So we had a feedback that the plugins are really old and they should be removed or updated. Any thoughts about what we should put in there?
What do you mean by plugins? Those that appear in the menu? Or the toolbar widgets?
Certainly some bear updating but I would argue the whole plugin access system should be refreshed. If anything the plugin menu should be a menu of its own rather than a submenu. Itās always annoying to access the submenu and even more as a beginner.
As for toolbar widgets pretty near - most of them need updating. The LC boffins felt the same and have updated most of the widgets although annoyingly you now have to refer to them as widget instead of field, button etc.
On that note have a look at this MIT licenced improvement on the standard button:
Itās what LC have based their new button on, but the source is MIT licensed and fully compatible with GPL3.
Ralf Bitter, the author, has released quite a few open source solutions that are worth looking at with the reboot of the platformā¦
There is an illustration on
I was referring to the old submenu āPluginsā. The widgets (in the Tools palette) need to be updated as well? Iāll take a look at the repo you posted and see what we can do with it.
For sure many of these need updating. At the very least the layouts, as pretty much none of these have a responsive layout which is inexcusable in this day and age for most cases.
The one that immediately springs to mind is the regex one. Itās really not that helpful and a lot could be done to make it more user friendly.
On that note Iād ask if it might be possible to change the flavour of regex from PCRE to JavaScript flavoured regex. The immediate benefit would be allowing the /g flag for global matches which PCRE lacks, as well as making it 100% compatible with web based resources, which PCRE isnāt quite, because of the lack of the /g (global) flag which allows you to find all the matches, not just the 1st one.
Iāve always had a soft spot for plugins. During my very long tenure with the other company, I built at least a couple dozen of them ā mostly little helpers to make life easier.
Things like a clone of the old Scripterās Scrapbook, where you can stash tiny scripts youāll definitely need againā¦ right after youāve forgotten how you wrote them 
.
And then there are the widgets on the left. In my opinion, incredibly powerful ā in practice, mostly admired from a distance. I remember someone once confidently said that weād have a thousand widgets by the end of the year. Fast forward several yearsā¦ and weāre proudly sitting at about a dozen. Ambition was not the problem.
That said, I do agree this is a good moment to clean house. Some of the old plugins still hanging around are about as useful as a screen door on a submarine ā either they donāt do much anymore, or they donāt work at all.
Iām sure as the app evolves, each of us has at least one plugin idea that could earn a spot. But for now, weāve got a few more pressing things to focus on before we start building the next thousand plugins 
I wonder what it would take to do this? I was going to update PCRE in the future, so might be a good idea to look into what needs to be done.
I do like the idea of promoting the Plugins menu to being a primary menu instead of a submenu.
One of the strengths of this branch of the xTalk family tree is the extent of customizability. Celebrating that is useful.
AFAIK few (none?) of the plugins there carry explicit declaration of compatible license, so Iām not sure how they wound up in the old LC Community Edition.
With LiveNet I didnāt mind, and likely the others didnāt either. But nice to have things tidy.
And speaking of LiveNet, thereās a place for it, but not on this earth at this time. My intention was to provide a digital watering hole to celebrate whatās going on around the community, colating the most recent activity from various LC RSS feeds, useful stacks, and community notes. But as the size of the community has shrunk, thereās less to aggregate, and at best it just draws attention to the shrinkage.
There was also a secret agenda with LiveNet: a living reminder that with this engine entire application experiences can be downloaded easily and quickly with no need to install.
In this regard an xTalk can serve as a sort of custom web browser, but instead of serving public resources it serves only stack files, a data type unparsable to search engines; a universe of possibilities living behind a velvet rope, like Hesseās Magic Theater, very intentionally āNot For Everybodyā; a discovery to be found by initiates, with all the richness of desktop integration, things beyond what web browsers can do.
But no one took the bait. To this day the potential for this sort of groupware app streaming is something Iāve only seen realized once, in Richard Herzā Reactor Lab.
Heād made that to share chemical reaction simulations for his students at a California college, and later it became somewhat federated, with other reaction simulations downloadable from colleaguesā servers across the country.
But RL sims were generally single-window designs and quite self-contained, and didnāt really benefit from being outside of a web browser. Indeed, it began before web browsers had XHR, but in later years was ported to the web.
I still think about the potential with xTalk streaming groupware, but over time it seems clear thereās a different way to promote that. And with todayās web browsers being so capable, the utility of such a thing should be more focused, take a different formā¦
So in short, I wouldnāt be butthurt if LiveNet was removed. If thereās a use for something like that in the future, other integrated options can be discussed then.
Okay, weāre going to need some new plugins, anyone up for the task of either revamping what we have or making new ones?
The Plugins submenu has been promoted to full menu now
1 Like
we should discuss if we care about any of these old ones (maybe except for navigator, i think), and then we should discuss what we really want for plugins, and whether we just want those to be part of the HXT or used to improve the existing pieces of HXT, OR if we want to have a store which carries you off to optional plugins.
Iām thinking things like
- bnGuides (draws guide lines)
- levureBakersAssistant (instead, update projectBrowser?)
- revApplicationOverview (instead, update projectBrowser?)
- tmAlign3 (updated alignment tools?)
A while back I open-sourced my Tools palette PowerTools with an MIT license, so feel free to incorporate it if desired.
The documentation is still online at PowerTools
Same goes for PowerDebug.
Thank you for the kind offer. Will have to look into it after the next release.
The guides I have preferentially always used is from the plugin DevGuides (GitHub - DTByte-Ltd/DevGuides: Extension to help LiveCode developers to design the UI. Ā· GitHub). This comes with a simple installer stack and works extremely well in HyperXTalk, highly recommended.
On a related note , I would also recommend Richardās (@FourthWorld) Devolution plugin.
As it stands, for this to work you have to create a folder called āPluginsā at the same level as the āExtensionsā folder.
@FourthWorld : Iām guessing āPluginsā is hardcoded in 4wDevo, as when I put it in āExtensionsā it kept throwing up errors and wasnāt showing in the plugins menu. Itās always been a bit unclear why weād need to have to have both an extensions and plugin folder in āMy LiveCodeā folder in ~/Documents, and it is no surprise to see this has been carried over to HyperXTalk.
Would it be sensible to rationalise this to a single folder?
I emailed them to see if we can include DevGuides directly into HyperXTalk.
The folders are used for different purposes. The āExtensionsā folder is for LCB extensions (aka widgets) and the āPluginsā folder is for code written in HyperXTalk
Itās MIT license - you can just use.
I prefer to get permission, I find it shows a level of respect 
Plus they might be interested in HyperXTalk too so good to get the word out.
3 Likes
I havenāt heard back from the guys who DevGuides. If I donāt hear back from them after the weekend, should we just go ahead and use it? As Stam pointed out, it is MIT license.