, possible to self inflict that pain.

I experience a physical revulsion just thinking about all the wasted years I spent with COM and DCOM.

Re: Announcing .NET Core 1.0

#124
post #78

I was really hoping for ARM support at least as far as RasPi goes for 1.0. But still very good news I personally love .Net it's nice to be able to use it cross platform without having to work specifically around mono.

From what I see Samsung has been contributing on that front. They said in the announcement that Samsung has joined the Dotnet Foundation.

Yeah, the only problem with Samsung is that they most likely going to make builds for the Exynos based SOC which isn't that wide spread. The ARM ecosystem is quite fragmented and depending on how you build your software it's not as portable as one would assume.

It's also not clear which instruction sets they are going to be aiming for Exynos started with ARMv7 + all the stuff that Samsung added to it, but the newer ships are ARMv8. The Raspberry Pi still uses ARMv6, 7 and 8 depending on the exact model with the newer RasPi compute, and Zero models still running the old BCM2835 ARMv6 CPU's.

So I'm hoping for a more or less clean stock ARMv7 and ARMv8 builds for the .NET core coming out sooner rather than later, I also hope they'll release the X86 version to more platforms than Windows since running it on something like an Intel Edison which comes with a 32bit ATOM CPU will be quite cool.

.NET is quite powerful and for my taste it's considerably better than Node.JS (I don't like Javascript, this isn't some technical observation just personal preference) but I do like C# and F# quite a bit and having the ability to run the same code across multiple platforms makes me excited for IOT/Embedded devices again especially considering that I don't have to work through some of the headaches that come with Mono (some pun indented).

Re: Announcing .NET Core 1.0

#125
post #111

Does this still use mono's jit and GC on mac and linux? Last I checked mono's GC is conservative and the jit isn't as good as the .NET one.

It does not; it uses the same JIT and GC as the Windows version of .NET Core, and the code is also shared with the (full) .NET Framework that ships inside Windows.

Actually, this has always been the case for .NET Core, but you may be thinking of our older preview versions which gave you the option to run your app entirely on mono.

(I work on .NET at Microsoft)

Re: Announcing .NET Core 1.0

#126
If I'm an experienced Python developer who develops a lot of websites and APIs using Django or Flask and SQLAlchemy, is there any reason to try this stuff? C# is a great language, but what about the libraries? What replaces Flask? What replaces SQLAlchemy? How do I deploy? Looking for some practical reasons to invest time on this if Windows development is not in my roadmap.

Re: Announcing .NET Core 1.0

#127

Earlier quoted context omitted.

I still can't shake the feeling that their naming is just confusing. I still am not sure what exactly is .NET and what's not.

.Net is 14 years old now - if you don't know what it is by now you're probably not ever going to know. It's a cross platform runtime and a group of programming languages that run on it, just like Java.

My point is that it's just hard to understand what exactly one means when they're writing ".Net". Do they mean the entire framework, a specific library, the CLR, one language, etc. For whatever reason, the name has a very contextual meaning, and people use them interchangeably (".Net Core" is clearly specific though).

Not that they're alone. Adobe did worse with "Flex" since it could refer to the compiler, a framework, and an editor at one point until they decided to make it a bit more standardized.

Re: Announcing .NET Core 1.0

#129

F# doesnt work ootb with sdk preview2. Work ok with preview1 of SDK (win/ubuntu/osx/docker), but preview2 has a bug, the fix is in progress ( ref https://github.com/dotnet/netcorecli-fsc/issues/12 ) and will be published a nuget package with fix soon.

You reported the issue 11 days ago. I'm surprised they just announced it with such a basic use case being broken. It doesn't bode well for F# as a first class citizen in their ecosystem.

Seconded. I saw the announcement and immediately went to grab the docker image and give it a shot with F# via "dotnet new -l F#" and go. It restored and built, but didn't run. So v1.0 is broken, I guess?

Re: Announcing .NET Core 1.0

#130

Interesting name choice considering they have already had a .Net 1.0. I mean I get it, I understand why they chose to make this more of a 1.0 release but for those already in the .Net ecosystem it seems a little confusing to me.

Excuse me if this was already obvious, but the word "core" designates a new product line, starting from 1.0. Obviously, it's informed heavily from the old .net framework, but it's not backwards compatible with it.

This is just a personal view, but my first instinct when I read about product "CoolApp" and "CoolApp Core" would be to assume core is a subset of the first. Not a new and re-engineered product line.

I could see why others might see .Net Core as a parallel product as opposed to a new one.

Post reply on HN