Why is LiveCode not recognized more as an app developer language?

[richmond62]

If Licecode isn't more widely recognized as an app developer language it's simply because it's not a programming language!

-

-
Indeed, 'Licecode' is not a programming language, but 'LiveCode' is a programming language.

All that this poster is doing is moving goalposts.

Why is LiveCode not recognized more? Because it's not really free for professional app development.

This is an extremely odd statement as there is probably no way of demonstrating that the "not recognized more" follows from the statement about it being a commercial product.

As a Free, Open Source version of LiveCode was available for a long time, continues to be available, and there is continuing development in that direction, and that seems to have done nothing whatsoever to increase its recognition I cannot see how that statement has any value.

Why do most Europeans drive cars with manual gearboxes when any car mechanic will tell you that cars with automatic gearboxes are better?

Why was the VHS videotape almost universally adopted when the Betamax format was demonstrably superior?

Why does 'the world' use Microsoft Windows when there are alternatives which are probably better?

Basically it probably comes down to nothing more than being in the right place at the right time, and how hard the thing was plugged at that time: and in all probability that time is past.

I am waiting for someone to go "professional app development" again . . . but "professional app development" does NOT have to be closed-source, and if you want to define "professional app development" as working only on closed-source apps I'd say go and say that to the developers of GIMP, Inkscape, LibreOffice, and so on.

[stam]

Stam.

I miss then moniker "TransScript".

Craig

You’re not wrong but alas script only stacks come with that moniker…

Arguably what LC should have done is create a free interpreter/engine, but charge for the IDE (still leaving you free to use the interpreter any other way, so for example could run code in VSCode).

While the two are tightly bound, they are not inseparable and providing a free interpreter would put LC on a similar footing with Python.

And the language could be free but to actually get a useful experience and build apps, you would still need a full licence. Of course other IDEs could be allowed but in theory at least the genuine LC one would always be the superior offering…
And by having a freely downloadable (or even online) interpreter, the size of the standalone would reduce greatly.

But that’s just musings of a non-professional programmer… I’m sure this would require a large amount of non-trivial work, but imagine if LC did something like this… could well set the cat among the pidgeons…

[dunbarx]

I am positive that Scotland reads all of this sort of thing, all the time. I know that they have never responded in any way to any of it.

This is not surprising. But I wonder if any of it triggers a conversation in the LC lunchroom. I would love to be a fly on the wall...

Craig

[dunbarx]

Stam.

pontificating the current “perfect” state

As simply as you can, what are the most important couple of issues with the IDE? I am only a hobbyist, and at the top of my wish list is that the "Align Controls" pane in the PE would stick around.

Craig

[stam]

As simply as you can, what are the most important couple of issues with the IDE? I am only a hobbyist, and at the top of my wish list is that the "Align Controls" pane in the PE would stick around.

Well there are no show-stoppers, but lots of little niggles... death by a thousand paper cuts perhaps. Not sure there is a 'simple' way to put these, so I'll limit these to a few.

You can address your wish to persistently align controls with the responsiveLayout library, but if you figure it out please let the rest of us know how lol...

Which brings me to 1 of the many niggles: poor documentation of the new widgets. ResponsiveLayout has a large number of settings that are presented as if we should know what they do - but I have failed miserably to interpret these...

Other niggles (from what springs to mind - many more exist):
Code-Folding - Bernd has produced a most excellent solution for this, but it requires developers to a) know that this exists and b) install it. In my mind, this would be part of the IDE already!

Error management - occasionally I encounter run-time errors that just point to the remote debugger or some such, even though I'm not using remote debugger. I think this is because some errors evade triggering an exception in the IDE and stuff just fails without a clear reason. Prime example: The Geometry Manager. The issue with the GM is that if anything at all (name, grouping etc) changes, it fails silently part way through resulting in objects positioned randomly off-screen and so on. It's a real bugger...

Basic objects need modernising: For example, the text field should have options in the PI to provide placeholder text, make it a password field etc. The tab panel should be an actual container than contains controls and allows you to tab through them in the IDE. And so on. The built-in guides work but not as well as the FerrusLogic devGuides, which I still use even though FerrusLogic is no more (available on GitHub). All of this is standard in any modern IDE in other languages.

Overall, the IDE lacks the polish you see in other IDEs like Xcode - if you're coming from Xcode the LC IDE feels much less professional and lifted from the 90s... (personally I hate the multiple palette's - but that's my pet peeve). Even if nothing substantial changes, improving looks will aid interest from developers in other languages. As it is, I'm sure they just take 1 look at it and say 'nope'....

