Some time ago we published an article about "Linux Sucks!" , a talk given annually by Bryan Lunduke in which he offers a profound critique of free software and the Linux environment, while also highlighting the inherent ingenuity of this software and the brilliant ideas behind it, despite all the existing problems. He also reflects on the difficulty of using and configuring a GNU/Linux system and how this difficulty has been gradually overcome over the years.
Well, a few days ago I came across an article from the FSF listing the most interesting projects urgently needed to further develop the GNU project , projects that either didn't yet exist or were in very early stages. This list was aimed at developers to encourage them to get to work. Combining this with the "Linux Sucks" idea, I'd like to share my personal opinion (I'm sure you'll have very different ones) on what Linux's weaknesses are compared to other operating systems and how it could improve.
Simply praising Linux as the best of the best is destructive to the project. Linux kernel or GNU developers don't need sycophants flattering them and telling them how well they do everything; they need critics who point them in the right direction. Starting with Linux users, and from this blog, we should be more critical, because as Kepler said, "I prefer the sharp criticism of an intelligent man to the unthinking approval of the masses."
My critique is this; please add yours in the comments :
- Few software and hardware companies involved: yes, there are more and more corporations that are interested in Linux, creating compatible video games, software and hardware for this platform. But it is still very far from the current situation that Microsoft has with Windows. If they exist multitude of alternatives for programs from other platforms, but it is not about having alternatives, it is about being able to have the same possibilities as on other platforms. For companies to start looking at Linux it is necessary to conquer a greater market share as it has Windows or Mac OS X, but this is quite difficult today. Therefore, the only solution I see is to accelerate the development of projects like Wine or Darling.
- Fragmentation: This is something that has been talked about at length and Linus Torvalds seems to agree with it for being "nurturing", but perhaps seeking a more universal development and not dispersing efforts so widely would solve many problems and improve others. That is to say, it is good to have several distros to choose from or different desktop environments to use the one that we like the most or that suits our needs, but from there to there are hundreds and hundreds of distributions or dozens of graphical environments ... On the other hand , this fragmentation also makes the previous point difficult (for example, by the number of packages RPM, DEB, ... and existing distros), a non-standardization makes many back down. In short, this translates into many developers interested in the GNU / Linux world but all dispersing their forces rather than joining them. Perhaps a development model more like that of other projects like FreeBSD is acceptable.
- Design and functionality: Windows and Mac OS X are masters at this, it is true that they seem systems designed for idiots, but if you want to expand the reach of Linux and bring it to the masses, you have to create more attractive and intuitive environments. Some programs lack GUI or are not very functional, you have to change this. Canonical has managed to capture this idea for Ubuntu and is doing an extraordinary job, that is why it is one of the most used distros (without detracting from other equally remarkable projects in this regard). We all want a system as beautiful as Mac OS X and as functional, without falling into Apple's mistake of neglecting its terminal.
- Network stack: Last year, Facebook tried to improve the Linux network stack by hiring experts to help develop the kernel in this regard. The Linux network stack isn't horrible, but it can be improved. FreeBSD is an example to follow, since it has an enviable network stack, and it is what Facebook was after, to match or improve it.
- Safety: that with GNU / Linux you can be more secure (some distributions more than others) than with other systems is real, but we cannot relax and say that Linux is the safest in the world, since it is not true. And in this case I go back to BSD to give an example of security with the OpenBSD project. For the Linux Foundation and the FSF to allocate resources to audit the security of their projects or dedicate a team of experts to polish the security of the system would not hurt.
- Pressure group: there is an anti-Linux "lobby" that comes from some companies like Microsoft, despite the passivity that Satya Nadella has highlighted in the era, and Apple. But in one way or another, the pressure they can exert due to their monopolistic situation means that drivers, software and technologies do not reach Linux sooner. Perhaps the FSF or the Linux Foundation could do something in this regard by pressing in some way so that, for example, things like UEFI Secure Boot do not happen, or ensure free standards as AMD does. Also, not only that, you could also do pro-Linux campaigns. Have you seen any commercials for Linux on TV or some other off-line media? And if I repeat that same question, changing "Linux" to "Microsoft Windows" or "Apple"? So the answer varies dramatically.
Perhaps instead of criticizing Linux's "enemies," we should learn from their strengths to improve. Taking ideas from OS X, Solaris, FreeBSD, Windows, etc., doesn't have to go against the Linux spirit, as long as the goal is to make Linux the best. Let's turn enemies into learning opportunities and disadvantages into advantages.
Do you see other things that need fixing? Feel free to share your feedback in the comments …