The collapse of the .NET ecosystem
-
Remember when WinForms was replaced by WPF? Only to be replaced by Silverlight?
So Silverlight, a framework for websites and phone apps, replaced WPF, a framework for desktop apps? The same WPF that forms the base of the latest versions of Office, Visual Studio, and SSMS?The .Net Framework has become fragmented and stagnated.
The recently-added Task Parallel Library isn't an advancement then?We’ve seen “forks” of the .net Base Class Library to Silverlight, Windows Phone, Windows Store apps and Xamarin each with their own unique flavor.
The BCL is essentially *identical* in all forms of the .NET Framework; *that's why it's called the **Base** Class Library!* .NET CF is different though, I'll grant that.Android development is done in Java
Except when it's not.
Tip for the author: if you're going to spread FUD, at least try and sound like you know what you're talking about.
-
What have they removed from .NET? ... ever?
I could be wrong, but I'm 99.999% sure everything from .NET 1.1 still works fine in 4.5. Each version is a superset of the version that came before.
That may be true. But, in a sense, you can say that they've removed (or at least started to deprecate) Windows Forms, replacing it with WPF. The first hints were when they stopped having certification tests for Windows Forms programmers.
-
Are the APIs marked deprecated?
-
I can tell you from experience, I have yet to find an internal app that worked better as a website. Ever.
I haven't found one more responsive.
I haven't found one more UI friendly.
I haven't found one more feature rich.That's not to say I haven't found USEFUL internal business apps coded as websites, but I haven't found better ones.
The ones that work the best tend to be the ones that you have to adapt your business to, like JIRA, or some task tracker with swimlanes.
But nothing that composites the needs of a business.
And as far as hammers go....
WPF and C# is a really damn good hammer.
-
That may be true. But, in a sense, you can say that they've removed (or at least started to deprecate) Windows Forms, replacing it with WPF.
Windows Forms is not a C# or .NET feature. So... good nonsense there, and yet totally irrelevant.
-
I suppose you can make anything go away if you just believe hard enough.
Sky is falling stuff about server-side vs client-side aside, the reason I don't expect c# to go away is it's just a fucking joy to work with. Give me any legacy c# or Java project and a good IDE, and I'll have that shit figured out before management can even get their requirements straight.
So maybe the tide is headed out for Microsoft and Oracle, but I don't see that as any reason to worry. I think you'll find though, that when it comes back in, all this mess of Ruby and JavaScript is going to get washed away.
-
I think the point is that even if things aren't being removed altogether, they are still being replaced, in the sense that we are being given better ways to do things. To say they aren't replacing older features just because they're still in the language would be like saying they didn't replace Windows XP because you can still install it if you have the disc.
-
-
that's going to bite people in the ass a few years along the line when they realize their team of hipster server-side Javascript code cowboys produced code that is fucking shit
To be fair, 90% of all code ever coded anywhere in any language is shit.
-
I can tell you from experience, I have yet to find an internal app that worked better as a website. Ever.
I haven't found one more responsive.
I haven't found one more UI friendly.
I haven't found one more feature rich.But the webby versions have colored rectangles with monochrome icons and/or short words inside! That's got to be worth something, right?
-
.NET isn't going anywhere; there's far too much invested in it globally
Is the best defense for .NET really just "we already use it so much that Microsoft could singlehandedly end the world by discontinuing it, so we should use it even more"?
-
@tharpa said:
That may be true. But, in a sense, you can say that they've removed (or at least started to deprecate) Windows Forms, replacing it with WPF.
Windows Forms is not a C# or .NET feature. So... good nonsense there, and yet totally irrelevant.
Point taken. It's totally irrelevant. It's not like there's any relationship whatsoever between Visual Studio and .Net.
-
Gee, I wish someone sent me that memo so I could have developed my physics-heavy multiplayer game client and server in Javascript instead of C++! While they're at it, they should send that same memo to all the other non-mobile game developers out there, as well as audio workstation and plugin developers, CAD developers, scientific application developers...
