[Gambas-user] Curious and interesting article.

[Brian G]

----- On Dec 5, 2021, at 3:41 PM, Benoît Minisini g4mba5 at gmail.com wrote:

> Le 05/12/2021 à 23:29, Antonio F.S. a écrit :
>> El 5/12/21 a las 22:04, Benoît Minisini escribió:
>>> Le 05/12/2021 à 11:31, Antonio F.S. a écrit :
>>>> Good morning.
>>>>
>>>> I'm the guy from the "Gambas Foundation" that created so much
>>>> controversy on this mailing list. :-) :-) :-) By the way, I would
>>>> like to say to the person who insinuated that my proposal was that of
>>>> a novice, that probably when I was programming in big companies with
>>>> Fortran and Cobol, maybe he would be very young but he is not of my
>>>> generation (I was born in 1962).
>>>>
>>>> I share with you an article that I found interesting and that, even
>>>> though it is not directly related to GAMBAS, it does belong to our
>>>> computing area.
>>>>
>>>> Best regards.
>>>> Antonio F.S.
>>>>
>>>>
>>>> URL:
>>>>
>>>> https://fadingeek.medium.com/the-era-of-coding-is-ending-why-this-is-very-important-bdaa926bdc4
>>>>
>>>
>>> I read the article : It looks like pure salesman bullshit. I wonder if
>>> the article has been written by someone who actually used the "magic"
>>> tools he's talking about and if that guy actually wrote and maintain
>>> real softwares...
>>>
>>> --
>>> Benoît Minisini
>> ---------------------------------------------------------
>> Hello.
>> 
>> Well, I remember in 1987 when at Burberrys Spain I was programming on an
>> HP 3000 on the MPE operating system with Fortran and hierarchical IMAGE
>> databases, creating code and more code in a brand new line editor (linux
>> nano is very modern compared to the one I was using then). The hammering
>> of impact printers and the mad spinning of tapes in those big machines
>> that looked like eyes and nothing else.  Our biggest hard disk was a 40
>> megabyte hard disk that weighed about three kilos and was two feet wide
>> and one foot deep. The connections on the terminals were made by Rs232
>> through huge multiplexers extending the communication signals with
>> cables, which resembled water pipes... At that time, thinking about the
>> Internet was like being told magic stories, at least in these IDEs and
>> programming languages...
>> 
>> Age has taught me to be cautious and humble...
>> 
>> Best regards.
>> Antonio F.S.
>> 
> 
> My comment may seem harsh, so I will try to explain myself better, even
> if English is a difficulty for me.
> 
> I understand what you answered. But do you see the difference with that
> article? Your are talking about your real and actual experience, whereas
> I found the article full of smoke not related to any real experience.
> 
> In other words: hardware has changed, but everybody still have to code.
> Maybe not on punch cards anymore, but did coding -really- and
> -fundamentally- changed?
> 
> Note that I have a similar experience of the hardware evolution than
> you, having learned computing programming on an Amstrad with just 48 Kb
> of free memory and a tape reader.
> 
> I play flute (orchestral flute) : the instrument was mainly a piece of
> wood with holes for a long time, until the mid of 19th century where it
> becomes a metal instrument with a complex key mechanism. But learning
> flute has not really changed, which is confirmed by the flute method
> written by Quantz in 1752. Most of what you read in it still apply. It
> was a bit of shock the first time I read it!
> 
> When you search and think a bit, blowing in a hollow tube with holes to
> make music is a human activity since the prehistoric time (we found
> flutes made by prehistoric people with bones about 30 000 years ago). We
> didn't change a lot since that time, and so playing a flute may still be
> the same art for thousand of years. It's just that we don't really use
> bones anymore. The instrument became more sophisticated with time, but
> the art didn't change fundamentally.
> 
> I think you have the same thing in "coding" activity. It's something
> that does not depend on computers, except that without a computer or
> something similar you can't actually code. It's a mental activity where
> you think about numeric objects produced by the hardware and the user,
> design how to represent them, which algorithm will structure them,
> decide how to present them to the human, and so on.
> 
> I think it's the same thing with architecture too. An architect can
> fully design a building in his head. Architecture techniques
> considerably changed with time, but I almost sure that the core of the
> architect art is the same for building a viaduct or the Aegyptians
> pyramids - except that if you ask a viaduct architect to build something
> like the Great Pyramid, he will tell you he will not know how to do. But
> you see what I mean.
> 
> I hope my explanations will make you excuse my initial harsh and proud
> comment. This kind of article about computing activity full of fuzziness
> tend to annoy me a lot!
> 
> Regards,
> 
> --
> Benoît Minisini
> 
> ----[ http://gambaswiki.org/wiki/doc/netiquette ]----

