Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.os.linux.misc > #80972 > unrolled thread

Re: AI-Based Coding Taking Over

Started byLawrence D’Oliveiro <ldo@nz.invalid>
First post2026-01-12 07:24 +0000
Last post2026-01-18 13:01 +0100
Articles 14 — 9 participants

Back to article view | Back to comp.os.linux.misc

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: AI-Based Coding Taking Over Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-01-12 07:24 +0000
    Re: AI-Based Coding Taking Over rbowman <bowman@montana.com> - 2026-01-12 07:43 +0000
      Re: AI-Based Coding Taking Over Richard Kettlewell <invalid@invalid.invalid> - 2026-01-12 08:39 +0000
        Re: AI-Based Coding Taking Over rbowman <bowman@montana.com> - 2026-01-12 19:39 +0000
          Re: AI-Based Coding Taking Over Nuno Silva <nunojsilva@invalid.invalid> - 2026-01-12 23:40 +0000
            Re: AI-Based Coding Taking Over ram@zedat.fu-berlin.de (Stefan Ram) - 2026-01-13 01:48 +0000
              Re: AI-Based Coding Taking Over c186282 <c186282@nnada.net> - 2026-01-12 23:15 -0500
            Re: AI-Based Coding Taking Over rbowman <bowman@montana.com> - 2026-01-13 07:54 +0000
              Re: AI-Based Coding Taking Over John Ames <commodorejohn@gmail.com> - 2026-01-13 08:07 -0800
                Re: AI-Based Coding Taking Over Chris Ahlstrom <OFeem1987@teleworm.us> - 2026-01-13 11:50 -0500
            Re: AI-Based Coding Taking Over "Carlos E.R." <robin_listas@es.invalid> - 2026-01-15 14:06 +0100
              Re: AI-Based Coding Taking Over Nuno Silva <nunojsilva@invalid.invalid> - 2026-01-17 12:11 +0000
                Re: AI-Based Coding Taking Over c186282 <c186282@nnada.net> - 2026-01-17 19:49 -0500
                  Re: AI-Based Coding Taking Over "Carlos E.R." <robin_listas@es.invalid> - 2026-01-18 13:01 +0100

#80972 — Re: AI-Based Coding Taking Over

FromLawrence D’Oliveiro <ldo@nz.invalid>
Date2026-01-12 07:24 +0000
SubjectRe: AI-Based Coding Taking Over
Message-ID<10k27ip$260ec$1@dont-email.me>
On Wed, 26 Nov 2025 13:23:41 +0200, Anssi Saari wrote:

> I wonder how well those things manage software lifecycle?

I know of no solution that successfully combines graphical-based
programming with effective version control. Even after decades of
experience with both, there is still no graphical equivalent to the
ability to do a diff on plain text files.

[toc] | [next] | [standalone]


#80973

Fromrbowman <bowman@montana.com>
Date2026-01-12 07:43 +0000
Message-ID<msjn12F990eU3@mid.individual.net>
In reply to#80972
On Mon, 12 Jan 2026 07:24:10 -0000 (UTC), Lawrence D’Oliveiro wrote:

> On Wed, 26 Nov 2025 13:23:41 +0200, Anssi Saari wrote:
> 
>> I wonder how well those things manage software lifecycle?
> 
> I know of no solution that successfully combines graphical-based
> programming with effective version control. Even after decades of
> experience with both, there is still no graphical equivalent to the
> ability to do a diff on plain text files.

I'm not sure what you mean.

https://en.wikipedia.org/wiki/Microsoft_Visual_SourceSafe

It sucked but you could easily do a diff.

https://en.wikipedia.org/wiki/TortoiseSVN
https://en.wikipedia.org/wiki/VisualSVN

I used subversion from the command line but either are okay.

[toc] | [prev] | [next] | [standalone]


#80976

FromRichard Kettlewell <invalid@invalid.invalid>
Date2026-01-12 08:39 +0000
Message-ID<wwvpl7f6x96.fsf@LkoBDZeT.terraraq.uk>
In reply to#80973
rbowman <bowman@montana.com> writes:
> Lawrence D’Oliveiro wrote:
>> Anssi Saari wrote:
>>> I wonder how well those things manage software lifecycle?
>> 
>> I know of no solution that successfully combines graphical-based
>> programming with effective version control. Even after decades of
>> experience with both, there is still no graphical equivalent to the
>> ability to do a diff on plain text files.
>
> I'm not sure what you mean.

