Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

.NET 10

devblogs.microsoft.com

Post reply on HN
Print view View original

181–190 of 605 posts

Re: .NET 10

#181

Post by fud101 »

CharlieDigital wrote: post #140
jitbit wrote: post #2

For us, every .NET upgrade since .NET 5 has gone surprisingly smoothly and reduced CPU/RAM usage by 10–15%. We were even able to downgrade our cloud servers to smaller instances, literally. I wish .NET was more popular among startups, if only C# could get rid of the "enterpisey" stigma.

I'm at a series-C, YC startup. We made a switch from TypeScript to C# two months back. Now we have a team of over a dozen backend engineers working on C# transitioning from TypeScript. 90% are working with C# for the first time. (We are still hiring backend C# engs!) I can say that it has gone waaaaaay smoother than anyone would have thought. This is a decision (language switch) that the team has been putting off for…

So should I learn C# by learning Typescript? Does that make sense?

Re: .NET 10

#182

Post by mrsmrtss »

Xelbair wrote: post #167
mrsmrtss wrote: post #105

Earlier quoted context omitted.

>If i remember right Java's webserver beats ASP.NET in performance benchmarks That's not the case anymore. Kestrel is one of the fastest servers there is, and it beats every Java server out there.

the famous benchmark that put ASP.NET in #1 spot was hyperoptimzied to hell while competing with normal implementations of other systems. https://web.archive.org/web/20250330225652/https://dusted.co... archived version becasue original one gives 404 currently according to techempower benchmarks ASP.net is 55th overall in minimal variant, while being 83 in normal one in Fortunes benchmark which is basically a normal u…

Look at plaintext results if you want to compare just servers. Fortunes benchmark has too many variables (including db queries etc) to say anything about server performance. As for fortunes benchmark, for more realistick Java performance numbers I would look at Spring. Also there are many shady things on those Java bencmarks.

Re: .NET 10

#183

Post by whoisthemachine »

Arisaka1 wrote: post #106
netdevphoenix wrote: post #94

Earlier quoted context omitted.

> I really, REALLY wish I was in another timeline where I could say in an interview "yes, I use Linux on my desktop and Rider for my IDE" without being seen as a traveler from outer space. Could you please elaborate? Are you referring to most .NET shops not straying away from Windowsland?

It's not about what the company uses, but how informed the technical people responsible for hiring candidates are around the ecosystem they claim they work with. Example: Expected: "Oh, you're on Linux? I heard about Rider. We use Windows and Visual Studio here for parity. You're okay with that, right?" (me: Obviously, tools are tools) Actual: "Does .NET run on Linux? What is Rider?" I mean, .NET has been running on…

There's still a lot of folks who consider themselves .Net experts who don't know how to program with async/await, so knowing about a niche IDE (which I also exclusively use) is asking a lot for those people.

Re: .NET 10

#184

Post by maccard »

bilekas wrote: post #101
high_na_euv wrote: post #87

Earlier quoted context omitted.

Evaluating complex expressions during debugging. In c# I can evaluate complex linq data transformation in watch window in visual studio during debug, at fly. In cpp I cannot. Not even nested evaluation is working.

Ah okay, yes, debugging tools are a bit more friendly with C# but it's again the nature of the beast of cpp, but to nitpick this is a compiler area moreso than a tooling issue . CPP will compile direct to binary, whereas C# will compile to machine Lang iirc, an intermediary step anyway, so you can imagine it makes debugging much easier.

> to nitpick this is a compiler area more so than a tooling issue

If there’s one thing c++ is good at it’s bikehsedding where the responsibility for fixing this stuff is. When I started programming in c++ 16 years ago, modules were “imminent” and reflection was coming soon.

Modules are still unusable because what we’ve standardised is insanity and the standards committee refuse to make a decision to support the backwards compatibility of tooling and code that will never be upgraded or even used in a modern setting.

The compiler vendors blame the standards committee, the committee say that tooling isn’t their problem,and standardise something that has no theoretical objections but in practice requires deep concessions to the spirit of the idea. At the same time, they’re perfectly content to assume that everything is developed the way they imagine - adding ranges to algorithm and forcing the compile time because modules technically exist.

Re: .NET 10

#185

Post by loldot_ »

CharlieDigital wrote: post #140
jitbit wrote: post #2

For us, every .NET upgrade since .NET 5 has gone surprisingly smoothly and reduced CPU/RAM usage by 10–15%. We were even able to downgrade our cloud servers to smaller instances, literally. I wish .NET was more popular among startups, if only C# could get rid of the "enterpisey" stigma.

I'm at a series-C, YC startup. We made a switch from TypeScript to C# two months back. Now we have a team of over a dozen backend engineers working on C# transitioning from TypeScript. 90% are working with C# for the first time. (We are still hiring backend C# engs!) I can say that it has gone waaaaaay smoother than anyone would have thought. This is a decision (language switch) that the team has been putting off for…

It is no coincidence that C# and TS are similar. They are created by the same person, Anders Hejlsberg. The C# language may have some baggage from back in the day, but at least it has a very good, non-fragmented ecosystem. While Typescript may have learned from some of C#'s mistakes, the js/ts ecosystem is a dumpster fire imho.

