[fpc-pascal] What to do to get new users

[Marco van de Voort]

Op 16/10/2024 om 02:54 schreef DougC via fpc-pascal:
>
>     ---- On Tue, 15 Oct 2024 19:07:58 -0400 *Rainer Stratmann via
>     fpc-pascal <fpc-pascal at lists.freepascal.org>* wrote ---
>     At the Lazarus Congress in Cologne in October 2024, it ended up
>     being very
>
>         interesting. An important question came up.
>
>         Why are no new users coming to Lazarus/Freepascal?
>         Why do we find it so difficult?
>         How can we get new, younger users to come to us?
>
>
>
> The most fundamental reason more people are not adopting Pascal as a 
> programming language is because in the mid-1990's Pascal lost out to 
> Java as the initial language that was taught in university settings. 
> Because that was the last place Pascal was widely used, it quickly 
> became obscure. Also, C was always thought to be the pinnacle of 
> achievement. That is, if you could program in C you were considered 
> elite. That continues today, although those promoting Rust are making 
> a commendable and stalwart effort to have it accepted into the Linux 
> kernel.

I agree roughly with you. It is simply a case of what brings in initial 
users. It used to be TP and Delphiers for either platform or cost but 
that is decreasing. We don't have big patron or other mechanism that 
substitutes.

All the navelgazing on what could be wrong with the presentation or 
initial user experience don't make up for those initial numbers. That 
mostly influences the conversion rate of initial impressions to users 
that try it, but we simply don't have the same number of initial 
impressions.  And those initial impressions are generated by major 
vendors or linux distributors or whatever adopting a language.


>
> Today, Python is probably the most widely used programming language. 
> Its appeal is based on ease of learning and relaxed rules that 
> increase ease of use. While it is as severely flawed language from a 
> strict design viewpoint, it has taken over.

Keep in mind that before Python became data scientist's darling, many 
distributions installed it by default for over a decade to replace 
default distro engineering scripting in Perl (the scripting darling 
before Python). It is not just features, exposure is also a factor.

>
> So, faced with this situation, what can be done? Short of a massive 
> and expensive marketing effort, not much, except to remove obvious 
> attributes that hinder adoption.

Not always.  Keep in mind that entice an initial user to stick around a 
bit longer is only one stage.   Core motivation to start and persist 
with anything beyond the first install, and all other stages till you 
are a core developer ;-) (and each stage has a progressively smaller 
conversion factor).

Overly focusing on the initial user experience is a luxury from projects 
with enormous initial impressions, where every effort for the initial 
user leads to masses of extra new converts, that will sooner or later 
permeate all aspects of the project. Either by paying the bills or 
taking part in development)

But we are not in that situation, and must keep an eye on all users, not 
just focus on the initial user conversion. Making sure the few initial 
converts stick is as important as getting more initial impressions.

> One such attribute is the messy, cluttered look of the IDE. An IDE 
> redesign with a cleaner, simpler, and visually appealing presentation 
> would help enormously. It looks like that is what is being proposed here.
I think the docked IDE by default will have its advantages. Redesign is 
probably not an option.
> I have always found that the self-contained nature of Pascal/Delphi 
> executables is a big advantage over other language systems. Just copy 
> the file and run it, even on a system that has never seen a 
> Pascal/Delphi executable before. If we could do that with the 
> IDE/compiler it would be magic! The install process is far from that 
> goal right now.
>
And the longevity of those binaries, and quick install.   Copies that 
are installed standalone can be copied with components and all. Here at 
work we use Delphi for the main delivered applications, but lazarus for 
everything else because of the limited number of Delphi installs.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freepascal.org/pipermail/fpc-pascal/attachments/20241016/67304a42/attachment.htm>