5 ms·
The History of .NET [video]
- ensei5459 9y agoThe mark of a good speaker is one who can hold an audience's attention. This man, is a very good speaker.
- ramenmeal 9y agoYeah I don't know why I just watched all that.
- akirofi 9y agoI attended this same talk last month at NDC London. Very interesting talk with lots of little details, and a brilliant speaker.
- manigandham 9y agoYoutube version (NDC Sydney 2017): https://www.youtube.com/watch?v=IqWar6cEWsA https://www.youtube.com/watch?v=IqWar6cEWsA Also a more recent one from NDC London Jan 2018: https://www.youtube.com/watch?v=trHTLFNFoWk https://www.youtube.com/watch?v=trHTLFNFoWk
- mistermann 9y agoThat channel is chock full of good content: https://www.youtube.com/channel/UCTdw38Cw6jcm0atBPA39a0Q/videos https://www.youtube.com/channel/UCTdw38Cw6jcm0atBPA39a0Q/vid...
- wangii 9y agoNitpick: the javascript engine of firefox is spidermonkey, not nitro.
- zyberzero 9y agoI saw this talk last September in Sweden, at DevIntersection Europe. Campbell is a really good speaker and I really enjoy listening to his podcast dot net rocks. Try it if you haven't!
- klausjensen 9y agoIf you enjoyed this talk, you might enjoy his (and Carl Franklin's) podcast Dotnet Rocks: https://www.dotnetrocks.com/ https://www.dotnetrocks.com/
- thom 9y agoBlimey, that's a blast from the past - I used to listen to that on my commute to my first .NET job out of university in 2002/3. Impressive run.
- cm2187 9y agoTo be honest it has sort of moved on from .net. The podcast is mostly about cloud, javascript frameworks, technologies and soft skills these days. The geek outs are the highlight in my opinion.
- pliftkl 9y agoI really enjoy their podcast despite the fact that I don't really do much .NET these days. They have excellent guests, a good range of topics, and makes for a very nice commute listen.
- hedonistbot 9y agoThis video can be called "Microsoft the Good Parts". Also WPF gets only one mention and the future is Azure? So is WPF heading the way of Silverlight? On a side note, I wasn't aware that the agreement with the US government was meant to expire after ten years. Is it a coincidence that right after that we got the Windows Store push?
- WorldMaker 9y ago> So is WPF heading the way of Silverlight? Yes and no. WPF has a long security support lifetime ahead of it (much longer than Silverlight got), but the present is and has been the UWP XAML stack with all new feature investment going to it (where both .NET and C/C++/"native" code may take advantage of it). (For what it is worth, to go ahead and answer some of the FAQ: With .NET Standard 2.0 the migration story from WPF to UWP XAML is the best it has ever been. UWP apps don't have to Store published and sideloading has been on by default and as easy [if not easier] than Android APK/Windows MSI installation for more than a year now. UWP apps can contain classic Win32 desktop apps using the "Centennial" Desktop Bridge and even communicate with those parts for an eventual/piecemeal migration strategy.)
- twoodfin 9y agoThere’s quite a bit factually wrong with this presentation, some of which is flagrant enough that I have to assume it’s intentional. For one thing, Bill Gates is pictured in his deposition for US v. Microsoft with David Boies, not “US Senators”. For another, Windows 2000 is described as being released contemporaneously with Intel’s Pentium, when they appeared nearly 7 years apart. There’s also a very strange take on US v. Microsoft generally. If you actually watch Bill Gates’ deposition (recommended!) it’s clear that the focus of the government’s case was alleged attempts to use the dominance of Windows to crush Sun’s Java and Netscape’s Navigator, not undocumented APIs as the speaker claims. And the description of the collapse of Longhorn is odd. I’ve never before heard the theory that the team was inadequately resourced due to the requirements of Windows XP SP2. Also an interesting take that Vista only got poor reviews because it was released too early in order to meet enterprise contract commitments. Finally, I have to question a supposedly insider history of .NET that isn’t rich enough in detail to include “COOL”.