That's all small potatoes in terms of the number of developer jobs. Most professional programmers are actually toiling on internal corporate projects. And they all moved from desktop apps to intranet web apps.
Maybe people have become familiar with it and don't need as much training anymore? Maybe it has a shallower learning curve than other stacks?
Maybe.
Maybe they are instead interested in learning about new technologies to supplement their bread and butter?
If these new technologies don't involve MS, that's a problem for MS and MS ecosystem in general.
Maybe they've internalized C# syntax, conventions, and much of the .NET framework so that they don't need to hit Google every five seconds?
Nope.
Maybe Microsoft land is less focused on open source and more on closed source, commercial solutions? Maybe most people who are in Microsoft land are developing closed source systems and have less incentive to share?
Open source is the engine in the heart of a vibrant programming ecosystem. If that stops, 5 years later you get your next MUMPS or FoxPro or whatever.
If we consider smartphone market saturation as the benchmark, then yes, Microsoft is in a distant 3rd place. However, developers aren't necessarily shut out because any device with a browser can access web apps. I don't consider it a terrible tragedy that chess apps, fart noise apps, etc. are less available on one platform or the other.
Real world usage disagrees. Sites can only be made bearable on phones. You want the real deal, you need an app, even if all it does is encapsulates a web view.
I can confirm this in my own experience. I always prefer to use an app, if available. If discourse wrapped itself in an app, I'd ditch web version in an instant.
In fact, many pundits predict the downfall of open internet as a commercial platform. They say, basically, when people want to buy something, instead of googling for it, they'll open up an app on their touch device and look there. See what happened with app stores. People used to love to download and install programs. Now, if you're not on Steam or AppStore or PlayStore, you get scraps at best.
Google is already feeling the hit in terms of reduced add revenue. Searching the internet and going to all sorts of different web sites is just so inconvenient on a touch device. It's so much nicer to open up a grid of colorful icons you can tap and buy. The next big war will not be over the shortest domain name, but over a place on people's home screen.
The point: Microsoft is excluded from all that, and is not likely to get in.
Maybe they're trying to lure LAMP/Oracle stack (that's a thing, right?) developers? Maybe they want to offer a choice rather than lead everyone down a single path?
Yeah but you have less of a reason to stay within the NET ecosystem. And if you don't stay, MS has less of a motivation to keep it updated.
So Silverlight, a framework for websites and phone apps, replaced WPF, a framework for desktop apps? The same WPF that forms the base of the latest versions of Office, Visual Studio, and SSMS?
"Framework for making internal corporate projects".
I can tell you from experience, I have yet to find an internal app that worked better as a website. Ever.I haven't found one more responsive.I haven't found one more UI friendly.I haven't found one more feature rich.
Web gives you three distinct advantages:
-
Copy / paste anything. I can't tell you how annoying is it I can't do that in native apps
-
Open in new tab. I can manipulate the state of various screens by keeping both the previous and next state. This is, BTW, one of the biggest problems with SPA-s like Ember. They sort of expect you to keep clicking links and moving from one screen to the next. I, instead, like to middle click and freely open stuff in new tabs, close them when I want. Which means I sometimes have like 10 instances of Discourse running in one Chrome window, each with its own complete multi-megabyte ember stack.
-
Navigate through URL. You can copy url and send it to someone. Or click a url in an email and find yourself exactly on the screen you need to be on. This is technically achievable in desktop apps, but rarely seen in reality.
And of course, there are all the other problems around local installation, updates etc...
-
-
the best defense
Not quoting more because mobile sucks.It's not a defence; it's simply a statement of fact.
-
-
I can tell you from experience, I have yet to find an internal app that worked better as a website. Ever.
I haven't found one more responsive.
I haven't found one more UI friendly.
I haven't found one more feature rich.All these are limitations that come with the platform. For similar reasons, I have yet to find a desktop app that worked better as a mobile app.
-
Is the best defense for .NET really just "we already use it so much that Microsoft could singlehandedly end the world by discontinuing it, so we should use it even more"?
