Suggestion: Non-Appbuilding Community Edition
daniel at digitalfauna.io
Sun Sep 5 09:39:29 EDT 2021
> Introduce breaking changes when it's necessary
> to move the language forward
We tried "cut off the old hair" memes already, remember?
I would argue that retaining legacy functionality and behaviours whilst
trying to modernise or improve a language introduces bugs and is a drain on
resources. Hence why nearly every successful language has had to bite the
bullet at points and break compatibility in order to move forward. That
didn't happen during the refactor (it probably should have). Maintaining a
stable code base whilst still supporting 10 year old stacks hasn't worked
On Sun, 5 Sep 2021, 15:32 Curry Kenworthy via use-livecode, <
use-livecode at lists.runrev.com> wrote:
> Hi Folks,
> This seems headed for trouble again if we're not careful.
> We must avoid repeating the same history:
> 1. Added work for LC Ltd without compensation*
> 2. Buggy struggling main product due to #1
> 3. Overcomplicating things
> 4. Burdening those who pay with the extra expense
> 5. *Added work for ourselves without compensation;
> (That was the previous "bright idea" remember?)
> Cutting out the free Community version is a smart move.
> $10/mo hobby is pretty darn cheap. Everyone can afford that.
> (Some end-clients want OSS, but only half of those know why.
> The other half are only repeating something they heard.)
> A demo is beneficial, and calendar-time-limited demos suck.
> Thus, unlimited-calendar-time demo could be the way to go.
> 10-line scripts suck too. Non-standalone might be the way.
> But as Alex said:
> > You can make it easy (or even trivial) for anyone
> > to install and run the stacks you create.
> Yep. Even easier than your example; no plugin necessary.
> A shortcut to your stack, and it launches the IDE.
> Not that much different from a full desktop app.
> I could make it near enough to please most users.
> Then we're still encouraging nonpayment for LC.
> So we need an additional limitation.
> But now we're getting into bad ideas...
> > what if only licensed versions of LC could produce
> > and run distributable/shareable stacks while the free version
> > could only run stacks produced by that particular instance of the app?
> That's getting nowhere. Two separate communities to support
> (plus the sucky problem of Community-can't-run-this-stack)
> so extra work for free, and it's begging to be gamed.
> I can probably still make a great "app" experience.
> This repeats all or most of the old problems,
> and even discourages using the $10 version.
> Worse still...
> > Introduce breaking changes when it's necessary
> > to move the language forward
> We tried "cut off the old hair" memes already, remember?
> That was part of the open source breathless refactoring excitement.
> Result: twice the bugs with a quarter of the performance for years.
> Plus tons of added work for us and our clients to update stacks.
> Some people went out of business, others used tons of time or money.
> Many of these misguided repeating memes simply need to die!
> Better to kill a meme than to see more people get hurt.
> LC is not an OS. Breaking changes have been a major pain in the rear.
> Almost as bad as the extra bugs and performance problems.
> To have a future, we need a firm stable foundation to build upon.
> Not encouraging an ever-shifting mire. Recompiles, yes. Rewrites, no.
> If maintaining is not easier in LC, people will go use other tools.
> We've seen that already. We need to learn from experience.
> > I'd also get rid of any existing lifetime
> > and lock in licenses (sorry, time to clean house)
> That would clean house all right! Hand grenade style.
> We'd be getting to the point of serious self-harm.
> This one is not so bad, though:
> > Nag screen with 5-10 second timeout in IDE and standalones
> Nix the standalones; free should be noncompiling, at least for desktop.
> But nags can be useful. And a possible mobile solution,
> but it would have to be combined with one more limitation.
> Today's users are quite willing to tolerate some nags.
> If LC Free competes against its own $10 version, nobody wins.
> We're back to a buggy underfunded main product.
> I've been too sick (good old Delta) to follow the whole thread,
> just read the last few messages, and will probably not be able
> to follow the rest of the discussion for a while either.
> This post is all I can muster.
> But let's learn from the old mistakes, eh?
> Many things which sound great ... aren't.
> Repeating them makes things ... worse.
> Let's not nuke ourselves again in the process. :)
> Happy coding, and hopefully I'll be back in action after a few days.
> Hoping this doesn't head right off the cliff while I'm down sick!
> Wish I could give this the full attention it deserves....
> Best wishes,
> Curry Kenworthy
> Custom Software Development
> "Better Methods, Better Results"
> LiveCode Training and Consulting
> use-livecode mailing list
> use-livecode at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
More information about the use-livecode