Why isn't Lazarus / Free Pascal more popular?

[marcov]

I cannot say that I ever had the pleasure of using punch cards but I bet nobody wrote bloated software back then. 

True. Minimalism for minimalism's sake is dead on arrival.

[JD]

Especially all these newfangled "solutions" like containerization, which are primitive workarounds for the problems of an old and poorly designed operating system. Containers and virtual machines are unfortunately only prostheses, not the optimal final solution.

Yes, but it also depends on what one wants to do. Containers solve the problem of "it works on my development machine, but it doesn't work on another machine". Before using containers, I had to think of what OS (Windows/Mac/Linux) I would deploy my application onto.

Since I started using containers, I only develop on Linux, test in a local Linux container AND deploy to any OS that can run a minimal Linux OS container like Alpine Linux. This example in the Wiki https://wiki.lazarus.freepascal.org/Docker_Containerization was a good starting point for me. It gave a new lease of life to my old Lazarus applications. It is hard to call Object Pascal outdated when a well-designed containerized Object Pascal server on legacy machines can handle considerable loads in real-time.

Another use case is multi-tenant database applications. Running databases for different tenants on the same machine but listening on different ports is trivial with containers. One can even mix and match different versions of the same database on the same machine in different containers. Testing the latest version in its own container before deployment, while the legacy versions remain untouched. Using tools like docker-compose also means one can deploy an entire full stack web application in under 10 seconds.

Are containers for everybody? No, but once you try them, especially for server-side development and get used to it, it is hard to not appreciate the convenience they bring.

[JD]

I’m not sure if I’m understanding this right, but do your containers work without an Internet connection?

Yes. You can decide whether to expose containers to the Internet or not. The ones on my development machine are not exposed to the Internet. They are just for development and testing. The only time I need an Internet connection is for building the initial image the container is created from. When I have to download Alpine Linux for example. But once it is done, I don't need to repeat it.

I have other ones in VMs and VPS that need an Internet connection to work because data has to be transferred.

[LV]

I had some free time this evening. I translated the optimized Pascal code (Reply #72) into C++. Here are the results averaged over five runs of the programs on Windows 11 (Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz).

The matrix multiplication benchmark for a size of 1000x1000.

Code: Text  [Select][+][-]

1. -----------------------------------------------------------------------------
2.                          |  FPC 3.2.2  |   C++ (GCC 14.2.0) | difference    |
3. -----------------------------------------------------------------------------
4. Naive Execution Time, ms |  1447       |   1322             |    9%         |
5. AVX Execution Time,   ms |  44         |   40               |    10%        |
6. -----------------------------------------------------------------------------
7.  

[duralast]

Do you mean "applications" created using "modern" technologies such as Python, JavaScript (node.js, Electron) and the like? Perhaps this will end at some point if governments and users will strongly push for reducing the energy consumption of electronic devices containing CPUs (from microcontrollers through smartphones, tablets, laptops, desktops to industrial controllers). Including energy consumption for their production and disposal. Of course, the boards of companies (corporations) are strongly opposed to this (because their profits will decrease). For now, this problem is ignored by governments and most ordinary users.

Although you did not mention AI, you did bring up energy consumption. The following may be of interest.

The IEA’s special report Energy and AI, out today, offers the most comprehensive, data-driven global analysis to date on the growing connections between energy and AI. The report draws on new datasets and extensive consultation with policy makers, the tech sector, the energy industry and international experts. It projects that electricity demand from data centres worldwide is set to more than double by 2030 to around 945 terawatt-hours (TWh), slightly more than the entire electricity consumption of Japan today. AI will be the most significant driver of this increase, with electricity demand from AI-optimised data centres projected to more than quadruple by 2030.

In the United States, power consumption by data centres is on course to account for almost half of the growth in electricity demand between now and 2030. Driven by AI use, the US economy is set to consume more electricity in 2030 for processing data than for manufacturing all energy-intensive goods combined, including aluminium, steel, cement and chemicals.

[TRon]

Although you did not mention AI, you did bring up energy consumption. The following may be of interest.

... and water footprint see f.e. https://itcc.ieee.org/blog/the-hidden-cost-of-ai-unpacking-its-energy-and-water-footprint/

[VisualLab]

I refuse to use modern windows. I agree with visuallab , operating systems are poorly written and allow things that they shouldn’t. After all this time i doubt that it’s happening by accident.

The current operating systems are simply old, i.e. based on old assumptions, old ideas and old solutions. This is not some kind of conspiracy, it is simply that the boards of IT corporations do not feel like spending money on a new solution, because developing a new one would require a lot of money and time. They prefer to use worn-out (archaic) software. Especially since its design (assumptions, solutions) comes from the times of mainframe computers. As long as something works, they are happy. And what is important for the boards, a crowd of open source programmers work on it for relatively little money (in various foundations). If something goes wrong with this software, it is temporarily patched and repaired using adhesive tape and a piece of string to make it work. And the profits are not invested but eaten away (usually on nonsense). However, everything comes to an end sometime. On the other hand, most computer users are laymen, for them it works, so they do not see a problem. Just like for people in the times of horse-drawn carriages, hand-held weaving machines, etc. things. They also thought back then that everything was working fine, why the hell do they need electricity?

[VisualLab]

Do you mean "applications" created using "modern" technologies such as Python, JavaScript (node.js, Electron) and the like? Perhaps this will end at some point if governments and users will strongly push for reducing the energy consumption of electronic devices containing CPUs (from microcontrollers through smartphones, tablets, laptops, desktops to industrial controllers). Including energy consumption for their production and disposal. Of course, the boards of companies (corporations) are strongly opposed to this (because their profits will decrease). For now, this problem is ignored by governments and most ordinary users.

