Livecode: observations

[snowfun]

I've been a user of Livecode since "Day 1" having travelled the well-trodden road of hyper-, meta- and super-.

I don't write "apps" for a living. But I do write the occasional small piece of software which runs on the web or desktop for use in my teaching environment (University). And by and large I have never failed to get a piece of software to do what I want it to do. Possibly not elegantly but if it works then (for my non-commercial purpose) then that is fine.

I have been battling to get wider adoption of Livecode within my academic institution (by coincidence not too far from Livecode HQ but that is indeed only a coincidence). If I want to write anything for wider adoption (i.e. more than me using the tool in my lectures) I am told to "Use Python".
I managed to get a server manager to look at the Livecode website... "it's childish..." (and in many ways it is hard to disagree)
A software developer... "not interested in 'easy, fast, free'..."
Another software developer... "it's a toy for kids to learn how to code..."
And (ad infinitum) "use Python..." (I confess I can't because I don't know and don't want to learn how to programme in Python but that's irrelevant).

I wish I had not mentioned that the new version "does" HTML5 (at least that is what I thought the web site said). Someone bothered to look... ooops.

And now we have yet another series of crowd sourcing appeals (dollars - sic). Is this really what a professional programming tool should be doing? If the language is "free" (which it isn't...) why isn't an enthusiastic talented pool of developers constantly adding to it? I think I have a badge to show my support for the open source campaign. I threw my cash at the HTML5 funding. I'm no longer sure I want to invest more cash...
But, and I don't particularly want to even start thinking this way, I can sense that my time with Livecode might be nearing an end. I even registered for a Python account (and - I admit - easily wrote a script to access a mySQL database!)
Depressing days indeed. But as the subject line states, just my tiny observations...
Any grounds to remain optimistic?
Tim

[paul_gr]

Any grounds to remain optimistic?
Tim

In my case, No.

After using LC for 8 years and trying to get others interested in Livecode, I have had similar experiences.
Sadly I am starting to agree with some of their opinions.
The best part of Livecode is the community of users; but IMO Livecode itself doesn't seem to know what direction to go in.
Maybe other users will respond in a positive manner...

Paul

[dunbarx]

I have been accused of being a "fanboy". I suppose I am.

Perhaps it is because I only write for desktop (and even there use little of LC's database, file gadgetry, video, sound or server capabilities) I might be forgiven for thinking LC is the greatest thing since sliced dice. Greatest because I can do anything I want to. Anything at all. Of course, did I mention that there are entire worlds that I do not deal with?

So is it weak support for those worlds that is disappointing? Or the time it is taking to get LC8 out? Or is mobile just not up to snuff, like having to pay for a remote debugger? Is the team too small to compete with Python development? This was the same issue with HC, lack of massive cash, marketing, support and effort.

Who was it who said "instant gratification isn't fast enough?" I hope it is not that simple magic just doesn't cut it anymore...

Craig Newman

[LCMark]

@snowfun: To be honest, I'm not entirely sure what exactly you are getting at in your post, however, I do think that you are purporting a number of things which suggest a slightly rose-tinted and perhaps not entirely representative view of the (open-source) software world...

If the language is "free" (which it isn't...) why isn't an enthusiastic talented pool of developers constantly adding to it?

Be careful not to confuse 'free-as-in-beer' with 'free-as-in-freedom' - code costs money to write as people have to be able to support themselves whilst they do. (I realize that Richard Stallman says that 'software engineers should live modestly', but they still have to live and that does tend to come down to having a income last time I checked).

In terms of free-as-in-freedom: I believe the source-code for the LiveCode system is available at https://github.com/livecode and has been since 2013.

You imply Python as a comparison - but I believe a substantial amount of development on Python was initially funded by a DARPA grant and then essentially funded by Google when they employed Guido van Rossum (the language designer) for a good number of years. This is quite typical - projects often start off out of research grants in universities and /or large organizations employ people to work on the projects they want to contribute to (thus essentially paying for their development), or contribute money to organizations which pay for people to work on the projects they need.

