Quick Question re livecode built websites

[bbalmerTotalFluency]

Hi:

My experience of that is zero. My knowledge and understanding extremely limited. I'm wondering if they are reasonably performant. I don't want to have to build one to find out. Hoping the gang here might know.

My extremely dodgy understanding is that on going to a livecode website home page - a whole load of stuff gets downloaded first, then the page renders. Which sounds like it would be rather slow.

Is LC's site built in LC (that's as performant as I need it to be)? Can anyone point me to any websites (public facing) that are built in LC? Can anyone simply point out that I am wrong in my belief above OR that it takes so little time that it doesn't matter?

I'm thinking of a simple website, very straightforward. But why not leverage my LC "fluency"? rather than tangle with js - which we all know and love but I find it hard to hold multiple programming languages in my head successfully without becoming at least slightly worse at both.

Bruce

[SparkOut]

Do you mean "website" to be an app built in the LiveCode IDE which is then published to html in the Standalone Builder?

Or do you mean "website" to be a structure of pages delivered by a server with mixed content including that generated by LiveCode from hosted files and libraries much the same as PHP?

If the former, I have no current experience but I believe that in more recent years there have been considerable advances in the initial load and performance over the original experience of early years. I am not competent to advise, however.

If equating with PHP, then I think you will find LiveCode Server is very responsive and performance is admirable. You would still need a combination of html, javascript and LiveCode to create much dynamic content, but treat it like a pure replacement of PHP and simple websites would be a breeze.

My own use has never really been to "create" websites so I can not direct to any public facing examples. In the past, I have frequently done a lot of back end work using LiveCode Server to enhance or automate functions and features for a client on their own site, often interacting with MySQL and/or Wordpress as well.

[SWEdeAndy]

Hi Bruce,

So, to load a website created with LiveCode 10, an instance of the engine needs to be downloaded once (per session). For each LC10 dp version this is getting lighter and faster. I cannot judge fairly what the "normal" download speed is, as I'm blessed with a gigabit/s connection, so for me it's always lightning fast, with no measurable difference compared to most other websites. You'll test and see how it works for you.

The LiveCode website is not built with LiveCode. (And yet, it has for some reason always been amazingly slow to load...)

Generally, if all you want to do is build a website, with mainly text and image content for example, I would recommend using Wordpress or any other website builder tool out there.

However, if you want to build a web app, that needs programming to perform things - and especially if you already have an LC stack/desktop app that you have developed - then it's in many cases ideal to try deploying it to web with LC 10.

I have a number of example deployments that you can test (NB! They work best in desktop browsers - some things work on mobile browsers as well, some not at all):

An early test of various functions:
https://wheninspace.com/portal/

Linking in external resources like Mermaid:
https://wheninspace.com/LCXcode/

Example of a simple stack deployed without any adjustment to web at all - it just works:
https://wheninspace.com/colourcalculator/

LC stack and web page interaction via simple JavaScript:
https://wheninspace.com/WebpageInteractionStack/

Example showing extended capabilities that can be implemented, such as local storage and file upload:
https://wheninspace.com/WebDeployHelperTest/

Game of Life simulator:
https://wheninspace.com/gameoflife/

This is a browser animation developer helper tool intended for use in the IDE, but works amazingly well on web too:
https://wheninspace.com/browseranimation/

And finally, not a web deployed stack itself, but the WebDeployHelper tool that facilitates testing and deployment, and enables several functions that would otherwise require JavaScript hacking to get:
https://github.com/wheninspace/WIS_WebDeployHelper

Enjoy!

[FourthWorld]

I'm thinking of a simple website, very straightforward.

Can you tell us a bit about what you want the site to do?

There are soooooo many ways to make so many types of web pages, and many don't require you to do any hand coding at all.

Knowing what you have in mind will help us narrow the range of options to the best fit.

[bbalmerTotalFluency]

Thanks everyone so far.

Richard - my thinking is not that specific yet. I can see many situations in which I might want to create a livecode app that functions over the web. Simple stuff. Collecting course registrations, automating actions, producing summaries, creating auto-course cut off points once full etc. One difference that might narrow down the question is that the site would definitely be public facing rather than within an enterprise - where I could see that big initial download being much less of an issue.

I just wondered if there was much point in exploring because people are apt to say - you can't keep users waiting more than 2 seconds these days or they lose focus and watch something boring on the telly. I wonder if that really applies to stuff like tiktok more than when someone is signing up for something in which they are interested.

