Evernote Abandons Microsoft .NET in Favour of Native Code

[esposito]

Please read this and come to your own conclusions:

Evernote's popular note-taking technology has recently moved to release version 4.0 with some major structural changes in terms of the technology's previous adherence to Microsoft .NET and the Windows Presentation Foundation (WPF), versus native code.

Writing on his "IT Writing" software application development blog, British journalist Tim Anderson says that the WPF as part of Windows Vista was originally intended to be the main GUI API for Windows, until the notorious "reset" midway through the Vista development cycle marked a retreat from managed code back to native code in the operating system.

So what are the lessons here? Is WPF no good asks Anderson?

"It is not so simple. WPF is brilliant in many ways, offering the productivity of .NET coding and a powerful layout framework.

However, it was a technology which Microsoft itself hardly used in its key products, Windows and Office — a warning sign," suggests Anderson.

By the way, you can read the entire article here:

http://drdobbs.com/cpp/228000453

[pukisoft]

Seems to me they did that because of performance considerations, something anyone would do when such considerations arrive.

.NET was never meant to be as fast or replace c++/native development, but to enhance productivity and portability between windows platforms and to mitigate .DLL hell. I don't see what's interesting about this.

[dilettante]

If this product is moving to native code there might also be other things driving it. One factor might be that the product is now successful enough to offset the cost of native code in order to gain its benefits.

Remember, C# and VB.Net are really RAD tools like VB, Delphi, or PowerBuilder before them. They were never meant to compete with C++ on the basis of flexibility and performance, but to reduce development and support costs. "Managed C++" offered a middle ground and if they made extensive use of it for this product a transition to native C++ might be fairly easy now.

So they may be adding up many factors to help justify a move away from .Net today.

[Max Peck]

So? I know people that have moved from Windows to MAC and others that have gone the other way. I don't consider that reason to declare either "dead" any more than switching development platforms by one vendor portrays the "end" of .Net.

You really seem to have an axe to grind about this. Are you spending as much time developing as you are worrying about this? Seriously!

-Max

[Nightwalker83]

Remember, C# and VB.Net are really RAD tools like VB, Delphi, or PowerBuilder before them. They were never meant to compete with C++ on the basis of flexibility and performance, but to reduce development and support costs.

I thought those languages were only supposed to make it easier for the developer rather than having to learn a lower level language such as C++?

[dilettante]

Actually they were meant to allow programming to be done quicker and by cheaper programmers. Since they were quite effective at this they became the "Cobol" of the PC and saw wide adoption.

But there is never a free lunch. A good programmer can almost always produce better programs working at a lower level. It just can't always be justified.

[NeedSomeAnswers]

Please read this and come to your own conclusions:

My Conclusion is that you are so keen to see Microsoft fail that you will see anything as a sign the .Net is going down the drain.

Esposito, i just dont get why you are so bothered? You have found an alternative language, and you can still program, why do you still bear such a grudge after all this time?

.Net may die at some point, just not yet and it will be still being used by development firms for probably the next 10 - 20 years to produce programs. This desire you have to see .Net dead will only disappoint you.

[ntg]

.Net may die at some point, just not yet and it will be still being used by development firms for probably the next 10 - 20 years to produce programs.

That's a really long time. Granted that the track record of .Net warrants a certain degree of optimism for its future...but still.

[Shaggy Hiker]

It's already had a longer run than VBx-VB6, and I would expect that some flavor will still be around 10 years from now, since VB6 is still around today. Still, I would be a bit surprised if whatever language fills the niche currently filled by VB still looks like the .NET of today in twenty years.

If the point to the post was to cause some kind of panic among .NET developers....good luck.

[ntg]

If the point to the post was to cause some kind of panic among .NET developers....good luck.

Not really. I'm a .Net developer since 2002 and not in a panic. I just think technologies are upgraded and the only constant is change. Kinda hard to think what we're going to use in five years, let alone ten or twenty.

[esposito]

I like your optimism. It certainly keeps sadness away.