In terms of open source programming languages, then:

• GCC - has always been heavily funded by industry for various purposes.
• LLVM/clang - started as a research project at University of Illinois (presumably some of our US compatriots tax-dollars were involved there) and Apple is a huge contributor these days
• Rust - a programming language funded by Mozilla in order to rewrite the FireFox browser
• Go / Dart - Google (I don't think anymore needs to be said here)
• Mono.NET - copyright held by Xamarin who dual-license and sell iOS / Android frameworks under a closed-source license

Programming languages and environments are some of the most technically involved projects that exist. It is not that surprising we don't have a huge number of contributors to the engine core - most of the people using our product on a day-to-day basis are doing so because they don't want to have to use a lower-level language (which is one of the prime motivations behind LCB in LC8 - we're trying to make it easier for more people to extend the LiveCode system). Obviously it would be nice if this situation changed, but it will only change as our user-base grows and we attract more people who are interested in getting their hands dirty with the internals - there are no magic wands we can wave to make that happen, it just takes time.

I'm not sure where the idea that 'open source projects are all written by people in their spare time' comes from, but I honestly don't think it is true for any open-source project of any significant size (certainly not any that evolve and develop at a rate commercial use tends to require). Sure, that's how many of them might start, and the many of the most popular projects have a lot of people contributing to them; but the reality is that all the open-source projects everyone relies on are heavily funded by large organisations (corporations in most cases) who have deep pockets. I think a recent survey done on Linux kernel contributions found that 90% of them were submitted by engineers working in paid-for roles in various companies around the world - free as in freedom, yes - free as in beer, certainly not as someone is footing the bill somewhere. (I'd also point out that fundamental open-source projects which don't receive enough funding tend to cause global problems - OpenSSL and the HeartBleed debacle is a very good example - http://arstechnica.com/information-tech ... d-openssl/).

And now we have yet another series of crowd sourcing appeals (dollars - sic). Is this really what a professional programming tool should be doing?

We have to get the money to fund development somehow (which is where the majority of our income goes). We have an open source product which anyone can download and use. Rather than have a undirected 'donate' button on the website (which many open source projects do), we decided it would be better to have a heavily directed 'donate' button. Perhaps you disagree - if you do have any magical money trees, I'll have a whole field full please

In terms of 'is this really what a professional programming tool should be doing?' then please do Google around the list of environments listed above and look at how they finance themselves. Unfortunately we don't have Google, Apple, or any other large organisation pouring money into us to fund development - we always have, and continue to stand on our own feet with the support (which, of course, includes financial) of our community and investors who see the value in what we do.

Finally, in regards to convincing others that LiveCode is worth looking at, then it sounds very much like the individuals you are referencing haven't actually taken the time to look at what LiveCode is and what it can actually do. This does, perhaps, say more about these individuals than us - in any endeavour it is always unwise to dismiss a potential tool for your toolbox without proper evaluation. That being said, unfortunately due to the nature of LiveCode and its heritage we do appear to have an innate uphill struggle in this regard and are always looking for ways to help improve our message and help people understand the value that we believe it brings - it is a problem which we constantly strive to find a solution to.

[snowfun]

Just an observation. Not "getting at" anything. Pleased to see some debate arising however.

Your comment "(T)hat being said, unfortunately due to the nature of LiveCode and its heritage we do appear to have an innate uphill struggle in this regard and are always looking for ways to help improve our message and help people understand the value that we believe it brings - it is a problem which we constantly strive to find a solution to." made me think.

Would you be prepared to assist me in trying to persuade my colleagues "up the road" (UoE) to implement Livecode on the servers and try to get it used across the institution?

[Lagi Pittas]

My twopenny worth.

@LCMARK I'm glad SOMEBODY from Livecode towers actually has responded here rather than lurking so here's my take for what it is worth.

I looked at Livecode many moons ago when it was metacard (I actually used Hypercard years ago when I was an Apple Dealer, and was programming in UCSD Pascal on the Apple 2 - but that's a long time ago.)

