Could there be a parallel effort on the Community Edition...

[micmac]

... and how would we know?

(it was only by chance that I noticed this forum)

Mic

[SethMorrow]

I suppose we could search through github for forks that are being worked on. If you find any, please do post them here and/or send them our way.

[richmond62]

(it was only by chance that I noticed this forum)

That's bloody bad: probably part of the problem is that people seem worried
about offending other people . . . these forums will wither and die without a concerted
effort to plug them all and everywhere.

[SethMorrow]

(it was only by chance that I noticed this forum)

That's bloody bad: probably part of the problem is that people seem worried
about offending other people . . . these forums will wither and die without a concerted
effort to plug them all and everywhere.

I think until there is something to plug, it's not really an issue. The previous maintainers definitely don't want us advertising our existence in any of their venues.

[richmond62]

I don't think "the previous maintainers", as you so coyly refer to LiveCode, care, as they believe that they are riding high
and any effort at continuing with an Open Source version is doomed.

[SethMorrow]

LC Mark was pretty clear about this.

I must also ask you not to use the LiveCode mailing list, bug reporting system or LiveCode forums for discussions surrounding your fork - particularly related to plans, ideas, developments and changes which are being or have been made.

[richmond62]

Ha, Ha: I already did in a covered fashion, and so far, no-one has made any noise about that.

HOWEVER, it can still be splattered all across the internet; in places like this:

https://livecode.fandom.com/wiki/Livecode_Wiki

[OpenXTalkPaul]

Well I think we should, eventually when there's something more tangible here, go out into the wilds, anywhere HC and other XTalks are still being talked about (I still read new lengthy articles every year lamenting no modern equivalent, you could often find me in the comments section plugging LC) and make some noise. And anywhere LC is still listed as OpenSource should be updated to point to OXT if possible.

https://github.com/serhii-londar/open-s ... ac-os-apps

[axwald]

Hi & hello, Greetings!

Idea how to promote this forum w/o being too intrusive:

Have a forum post (section, thread, whatever ...) that collects, displays & keeps up to date all the download links for the old Community Editions.
Make it prominent, pinned & usable w/o registration.

Such a link we could use nearly anywhere:

... you know, I always used the community edition 'cause ...

... If you're searching for the openSource version, look here ...

--
All code published by me here was created with Community Editions of LC (thus is GPL v3).
If you use it in closed source projects, for the Apple App Store or with XCode
you will violate some license terms - read your EULAs & license informations!

Quite innocent, right? But should give google rankings, and might direct ppl to have a look here.

Quite sure I could smuggle it even into the old forum, in my signature there ;-)
And if we had such, it could be used on livecode.fandom.com, too. Maybe even in the wikipedia.

Have fun!

[OpenXTalkPaul]

Good suggestions, I personally will not be promoting this forum on anything run by LC Ltd. but I can't control what other people do so...

But I think there's ample opportunities out in the wild, that never seem to have been taken in the past, to promote an xTalk language. I think that is something anyone could contribute to, a sort of "Gorilla Grow" marketing campaign. A "street team" ya know?
https://hackaday.com/2020/06/12/program ... n-english/

It's interesting, reading over these GPL3 license(s), it would seem to me, and I could be wrong about this as I'm no lawyer, that one would have to provide a copy to people who ask for it, any software/source where they've used other GPL things that are required for the software to function, for like 3 years (can charge reasonable fees for storage medium and such). So now I'm wondering was it REALLY such a big favor to "The Community" to put those installers back online?

[axwald]

Hi,

[...] I personally will not be promoting this forum on anything run by LC Ltd. [...]

I agree. It doesn't make sense to be confrontational, for nobody.
But there's a wide grey area between "keep silence" and "promoting". This is where a link to a "collection of download links for the community version" could come handy, IMHO. Might be innocent enough to avoid the "promoting". (as in my fictional signature example)

But I think there's ample opportunities out in the wild, that never seem to have been taken in the past, to promote an xTalk language.

One major handicap in promoting XTalk is that there seems to be no short, comprehensive explanation of it. See the discussion in hackaday, or in the Register - most ppl clearly don't have any idea of it & so assume imagined shortcomings that they criticize than.
I might have missed something, but this is another thing that would be quite helpful, IMHO. A possibility to link to a "short comprehensive summary of XTalk", hosted here. (I started such a while ago - could be finished here)

It's interesting, reading over these GPL3 license(s), it would seem to me [...] that one would have to provide a copy to people who ask for it, any software/source where they've used other GPL things [...]

