To wait or not to wait, that is the question....

So,

I’ve never used the web framework before, I came to Xojo last November and as there was already lots of buzz about the next framework I decided to wait.

However, now I’m at home all day I’ve completed my workload that was outstanding and am thinking about starting this project. It’s a project for a local non-profit and is taking an existing VB.net application that they have and moving it to the web.

The question that I have is this, if I was to start a project now using Framework 1 and designed pages with controls and the code behind them and then move it to Web 2 when it’s released what problems might I come across? What is the most time consuming issue that I am going to have? Should I wait or should I start now?

i choose to wait for api2 and new projects
and i made projects with api1.
rewrite a project from 1 to 2 is no fun and waste of time.
because your project is vb .net you think about asp .net webforms as solution?

[quote=481421:@Markus Rauch]i choose to wait for api2 and new projects
and i made projects with api1.
rewrite a project from 1 to 2 is no fun and waste of time.
because your project is vb .net you think about asp .net webforms as solution?[/quote]

Thanks Markus, good advice.

The project is a complete rewrite so language choice is personal preference.

From what I understand in Greg’s presentation, it will be possible to migrate current web projects to Web 2.00 when it becomes available, with the same kind of one way characteristic as API 2.00. Once the project is updated, it cannot be modified in versions previous to 2020R1.

But 2020R1 is not even in alpha yet, so you should not expect it right away.

Web projects have their own limitations, though, so moving a desktop app to Web has its own learning curve, irrelevant of the tool.

You may want to start experimenting with the current version, knowing that you will be able to move to Web 2.00 when available.

I have a web to do for one customer. I’ve started to play with it for some time now, and have to go further now.
but I will wait for web2 because it will imply too many rewrite I think. not worth it.

I have the feeling that 2020R1 will not be here until at least the end of April. Perhaps even mid June. Or later, as any such major version will probably have lots of work to be ready for prime time.

If your customer can wait, why not. But you don’t want to lose the gig.

Start now unless you can wait 2-6 months.

I would also start with web 1. Even if Web 2.0 gets in our hands very soon, it will be several months before it is ready to handle production work. Probably as mush as the 6 months @ suggests. It is a change as large as API 2.0 was, possibly more disruptive in comparison. Based on everything we heard so far, I would not expect my third party controls to still work with Web 2.0 (it will be a pleasant surprise if they do), adding to the disruption.

For the foreseeable future, I would reserve Web 2.0 to tests and prototypes only. I will do production development on 1.0 for a while, until bugs and various other potential issues are ironed out. I am hoping to be pleasantly surprised, just not planning on it.

For RubberViewsWE Version 1, is is most probable that it won’t work in Web 2.00, since it needs to identify each type of WebControl present on the WebPage, and manage each position and size. Since by definition all new controls are unknown, I cannot assume that RubberViewsWE will recognize them. Also, from what I understand Web 2.00 is API 2.00, and it is unclear if it will even be backward compatible with API 1.00.

From what I understand in Greg’s presentation as well, is that three modes will be available in Web 2.00: Fixed (present behavior), Fluid, and a future auto layout.

So RubberViewsWE for Web 2.00, if I can find time to do it, would only support Fixed. Hopefully, with the present auto layout implementation Version already provides.

exact this is why i am waiting for api 2, because of new features.

but my decision would also be the vb .net app size / complexity.

To answer a few questions from above…

  • Web 2.0 is only API 2.0. There are no API 1 methods, properties or events because it’s an entirely new framework.
  • When converting an old project, the IDE will attempt to move things over to the new event names where possible but code transition will largely be your responsibility. It is almost always problematic for us to adjust user code, so we only do the stuff that we deem completely safe.
  • Something important to note about layouts… the new controls are taller than the old ones by about half. That is, they were all about 22px tall and are now 34 for the most part. So while the layouts do convert, you’ll still need to tweak them.
  • We are doing some experiments to see if the old WebSDK controls can be even partially supported, but there is no word on this either way so far.

Ooooh… That means a complete redesign of some of my heavier master data forms (pages/containers). Some of my forms are already a bit crowded at 22 pixels! Better start now in Web 1.0 and review the data entry process flow. The transition will be easier.

Won’t there be a Theme that will match the Web 1.0 design to ease the transition?
Will new controls work with old sizing or will they spill out/overflow?

It’s not something we had planned on. Bootstrap themes tend to have larger controls overall by default, but there are smaller variants of some of their controls… so maybe?

Some yes, some no.

The current plan is to let people use the default bootstrap theme as it’s modern, clean and the controls all look good together.

FWIW, with all of the internal projects I have converted, layouts have typically taken 3-5 minutes to resolve and then you’re done.

Anyway, to correctly support mobile, 22px is way too small.

In the past, there was a 90-days period cycle (even earlier, it wasn’t the case). Has this changed, did we return to the “first” way? What’s the “rule” now?

The policy is still “All Xojo licenses include 1 year of updates and are covered by a 90-day money back guarantee” as stated on the Store page. Questions like that are best directed at hello@xojo.com though.

Yes, I’m sure this hasn’t changed. But if they release a new version, say, only 1 per year, you don’t get that much…
Until “now” (whenever that is), each 3 months was a new version available; looks like this has changed.

[quote=481647:@Arnaud Nicolet]Yes, I’m sure this hasn’t changed. But if they release a new version, say, only 1 per year, you don’t get that much…
Until “now” (whenever that is), each 3 months was a new version available; looks like this has changed.[/quote]
Sorry, I misread your question, I thought you were asking about the 90-day money back guarantee. The number of releases per year is typically 3 or 4 but it varies. FWIW, the most recent update was Xojo 2019 R3.1 released Jan 23 2020.

No problem. I understood you did perhaps misread it, so I re-wrote it in my answer.

It used to be said, by Xojo themselves (RealSoftware, at that time (not sure the exact period)) that every release had a 90 days cycle (many complained that releases were not “finalised” well enough because of this).
So, yes, a release each 90 days would result of 3 or 4 per year, but I’m curious about the specific statement (90 days per release) made years ago which, based on the observation of Michel that the Alpha hasn’t even started yet, looks to have been withdrawn.

Thank you.