Although you did not mention AI, you did bring up energy consumption. The following may be of interest.

The IEA’s special report Energy and AI, out today, offers the most comprehensive, data-driven global analysis to date on the growing connections between energy and AI. The report draws on new datasets and extensive consultation with policy makers, the tech sector, the energy industry and international experts. It projects that electricity demand from data centres worldwide is set to more than double by 2030 to around 945 terawatt-hours (TWh), slightly more than the entire electricity consumption of Japan today. AI will be the most significant driver of this increase, with electricity demand from AI-optimised data centres projected to more than quadruple by 2030.

In the United States, power consumption by data centres is on course to account for almost half of the growth in electricity demand between now and 2030. Driven by AI use, the US economy is set to consume more electricity in 2030 for processing data than for manufacturing all energy-intensive goods combined, including aluminium, steel, cement and chemicals.

Although you did not mention AI, you did bring up energy consumption. The following may be of interest.

... and water footprint see f.e. https://itcc.ieee.org/blog/the-hidden-cost-of-ai-unpacking-its-energy-and-water-footprint/

You are both absolutely right. This is definitely part of the problem of excessive resource consumption. Perhaps it is the excessive energy consumption (and consequently the need to cool the equipment) that will be the wall that the current stage of AI development will reach, which in turn will cause another AI winter. And until the energy consumption problem is solved, this AI winter will continue. In any case, overcoming excessive energy consumption will require a change in the technical solutions used to produce computers (at the same time: materials other than silicon, a larger scale of integration, a changed CPU architecture, a return to optimised machine code, etc.).

[440bx]

After all of the fossil fuels and fossil water from deep wells is squandered on ai and other nonsense,

if you believe Einstein (I do), human's "other nonsense" will greatly exceed the damage caused by A.I.  A.I has a whole lot of catching up to do before it matches the damage the "other nonsense" has already caused.

There is no end to the nonsense in sight.  On the contrary, the indications are it's going to get a lot worse.  It will still be a while but, the "talented leaders" we elect will eventually accomplish "Soylent Green".  It might get worse but, I'm an incurable optimist.

IOW, artificial intelligence cannot beat genuine infinite stupidity.  Not a chance!

[LV]

if you believe Einstein (I do)

Einstein was also mistaken. 

https://interestingengineering.com/culture/5-of-einsteins-most-epic-fails-that-prove-he-was-human

[440bx]

Einstein was also mistaken. 

Nobody is perfect.  Even Leonhard Euler, likely the smartest human being who ever lived, was occasionally mistaken but, that does not mean they weren't right most of the time.

[440bx]

440bx, while Einstein and his famous phrase is fairly accurate. The consequences of someone doing something stupid before modern technology used to be mostly limited to themselves, their household or maybe their village. It had no affect upon anything thousands of miles away. This is no longer the case.

That's hardly the case.

I'll offer two examples among tens of thousands.

1. The addition of lead to gasoline to prevent knocking.  One heck of a big "village" got affected.
2. Shortly after radium was discovered, all kinds of products appeared with radium, lipsticks, soaps, perfumes, toothpaste, you name it.  That caused all kinds of deformities and sicknesses in people (not to mention those in unborn babies.)

Bonus:

For centuries one extremely popular method of treating diseases was to bleed the patient.  No one knows how many hundreds of thousands of people died as a result of this "tried and true" treatment.

[JD]

Science is founded on experimentation and trials. They are not without consequences, and it has been like that throughout history, even to this day (think clinical trials).   

As programmers, many of us have lost data, destroyed computers etc. in our testing (not production, I hope) environments too. That is why languages like Java and C# were introduced. The safety of managed code was introduced to prevent disastrous unmanaged code from destroying things. Sometimes we need to be saved from ourselves. 

[marcov]

This thread is diverging, let's get back ontopic.

[VisualLab]

. Just like for people in the times of horse-drawn carriages, hand-held weaving machines, etc. things. They also thought back then that everything was working fine, why the hell do they need electricity?

After all of the fossil fuels and fossil water from deep wells is squandered on ai and other nonsense, we will be lucky if we can go back to horse drawn carriages.

I didn't mean to complain about AI (or the fact that it exists and is used). Nor do I intend to "excommunicate" her IT-ically. It is obvious that AI has its uses. The problem is people who believe that it is the golden mean to solving all of humanity's problems. And they fuel this hype about AI. And we know that such miraculous tools do not exist. Every tool has its limitations and scope of applications. There is no room for faith in technical fields. Either it is this way or it is another way - there is only knowledge. Moreover, apart from the laymen, this hype is fueled by the media, because they get money from it (and by corporations, because they want to earn "big money" from it, which so far is not very good). On the other hand, I complained about old IT solutions that should be replaced with something newer.

What I wrote about AI (and the next "winter") concerns the stages in the development of AI, computer science and computers in the context of energy consumption. Energy efficiency is desirable not only because of AI. The number of electrical (and electronic) devices is growing. And each of them generates heat (which we do not want). So as you know, reducing energy consumption will also reduce the amount of heat generated. On the other hand, we want efficiency in operation. Any way to reduce energy consumption while sacrificing performance as little as possible is beneficial. Programming languages ​​are one of many ways to reduce energy consumption. And here Pascal and Object Pascal are a plus.