Imagine something simple . . .

All sorts of amusements and nonsense unrelated to xTalk
User avatar
OpenXTalkPaul
Posts: 1485
Joined: Sat Sep 11, 2021 4:19 pm
Contact:

Re: Imagine something simple . . .

Post by OpenXTalkPaul »

astu wrote: Sat Nov 27, 2021 7:00 am As open source, the possibility is limitless.
Yes, maybe it is still a "cheap copy" of the LC9 community version. But it will evolve with time.

To discuss now if you should stay with an outdated community version or if you should do the work to make a phoenix out of it and develop something new out of it is as unnecessary as a pimple on your butt.

LC has shot down the OpenSource version only recently. For the fact that you are already working on an OpenXTalk based on this version, you are on the right way. And if that gets around, then OpenXTalk can also become something. One should not underestimate the OpenSource community. It starts slowly there, but once it has taken off, then it can become a high-flyer.
Welcome astu!

I agree, some open source projects linger around as zombies for a long time and then all of the sudden the right person gets involved and it bursts back into heavily active development! But there is already a LOT that can be done to enhance LCC 9.6.3 as it is!

One functionality I would like to add to OXT on all platforms, that is available in the commercial versions but never in the Community versions is A PDF Widget. We can wrap the same PDFium lib that's they wrapped for the Indy version. I can't imagine it's much more complicated than wrapping Apple's PDFKit as a Widget/Library (which I've already done). PDFium is open-source a spin-off from Chromium. There's a few others FOSS PS/PDF libraries, one in particular that I'd like to look at, that's based on GhostScript and supports things like Overprint inks inside a PDF (something Apple's PDF kit doesn't support).

Personally, I'm open to just about anything xTalk, even if it's some entirely different xTalk engine running in Web.0.x JS interpreter in a browser window, it doesn't really matter to me if you call it a stack window or a browser window.
FourthWorld
Posts: 280
Joined: Sat Sep 11, 2021 4:37 pm
Contact:

Re: Imagine something simple . . .

Post by FourthWorld »

astu wrote: Sat Nov 27, 2021 7:00 am As open source, the possibility is limitless.
Nothing is limitless. Everything in the universe is a balance between forces of limitation.

With software, the key constraint is time.

Paul's work shows amazing dedication, but the path ahead is long and arduous: Mac is a time bomb that will eventually require updating for M1; Windows has performance issues bordering on prohibitive for some uses; Linux hasn't seen solid GTK integration for years, and it shows; Android is a nonstarter due to the packaging change; iOS has always been a nonstarter.

Overcoming these will take tremendous time.

Paul's dedication is beyond laudable, but it'll take a team with a wide range of deep expertise to keep this engine afloat beyond a year or two.

To attract sufficient dev resources places this project in the same boat with LC: it'll require revenue to attract and keep the breadth of talent needed.

Will donations be sufficient? Are grants available? How much time needs to be spent on finding funding to cover all the other expenses?

This is not a simple mission, certainly far from limitless.
User avatar
OpenXTalkPaul
Posts: 1485
Joined: Sat Sep 11, 2021 4:19 pm
Contact:

Re: Imagine something simple . . .

Post by OpenXTalkPaul »

FourthWorld wrote: Tue Nov 30, 2021 5:06 pm Paul's dedication is beyond laudable, but it'll take a team with a wide range of deep expertise to keep this engine afloat beyond a year or two.
That's assuming that LCC is the only FOSS XTALK Engine that exists, which it isn't!
The LCC engine could be by something else entirely! Perhaps one like StackSmith (C++/ObjC and compiles LLVM bytecode, just like LCC) that's not built from a 30yr old MetaCard code base. Or something that runs on Java or Node.js (like WebCard or WyldCard).

Obviously we could use some more help here though.
With software, the key constraint is time.
I got 17+ years, if Social Security isn't emptied by then, maybe I'll work until I'm 105, who knows.
Mac is a time bomb that will eventually require updating for M1; Windows has performance issues bordering on prohibitive for some uses; Linux hasn't seen solid GTK integration for years, and it shows; Android is a nonstarter due to the packaging change; iOS has always been a nonstarter.