I would go with:
- Arguably the best developer tools that money can buy
- A well-documented, robust runtime library
- The choice of imperative, functional, or mixed-paradigm languages
to name a few.
-
In fact, many pundits predict the downfall of open internet as a commercial platform. They say, basically, when people want to buy something, instead of googling for it, they'll open up an app on their touch device and look there. See what happened with app stores. People used to love to download and install programs. Now, if you're not on Steam or AppStore or PlayStore, you get scraps at best.
[...]
The point: Microsoft is excluded from all that, and is not likely to get in.
Read up on the app discoverability problem and what it involves. Same problem as discoverability on the web. Only worse; put into an isolated walled garden.
Microsoft is betting big: they're skipping apps and going straight for intelligent devices that will go out onto the web for you, accessing cloud storage and computation services to get what you're looking for.
Why do you think they've been focusing so much effort on Azure?
Also; if their predictions for the future are right, then it's basically going to maim Google and leave them near-death. Google gets a very big part (most perhaps?) of its revenue from search-related ad impressions and their app-store.
-
Point taken. It's totally irrelevant. It's not like there's any relationship whatsoever between Visual Studio and .Net.
Windows Forms is not a .NET feature, and it's not a Visual Studio feature. You are proving to be too stupid to participate in this conversation.
Visual Studio does contain helpful tools for dealing with WinForms projects. But Visual Studio is not WinForms, and Visual Studio is not the only way of creating, editing, or building a WinForms project.
Visual Studio also contains helpful tools for dealing with DirectX HLSL shaders and about a trillion other things.
Open source is the engine in the heart of a vibrant programming ecosystem.
Utter crap. Open source was almost non-existent on Mac Classic, and it was an extremely vibrant programming ecosystem for many, many years.
-
Read up on the app discoverability problem and what it involves. Same problem as discoverability on the web. Only worse; put into an isolated walled garden.
That's a problem for the users and the little guys trying to break in. Not for the platform owners and big chains. Guess who controls the policy.
Microsoft is betting big: they're skipping apps and going straight for intelligent devices that will go out onto the web for you, accessing cloud storage and computation services to get what you're looking for.
After failing to break into the mobile sphere like five times?
So Microsoft's like: (Sniff) FINE! We don't even need stupid phones and apps! We'll make our own thing! It's gonna be like... a robot.. No! An AI! Yeah! (Sniff) And it's gonna be super cool and smart and everyone will want one! (Sniff) And Apple and Google will be like "Microsoft, you're soooo cool! Please let us use your cool AI!" (Sniff) And we'll be like "NO! YOU CANT HAVE IT! NYAH NYAH NYAH!!!" YOU'LL SEE!
Why do you think they've been focusing so much effort on Azure?
Because that's like the only part of the company that's doing great?
Also; if their predictions for the future are right, then it's basically going to maim Google and leave them near-death. Google gets a very big part (most perhaps?) of its revenue from search-related ad impressions and their app-store.
Yup. GOOG has been stagnant for years. Pretty soon shareholders are gonna start knocking on the door and demanding less crap like self-driving cars and fiberoptic internet and more money-makers.
-
Utter crap. Open source was almost non-existent on Mac Classic, and it was an extremely vibrant programming ecosystem for many, many years.
What was that like in the 70-ies? With punch cards flying through mail and the biggest danger being a mullet getting stuck in a tape drive?
-
No. The 80s and early 90s.
It was 3.5" disks filled with .MOD files of Nirvana songs. And the biggest danger was a pog getting stuck in the cooling vents.
-
No. The 80s and early 90s.
It was 3.5" disks filled with .MOD files of Nirvana songs. And the biggest danger was a pog getting stuck in the cooling vents.
The point is that, since the advent of the broadband internet, open source has.... I said, since a lot of people got internet and... What? I'm not whispering. Did you- (Sign) Turn on... TURN ON YOUR HEARNING AID, BLAKEY. YES, I SAID TURN ON-... No, I'm not mad. I'm only yelling because-.... (Sigh) Yes, you can have jello after dinner.