I'm also aware, or think I am aware, that the U.S. still have a lot of people who have truly dreadful internet connections. So having to load a 10MB file once per session might be a bit heart breaking for them.

I know I can do all this with PHP, JS, HTML and SQLITE. Which might be one standard solution - just wondering if my love of all things Livecode could be usefully extended to the web or if, for now, I should NOT involve myself in it.

[stam]

Sounds like you may best be served using a host that can run LC server to handle logic and build websites as per usual, just integrating LC 'brain' as needed. This should be as fast as anything else, not needing to download an LC web app first.

[FourthWorld]

Richard - my thinking is not that specific yet. I can see many situations in which I might want to create a livecode app that functions over the web. Simple stuff. Collecting course registrations, automating actions, producing summaries, creating auto-course cut off points once full etc. One difference that might narrow down the question is that the site would definitely be public facing...

This sounds like building a business, rather than a tool for use only in your organization. Is that correct?

There are soooo many ways to scratch that itch...

If I were building a business for course scheduling and registration, the first thing I'd do is see if there's a SaaS that already does what I'm looking for, and if they have an affiliate program or other partnering options so I could focus on my own value-add and leave everything else up to them. Some may offer whitelabeling so you can custom brand it.

This would mean less per-seat revenue, but at a cost of near-zero, so it's likely an unbeatable ROI.

Building systems like this in any tool stack is a lot of work. In addition to nailing the critical path of the requirements themselves, there are privacy and security considerations that will impact the development budget. And then there's all the stuff outside of development which usually totals more, like marketing and sales, which will be the same regardless how you build it.

Edu is a big arena, and has attracted a lot of players over the years. You may have a fresh take that requires custom implementation, but maybe not. Survey the landscape to determine options.

If you pursue this, there are two sides: client and server.

Your question was about the client side that LC's experimental HTML export aspires to provide.

And as Stam noted, LC Server can help with the server side.

Both sides will be needed.

A thought about each:

I'm also aware, or think I am aware, that the U.S. still have a lot of people who have truly dreadful internet connections. So having to load a 10MB file once per session might be a bit heart breaking for them.

Is it necessary that the app appear specifically in a browser window?

You could deliver a standalone that offers all the connectivity benefits, but with the UI benefits of an OS-native app.

You'd be in good company. Microsoft, Adobe, and others have been successfully doing this on the desktop for years, and most mobile apps use this model.

Then it's a one-time download opening the door to a lifetime of collaboration support.

I know I can do all this with PHP, JS, HTML and SQLITE. Which might be one standard solution...

I love my bicycle. I use it for nearly all my errands. I've even taken it to LiveCode conferences so I can get around town easily and enjoyably.

But for the long haul of getting to a conference I put my bicycle on a train and let someone else do the driving while I enjoy the scenery.

Even where we can do something ourselves, sometimes the entire world isn't wrong.

I can cycle from Los Angeles to San Diego. But will that provide the ROI I'm looking for?

If my goal is the exercise itself, bicycling beats the train.

But if my goal is to be in San Diego, the train does all the hard parts for me.

[bbalmerTotalFluency]

Richard:

I cannot contest the tsunami of flawless logic that you just espoused.

Thanks. I'll change direction. I like the train. I like high ROI.

Bruce

[FourthWorld]

Thanks. I'll change direction. I like the train. I like high ROI.

I'm intrigued, Bruce. Are you considering an existing courseware SaaS, or a standalone?

And there may be value in combining them, depending on the richness of your presentations.

[bbalmerTotalFluency]

I don't understand your question. Is that a bad sign?

I have a special way of teaching Spanish that drops learning time per level 0-A1, A1-A2 from 300 hours to 30. I've been doing that for years. I use a standalone to FACILITATE that method. I've rewritten the standalone recently. It works like a charm (except for crashing occasionally).

I am transitioning from private tuition to group tuition, hoping to end up with large groups like 100 at a time.

The website will do nothing more than explain what I do. At least at first. But then it would be nice to automate course availability, emails that say "sorry, this course is full, next one starts....", reminder emails etc. I could have it hand administered but computers are good at that sort of stuff. I think from what people are telling me, I want a standard website with livecode subbing from PHP - mostly because I prefer one language over several (of course the unavoidable js will creep into my project too) unless I can find something that already does that, which can be integrated into my website - as you said, that is probable.

I would have used livecode to substitute for html if that had been an option. It appears not to be.

So I think I have a path forwards. In fact, having thought about what you said earlier in this thread, I've decided on a super-simple website for now, see if I'm able to make the transition successfully.