He’s asking how you show the difference between two versions of a
program written in a
https://en.wikipedia.org/wiki/Visual_programming_language.

> https://en.wikipedia.org/wiki/Microsoft_Visual_SourceSafe
>
> It sucked but you could easily do a diff.
>
> https://en.wikipedia.org/wiki/TortoiseSVN
> https://en.wikipedia.org/wiki/VisualSVN
>
> I used subversion from the command line but either are okay.

Those are diffs of text files. If you applied them to something written
in a visual programming language the best you’d get is a diff of
whatever internal representation the language implementation used.

-- 
https://www.greenend.org.uk/rjk/

[toc] | [prev] | [next] | [standalone]


#81004

Fromrbowman <bowman@montana.com>
Date2026-01-12 19:39 +0000
Message-ID<msl0vnFgebrU4@mid.individual.net>
In reply to#80976
On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:

> Those are diffs of text files. If you applied them to something written
> in a visual programming language the best you’d get is a diff of
> whatever internal representation the language implementation used.

Okay, got it. I was associating 'visual' with Visual C++ etc, not Scratch. 
I didn't know that was used outside of grade school. 

[toc] | [prev] | [next] | [standalone]


#81027

FromNuno Silva <nunojsilva@invalid.invalid>
Date2026-01-12 23:40 +0000
Message-ID<10k40os$2o9fl$1@dont-email.me>
In reply to#81004
On 2026-01-12, rbowman wrote:

> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>
>> Those are diffs of text files. If you applied them to something written
>> in a visual programming language the best you’d get is a diff of
>> whatever internal representation the language implementation used.
>
> Okay, got it. I was associating 'visual' with Visual C++ etc, not Scratch. 
> I didn't know that was used outside of grade school. 

While I do not have much experience with this kind of programming, I've
at least encountered Simulink and Grafcet.

-- 
Nuno Silva

[toc] | [prev] | [next] | [standalone]


#81033

Fromram@zedat.fu-berlin.de (Stefan Ram)
Date2026-01-13 01:48 +0000
Message-ID<visual-20260113024406@ram.dialup.fu-berlin.de>
In reply to#81027
Nuno Silva <nunojsilva@invalid.invalid> wrote or quoted:
>While I do not have much experience with this kind of programming, I've
>at least encountered Simulink and Grafcet.

  The cover of the September 1984 issue of Scientific American
  featured a program in a visual programming language by Lanier
  called "Mandala". (This issue contains articles by authors
  such as Alan Kay, Niklaus Wirth, Terry Winograd, Stephen
  Wolfram, and Douglas B. Lenat.)

  Here's a program I wrote in my visual programming language.
  (I never implemented that language, though.)

                                .--------------------------------------------.
                                |    .---------------.                       |
                                |    |     .----.    |                       |
                                |    |   ->|    |    |                       |
       .-------.                |    |     | <= | -> |                       |
[] --->| qsort |---> []         |    | x ->|    |    |                       |
       '-------'                |    |     '----'    |                       |
                                |    '---------------'                       |
                                |            |                               |
                                |            v                    .---.      |
                                |        .------.   .-------.     |   |      |
.---------------.               |  xs -->| grep | ->| qsort | --->|   |      |
|      .---.    |               |        '------'   '-------'     | j |      |
| x -->|   |    |   .-------.   |        .------.                 | o |      |
|      | : | -> |-->| qsort | ->|  x --->| list | --------------->| i | -->  |
| xs ->|   |    |   '-------'   |        '------'                 | n |      |
|      '___'    |               |        .------.   .-------.     |   |      |
'---------------'               |  xs -->| grep | ->| qsort | --->|   |      |
                                |        '------'   '-------'     |   |      |
                                |            ^                    '---'      |
                                |            |                               |
                                |    .---------------.                       |
                                |    |     .----.    |                       |
                                |    | x ->|    |    |                       |
                                |    |     | <  | -> |                       |
                                |    |   ->|    |    |                       |
                                |    |     '----'    |                       |
                                |    '---------------'                       |
                                '--------------------------------------------'



[toc] | [prev] | [next] | [standalone]