-
The point is that, since the advent of the broadband internet,
Right and I'm saying you're full of shit.
Broadband internet is not a necessary ingredient. Nor is open source software.
All you're communicating to us is your lack of experience with different developer ecosystems, not some brand new universally-applicable "law of nature".
-
Open source is the engine in the heart of a vibrant programming ecosystem.
Not necessarily true. There are plenty of frameworks that are not open source and still have thriving developer communities. Lots of them, actually.
-
Not necessarily true. There are plenty of frameworks that are not open source and still have thriving developer communities. Lots of them, actually.
Yeah, iOS is an exception. But even Apple is going open source these days.
-
Article makes poor argument by inducing the correlation vs. causation problem of random data.
We would probably correlate the drop in C# jobs with the a drop in the number of ISIS members alive.
Especially with the stackoverflow list of top questions per language. It's completely understandable that swift is at the top because its a new fucking language that people are trying to learn. Likewise the C# jobs market can just be saturated as the stupid statistics are about advertised jobs not real time count of programmers current job.
-
Open source was almost non-existent on Mac Classic, and it was an extremely vibrant programming ecosystem for many, many years.
Bollocks. I never owned a mac, let alone a mac classic, but you're saying there was effectively no exchange of classic mac source code without paid licenses attached? I find that hard to believe.
-
I can't say for sure there wasn't, I can say I never ran across it.
There was tons and tons and tons of shareware. There were the rare open source programs (POV-Ray is, IIRC, the first one I personally found/used.) That's... pretty much it. Until OS X came around.
If by "exchange of Classic Mac source code" you mean tutorials and examples, yeah, there were plenty of those. Much of it was in those big 8.5x11" 400-page "how to" books; I still have some in my spare bedroom. It also helped that the Classic Mac API was super-easy to use in a way nothing else was until .NET came along. And people could make Classic Mac apps in APIs even easier than the Mac Toolbox like, for example, HyperCard or RealBasic.
-
That's a problem for the users and the little guys trying to break in. Not for the platform owners and big chains. Guess who controls the policy.
You made a statement regarding the open internet dying as a commercial platform and apps being a better replacement for it. The app discoverability problem is a very real problem for smaller companies trying to leverage apps as a commercial platform vs. trying to leverage the web. Moreover, the open internet has no single platform owner setting policies; one of its big strengths.
Your point is quite moot.
-
You made a statement regarding the open internet dying as a commercial platform and apps being a better replacement for it. The app discoverability problem is a very real problem for smaller companies trying to leverage apps as a commercial platform vs. trying to leverage the web. Moreover, the open internet has no single platform owner setting policies; one of its big strengths.
Your point is quite moot.
What the hell are you talking about? All your arguments come down to "but that won't be fair for the little guy".
Yeah, so? I'm not saying which direction I prefer, or which I think is better. Just the direction where market forces, technology and the big shakers are taking us, whether we like it or not.
-
how .NET will be running singularity space drives in year 3001
One BSOD, and the ship becomes lost in hyperspace forever.
I think I'll pass. :P
-
@cartman82 said:
how .NET will be running singularity space drives in year 3001
One BSOD, and the ship becomes lost in hyperspace forever.
I think I'll pass. :P
Singularity doesn't have those.
-
Article makes poor argument by inducing the correlation vs. causation problem of random data.
Thank you for expressing my point more eloquently.
We would probably correlate the drop in C# jobs with the a drop in the number of ISIS members alive.
So if they die, we're all out of jobs? I thought only @morbiuswilters was the resident drone programmer around these parts!
-
If there's a decline of .Net/C# is because of two reasons:
- Linux servers everywhere, specially web
- SPA
- Windows phone failure (because of Android you don't see this decline in Java)
Anyway, I don't see .Net going kaput, but not growing either. It looks a lot like JEE. There aren't as many jobs out there, and many are for backend stuff. And Node although you wont see Node in any enterprise in a long time, that doesn't mean those system in .Net/JEE won't become simple APIs for SPA.
