Object Pascal decline?

[avra]

But I think what we did wrong was marketing. We have yet to have any substancial marketing at all. Like most programmers we tend to think that good coders will see just how awesome our system is -- but fact is people stick to cool websites like glue.
jQuery is trivial compared to our RTL. Yet they enjoy millions of users for something I could slap together over the weekend using object pascal. Its very sad to see -- especially when you consider the potential for evolution of our language and the way we write programs.

Your SMS marketing efforts were focused on pascal community, which was good for start but not so good in long run. You should think where all your zillion target coders hang out (forums, blogs...), and then you should hit them there showing small code snippets that tell them how SMS is superior to jQuery or whatever else they are using. These snippets should be just short teasers with links to your articles (where you already excel) or youtube videos and tutorials. Being consistent, with properly chosen snippets you get new customers. I think this is a full time job for one person, and that work is much more important then having another developer. Besides marketing tips, I also think that your business model should change so SMS becomes free for all open source projects, universities and students, and you charge only for commercial use.

[mai]

But we are banned by embarcadero, so we must have done something right
They even threatened to poll their funding for PASCon if we were allowed to present out product. I dont know what is wrong with them, but that's how they treat us. As a threat.

They are aholes just like most of team b....
I quit using delphi in 2012 and don't miss it at all.

you're both right.

Laz can do it.

phugg Team 'B'.

[BSaidus]

Me too !
I quit using DElphi & I fell Free !!
Good work FPC/Lazarus teams & Thanks a lot !

[ArtLogi]

The problem with such discussions is that they overly focus on only the language as driving factor for adoption.

That only works if you actually expect the users for your tool to come from FPC/Lazarus/Delphi ranks. But those are mostly users that are making native apps.

For most users the attraction of Delphi (and by extension, Lazarus) is not just "Pascal". It is a combination of GUI design, standalone applications, and good low level connectivity.

Trying to make a pascal for the web (or any other radical different environment) means that you will only moderately attract existing Pascal users (which are generally biassed towards using native standalone binaries). Some might even use C# and Java mainly, and only use Pascal to make services, conversion and migration tools and hardware interfacing. At least that is what I see at our customers.

That means that you must get the bulk of your users from non-pascal users. And they already know the languages you derive your concepts from directly, so for them there is less of a plus.

FPC could copy umpteen features from C#/Java/python/javascript, and that picture still would be the same.

The documentation. Why none (well someone did mention .. applauses) mention the documentation, if there is no documentation the next generation of programmers will not jump to the board as easily (since no easy way to learn compared to "competitors"). Yes, many of the other languages have plenty of documentation, but FreePascal itself do not have good quality non-technical documentation which I mean by that is it lacks the user level documentation and by user I do not mean a computer/software majors or old professionals with 40 years experience in programmin and last 20 years of Delphi (you don't need the documentation (only reference) with your own build ecosystem). It is also a crime for the language dialect (FP) itself to blindly say that there is plenty of external resources out there, there is not, there is plenty of documentation for other OP dialects from the past. This is something that the FPC (and in some extend Lazarus) is missing, the documentation part of the project is in sidetrack. There is no good web infrastructure to easily maintain and publish the documentation, code examples, structure snippets etc. The Wiki is cluttered and have hit and miss google UI atm. The Lazarus code filling or what ever it is called (the thing what pops out after you press dot after some objects etc) is nice, but with it you need to already know something. There is no clear path of documentation from novice to advanced user. It is unbelievable force factor, the word I mean.

[ArtLogi]

Smart Mobile Studio is completely independent.

It is not in LinkedIn?? I did found out the Facebook page.

[vicot]

The documentation. Why none (well someone did mention .. applauses) mention the documentation, if there is no documentation the next generation of programmers will not jump to the board as easily (since no easy way to learn compared to "competitors"). Yes, many of the other languages have plenty of documentation, but FreePascal itself do not have good quality non-technical documentation which I mean by that is it lacks the user level documentation and by user I do not mean a computer/software majors or old professionals with 40 years experience in programmin and last 20 years of Delphi (you don't need the documentation (only reference) with your own build ecosystem). It is also a crime for the language dialect (FP) itself to blindly say that there is plenty of external resources out there, there is not, there is plenty of documentation for other OP dialects from the past. This is something that the FPC (and in some extend Lazarus) is missing, the documentation part of the project is in sidetrack. There is no good web infrastructure to easily maintain and publish the documentation, code examples, structure snippets etc. The Wiki is cluttered and have hit and miss google UI atm. The Lazarus code filling or what ever it is called (the thing what pops out after you press dot after some objects etc) is nice, but with it you need to already know something. There is no clear path of documentation from novice to advanced user. It is unbelievable force factor, the word I mean.