#81037

Fromc186282 <c186282@nnada.net>
Date2026-01-12 23:15 -0500
Message-ID<WmKdnWNEkZaYWvj0nZ2dnZfqnPidnZ2d@giganews.com>
In reply to#81033
On 1/12/26 20:48, Stefan Ram wrote:
> Nuno Silva <nunojsilva@invalid.invalid> wrote or quoted:
>> While I do not have much experience with this kind of programming, I've
>> at least encountered Simulink and Grafcet.
> 
>    The cover of the September 1984 issue of Scientific American
>    featured a program in a visual programming language by Lanier
>    called "Mandala". (This issue contains articles by authors
>    such as Alan Kay, Niklaus Wirth, Terry Winograd, Stephen
>    Wolfram, and Douglas B. Lenat.)
> 
>    Here's a program I wrote in my visual programming language.
>    (I never implemented that language, though.)
> 
>                                  .--------------------------------------------.
>                                  |    .---------------.                       |
>                                  |    |     .----.    |                       |
>                                  |    |   ->|    |    |                       |
>         .-------.                |    |     | <= | -> |                       |
> [] --->| qsort |---> []         |    | x ->|    |    |                       |
>         '-------'                |    |     '----'    |                       |
>                                  |    '---------------'                       |
>                                  |            |                               |
>                                  |            v                    .---.      |
>                                  |        .------.   .-------.     |   |      |
> .---------------.               |  xs -->| grep | ->| qsort | --->|   |      |
> |      .---.    |               |        '------'   '-------'     | j |      |
> | x -->|   |    |   .-------.   |        .------.                 | o |      |
> |      | : | -> |-->| qsort | ->|  x --->| list | --------------->| i | -->  |
> | xs ->|   |    |   '-------'   |        '------'                 | n |      |
> |      '___'    |               |        .------.   .-------.     |   |      |
> '---------------'               |  xs -->| grep | ->| qsort | --->|   |      |
>                                  |        '------'   '-------'     |   |      |
>                                  |            ^                    '---'      |
>                                  |            |                               |
>                                  |    .---------------.                       |
>                                  |    |     .----.    |                       |
>                                  |    | x ->|    |    |                       |
>                                  |    |     | <  | -> |                       |
>                                  |    |   ->|    |    |                       |
>                                  |    |     '----'    |                       |
>                                  |    '---------------'                       |
>                                  '--------------------------------------------'
> 

   Such 'languages' CAN work ... but note that
   it's almost impossible to represent a lot
   of the fine-grained incidentals with them.

   'AI' programming ... it's kind of 'modular',
   similar to yer 'graphical' programming. How
   well can IT cope with the fine-grained
   considerations ? Stupid-user issues ? Security
   issues ???

[toc] | [prev] | [next] | [standalone]


#81054

Fromrbowman <bowman@montana.com>
Date2026-01-13 07:54 +0000
Message-ID<msmc0sFmv52U3@mid.individual.net>
In reply to#81027
On Mon, 12 Jan 2026 23:40:12 +0000, Nuno Silva wrote:

> On 2026-01-12, rbowman wrote:
> 
>> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>>
>>> Those are diffs of text files. If you applied them to something
>>> written in a visual programming language the best you’d get is a diff
>>> of whatever internal representation the language implementation used.
>>
>> Okay, got it. I was associating 'visual' with Visual C++ etc, not
>> Scratch.
>> I didn't know that was used outside of grade school.
> 
> While I do not have much experience with this kind of programming, I've
> at least encountered Simulink and Grafcet.

I repress really well. A few years ago I was tasked with evaluating Talend 
Open Studio, a graphical ETL tool where you dragged stuff around, 
connected it, and it was supposed to generate Java code. It sucked 
mightily. That was the general opinion, not mine alone. They were bought 
by Qlik and the Open Studio product was dropped. 

The dream in the ETL and BI world is pointy haired bosses can do something 
with big data without talking to those snotty programmers. 

[toc] | [prev] | [next] | [standalone]


#81075

FromJohn Ames <commodorejohn@gmail.com>
Date2026-01-13 08:07 -0800
Message-ID<20260113080706.0000472c@gmail.com>
In reply to#81054
On 13 Jan 2026 07:54:04 GMT
rbowman <bowman@montana.com> wrote:

> The dream in the ETL and BI world is pointy haired bosses can do
> something with big data without talking to those snotty programmers.