AFAIK (after reading the "Open Source Licenses.txt" in my program folders) they avoided to use any GPL v3 code. All licenses I found are more permissive, means: don't have the "leech protection" of the GPL v3.

What I'd like to know is: Is there the source code for LC 6.7.10 Community anywhere (Github?)? Since this is, in my opinion, the last functional & professionally usable version of LC *1). I'd strongly support a fork of this one (with 64bit conversion), and I might throw in considerable effort. Anybody knows about?


Have fun!

*1) Forking 9.*.* is, in my opinion, not a good idea: during the changes in v7 & 8 they messed up their code & development cycle in a way that left, judged by the results, only a ramshackle hut with lean-to sheds at every edge - that they now try to keep from collapsing.
This is most visible in the Windows versions, where not only the script editor became intolerable, but even button hilites ceased to function properly, and the (even compiled) execution speed is magnitudes below what Linux & Mac offer in the same version. Getting such failures repaired in a fork looks quite unpleasant to me.

[richmond62]

Please can it be a later version as we need the Unicode functionality.

[OpenXTalkPaul]

Please can it be a later version as we need the Unicode functionality.

This is a big part of why I'd want to keep the engine based on 9.x or even 8.x.

I'm all for reorganizing the repos and dev cycles or whatever in order to make it easier for compiling from source and fixing problems though.

I personally would want 9 or 8.x+ for the Builder language part, for easier expandability.

I've primarily used it only on macOS. I have only run on Windows to test cross platform LCB extensions of libraries that I've wrapped (FluidSynth and HIDAPI), and only fairly recent hardware (7th gen Core i7 laptop), so I'm not familiar with GUI bugs or sluggishness there.

There is source available for LC 6.x here Download #4 (labeled PowerPC but I assume it's the full source):
https://macintoshgarden.org/apps/livecode-powerpc

[axwald]

Hi,

Please can it be a later version as we need the Unicode functionality.

This is a big part of why I'd want to keep the engine based on 9.x or even 8.x.
[...]
I personally would want 9 or 8.x+ for the Builder language part, for easier expandability.

well, one must set priorities.
I'd contribute to a functional software, most of all. And the versions you want to use are, in my opinion, far from it.

It seems they work on a Mac. But:

• On Windows & Linux the IDEs are intolerably buggy, slow & unreliable.
• And the Win StandAlones have massive engine problems, additionally.
• Add that the so-called "server versions" will never run on a "real server" due to exceeding dependency requirements, and, even when used as CGI, will load a complete font subsystem at every call (!) & thus are slow & CPU hogging.
• Don't forget that the Android implementation is quite buggy & years behind what's possible in iOS.
• And that even the latest Community edition for Linux doesn't support a single Linux distribution that still is in its support cycle.

For your entertainment I'll link you a few things you'll have to cope with:
Windows:
serious engine flaw, "lauch file with app" broken, button hilite broken, font weights broken, text rendering broken
Android:
URLlib broken in 9.6.3, button hilite broken,
font weights broken, orientation switching flawed, system date bugged
Server:
CGI use messed up, DB connection broken, dependency overload
(The server problems only hit on "real servers" in productive environments. On a hobby linux desktop you'll not notice 'em - that's what's these "server versions" obviously are made for/ tested on ...)

These are just a few of the "really game-breaking" issues. There's more of it. And then there's a plethora of minor bugs.


Don't fall prey to the same reality distortion field that LC Ltd. obviously cannot escape, and that may be their doom sooner or later: Among the target audience of a cross platform IDE "ppl coding on Mac" are only a minority!
A small fraction of the Win & Linux coders, surpassed quite probably even by the ChromeOS guys.

Why do you think LC Ltd. is on the verge of bankruptcy, since years? Delivering software optimized for Mac/iOS only, and buggy as hell for anything else (for insane prices ...)?
Why do you think so many serious coders have left the other forum during the recent years?

If you want to create a successful fork you'll need to attract contributors from outside the Apple bubble. But show any serious developer (not on a Mac) a LC 9.*, and they'll run, screaming.