 Benoît that is a wonderful analogy, I agree with your thoughts,
 I would like to add a few of my own here, though perhaps not as eloquent as yours.

 I see the codeless development as a real thing, but I see the separation of those who create the
 tools and those that use the tools that were created by the engineers.

 With the development of the flute, At each stage there was probably an artist who saw a way to improve
 the way they could perform/extend their art in a new way , They went to the person who created the instrument and told them
 what they would like to see, the (engineer) then changes the artists tool. The artist makes the most beautiful music.
 
 While the artist may not be able to make the tool, they could produce renowned results using it.

 Of course the engineer(artist in their own right) goes on to become wealthy with the sale of the new more expressive flute... lol

 We could view the trend towards codeless development in the same way, the artist/developer is free to use the tool to create some amazing
and useful software. When he/she sees a need that is not provided by the tool they reach back to the engineer to change the tool to more correctly serve their need. It is a close analogy to the flute, over time the codeless tools become amazing ways to produce some incredible products. But in the end there is still that engineer creating and managing(coding)the tool!

While the artists(human interface developers) seems to take the credit for the amazing products created with these tools. There is still those few who understand the roots, and ways of the machine. Those are the engineers they will continue to in there own way create ways to use the machine. A.I is important to the tool the artists use a in a way to simplify the interface. But A.I can only go so far. It builds and applies on what it has been shown. But there is always new technology that does not fit into the learned process..

I think You  Benoît are one of those engineers. You have given us an environment the hides many of the complexities of programming and allows us to apply our more creative and explorational ideas and thoughts in the development cycle without much thought to the real machines complexity.

Will codeless creation continue to grow and flourish Yes of course because it allows so many more artists to participate and allows the more expressive affirmation of the artists endeavours. While these interfaces seem to be codeless, are they actually? Ummm to those of us that code, yes but to the artist yes!! of course it is they need to learn how to use the tool, how to encapsulate their ideas into the terms that the tool understands, in effect codifying their thoughts into concrete terms the machine understands. 

While Gambas does not remove what would be defined as coding to a programmer, it does allow use to spend time being more creative less time understanding the environment and less time "Coding the difficult stuff". Thirty years ago Gambas may have been the magic tool...


Personal example for AI fail which was going to be an amazing tool:

    I was working on a project at Xerox Parc a few years ago building an AI that could reduce the number of staff technicians required by a Company called Lucent Technologies, Inc. The process went fairly well technicians would enter into a laptop the resolution to any issues solved each day for a couple of years, building the base information used by the A.I. Of course all the Tech Manuals were also learned by the system.

The Techs were let go retaining only a minimum staff that relied on the experience of the original techs and AI, things went well for a while, until a new and very much different type of equipment was introduced.... While the AI learned all the info from the new service manuals..

Of course it failed as there were to many ways a complex system could fail, the few techs were overwhelmed and I am Glad to say all the old Techs were rehired ... needless to say they never entered any solutions into the AI(The human Factor).. lol

Lucent is no more...  Point here is AI knows what it knows, can extend it some, but still lacks the creativity in many cases.

This applies to coding as well. AI is much better now , but still lacking in many ways!


My two cents, for what it is worth... lol

Brian G.