All of these niggles may well be addressed in the yet-to-appear version 10, and I live in hope that we'll get all the above in version 10.

But I wish they'd go some steps further by separating the engine from the IDE, providing the engine for free/opensource but licensing the IDE as a premium product... this would help put LC on equal footing with the likes of Python inasmuch as anyone could run code with a console output for example, and possibly build apps with homespun IDEs, but to get premium developer experience with GitHub integration and the niceties mentioned above you'd licence the official IDE.

But who am I to say what is best and what is feasible...

[Alfanso]

LiveCode isn't as widely recognized as other app development languages because it's less popular and lacks the extensive community and resources that languages like Swift, Java, and JavaScript have. Its niche market, higher learning curve for beginners, and limited promotion contribute to its lower visibility in the mainstream app development world.
Regards: Alfanso

[stam]

LiveCode isn't as widely recognized as other app development languages because it's less popular and lacks the extensive community and resources that languages like Swift, Java, and JavaScript have. Its niche market, higher learning curve for beginners, and limited promotion contribute to its lower visibility in the mainstream app development world.
Regards: Alfanso

Hi Alfanso - how do you use LiveCode?

I extremely strongly disagree with your point that LC has a higher learning curve for beginners, and you making this statement makes me think you're probably not a LiveCode user (also noticing this is your first post and your text reads like AI-derived drivel) - but please do prove me wrong.

It's also not a "niche market" - it's a general purpose language. But LC has been riding HyperCard/MetaCard's coattails for too long and that makes it look niche.

Your post basically says: LiveCode isn't recognised because LiveCode isn't recognised. Not sure that contributes anything.

But even so: Assuming the aim is to increase uptake, the first step is to get this recognised as language on GitHub (which it currently isn't) and the IDE needs a high grade polish if it is to compete with free alternatives (which it does not have). There are plenty of resources online but these again need updating and polish. LC cannot rely on 20 year old largely enthusiast-driven resources.

These things are doable in the short term. LiveCode is a great general purpose language and deserves recognition, but this won't happen with blue-sky thinking, it requires a lot of work from LC Ltd and this is what I expect when I pay premium prices for a licence.

[FourthWorld]

...what are the most important couple of issues with the IDE?

What's there is worth looking at, but also what's not there yet widely available in other tools.

E g. this recent thread: currency and other numeric formatting should be a field property, not a script.
viewtopic.php?f=9&t=39175

[dunbarx]

@Richard.

E g. this recent thread: currency and other numeric formatting should be a field property, not a script.

I know what you mean, but there must be a million "small" examples that any commonly published wish list on any particular subject would contain. For LC IDE, those seem of a much lower level of importance than the tone of that discussion implied. In fact, mine, the align objects thing, was the biggest, and it is only a convenience, not a hinderance.

And anyway, would being able to set the decimal point in a field property also manage the triple (or double, as per Richmond's Indian friends) "order of magnitude" separators?

Again, my point is that these do not nearly seem like deficiencies, rather modest improvements. What aspects of human endeavor does not need those?

@Richmond.
It seems that Indians do not simply separate into doubles, The rightmost group is a triple, and only then are doubles isolated as one moves left.

Craig

[richmond62]

@Richmond.
It seems that Indians do not simply separate into doubles, The rightmost group is a triple, and only then are doubles isolated as one moves left.

Quite possibly.

Never could work out Crores and Lakhs, nor a Yojana come to think of things.

Erm; "3.3 to 15 kilometers": https://en.wikipedia.org/wiki/Yojana

This reminds me of what my Mother found out when she was in Iran in 1952: the basic distance measurement was the Farsach ( = how far a donkey walks in a day).

"the local 2,2,3 style of digit group separators": https://en.wikipedia.org/wiki/Crore

My response is 100% pictorial:
-

digits.jpg (8.18 KiB) Viewed 6614 times

[FourthWorld]

@Richard.

E g. this recent thread: currency and other numeric formatting should be a field property, not a script.

I know what you mean, but there must be a million "small" examples that any commonly published wish list on any particular subject would contain. For LC IDE, those seem of a much lower level of importance than the tone of that discussion implied.

I can't speak for another forum member, but I can report my decades of experience consulting with a wide range of orgs from solopreneurs to Fortune 1000 companies.