So prepare for some years of serious bug squishing & optimization, BEFORE you can start to attract any contributors. Nobody will join a new project if all there's to do for quite a while is "cleaning up the mess".

Or just make a Mac-only XTalk fork.


Reminder: This is just my opinion, I know your view will differ.
I'd really love to see a thriving OpenSource XTalk. But I'm very sure that the task you're considering is ways beyond what's practical. And I kindly ask to stop the "roaring & hitting the monitor" for a moment & think about my reasons for this. Thank you!

Another idea to consider: As long as a fork follows too closely the current development of LC Ltd.'s product they cannot but regard you as "a danger to our existence". A fork of a late LC 6 however, lacking many of the "unique points of selling" LC Ltd. is so proud of, might not be considered a threat. There might even be possible a "friendly coexistence".

And have fun!

[richmond62]

Oddly enough I have never had a problem with any LC on Linux from RR 2.2.1 to LC 9.6.3

[OpenXTalkPaul]

>> any serious developer
The "serious developers" I know already scoff at the idea of a human friendly language and user centric IDE.

Since this Plain C changed to C++ code base dates back to MetaCard on BSD Unix in the early 1990s, and has been ported to macOS classic, macOSX, iOS,Win16,Win32,Win64, Linux, Android (which is Linux underneath), etc.etc., and has run on Motorola, PowerPC, Intel, ARM, etc, etc. for close to three decades now, there's always going to be compatibility issues, particularly if everyone working on it is mostly only using just one of the platforms from it's cross-platform support list. So it sure would be good to have some platform diverse involvement.

I can only speak for myself, but I'm personally not all that interested in things like "real servers", I AM a hobbyist coder. And I'm a fan of xTalks since the original scripting language, HyperTalk (in the late 1980s) and I want an xTalk to remain available to as many people as possible.

I've been making music/musician toys/tools for a target audience of mostly just me, that's part of why I'm here. If your ultimate goal would be to have commercial production quality anything for free then you should probably not waste your time on this and either stick with LC Ltd. or go with some other commercial entity or tools backed by the big players or that has an already large community... something like Flutter/DART (Google) or Visual Studio/C#/.NET/Mono (Microsoft) or even Python or something like that. I don't think any xTalk will ever be able to compete with those at this point, That's probably a cause that was lost way back when JavaScript became the dominant "scripting" language for HyperText interactivity in the mid-90s.

I've personally been clearing space on my laptop for installing some Debian distro (probably with Enlightenment and GNUStep). I'm not the Apple fanboy I once was and I may just drop macOS all together at some point (after I retire since my day job is in graphic arts and so I need to run Adobe apps as well as additional proprietary software).

I did read about this one on the list:
https://forums.livecode.com/viewtopic.php?f=18&t=36243
Seems to have something to do with Windows versions not being good at expanding it's buffers/memory allocation issue.
I did see there was script work arounds for that. I'm interested in this sort of thing because of MIDI message buffers.

Not sure about this, but GUI issues might be easier to fix than engine problems, certainly on mobile it's probably best to use widgets, either native or substitute custom built. Android engine is using Linux / Motif by default, which looks old and ugly to me, by default. I've made some pretty good looking / responsive Android apps with custom data grids and and such, have not hit this issue mentioned though:
https://quality.livecode.com/show_bug.cgi?id=22965

Launch command might be easily fixable by adding a check that the exe actually exists
https://forums.livecode.com/viewtopic.php?f=104&t=35765

Fonts!!! I'm a pre-press guy, don't get me started!...ugh!
The IDE has some system font names as string literals in it's scripts ...
I wrote my own LCB to use system font manager, because I wanted to retrieve more information about the fonts being used than the engine's minimal font functions could deliver.
I don't know, neither of the images look bad, just different from one version to the next, obviously something with pixel density changed between 8/9. I wouldn't worry about it since that's more of a backward compatibility thing. I wouldn't count on any app to being graphically consistent from one version to the next.
https://quality.livecode.com/show_bug.cgi?id=23308
https://quality.livecode.com/show_bug.cgi?id=21804

Date and time, localized on mobile...seems that could be fixed or could make an LCB library for that.
https://quality.livecode.com/show_bug.cgi?id=11726

Mobile URL problem I think could be fixed, or substituted
https://quality.livecode.com/show_bug.cgi?id=23308

I think this could be fixed in the LC CLI build, might just be adding a flag and some IF control structure... or could just be commenting out some font lib stuff to make a "no fonts" version
https://quality.livecode.com/show_bug.cgi?id=23163

I think there's probably a LOT of bugs that could be fixed fairly easily and they just don't have the manpower/time to do it all and have to spend time on priority things like customer demand... I know I've just killed an hour replying to this thread, and now I have to get ready to pick up my son, drop him off, and go to my real job...so i get it! But I'm not running a business here so I can spend my time on whatever I want to.

[richmond62]

Those effing 'serious developers' get right up my nose, but, hey, so those
poncy people who go on about classical music purism get up my son's nose:

https://youtu.be/eAUjxDGtr2o

And he has more talent in his pinky than those types have altogether:

https://youtu.be/MwI6Qnu2AAA

My Devawriter Pro, which features half a million lines of code and does "clever things" with
historical Indian languages is serious shit and was written in LiveCode over a 10 year period:

https://www.facebook.com/devawriter

[OpenXTalkPaul]

BTW, I did manage to squeeze in de/re branding updates to three IDE files this morning. Was looking at the included users guide PDF to see if that can be replaced.... I sure hope that PDF is not anything like the file that they're using for the printed manual, for their customers sake and the rainforests sake!

Screen Shot 2021-09-30 at 11.41.12 AM.png (367.57 KiB) Viewed 13142 times

Tiny 1bit icons interpolated to full letter size pages! Oh my!

[OpenXTalkPaul]

Those effing 'serious developers' get right up my nose, but, hey, so those
poncy people who go on about classical music purism get up my son's nose:

https://youtu.be/eAUjxDGtr2o

And he has more talent in his pinky than those types have altogether:

https://youtu.be/MwI6Qnu2AAA

Ya man! Your son is clearly into it, getting into that euphoric "zone" that I think artists long to be in all of the time!

Those "serious" programmers that I know (which includes one of my two brothers) do really boring and serious things for a living (in my opinion). I love xTalk because it's less like "serious programming" and more like crafting a story "script" (which is exactly why they called HyperTalk code a script and not a program). A language that tries to anticipate the way humans think, and doesn't try to make humans think like the programming language authors or computer wants them to think, is best for creativity in my opinion. That's likely why so many artists, musician, writers were (and some still are) fans of HyperCard.

My Devawriter Pro, which features half a million lines of code and does "clever things" with
historical Indian languages is serious shit and was written in LiveCode over a 10 year period:

https://www.facebook.com/devawriter

It does look very nice, like a lot of love has been put into it!

And that's the thing that being serious all the time seems to drain from some people, they don't put "the love" into it, because, it seems to me, that they don't actually love it! What they often do seem to love is having some feeling of superiority over others. I'm too old and too satisfied with where I've gotten to from where I've been, to care about other peoples attitudes like that.

Obviously some work that needs to be done is unavoidably serious, but if you've got 'the love' that can make it easier to do the hard work.

[axwald]

Hi,

"serious coder" here is the equivalent to the mailing list's "cheese"? Didn't know, apologize!

• For me a "serious coder" is someone that does its coding with love & considerable time & labour, and has fun doing it. Someone who, for instance, spends most of the time coding by choice - no matter if payed for or not.
• In contrast to someone that only codes for money, and wouldn't touch an IDE in private.
• In contrast to someone that occasionally does some coding, but has many other, more important, things occupying his time.

Too be honest, I felt a bit startled by your reactions. It really seems as if I triggered a reflex.

This explained, I understand that we have quite different opinions & goals, and that I don't really suit in here. So my best wishes for your project, may it (and you!) live long & prosper!

Have fun!

PS:
Richmond62: Sorry, I don't click random youtubes, and I don't have time nor indulgence to burn my hours on facebook.
Paul: Obviously you have just skimmed the links I provided. Additional hints:
The libURL problem was observed in Win, too - but only randomly!
The "Windows not being good at expanding it's buffers/memory allocation" is noticed as solved in LC 9.6.4
Both Win & Android show unpredictable fonts & sizes, different from any other apps on these devices.
SQLite on newer Linux versions doesn't work - code-created dbs fail with "invalid database type" ...

Edit: Misspelled a name, sorry ...