"Visual programming" has always seemed like a solution to a problem
nobody has, dreamed up by people who don't really understand what the
*actually* challenging parts of software development are.

[toc] | [prev] | [next] | [standalone]


#81077

FromChris Ahlstrom <OFeem1987@teleworm.us>
Date2026-01-13 11:50 -0500
Message-ID<10k5t51$388ul$2@dont-email.me>
In reply to#81075
John Ames wrote this post by blinking in Morse code:

> On 13 Jan 2026 07:54:04 GMT
> rbowman <bowman@montana.com> wrote:
>
>> The dream in the ETL and BI world is pointy haired bosses can do
>> something with big data without talking to those snotty programmers.

Now they'll talk to snotty AI.

> "Visual programming" has always seemed like a solution to a problem
> nobody has, dreamed up by people who don't really understand what the
> *actually* challenging parts of software development are.

At work a few years ago I was forced by funding and then wanting
to get off a C#/.NET project (working for an arrogant cowboy) to
delve into the wonderful world of SysML/UAF. With Cameo Systems
Modeler (MagicDraw) one could use diagrams to create programs and
simulations.

Fairly clumsy and difficult to master. Plus some things couldn't
be done without paying for a plugin [e.g. ALF, the Action Language
for Foundational UML (fUML)]. Some calculations could be done
using Octave (free) or MATLAB ($$$).

One could also created documentation by drawing diagrams with
<<view>> elements (IIRC), <<viewpoints>>, <<blocks>>, and
connecting them. An example here, too fuzzy to read though:

    <https://youtu.be/P7oTrUjLKU8?t=28>

I generated some fairly large documents (HTML, PDF) using it.
Again, clumsy.

There's also the Report Wizard, which uses the Velocity Template
Language (Apache license) to lay out page structures to fill in
with model data. Again, clumsy but workable, for generating
LibreOffice and Microsoft documents.

I was quite happy to retire and get back to C++, LaTeX, etc.

-- 
The grass is always greener on the other side of your sunglasses.

[toc] | [prev] | [next] | [standalone]


#81155

From"Carlos E.R." <robin_listas@es.invalid>
Date2026-01-15 14:06 +0100
Message-ID<va5n3mxurk.ln2@Telcontar.valinor>
In reply to#81027
On 2026-01-13 00:40, Nuno Silva wrote:
> On 2026-01-12, rbowman wrote:
> 
>> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>>
>>> Those are diffs of text files. If you applied them to something written
>>> in a visual programming language the best you’d get is a diff of
>>> whatever internal representation the language implementation used.
>>
>> Okay, got it. I was associating 'visual' with Visual C++ etc, not Scratch.
>> I didn't know that was used outside of grade school.
> 
> While I do not have much experience with this kind of programming, I've
> at least encountered Simulink and Grafcet.
> 

<https://en.wikipedia.org/wiki/LabVIEW>

You can see here how the code looks, it is a box graphic:

<https://en.wikipedia.org/wiki/LabVIEW#/media/File:Labview_code_example.png>


Despite the name, it is not only for use in labs, we did production code 
for factories.

-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]


#81239

FromNuno Silva <nunojsilva@invalid.invalid>
Date2026-01-17 12:11 +0000
Message-ID<10kfu9f$2ekp2$3@dont-email.me>
In reply to#81155
On 2026-01-15, Carlos E.R. wrote:

> On 2026-01-13 00:40, Nuno Silva wrote:
>> On 2026-01-12, rbowman wrote:
>>
>>> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>>>
>>>> Those are diffs of text files. If you applied them to something written
>>>> in a visual programming language the best you’d get is a diff of
>>>> whatever internal representation the language implementation used.
>>>
>>> Okay, got it. I was associating 'visual' with Visual C++ etc, not Scratch.
>>> I didn't know that was used outside of grade school.
>>
>> While I do not have much experience with this kind of programming, I've
>> at least encountered Simulink and Grafcet.
>>
>
> <https://en.wikipedia.org/wiki/LabVIEW>
>
> You can see here how the code looks, it is a box graphic:
>
> <https://en.wikipedia.org/wiki/LabVIEW#/media/File:Labview_code_example.png>
>
>
> Despite the name, it is not only for use in labs, we did production
> code for factories.