Re: .NET 10

#186

Post by whoisthemachine »

lolive wrote: post #60
cies wrote: post #28

That's a lot of goodies in a new release! It seems it is outpacing the JVM's development...

Apart from [the equivalent of] records, I see nothing big. Except... this '''let! a = fetchA() and! b = fetchB()''' really puzzles me. Does C# have a high-level syntax for concurrency timing? [something that Java is strongly lacking, and that Typescript did solve with Promise.all(), which is an ugly syntax, from my perspective] Any elaboration on this is very welcome.

As others point out, that's F#, but yes C# has `async`/`await`, and has all the `Promise` methods, just under the `Task` class instead (and with slightly different names/calling conventions through out).

Re: .NET 10

#187

Post by array_key_first »

mexicocitinluez wrote: post #151
christophilus wrote: post #146

Earlier quoted context omitted.

I really liked working with C#. I spent 15 years or so with it and found it very productive. But no; I don’t miss the culture of C# / Microsoft shops at all.

> culture of C# / Microsoft shops at al What do you mean?

I worked at a Microsoft shop, and this was my experience.

1. Process, process, and more process. Doing anything required layers of management approval. Trivial tasks become month long, or even years long, processes.

2. You have no power or agency. Something is broken? You're a developer, you should be able to fix it right? No. Broken things stay broken. You swim in your lane and keep your head down. Mediocrity is the goal.

3. Optimization doesn't exist. If a process is manual and takes you, a developer, 10 hours, then that's what it is. Nobody gives a flying fuck about tooling. Nobody cares if you spend 50% of your dev time doing random stuff. And if you even dare try to fix it, you will be told it's impossible and you're wasting your time.

4. Management is king. You will have to lie to them. You will have to spend time re-entering the same data in 5 different places so they can read it conveniently. You will have to make Excel workbooks. You will have to dumb things down, and then dumb them down again, and again. Everything is about Jira... Unless they're a really high up manager, in which case you have to take whatever is in Jira and put it in a word doc and send it to them, because they don't know how to open Jira.

Re: .NET 10

#188

Post by jcmontx »

deburo wrote: post #175
jcmontx wrote: post #136

I've worked with .NET for over 10 years and we built our startup on top of it. Here are my thoughts: Pros: * Stability * Very decent standard library * Good balance between productivity and robustness * Great package management * Extremely easy to upgrade, so essentially free goodies (performance) every year or so Cons: * Very MSFT dominated (obviously) * Subpar tooling outside of Windows (I'm looking at you C# Dev K…

* C# has way to many features, it feels bloated Most of the newer features make C# less cumbersome to use. I can't imagine using the language from ten years ago, in fact the first thing I do when working on a legacy app is to set langVersion=latest.

Sometimes less is more. It very easy for devs to use the language in completely different ways. I lean towards Go style in that regard. My ideal language would be like Go + Linq.

Re: .NET 10

#189

Post by danielbarla »

deburo wrote: post #175
jcmontx wrote: post #136

I've worked with .NET for over 10 years and we built our startup on top of it. Here are my thoughts: Pros: * Stability * Very decent standard library * Good balance between productivity and robustness * Great package management * Extremely easy to upgrade, so essentially free goodies (performance) every year or so Cons: * Very MSFT dominated (obviously) * Subpar tooling outside of Windows (I'm looking at you C# Dev K…

* C# has way to many features, it feels bloated Most of the newer features make C# less cumbersome to use. I can't imagine using the language from ten years ago, in fact the first thing I do when working on a legacy app is to set langVersion=latest.

I've heard this argument before from the perspective of C# having more keywords and language features to be aware of than something else (in my particular argument, the other side was Java).

From this perspective, I can't say I disagree as such. If you look at the full set of language features, it sure is a lot of stuff to know about. The argument that it is too much, and that we should sacrifice expressiveness and signal to noise ratio in the code to keep the language simpler, I don't agree with.

Re: .NET 10

#190

Post by dustedcodes »

Xelbair wrote: post #167
mrsmrtss wrote: post #105

Earlier quoted context omitted.

>If i remember right Java's webserver beats ASP.NET in performance benchmarks That's not the case anymore. Kestrel is one of the fastest servers there is, and it beats every Java server out there.

the famous benchmark that put ASP.NET in #1 spot was hyperoptimzied to hell while competing with normal implementations of other systems. https://web.archive.org/web/20250330225652/https://dusted.co... archived version becasue original one gives 404 currently according to techempower benchmarks ASP.net is 55th overall in minimal variant, while being 83 in normal one in Fortunes benchmark which is basically a normal u…

> archived version becasue original one gives 404

Thanks, fixed now. I literally just migrated from GCP to Hetzner over the weekend and was about to finish the migration today :)

Post reply on HN