Overcoming these will take tremendous time.
I'm not sure what you mean by some of this. M1 requires recompiling possibly with some compatibility issue fixes, should not require a bomb squad. BTW, I just upgraded to BigSur, so I'll be looking at compiling on Mac again. The App Store, that's always been a moving target, but if you Jailbreak (run unsigned apps) or Side-Load with a Developer account (free dev account allows 3 or 5 apps signings, Dev account is $99/year but allows many more app signings), then you can build apps for yourself on your own private devices pretty much to your hearts content. Who knows maybe the EU courts do something based on anti-competitive laws enforcement or something else that makes 3rd Party Apps Stores are much more available.

Windows has performance issues, I don't know much about (there was some recent improvements). Android engine works great for me and, APK can be rebounded to APK/AAB Bundles, although the build scripts could probably be reworked to support AAB (which is are all script that use the Android SDK CLI tools). Linux...well Linux needs some love for sure, I think some of the problems are fixable / replaceable.
To attract sufficient dev resources places this project in the same boat with LC: it'll require revenue to attract and keep the breadth of talent needed.

Will donations be sufficient? Are grants available? How much time needs to be spent on finding funding to cover all the other expenses?
Grants/Funding is certainly something to consider for the future.

Oh, and as far as we know the universe is limitless, as far as we know this Universe may not even be the only universe that exists! Don't be a pessimist, anything is possible! Hell, maybe someday Apple will even allow cool stuff like Emulators on their crApp store.
User avatar
richmond62
Posts: 2621
Joined: Sun Sep 12, 2021 11:03 am
Location: Bulgaria
Contact:

Re: Imagine something simple . . .

Post by richmond62 »

Mac is a time bomb that will eventually require updating for M1
Well, Yes, BUT . . .

I would assume that a lot of people will continue using INTEL Macs they have recently invested in as long as possible.

Fourthworld (who lives in the First World) is always looking forward, while I (who live in the Second World [if one wants
to be polite about Bulgaria]) am always looking backwards: of course reality [Mind you, I am never quite sure if any sort of
objective reality exists, and even if it does it is not hidden by the maya tattva] probably lies somewhere amidst these
2 extremes.
https://richmondmathewson.owlstown.net/
User avatar
OpenXTalkPaul
Posts: 1485
Joined: Sat Sep 11, 2021 4:19 pm
Contact:

Re: Imagine something simple . . .

Post by OpenXTalkPaul »

I think the objective truth is somewhere towards the middle.

Of course there's always going to be people interested in running old hardware for various reasons.

I'd point you to the "CustomMacPro" crowd of people who have pushed 2008 to 2012 MacPro tower computers so far that the M1 is just recently starting to surpass some of these rigs! They've hacked newer XEONs into the socket and thanks to OpenCore (originally a Hackintosh Bootloader, it can run on real Macs now too) people are running BigSur+ on 13-14 year old hardware (albeit heavily upgraded). So as long as you can get it to run on Core i3 to Core i9 / XEON at pleasant speeds, there's going to be some people still doing that for quite a few years to come.

Nice thing about old rigs is that the parts to max them out, get them to their fullest potential, tend to get cheaper on eBay as the years go by. You can buy REALLY high-end sound cards for PPC Macs for example for $100 now that originally cost thousands of dollars, Musicians seem to go for that sort of thing (love old synths).
FourthWorld
Posts: 280
Joined: Sat Sep 11, 2021 4:37 pm
Contact:

Re: Imagine something simple . . .

Post by FourthWorld »

Scouring the world to look for folks using hardware old enough that its maker no longer supports it seems a less interesting question than what's required to recompile for M1.
User avatar
OpenXTalkPaul
Posts: 1485
Joined: Sat Sep 11, 2021 4:19 pm
Contact:

Re: Imagine something simple . . .

Post by OpenXTalkPaul »

FourthWorld wrote: Sat Dec 04, 2021 11:47 pm Scouring the world to look for folks using hardware old enough that its maker no longer supports it seems a less interesting question than what's required to recompile for M1.
I know what's minimally required for me to be able to compile for M1, that would be a computer running BigSur (check), and optimaly to be able to test it on actual M1 hardware (not check, yet, maybe I'll pick up a new Mac mini down the road).

My understanding is that Xcode has very good cross-compiling on either an Intel or M1 cpu on BigSur+
Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests