Wondering about LC and HTML5

Terry Vogelaar tvogelaar at de-mare.nl
Tue Jun 21 00:47:18 EDT 2011


Hi Chipp,

You bring up some interesting points. But still, in my opinion, HTML5 is (as of now) too limited. In spite of everything Apple is trying to say about it, it is not a standard. And considering the approach it is taking, it never will be. What I mean is that there is still room to add features to it; a stage that would have been over if it was in beta. Still, the entire industry wants to use it. Which results in a muddy, half-baked product. 

Yes, I agree that the puck is going to be there eventually. But what needs to happen first is some decision-making about what gets included in the standard, and what is shifted to HTML6. And that is a scary step to take, because it means excluding functionality, and thus limiting possibilities for web app developers.

Sorry, I love HTML5 and use it as much as I can. But I also passionately hate it. It is all going too slow and too fast at the same time, if you get what I mean.

Back to the subject: should LiveCode move towards being a web app development environment? I truly don't know. I like the idea. But it also means there is a major change in the way LiveCode works. It is proud to be compile-free, breaking the development cycle of editing, compiling, running and debugging. If however the code is being translated to something else like HTML5, there HAS TO BE a compile step. The code that is used (HTML, CSS, JS, PHP) differs from the code that is written (LiveCode). Which makes debugging a living hell. You'll end up changing the HTML5 files manually, and that is exactly what you tried to avoid. Otherwise you would have started using HTML5 right away instead of using LiveCode first. 

If there are people who pursued this dream further than any of us, it would be Jerry Daniels and Sarah Reichelt. With their Rodeo product they accomplished exactly what you describe: a development environment to use LiveCode-like input and convert it to a fully functional web app. It isn't my cup of tea, for exactly the reasons I mentioned above. But they made a fine product. I agree with what Andre Garcia wrote: "Desktop code is not portable to server environment but if you start from the server point of view from the beginning then it is possible." Jerry and Sarah started from the server point of view; they succeeded. Does this mean that every LiveCode project can become a web app? Absolutely not.

If I judge based on the keynote on WWDC'11, Apple seems to actually be moving away from web apps, moving towards standalone apps. For me, that is indicating that, although Apple is pushing HTML5 forward, the puck is not there yet. It will be. With a time span of a few years and a puck constantly moving, the puck will be on every spot of the playing field at some point in time. Does that mean it is wise to stand still or perhaps move in whatever direction we feel like? I don't think so; it is wise to go to the spot where it will be when we can hit it the soonest. Translated to our situation, it means that we envision how long it will take for us to get the job done, and figure out whether HTML5 will be ready enough to deliver. If your project is a stunning website, the answer is: go for it, you will love HTML5. If it is an application, I wouldn't hold my breath.

Terry


Op 21 jun. 2011, om 04:10 heeft use-livecode-request at lists.runrev.com het volgende geschreven:

> Perhaps it's wise to consider an oft-quoted famous statement by
> Wayne Gretzky, "I don't skate to where the puck is. I skate to where it's
> going to be."
> 
> Let's count the number of OS'es now in need of support from multi plaftform
> IDEs:
> 
>   1. MacOS Tiger, Lion and who knows what previous versions
>   2. Windows 7 and soon 8, not to mention 2000, XP and Vista still in use
>   3. iOS (iPad and iPhone)
>   4. Android and the many different flavors and versions it has for both
>   phone and tablet.
>   5. HP's new WebOS
>   6. Doesn't Blackberry have a tablet OS?
>   7. ChromeOS and Chromebook
>   8. Linux and it's many different flavors
>   9. I'm sure I'm missing some
> 
> Now the number of mainstream browsers:
> 
>   1. Internet Explorer (I see where Google just end of lifed support of it
>   in Gmail and Google Apps)
>   2. Chrome
>   3. Safari
>   4. Firefox
> 
> It appears to me, the browsers are consolidating much quicker than any of
> the OS'es. In fact, the OS'es are all competing with each other by
> highlighting the differences, something the browsers have to be much more
> careful about.
> 
> Assuming HTML5 becomes decently stable and robust AND capable, doesn't it
> make more sense to target it as a delivery platform rather than having to
> contingency plan for the constantly moving target which the various OS'es
> represent? And, that's the real problem with most OS'es today-- they are all
> basically the same. Sure, some are prettier, others more safe, but they all
> provide the basic same functions.
> 
> In fact, what is more interesting to me is seeing how Apple had to actually
> reset OS expectations with the iPad. Let's be honest, the OS on iPad is way
> behind in terms of functionality. No true multi-tasking. The interface has
> been dumbed down considerably yet see how EASY they are to use. And Jobs
> knows that EASE OF USE translates into sales. The first Mac 128 cold booted
> in under 30 seconds. My Mac and Windows machines take many minutes. Don't
> get me started about how feature creep has ruined countless applications--
> MS Word primary among them. I can get 99% of what I need done word
> processing-wise using the MacWrite like GDocs. Adding features on top of
> features on top of features is just confusing everyone.
> 
> One thing about Chromebook-- EVERYONE knows instantly how to use it. Turn it
> on and it's just a browser. I don't care whether it's Linux, Debian, MacOS,
> Windows or your mothers cupcake under the hood-- it just works.
> 
> So, if we focus on where the puck WILL be-- is it fair to say it's headed in
> the direction of HTML5 web apps? I dunno, but it is interesting to
> consider....




More information about the use-livecode mailing list