lua-users home
lua-l archive

[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index] [Thread Index]




Only for Lua as a possible replacement for Python. The future of Lua
lies where the past of Lua also lies, as a scripting language embedded
in an application, not as the backbone of standalone applications.

I think this may be a false hope. Even though Lua is extremely good at that I have failed to convince popular (in my domain) open source software to use that as their scripting language when they decided they want to add one.

I am an Electrical Engineer designing Power ICs and 90% of my colleagues  don't need to know any programming in any language. I use Lua as an application language to manage my project, generate corner simulation scripts and data analysis on all my chips. I generate and serve my website using Lua[1]. I even used it with IUP and IM to generate Bingo game cards with cartoons for my Son's Birthday.
    The reason I chose Lua was because it was much more elegant and clean than Python. And slowly I realized that I can do anything in it and I dont need to learn any other language. I can use it in embedded projects to making a full website, to making a full blown GUI application to doing complex math and data analysis. I had come to Lua after riding on Basic/Qbasic/Visual Basic/C/Matlab and very briefly Python. After finding Lua haven't looked at any other language but now sometimes get worried that it may die out. 
     So which open source software? I tried to convince KiCAD[2], QUCS[3], Klayout[4] teams to embed Lua but they rejected my proposal with a majority of their developers favoring Python (Klayout used Ruby). Main reason is since it is more popular and better ecosystem. 
     I tried convincing a Semiconductor IP startup I was advising to base their web platform on Lua but they didnt want to take up a lesser used language and risk not finding developers for it plus I was the only one that seemed to think it was the good idea.

Google just recently announced their TPU chipset [5] which makes Tensorflow and Python more desirable.

I am afraid that just having the best technology is not enough. If nobody uses it then it is simply useless. So marketing the technology and removing any pain points even if they are not at all related to the technology (In Lua's case being it's ecosystem and a standardisation path and guidance) is sometimes critical for the technology to survive in the long term and benefit everyone. 

  
[1] - https://milindsweb.amved.com
]2] - http://kicad-pcb.org
[3] - http://qucs.sourceforge.net
[4] - https://www.klayout.de
[5] - https://cloud.google.com/blog/big-data/2017/05/an-in-depth-look-at-googles-first-tensor-processing-unit-tpu