Loading User Information from Channel 9

Something went wrong getting user information from Channel 9

Latest Achievement:

Loading User Information from MSDN

Something went wrong getting user information from MSDN

Visual Studio Achievements

Latest Achievement:

Loading Visual Studio Achievements

Something went wrong getting the Visual Studio Achievements


evildictaitor evildictait​or Devil's advocate
  • A UWP Web Browser written in HTML, JavaScript, and CSS (and some C++/CX)

    It's not a browser "built in JavaScript and CSS" any more than Firefox is. The Chrome around the edge of the WebView control is scripted with JavaScript (just like it is in Firefox).

    All of the logic of the heavy lifting, rendering, parsing and so on of this "new browser" is done by the WebView control, and that's just IE written in C++ under the hood.

  • Why does Windows think it knows better than me?

    , cheong wrote

    The force reboot part sounds particularly lame to me.

    For server SKUs, each process has right to delay shutdown for 5 minute, they may get away from it with clean shutdown. Since Vista each process on Consumer SKUs has only 2 seconds to do cleanup. That means applications with lots to write when quit will do incomplete write on shutdown, and with transactionaly write on NTFS, it means your previous work is gone.

    Realistically, I think it's the opposite. Servers have much more exposure to hackers than ordinary users. For example, it took hackers about 4 hours after Heartbleed was announced to have scanned the entire IPv4 range for servers that were vulnerable and begin exploiting them.

    Servers also have no excuse for not building in redundancy. If your server can't handle an unexpected reboot, it can't handle a server crash either. All servers realistically should be at least pairs of machines as a clustered availability set so that one can handle the requests when the other goes down. In such a scenario, Windows Updates hard-rebooting the machine isn't even a problem so long as it doesn't hard-reboot all of the machines in the availability set all at the same time.

  • A new Crockford talk

    , bondsbw wrote

    @evildictaitor:  I agree that iOS was the reason both Flash and Silverlight died.  (That was actually what I meant in my first response on the matter... I just didn't mention it explicitly.)  But I feel that HTML5 put the nail in the coffin... otherwise there would have likely been a need for something like Flash or SL at some not-insignificant level.

    I think that's the wrong way round. HTML5 was invented because Flash needed to be killed for iOS. On desktop machines, HTML's lack of slick graphics, video, audio or precision between browsers wasn't as big a deal, because you got those features via Flash. Apple's decision to remove flash necessitated adding to HTML all of the features that Flash was previously providing.

    It's not that HTML5 killed Flash. It's that killing Flash created HTML5.

  • A new Crockford talk

    , Harlock123n​ew wrote

    I did not view the abandonment of SL as a failure of SL more a casualty of a political war. Either way I just see this fellows IDEA ending up in the same waste basket as SL after countless hours spent...

    A Superior/more Secure Technology is not necessarily a Successful Technology....


    Silverlight didn't die because of HTML. It died because of iOS.

  • A new Crockford talk

    , fanbaby wrote

    @Dr Herbie:

    I'm not employed, let alone employing anyone, but if i was interviewing someone for a C# job, and he didn't know who Crockfod is...

    Then you'd ask them some relevant C# questions because knowing who Crockford is isn't at all correlated to their ability to perform the job you're potentially hiring them for?

    Or was that not what you meant?

  • Did someone here order a binary byecode for the web?

    , cheong wrote

    Btw, since Java Applets didn't make it to land on iOS landscape, I doubt this WASM thing can, but we'll see...

    WASM is (for now) compatible with JavaScript, so anything that has a JavaScript engine will run it. Browsers that support it natively will just run it faster.

  • Did someone here order a binary byecode for the web?

    To be clear: I work with a lot of people day-to-day whose full-time job is killing Flash. But before Flash existed, the web was documents, not applications. Flash showed the world how the web could show us movies, slick animations, web-casts, awesome interactive games, online-radio, pod-casts, YouTube and Netflix.

    Under the hood, Flash isn't great. But it enabled millions of web-developers to express themselves more freely, it drove forward the web as a place where you could write awesome applications that work on every platform, and made billions of dollars for Adobe in the process. And people forget that before Flash, any interactive content required downloading a new ActiveX-control, and all embedded media was embedded Windows Media Player.

    And it's perhaps also a surprise that despite how many security vulnerabilities have been found - and continue to be found - in Flash, Flash made the world safer by making the overwhelming majority of Active-X controls obsolete.

    Maybe Flash lived too long. But there is no question the web is better because it existed at all.

  • Did someone here order a binary byecode for the web?

    , Bass wrote

    The only way [Flash could have succeeded] is if you become literally the dictator of the world and forced all these random adversaries with competing interests to use your pet technology on pain of death.

    Flash totally dominated the browser market (on all browsers, including open-source ones) up until it was killed by Steve Jobs because he didn't like the fact Flash was draining too much battery on his proprietary iPhone devices. Flash's life and subsequent prolonged death have absolutely nothing to do with whether it was open-source or not.

    I've always said that for Silverlight, Microsoft made a strategic error in killing it and burning the developers who had relied on it. Instead, it would have been really cool if it cross-compiled your C#/XAML code to Javascript/Canvas. That way people would have been able to continue using the Silverlight tool-chain, and would have automatically extended the reach of Silverlight apps to Mobile browsers, and would still have a strong community even now.

    Hopefully with the introduction of WASM, we'll be able to get more high-end toolchains for development of browser applications in languages that scale well to teams of developers and millions of lines of source-code. When that happens, you won't need to be a billion dollar Silicon Valley company with your own proprietary "compile-to-Javascript" solution to be able to build compelling high-end applications for the web.

  • Did someone here order a binary byecode for the web?

    Actually, it's really better news for people that don't like JavaScript.

    If Microsoft can compile other languages down to asm.js it will mean people will be able to program their websites in languages that they are more used to, and that scale better with large teams of developers and high code-churn.

  • Windows to support SSH natively

    , fanbaby wrote

    Win32 = monster of an  API, a mishmash of 2 quite unrelated parts, that only now, with nano are being rightly split.

    Win32 isn't an API. It's several different collections of APIs. The graphical/user parts have always been exposed via GDI32/USER32, and the file-system bits are exposed through KERNEL32, and the networking stack is exposed through WS_32. Other bits are exposed via other DLLs.

    Nano doesn't have support for GDI32/USER32 (because nano runs "headless"), but saying it doesn't support Win32 is just a straight-up category error.