JavaScript ubiquitousness doesn't help either. Hell, you can even write desktop and XBox apps with it now.
-
Linux servers everywhere, specially web
Irrelevant; that was the case before .NET ;)
-
If there's a decline of .Net/C# is because of two reasons:
- Linux servers everywhere, specially web
- SPA
- Windows phone failure (because of Android you don't see this decline in Java)
Good guy @Eldelshell
-
So what if .NET apps become APIs for SPAs? It's not a bad use case. And I don't see it making .NET less popular.
As for Linux servers, soon you'll be able to run ASP on them too. The question is whether people will go for it.
Windows Phone... Yeah, they blew it. Shame, because it doesn't seem to be that bad of a system (now - AFAIK pre-8 WP wasn't that good), but the chicken-and-egg problem is biting them hard
-
So what if .NET apps become APIs for SPAs? It's not a bad use case. And I don't see it making .NET less popular
That an API can be maintained by many less people than a full web application. And if the FE team starts moving more and more business rules to the frontend, your backend becomes a HTTP shell for your database and some other systems. Now your .Net/JEE devs don't have to build the UI, where usually, most of the work effort is done.
-
But detaching frontend from backend has been a thing since fuck-knows-when. The whole MVC paradigm works like that - you can hand out writing views to your Javascript/HTML crack coders, and work on the backend.
And most business rules you can't move, because it makes your application either ridiculously insecure, or ridiculously underperforming.
-
And if the FE team starts moving more and more business rules to the frontend
The urge to assault the FE team with ClueBats™ rises
-
Linux servers everywhere, specially web
? Are there more (proportionally) now than there were before .NET became popular? Because I highly doubt that's true.
SPA
Which still need a server written in, for example, .NET.
Windows phone failure (because of Android you don't see this decline in Java)
Windows Phone hasn't failed yet. That said, I hope Microsoft doesn't pull a "Zune" and give up on the product line WAY too early.
Oh and look, .NET criticism from a guy who can't count to three. Awesome. Added bonus: you're a moron.
As for Linux servers, soon you'll be able to run ASP on them too.
ASPX I hope you mean. Otherwise, *shudder*
Windows Phone... Yeah, they blew it. Shame, because it doesn't seem to be that bad of a system (now - AFAIK pre-8 WP wasn't that good), but the chicken-and-egg problem is biting them hard
Let's say Windows Phone was an utter failure, sold 10 models in the last 3 years-- what percentage of .NET development is done specifically for Windows Phone? Is this really a significant data point?
Now your .Net/JEE devs don't have to build the UI, where usually, most of the work effort is done.
... did they ever? How is this new?
-
-
Windows Forms is not a .NET feature, and it's not a Visual Studio feature. You are proving to be too stupid to participate in this conversation.
Developing Windows-based client applications with the .NET Framework - .NET Framework
Develop Windows-based applications with .NET. You can use Universal Windows Platform (UWP), Windows Presentation Foundation (WPF), or Windows Forms.
There are multiple ways to develop Windows-based applications with the .NET framework that run locally on users' computers or devices. This section contains topics that describe how to create Windows-based applications by using Windows Presentation Foundation (WPF) or by using Windows Forms.
Your argument is retarded because whatever pedantic dickweedery you want to bring up about its relation to .NET, it's certainly part of the ecosystem.
You areI am proving to be too stupid to participate in this conversation.FTFY
-
I can tell you from experience, I have yet to find an internal app that worked better as a website. Ever.
I haven't found one more responsive.
I haven't found one more UI friendly.
I haven't found one more feature rich.The internal web app I work on meets all of those conditions.
- More responsive. The original desktop app was a thick client. Everything was loaded into memory on the client. The web app is a thin client. Even on the server side, only the needed data is persisted in memory. In some areas, response time is 10x better in the web app then in the old desktop app.