Ah yes, definitely heard of it, although I think I've never used it
myself.

-- 
Nuno Silva

[toc] | [prev] | [next] | [standalone]


#81256

Fromc186282 <c186282@nnada.net>
Date2026-01-17 19:49 -0500
Message-ID<6hednXZDWd67s_H0nZ2dnZfqnPqdnZ2d@giganews.com>
In reply to#81239
On 1/17/26 07:11, Nuno Silva wrote:
> On 2026-01-15, Carlos E.R. wrote:
> 
>> On 2026-01-13 00:40, Nuno Silva wrote:
>>> On 2026-01-12, rbowman wrote:
>>>
>>>> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>>>>
>>>>> Those are diffs of text files. If you applied them to something written
>>>>> in a visual programming language the best you’d get is a diff of
>>>>> whatever internal representation the language implementation used.
>>>>
>>>> Okay, got it. I was associating 'visual' with Visual C++ etc, not Scratch.
>>>> I didn't know that was used outside of grade school.
>>>
>>> While I do not have much experience with this kind of programming, I've
>>> at least encountered Simulink and Grafcet.
>>>
>>
>> <https://en.wikipedia.org/wiki/LabVIEW>
>>
>> You can see here how the code looks, it is a box graphic:
>>
>> <https://en.wikipedia.org/wiki/LabVIEW#/media/File:Labview_code_example.png>
>>
>>
>> Despite the name, it is not only for use in labs, we did production
>> code for factories.
> 
> Ah yes, definitely heard of it, although I think I've never used it
> myself.

   If it can run a gas chromatograph machine it
   can run a giant production reactor for making
   polyethylene. LabView had a pretty good rep
   and I have heard of it being used for "factory
   stuff" as well.

   Old old days, banks of clickey relays and analog
   I/O. Then transistors. Now, software & digital.
   Tomorrow, digital 'AI' and humans will be mostly
   out of the loop.

[toc] | [prev] | [next] | [standalone]


#81268

From"Carlos E.R." <robin_listas@es.invalid>
Date2026-01-18 13:01 +0100
Message-ID<5muu3mxkli.ln2@Telcontar.valinor>
In reply to#81256
On 2026-01-18 01:49, c186282 wrote:
> On 1/17/26 07:11, Nuno Silva wrote:
>> On 2026-01-15, Carlos E.R. wrote:
>>
>>> On 2026-01-13 00:40, Nuno Silva wrote:
>>>> On 2026-01-12, rbowman wrote:
>>>>
>>>>> On Mon, 12 Jan 2026 08:39:01 +0000, Richard Kettlewell wrote:
>>>>>
>>>>>> Those are diffs of text files. If you applied them to something 
>>>>>> written
>>>>>> in a visual programming language the best you’d get is a diff of
>>>>>> whatever internal representation the language implementation used.
>>>>>
>>>>> Okay, got it. I was associating 'visual' with Visual C++ etc, not 
>>>>> Scratch.
>>>>> I didn't know that was used outside of grade school.
>>>>
>>>> While I do not have much experience with this kind of programming, I've
>>>> at least encountered Simulink and Grafcet.
>>>>
>>>
>>> <https://en.wikipedia.org/wiki/LabVIEW>
>>>
>>> You can see here how the code looks, it is a box graphic:
>>>
>>> <https://en.wikipedia.org/wiki/LabVIEW#/media/ 
>>> File:Labview_code_example.png>
>>>
>>>
>>> Despite the name, it is not only for use in labs, we did production
>>> code for factories.
>>
>> Ah yes, definitely heard of it, although I think I've never used it
>> myself.
> 
>    If it can run a gas chromatograph machine it
>    can run a giant production reactor for making
>    polyethylene. LabView had a pretty good rep
>    and I have heard of it being used for "factory
>    stuff" as well.
> 
>    Old old days, banks of clickey relays and analog
>    I/O. Then transistors. Now, software & digital.
>    Tomorrow, digital 'AI' and humans will be mostly
>    out of the loop.

The hurdle was that we were using Windows 95, and it could crash.


-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [standalone]


Back to top | Article view | comp.os.linux.misc


csiph-web