1090
Update: Unity office death threat was made by a Unity employee
(www.polygon.com)
This is a most excellent place for technology news and articles.
Three years ago after trying Unity for a month I chose to learn Godot instead. I see now how right that decision was. Well done past self. Have a future cookie.
For me the rule that has always worked is "bet everything on open-source". It has always paid off.
When people at uni used Matlab, I learned R (before R-studio even existed) and python. I moved to linux as soon as I could. I never wanted to learn anything MS or Apple specific, or proprietary technologies such as visual studio, excel, vba, c#, SAS. I went on docker ASAP...
Now the world in my field runs on open source tecnologies, and I am the leaders of the "new stuff" wherever company I go.
On the long term learning open source solutions is always a win. Best case scenario it becomes the industry standard, worst case scenario it gives you the know how to master proprietary tools
C# and Visual Studio are pretty great now, and they don't lock you into Windows at all. Most of C# is open source.
My experience is very different. I know a lot of c# developers, they are locked, even if c# now looks open source. They are locked as a mac user is locked to mac. C# is the most monopolizing language I know. Usually people know more languages, they easily move from one language to the other, from one programming style to the other depending on the task, they can easily learn different tools, different ways of doing stuff. All c# developers I know seriously struggle to move out of their conform zone, that is visual studio. To the level that many even struggle with vscode. And the way of doing things of visual studio is usually good for windows but it is the worst when doing more "modern" things, from ai to kubernetes
So your experience is that you've never heard of Swift and you haven't heard anything about dotnet since Visual Basic in 2002.
Classic example of experience bias. "Well in my experience this is how the world works"
Preach it! One of my colleagues writes all his machine learning code in Matlab. Brilliant person, has done some incredible research, but can do anything with the code because no organization is going to bring Matlab into its clusters and pay for all the licenses needed to run it. So while plenty of presentations and papers have been written of this research, the actual process of letting people use it takes an additional army of Python developers to translate and test every new feature and enhancement.
This is what happens when you build your career around walled garden platforms. Inevitably, you'll reach a dead end. Focus on learning tools that enable you the most. Open source will always win in the end, because it will never come with this very heavy piece of baggage that proprietary tools have. This is why the internet is built on Linux and not Windows.
Unity is the same way. When you build your career on a technology that a single company can strip from you on a whim, that's a big risk. I really hope that Godot and other open source engines take off after this. It will be a painful transition for many developers, but hopefully it's a lesson very well learned.
For what it's worth, C# is a ECMA standardized language (https://www.ecma-international.org/publications-and-standards/standards/ecma-334/) and has a linux-based implementation (mono -- https://www.mono-project.com/docs/about-mono/).
Though it is hard to overcome the obvious Windows origins of the first implementation.
Mono is becoming outdated now that dotnet just supports Linux.
(It took a lot from mono to do so.)
I was completely unaware! I haven't been keeping up with the .net / c# ecosystem mostly because my job doesn't make me. That's both good and bad, I guess.
For me, it’s “learn everything”.
The best devs in XYZ language/framework aren’t the ones who are experts in XYZ, but the ones who are just good enough in XYZ and 15 other things that they see what XYZ excels at, and lacks, and how patterns from elsewhere could be adapted to supercharge XYZ’s strengths and mitigate its weaknesses.
Good move. MATLAB is trash.
Eh, depending on your career Excel is worth a tiny bit of time given its pervasiveness and how powerful it is. But like you say, learning open source will make Excel a piece of cake.
If you think MATLAB is trash, wait til you see Octave!
That's a good rule. I only accidentally got into open-source, but now that I know what it is and what it's all about, I am totally sold on it and will almost always choose open-source over proprietary alternatives.
I agree for the most part, but Excel is just so good.
They say “Excel excels at excel”.
I prefer Google Sheets over Excel but cannot tell you why.
Is it the UI? We use Google sheets at my work and I hate it. Missing formulas & formatting options that I like.
It works fine if you do the basics, but its not as full featured as I want it.
Really? My Google Sheets has so many formulas and conditional formatting. There's so much I can't do in Excel that I can do in Sheets :/
Also, adding checkboxes in Excel is a pain lol. They made it so difficult.
I don't use excel other than as a glorified calculator. I don't use word as well. My department knows and I am pretty open when I do interviews. If the job requires to open more than 1 file Excel every 2 months, I am out. If I need to open a single excel sheet with VBA, they wasted my time.
Excel is fine, is what people do with excel that is not fine
What do you use for spreadsheets, libreoffice? I could see not liking a specific program but I love a spreadsheet and use them constantly. I use libre for ideological reasons but don't find it as convenient for certain tasks as excel or google sheets.
Python/pandas, R or a real database depending on the task.
I don't dislike excel. I dislike what people do with excel. And I dislike vba
I'm curious why you chose R as an alternative to Matlab instead of Scilab. Scilab is specifically designed to be a free and open source alternative to Matlab.
For my thesis I was writing some test software and when deciding which language to use Matlab was immediately ruled out due to the cost (and also the extra cost for the toolkits I'd need). I instead went with Scilab which now means that anybody wanting to reproduce my results can do so freely.
Because at the time I needed more the statistical and plotting part. Ggplot was not yet a thing, but R was already pretty nice for plotting and stat.
I was using other, lower-level languages for more intensive tasks, as I was working in high performance computing.
Makes sense, thanks for the response! It is kind of fun to have a mix of the higher level (like R/Scilab) and lower level (which I used Fortran for mine).
We all started with fortran the old times in hpc and scientific computing. Kids nowadays don't know the thrill of retro engineering fortran code! /s
I wish I could bribe past me with future cookies...
Leave one out overnight and tell your present self that they can have it in the future if they do x before tomorrow. If you succeed, then you get a cookie. If you fail, eat the cookie anyway. At least you tried.