- More UI friendly. The desktop app was written to cram the up into as small a space as possible. Even as larger monitors became the norm, no accomodations were made to de-clutter the app. In the web app, each page is assigned a function, navigation is clear and labelled. Aside from 2 or 3 problem users who have trouble even rebooting (sadly not an exaggeration) users require minimal on the job training to be able to do their jobs. Additional training just shows them how to use the advanced features.
- Feature richness. I have personally added 2 entirely new modules to the web app that we're not in the desktop version. This has extended the use of the software to other departments, aiding the communication of data. That's not even counting the additional features I've designed for existing modules based on tickets that would come in asking for help with one issue or another.
But you won't ever see this software. And even if you did, you'll never have a chance to see the desktop version because it's been retired in the past year. So your anecdote sucks because it relies on you being able to see every piece of internal software that has been converted from a desktop app to a web app.
-
desktop app to a web app.
Which is only happening because it's trendy.
People get a chance to start over and they make a web app, because they have a hammer. Each of those advantages can be conveyed in a desktop app too. But your anecdote sucks because it relies on converting an existing app.
If you take a winforms app, and a webforms app, which one wins, every time?
Because people were trying to program a webpage like a desktop app, and that failed.And now people are still trying to program a webpage like a desktop app. They just have better tools to do it with.
There's nothing that has addressed the elephant in the room. Web was designed for document style UI, and it fails hard when it tries to do app UI. The big blanket used to cover it all up? Predesigned CSS like bootstrap, so people stop making the same mistakes.
And don't think I'm saying all this because I like my hammer more. I've done both, and I love KO much more than WPF data binding.
It's just that I don't think large data manipulation is better in a web app. Both have their uses. And because of that, C# is still useful.
-
Copy / paste anything. I can't tell you how annoying is it I can't do that in native apps
That's because people disable a field rather than making it read only. I have mixed feelings about making all static text selectable, and whether that just adds click points that you don't need to a UI. Don't know how many times I've been annoyed by clicking on the wrong thing in a web UI.
Open in new tab.
You're going to have the same session state problems whether you implement this in web or desktop app. How many websites totally screw up new-tab-navigation?
This is technically achievable in desktop apps, but rarely seen in reality.
The two enterprise desktop apps I've worked on supported just this thing. By making every screen be able to dynamically load its context, and having every screen loadable on demand. Sure, some screens wouldn't work, because they needed parent context, but the same would be true for certain links.
there are all the other problems around local installation, updates etc...
Which would all be present if you were installing a local internal web-app from a third party source, or distributing from a test environment.
But I think that gives you one advantage. You get a buffer on your users adjusting to the new software.
How many times have you seen, "Welcome to the new version of our site." and thought, "crap, I didn't want this to change right now, I just needed to quickly get to something".
-
-
Feel free not to, but I'm only arguing that destop apps are still useful, not that web apps are inferior. I just believe that desktop apps are superior for internal business applications.
I don't see everyone making a switch to web app document editing or email client.
What you've done is say
"When I converted this old desktop app to web app, it was suddenly better."
Which I'm saying,
"You could have simply redesigned it as a desktop app too, and also seen similar benefits."
So it's not fair to say that when you refactor things, they get better, because web app.
-
You're going to have the same session state problems whether you implement this in web or desktop app. How many websites totally screw up new-tab-navigation?
Quite few actually.
The two enterprise desktop apps I've worked on supported just this thing. By making every screen be able to dynamically load its context, and having every screen loadable on demand. Sure, some screens wouldn't work, because they needed parent context, but the same would be true for certain links.
So you laboriously recreate a feature browser gives you for free.
But I think that gives you one advantage. You get a buffer on your users adjusting to the new software.
How many times have you seen, "Welcome to the new version of our site." and thought, "crap, I didn't want this to change right now, I just needed to quickly get to something".
I don't know how many hot-fixes I have had to push after big updates. Users didn't even notice. That would have been a nightmare if I actually had to deploy software to peoples' devices.
Also, as far as users are concerned, nothing should ever get updated, ever. Unless it's that one thing they asked for. Maybe.
Singularity (operating system) - Wikipedia
Windows Forms - Wikipedia