Anyway, now I would like you to read this, hoping that it will make you aware that I'm not the only person in the world who believes that those who invest in whatever product a company proposes may sometimes be wrong:

Developers often chose Microsoft platforms because they feel they are the safest bet. "You don't get fired for buying Microsoft," the saying goes. Interestingly, despite how Microsoft has left Visual Basic and FoxPro developers in the dust, most IT development teams still subscribe to that myth.

This needs to change. We produced a white paper on this flawed risk assessment mentality topic a few years ago, and its premise remains relevant today. I encourage you to give it a read and share it around.

Microsoft .Net developers probably don't see this coming, and assume that their .Net apps will be upgraded so they can natively exploit the new operating system. My advice: Don't hold your breath. Prepare to write a ton of new code in JavaScript and HTML5. Don't expect your .Net apps to inherit any awareness of the new Windows 8 UI.

The rest of the article is here:

http://blog.alphasoftware.com/2011/0...andon-net.html

[ntg]

Actually it used to be "IBM" instead of "Microsoft" as I'm sure you know. Currently, IBM provides the only platform I know of where you can still use the same stable, obsolete, six-weeks-for-a-hello-world development tools that you've been using singe the 80's.

You need to be realistic esposito. I've gone through more changes than I'd like to remember in my time and I'm not under the illusion that this will someday change. In the Wintel environment in particular, the only sure thing is rapid change. If you expect the tools generated during the earlier century to keep working in this century and beyond you can stop holding your breath. It's never going to happen.

If you think that Microsoft is the wrong horse or that Microsoft dev tools are the wrong bet, that's fine. There are other options available and you can use them. But an InvalidExpectation() exception should be thrown at you if you expect that Microsoft will not make drastic changes here and there every few years. The one thing you cannot do is adopt a Microsoft technology and hope that it will be evangelized forever with incremental and smooth updates going from version 1.0 through version 99.1, while everything in 1.0 keeps being supported.

Some changes may appear to be bad ones. I'm having trouble making up my mind about 8 myself, I guess we'll just see how it all plays out. Other changes may be good ones. If you remember some of my older posts, I understand the pain of VB developers when the whole VB6 train was thrown out in the cold. But VB6 did not have a wide enough or solid enough foundation to support a lot of things that are child's play with VB.Net. At some point, it had to go.

You may even get the impression that Microsoft is changing things so fast in order to keep everyone running like crazy just trying to play catch-up while they get the upper hand. Even if that's so, that's Microsoft. No point bickering about it.

[esposito]

If you remember some of my older posts, I understand the pain of VB developers when the whole VB6 train was thrown out in the cold. But VB6 did not have a wide enough or solid enough foundation to support a lot of things that are child's play with VB.Net. At some point, it had to go.

I'm not so sure. I believe that the reason why they abandoned VB6 was just because they wanted to create a Java-like platform and a new version of their OS based on that platform. As soon as .NET came out, MS started to deprecate the development of native programs, not because they were not as powerful as .NET software but for purely commercial reasons: they only wanted to beat Java and it didn't matter if millions of developers had to put thousands of lines of code in the bin. Today we all know that their project has failed miserably and the question is whether or not we should follow their next bet.

Delphi was extremely similar to VB6 and it has not died out. Why?

Now MS seems to take back what they said and I'm sure that, if you asked them what language to choose, they would tell you that .NET has no future so you can either address your attention to HTML5 or resume developing native code.

[esposito]

Delphi was extremely similar to VB6 and it has not died out. Why?

Not only has Delphi not died out but apparently it still allows some software houses to develop programs that they then sell to Microsoft for millions of dollars. Just think of Skype:

http://delphi-insider.blogspot.com/2...microsoft.html

P.S. If I think that a few weeks ago some people assumed that Microsoft would rewrite Skype in .NET it just makes me laugh.

[ntg]

Delphi was extremely similar to VB6 and it has not die out. Why?