It was fine for media and graphics applications but I wrote "serious" programs Ledgers, Stock Controls, Property Management , Payrolls etc etc and I had been using "proper" languages like Pascal, Delphi, "C" for Telephony Voice Applications (Use the best tool for the JOB). WE programmers have a habit of assuming our day to day most used language is the best one for everything - it ain't - it's the comfortable shoes syndrome, the language is scuffed, short on tread and the colour has faded but it.s comfortable and we can program without looking up in the dictionary for every 3rd statement.

Over the years I looked at runrev and it still couldn't hold a candle to Visual Foxpro (not much can - for windows based Database and Financial programs - Or the Channel Tunnel software). It came on the radar again when the crossplatform ie IOS and Android options were included because Microsoft stopped developing Visual Foxpro in 2004 - and yet it still works if I'm not bothered about Mac, IOS or Android.

I gave it another chance again but still hated the put x into Y and the non dot notation and lack of object orientation with classes but the fact that I could use real Databases and sqlite easily was what made me decide that I had in the words of Sun Tzu to "Burn my Bridges" at least for my current application.

Three months down the road when one of my clients had a problem with their Sage Accounting system I used Livecode to create a oneoff utility that ordinarily I would have done in Foxpro (or Delphi) that's when I knew it was the right decision.

I now very rarely use a = b + 25, the verbose object naming is still verbose but doesn't have me swearing - cut and paste works.

I introduced a programmer friend of mine to Livecode after he had derided it as a Toy language by telling him give it chance (the put x in Y and non dot notation was his bugbear as well)

His problem now (he is working in piece of software with me) is the inconsistencies and bugs in the CEF browser he's got used to the syntax (although we would both jump at the chance of using dot notation or a javascript prototype object system) - these two mods would bring over the people you need to bring over the delphi, clipper, foxpro, vb, alpha five programmers (yes I've programmed in every one when needed but Foxpro was still the best for my bread and butter). Notice I didn't mention C which I used for telephony applications and CNC Alpha machines - when I would have been stupid to use Clipper or Foxpro - C programmers usually see everything as a nail because "C" is their Hammer.


Suggestions
1. Your website is a Joke - and I'm being nice here stop farting around and pay tactile media to redo it please!!!
2. get the html5 and cefbrowser stuff working consistently asap - widgets can wait a bit - we never had them before what we never had we wont miss for a year - use it internally to develop LC so you are eating your own S**T (or dogfood in Microsoft Parlance) - you will find all the bugs without alienation anybody.
3. Talk to the Business Coders about what they see as the problems on getting other "serious" programmers on board - i'm not being derogatory it's just that LC looks like its just for Puzzle games and "Apps" not software solution at a cursory look.
4. license Scott Mcdonald's Grid control ASAP - Theres a Pascal programmer who knows his onions and burnt his bridges from Delphi because of the CrossPlatform options in Delphi - ask him for his suggestions , and take note he'll have a good few i'm sure
5. Push Monty's stuff as well
6. Get Databinding in there as soon as possible
7. How about tying in a Javascript interpreter into the Engine so you can ease people in - Python has it's own problems, If you want elegance go for Ruby but I can find fault even with an orthogonal language.
8. Get a round table setup with the Usual Suspects (Richard, Klaus, Richmond, Dunbarx , Craig, Simon, Monty, MWeider - sorry if I missed you off - you know the ones) and have a skype meeting every couple of months to air the dirty laundry
9. Sort out theDocumentation
10. Did I say Sort out the Website? - stop promising Jam Tomorrow !!

I hope LC Towers takes this as constructive criticism - I'm in for the long haul even though there are bugs/undocumented features and plain old no attention to detail - graphics that look thrown together or the real bugbear checkboxes that look ugly in the IDE, Tabs that look ugly on windows - black instead of defined with a checkmark or dot.


Sorry for the long stream of consciousness and for the Smelling Pistakes - I really want LC to succeed - FIX THE BL***Y WEBSITE!!!!!

Lagi