You nailed it. A better documentation is badly needed.
To that I would add: better tutorials, especially video tutorials. I find it ironic that (as is the case with many other languages) you find lots of tutorials that cover the basics (syntax, control structures, etc), but not much beyond that.

Something that would be *greatly appreciated* is a proper video tutorial for Lazarus development. Again, covering not only about isolated aspects of the language, but covering especially the program building progress. That is, how the programmer actually develops a whole program, phase by phase. You want to see how he jumps through the hoops.
Such a video tutorial would be a *huge* success, I am sure. It would pull in a large number of programmers from other languages, especially those who are sitting on the fence.

[sysrpl]

vicot,

I am the developer behind http://www.getlazaurs.org. You mentioned video tutorials which provide a step by step of the applications design and construction process with Free Pascal and Lazarus. Can you clarify more precisely what you mean by this?

Myself I imagine the process of designing and building applications with Lazarus to be not particularly difficult or complex. I do not see a large number of steps involved.

For example:

Create a new project and save the files
Drop components on the form
Use the object inspector to alter component properties and create events
Write code in the event handlers
Press F9 and test your application

What more would you like to be covered? How to use the Free Pascal language? How to use some of the more common classes or functions? What exactly? Please be specific.

And regarding documentation, I've made the effort to at least clean it up. I've also added incremental searching:

http://docs.getlazarus.org

I believe I understand your issue with documenting how to use the Free Pascal language, but many books have been written on the subject some twenty years ago when it was called Delphi. It hasn't changed that much. Perhaps someone could rewrite a summary of the basics of Pascal using one or two of the better books as writing style guides. If anyone wants to take on this task I can host it using a new prominently featured section at getlazarus.org, possibly linked into the documentation pages.

[marcov]

It is also a crime for the language dialect (FP) itself to blindly say that there is plenty of external resources out there, there is not, there is plenty of documentation for other OP dialects from the past.

I do think we could use a good introductionary book.

I don't see much use in trying to do this ad-hoc/grassroots.  Nearly all the language mostly have the core docs written by one hand or a small team (and often by O'Reilly) . The book is then converted and published on the web

All the smaller languages that use the web/wiki method end up in infinite  clicking around in wikis chasing your own tail.  Terms are never defined before use etc etc.

So it should be a document with head and tails, a narrative from chapter to chapter each chapter adding a new concept.  Videos can be linked from it to give more insight in the manual IDE handling.

(personally I only like short videos on specific IDE topics, and sometimes a overview videos. I hate videos as main instruction tool, they are always too fast or top slow)

This is something that the FPC (and in some extend Lazarus) is missing, the documentation part of the project is in sidetrack. There is no good web infrastructure to easily maintain and publish the documentation, code examples, structure snippets etc. The Wiki is cluttered and have hit and miss google UI atm.

IMHO this should not be webinfra at all, but a dedicated documentation team.  Wikis are only part of the documentation as  the filing cabinet for everything that is not covered by other means.

The Lazarus code filling or what ever it is called (the thing what pops out after you press dot after some objects etc) is nice, but with it you need to already know something. There is no clear path of documentation from novice to advanced user. It is unbelievable force factor, the word I mean.

No, there already are reference docs etc for that. They could always use more manpower, but that is not the biggest missing piece.

What we need is a good language tutorial book, a good introduction into the LCL concepts etc. For that we indeed rely on Delphi book availability (and a sizable of new entrants having Delphi knowledge), that is a known problem, and is needed for all users, also intermediate users that need to read up on a new feature (e.g. introduced by a newer delphi).  Jeff Duntemann was working on a book, but afaik he was more into the TP dialects.

I don't think the current docs even mention e.g. "event driven", and shortly explain the concepts.

When you have base tutorial (also as a framework to add more in the future), you can start making parts of it more accessable to total newbies when you sense problems.

BUT, coming back to the thread title, I don't think that is the main reason for why we are not as popular as javascript. A project driven by what, 30 parttime volonteers can't match something that has whole industry conglomerates dedicated to it.   You can't solve that with a couple of videos.

I think it is more realistic to think about the next step in docs (and IMHO it is expanding what is there, and the missing tutorial category), then to pretend we can match the umpteen PHP tutorial sites. Because they can always pick another one, and dive into the ref docs or buy a book if something goes wrong (which encourages a smartshop mentality). But for FPC that would be the ONE resource, with only the wiki as minor addition.

[Nebula]

This thread has split off to "Documentation / tutorials need improvement (was: Object Pascal decline?)"
http://forum.lazarus.freepascal.org/index.php/topic,31272.0.html

[kveroneau]

I don't like bringing back old posts usually, but a Google search on why Pascal declined brought me to this post, and I found some of the replies very useful.  Also, in the context of my additions to this topic, I think having the rest of this thread available is useful.

Firstly, I'd like to bring up an interesting case where a not so known language magically gained popularity by the masses, I am still researching on why Ruby on Rails soared to the heights it did, Ruby the langauge has been around for a very long time before Rails, and Ruby itself wasn't really used much until Rails came onto the scene...  I almost feel if someone with enough marketing power could create a very powerful and easy to use Web framework that could blow up like Rails did, Object Pascal could potentially make a huge comeback.  Pascal is unfortunately in the back of most peoples minds, as it has the stigma as the language one might have learned in school to initially learn programming.  Pascal has come a long way since it's academic roots, and today has very similar features to powerful langauges like C, and has many advantages over C as well.  For example, in Pascal handling external dependencies and building a program from source is much less of an effort than it is in C.  In Pascal, a Makefile is more or less optional, but in C, a Makefile is a huge requirement or your project will be very difficult to build.  Pascal has speeds comparable to C, and when compiling, Pascal blows C out of the water.

The main issue I see for the lack of Object Pascal gaining widespread popularity, is that nobody really knows what programming in Object Pascal is like in 2018, and they have all grown used to programming in Java, C, and C# to care about changing over.  Then there's the stigma behind the language itself.  When I tell someone in my workplace that I use Pascal for some personal projects, or that I build a custom tool at work in Pascal, they give me a weird stare, and sometimes say, "I haven't heard of that being used since the 90s".  People see Pascal as something old, like COBOL or FORTRAN, something they assumed has long been dead and is no longer modern or "cool" to use.

I feel that if people give Object Pascal a chance and see how modern it is when compared to some other so-called modern languages, I believe most people will be very pleasently surprised.  I find Lazarus very modern and flexible, while I do not use it for every single project, I do use it when I want to quickly build a desktop application, as I see that as one of Lazarus' largest appeals, it's fully cross-platform, and works really nicely on all platforms it does support.

If I had the spare time in my day, I'd spend time attempting to advertise Object Pascal a bit more in a blog, YouTube videos, etc...  However, I currently lack the extra time, as I have other projects I am working on at the moment.

These are just my 2 cents, and I personally believe Object Pascal has a lot of potential in modern software development, Lazarus just needs to show off some sort of killer framework or tool, much like how Ruby got Rails.

[Almir Lima]

The programmers of the "new generation" love a Virtual Machine.

[Paul Breneman]

The programmers of the "new generation" love a Virtual Machine.

http://wiki.freepascal.org/Small_Virtual_Machines   

[kveroneau]

The programmers of the "new generation" love a Virtual Machine.

http://wiki.freepascal.org/Small_Virtual_Machines   

I'm not sure that's what he meant by Virtual Machine.  I believe he was referring to the JVM, CLI(.NET), among others.  New generation programmers are essentially too lazy to manage memory resources themselves, and allocate that task to a VMs garbage collector.

[Almir Lima]

The programmers of the "new generation" love a Virtual Machine.

http://wiki.freepascal.org/Small_Virtual_Machines   

I'm not sure that's what he meant by Virtual Machine.  I believe he was referring to the JVM, CLI(.NET), among others.  New generation programmers are essentially too lazy to manage memory resources themselves, and allocate that task to a VMs garbage collector.

Exactly...
I've heard that Pascal procedural runs as fast as a program written in C.
I do not know if this is true, but I would not be surprised if it were true.

[Leledumbo]

Exactly...
I've heard that Pascal procedural runs as fast as a program written in C.
I do not know if this is true, but I would not be surprised if it were true.

FPC can surely produce code that's magnitude faster than TCC, maybe DCC too, probably on par with MSVC, but it's still quite behind GCC and Clang. It's language implementation (e.g. compiler) comparison that we're talking about when talking about speed, rarely about the language itself.