Microsoft buried the typescript – and reinforces it: Why?

Since the Microsoft Typscript was presented and published in October 2012, the language has become an integral part from modern web development. Hardly anyone works on a web-based UIS, the typescript is likely to pass in the last ten years. But the criticism of the typescript is increasing: noises are becoming more and more voices that criticize the poor performance of the type script compiler.

After all, it was originally an essential aspect of web development that you did not have to wait continuously for a compiler. Typscript has successfully destroyed it. Typescript has greatly improved how to deal with JavaScript – especially in large teams and complex applications – but gradually the speed and ease of web development.

HashiCorp: New names and features for Terraform, Packer, and Vault

But this is the end! On 11 March, Microsoft announced that the performance of the Typescript compiler was intensified by a factor of ten. And it is completely re -written or better, the port. Surprisingly, however, not according to Jung, C# or C ++, but according to GO. This question raises: Why do all the people go?

Microsoft announces 10x fast script

Microsoft was entitled to “a 10x sharp typescript” about a week ago A blog post Even more a video Published, in which Anders Hijilsburg explains what Microsoft Pass is planning with the Typescript compiler in the passage of passes. Anders Hijalsberg is not only a developer in Microsoft, but a major language designer for Typescript and C#. Hence they have widespread experience in the development of programming languages ​​and compiler construction. He announced that the typescript compiler, as we know today, would be replaced by a new implementation in Go. However, Microsoft will not write everything from scratches, but the current code base – which is currently written in the typskript – at the port.

Yes, you read correctly: Typescript is currently written in Typescript, in future it will be written in Go. Microsoft promises a factor of better stability and reliability with ten better performance, too much storage efficiency. It immediately raises the question: what does it mean to me? Why go? What about renewal with existing devices and usually ecosystems?

Why is the typescript slow

To answer these questions, the previous situation must be considered first. The Typescript compiler is currently written in the Typescript. On the one hand, this is an advantage because a compiler almost proves that it works by translating itself. On the other hand, it also means that the typescript compiler, such as any application written in the typscript, is translated according to the JavaScript and should be executed in the same word environment like Node.JS. It has not been explained, but at least JIT-Compied, but not as soon as it can happen with a native code.

Especially with large and complex applications, this means that the compilation slows down unnecessarily because the compiler itself is slow or not at least as soon as possible as it can be theoretically. This is clearly clear in the visual studio code: Microsoft specifies a compilation time of about 78 seconds for a reference system – about a minute and a half. Visual Studio Code is a major project, but still “only” includes 1.5 million lines code, which is not too broad for a real world project.

Aurous acceleration

With the new compiler, which is currently available as an early version, the same process takes only 7.5 seconds – one to one acceleration. This effect is reproducible to be reproducible: the playwright requires the need to compile only a second instead of 11 seconds. With date-FNS, the time falls from 6.5 to 0.7 seconds. Sometimes the effect is slightly strong, sometimes slightly weak, but on average it matches the ten factor. And this is only the first preview – so there may be other capacity for adaptation.

In addition, RAM requires about 50 percent due to the new compiler. This local development may not be particularly relevant on calculator, but it has a massive benefit for CI/CD pipelines. The new compiler is not only sharp, but also very economical. It is impressive.

But how does Microsoft achieve this improvement? The answer is simple: the natural code runs much faster and more economical than the code, which is to be translated by a JIT compiler on runtime.

Why go?

Nevertheless, the announcement was amazing: Microsoft Typescript in battle required many. Hence the option of GO causes discussion. Microsoft clearly explains the decision: the rust provides excellent performance, but its memory management with concepts such as borrowing and ownership will be an unnecessary barrier to 1-1 port of the current code. The position is the same with C ++: Although it is difficult to defeat C ++ in terms of performance, language is considered prone to old -fashion, errors and hard. C# In turn will be the disadvantage that the compiler will depend on .NET-Rundiime, which is impractical, as a compiler should ideally run as a stable binary. AOT compiler from .NET will not support all desired target platforms. Therefore, C fell down from selection.

So why go? Above all, because it is ideologically similar to typskript and therefore simplifies porting. Memory management occurs through a waste collection, so that no manual memory management is required. GO can also create stable binergies for all normal platforms, which have no further dependence over the period. GO is also adapted for performance and parallelization – the required aspect for a compiler.

Microsoft did not choose Go because it would be the “best” or “fastest” language, but because it offers the best balance between performance, maintenance and simple port. This is a sensible decision that is fairly sound.

What does it mean for the future

What does it mean to you as a typist developer? Everything in short term is as always. The type script 5.9 appears next. The following 6 series will gradually mark certain characteristics as “Per Phanavanta” and introduce a change-like change for the new compiler. It is published with Typescript 7.0. There is no specific date yet, but so far it has been two to three years between two major releases. Therefore, publication can be expected as soon as possible in 2027. It is still a long time, but gives enough time for gradual stay. Provided that you do not ignore all the upcoming versions up to 7.0 and only work when the change is unavoidable.

What does change bring? First and rapidly, both locally and in the CI/CD pipeline are formed. The type of script-length server response is also likely to be faster, which enables better IDE performance. In addition, from the Typescript 7, no installed node.JS is required for the compiler – a theoretical advantage, which can actually be helpful in some situations.

It’s exciting

However, the questions are still open. How will the typescript be integrated into other applications and web browser in the future? So far, it has not been a problem because the compiler was compiled in JavaScript. Theoretically, this can be done through the Viv assembly, which supports GO, but practical implementation remains to be seen. In addition, the developers of the extension for the type script compiler will have to adapt their tools. It is still not clear whether the JavaScript-based version of the compiler after version 7 or whether Microsoft makes a hard change.

The declaration was amazing, especially steps to go. However, Microsoft clarified that this decision is final. The GO’s choice is a wise agreement between performance and stability.

What do you think about this development? Do you welcome porting? And how do you feel about the choice of leaving? Write your opinion in the comments!

(RME)