Because it wasn't. I've gone through Turbo Pascal 3 to Delphi 1 and left at Delphi 3 (or 5, not sure anymore, must be getting older). Even back then, it was clear that for serious server-side work VB6 wasn't a real choice but Delphi could hack it after a fashion.

For me it turned out that front-ends could be developed much faster with VB6 than with anything else. But when you hit a brick wall, that was it. End of the line. You had to resort to extremely ugly, time-consuming and difficult to support hacks to get around VB6's limitations or you had to go to C++ or Delphi and do what you really wanted in a separate COM object, then use that from VB6.

Delphi was in a different league than VB6. It just never had the user base of VB6 and that's why it never gained the popularity that it deserved. VB6 was ultimately a hack from start to finish. Microsoft would have to tear it apart and change it so much to play with the big boys that it wouldn't be VB6 anyway. That's why VB.Net felt so alien to VB6 users from the start.

Of course you're right in saying that a lot of other reasons were behind the making of .Net. That just doesn't mean that there everything was perfect with VB6. Far from it.

[pukisoft]

This whole argument is getting silly. for one as numerous people have reiterated, the new OS will have two UIs, one for classic applications and the new UI.

To say it won't support .NET applications is the same to say it won't support the thousands of applications not written for the new UI, and that's just not going to happen. One of the things that defines windows is its compatibility with older versions of the O.S. applications, and to phase .NET out they'd have wait at least two iterations of the O.S, and that's more than 7 years right there. Add to that the fact that they've invested heavily in their Asp.Net platform, and the framework seems to be going nowhere but forward.

[esposito]

At this point, I advise everybody to switch to Delphi.

[Max Peck]

At this point, I advise everybody to switch to Delphi.

Yeah, RIGHT ... sure thing. I'll put on the brakes right now. Just because you say so, VS2008 / SQL Server 2008R2 and Active Reports 6 (a platform I carefully assembled for it's incredible functionality and potential longevity) is now no longer relevant. Thank you so much! I can't wait to begin coding in Pascal tomorrow! All 200,000 lines of code I've written in C# will have to be re-coded. At least now I've got a future! ;-)

You're getting SILLY esposito! What seemed to start out as a developer just a bit concerned about his career path has turned into a real troll session. Knock it off, please.

-Max

[Shaggy Hiker]

Not really. I'm a .Net developer since 2002 and not in a panic. I just think technologies are upgraded and the only constant is change. Kinda hard to think what we're going to use in five years, let alone ten or twenty.

Sorry about that ntg, I meant to be more clear. My comment was about the reason for the original post, not your post.

At this point, I advise everybody to switch to Delphi.

Why Delphi? You should have switched to C++ years ago. With Delphi, you are still hanging onto one company's product. You sink or swim with the whims of one company, a situation that has left you quite unhappy in the past. C++ will outlast Delphi, .NET, MS, VB6, and pretty nearly everything else. Better yet, you won't be tied to only one dev environment, as there are plenty of good ones out there, even free good ones. All that, and you get a HUGE community of coders supporting it, and you can easily port your knowledge into other languages that are derived from C++, such as Java, C#, and others.

As for .NET being a miserable failure, it has a massive installed user base, has existed longer than VBx (1-6), and still has new versions coming out. For an industry that changes as fast as the computer industry, most companies would love to have a 'miserable failure' of that nature. After all, it took you ten years before you could find articles to cite (and mighty few of them) that bashed .NET to any significant degree.

Of course, most of us don't work for ourselves, so there is somebody else dictating which language we will use. If MS decided to stop supporting .NET today, I would only move away from that language when my employer decided to, as I have some input, but no choice, in the matter.

[FunkyDexter]

Personally I think we should all revert to filing cabinets. They almost never crash on you (though it's far more painful when they do). I never understood what was wrong with a pen and paper and I just wish things would remain stationary.

[Max Peck]

Personally I think we should all revert to filing cabinets. They almost never crash on you (though it's far more painful when they do). I never understood what was wrong with a pen and paper and I just wish things would remain stationary.

You mean "stationery". LOL!