In my line of work, I can't limit my awareness of tooling options to the offerings from any single vendor.

Since the dawn of graphical computing, spreadsheets offered formatting as a field property. The expectation of similar ease among GUI tools has become well established.

The bar was lower back when HC premiered in 1987. Software tooling has exploded over these 34 years. The universe of options for crafting calculations in software today is nearly uncountable.

In this rapidly-evolving ecosystem, standing still means falling behind.

It's not easy being making dev tools in the 21st century. It's brutal, really. But if success for a small business is possible at all, the opportunity must be embraced soberly, with open eyes, no matter how daunting what we see may be. Sometimes tough choices focusing fiercely on unglamorous but key objectives keeps a product in the game long enough to later explore more exotic options outside the primary revenue stream

And anyway, would being able to set the decimal point in a field property also manage the triple (or double, as per Richmond's Indian friends) "order of magnitude" separators?

That's another good argument for handling this in the engine, where it can take advantage of localization tools provided by the host OS.

Again, my point is that these do not nearly seem like deficiencies, rather modest improvements. What aspects of human endeavor does not need those?

Precisely. When a need is nearly universal, and nearly universally supported in alternative solutions, it's an "enhancement" only to those to have already committed to the tool, and a "deficiency" among those evaluating all relevant options.

[richmond62]

AI-derived drivel

I thought I'd see what drivel an A.I. thing would come up with:

https://www.perplexity.ai/

"Why is LiveCode not recognized more as an app developer language?"

"LiveCode is not as widely recognised as other app development languages for a few key reasons:
Lack of popularity and community: LiveCode has a smaller user base and community compared to more established languages like Java, Swift, and JavaScript. This means fewer resources, tutorials, and third-party tools are available for LiveCode.
Steeper learning curve for beginners: While LiveCode is touted as an easy-to-learn language, the initial learning curve can still be challenging for new users, especially compared to no-code/low-code tools that require less programming knowledge.
Competition from more sophisticated tools: Other app development platforms and frameworks offer a more polished and feature-rich experience that allows users to create apps quickly, which can be more appealing than the LiveCode approach.
Branding and positioning challenges: The name "LiveCode" may not clearly communicate that it is a full-fledged app development language, and the company has struggled to position it as a viable alternative to more popular options.
Lack of widespread enterprise/industry adoption: LiveCode is not as widely used in enterprise and industry settings compared to languages like Java, C#, and Swift, which reduces its overall recognition and adoption.
In summary, LiveCode's relative obscurity, steeper learning curve, competition from more user-friendly tools, and branding challenges have all contributed to it not being as widely recognised as other app development languages in the industry."

Now we all know, despite all the crapulous blehh about robots thinking for themselves this is just a spot of pattern-matching and an internet trawl.

I can see 1 thing I strongly DISAGREE with, and 1 thing I strongly AGREE with: in that order:

"steeper learning curve"

This does not begin to make sense having just had my 'annual peep' at Java and Python.

"Branding and positioning challenges: The name "LiveCode" may not clearly communicate that it is a full-fledged app development language"

As I have seen almost no 'communication' outside a charmed circle for many years . . .

[dunbarx]

Fair enough, but for someone learning (and evaluating) LC, would any of these points mightily discourage that person? It seems to me that only when one becomes fairly proficient would these even show up at all.

And by that time, if the short learning curve and astonishing power available at that early stage had not already won them them over, would these much smaller issues discourage them? They would not for me, not nearly. Perhaps for those coming from the languages these would weigh more heavily? They would already have their own "versions" of whatever thing they encountered in LC to compare.

The task is to allow new users to "get" the advantages early. That ought to be more than enough. But even for experienced new users, those advantages ought to be enough as well, at least to evaluate fairly.

I remember the back and fourth thirty years ago with HC. Programmers dismissed it as a toy on its looks alone. I got that, and called it a prejudice. That word alone got me in big trouble with Amthony Black.

Craig

[richmond62]

The whole thing makes me think of beer.

Throughout my childhood people were telling me that beer was bad (my Dad actually told me this once when he had beer froth on his top lip. LOL).

When I was about 14 I tried some beer (mainly because I am obtuse and when everybody tells me something is crap . . .) and I have never looked back.

So the first 'hurdle' is getting people to sit down in front of a computer and show them the basics of LiveCode.

After that there really should not be any obvious problems.

[richmond62]

Ambiguous name:

https://marketplace.visualstudio.com/it ... r.livecode