Hey ... I finally reached 300 posts! Not much compared to some of you guys but anyway ... ;-)

-Max

[techgnome]

Funky - oddly just the other night a bunch of us was talking about the card catalog system and looking things up on microfiche ... then we had to explain microfiche to the "baby" of the bunch... :P

esposito - ok ... we could switch to Delphi... thing is, it's changed hands so many times... who makes it any more? I've looked off and on over the years... then I look at the price... and pass. It's cost prohibitive for a one-person shop who just wants to play with it as a hobby. Also... there's no telling who will buy Delphi next month, then decide to kill it too... Then what would you do? At least VB/.NET has had a single owner over the years.

I've been blessed with the ability to pick up virtually any language I set out to learn (except Perl... for some reason that just confuses me) ... I use what my employer needs me to use. For the last 16 years, it's been VB in one form or another. I've also had to do COBOL, C++, C# and even Ada. What ever. Should .NET die... fine so be it, I'll just simply pick up what ever new language my employer decides is the next generation... if it's HTML & JS... so be it... if it's Java, fine... what ever. Or maybe I'll decide that's the time to enroll in culinary school.

In the meantime, there's no point in worrying about the unknown. Esposito, I realize that you're on this Quixotic crusade to convert the world to Delphi... as commendable as it is, using scare tactics like this from seeming biased writers (imho there's no such thing as unbiased writers) doesn't help the case but only increases the push back that you're going to get. True, most of us that have known you (as well as one can through the forums here) know your opinions about .NET and understand that your livelyhood depends on these apps and components that you churn out, and that you have customers that have connectivity issues that limit their ability to get something like the framework... and that's fine... the .NET tool doesn't work for you... the Delphi one does... fine. The problem as I see it... you come off as being arrogant and condescending which doesn't help your case. People like me will simply push back and outright reject you just on principle alone when encountering things like that... no matter how much we might agree (or disagree). I have the same problem with political or religious discussions.

Anyways... what was my point? I don't remember now... I think I strayed from what I was trying to say....

-tg

[baja_yu]

Anyways... what was my point? I don't remember now... I think I strayed from what I was trying to say....

-tg

You were about to ask if we knew any good schools with culinary courses.

[JuggaloBrotha]

I was chatting with a few friends who work for different Delphi shops here in my local area and they mentioned their companies just stated they'll be completely switched from Delphi to C# (.Net) within 14 months. They mentioned it had to do with support & longevity issues Delphi has a lack of, the companies want to make sure their software will run on Windows systems with ease for more years than they can see Delphi handling.

Since I work for IBM and program in .Net, I have no idea how well off Delphi actually is.

[Max Peck]

... Or maybe I'll decide that's the time to enroll in culinary school.

You know ... I've thought about truck driving school myself. There's one down here in Sanford ...

-Max

[techgnome]

You know ... I've thought about truck driving school myself. There's one down here in Sanford ...

-Max

It's like that scene from Top Gun "Hey, Mav, you still have that number for the truck driving school?"

-tg

[Max Peck]

It's like that scene from Top Gun "Hey, Mav, you still have that number for the truck driving school?"

-tg

Zackly!

-Max

[formlesstree4]

So...why is the OP so hateful towards .NET? It doesn't really make any sense to me. It's a personal preference to use a language and the OP apparently is in love with Delphi.

[dilettante]

Part of it is related to the damage the .Net juggernaut did in terms of killing other development tools some people preferred. Part of it is the insecurity of commercial software written in .Net, which is trivially reverse-engineered. Part of it is being a member of the Delphi community, which has always been kind of toxic.

The funny part of this is that Anders was (a.) the creator of Delphi, (b.) vowed to kill Visual Basic, and (c.) was a prime force behind .Net after he caused Microsoft to be sued by Sun and lose $20M and a lot of less tangible things.


So true Delphi fanboys ought to be just in love with .Net.

[formlesstree4]

...Part of it is the insecurity of commercial software written in .Net, which is trivially reverse-engineered...

Xenocode had a great Obfuscator that seemed to eliminate the usage of a program like .NET Reflector.