It's a bit like PNGs back when all we used in Windows were ICO files.

Anyway, also added the icon justification option I 100% know someone would eventually ask for later when "extend full screen width" is enabled for a dock. It's only 2 more lines of code and might as well add that option now

Small but MAJOR change I need to do:

Many, many, eons ago, the "Apply" button in Preferences was actually called "Preview". There was also a "Restore" button.

In the new GUI the "Restore" button was removed and the "Preview" button, For some reason I can no longer remember, was renamed to "Apply".


Here is the problem: what the Apply button in Preferences does is not what a user expects an Apply button would do (i.e. apply changes permanently but keep dialog open), it is still what a Preview button would do: it applies the changes, and you can then make them permanent by clicking OK or revert them by clicking Cancel.

I've also seen many users thinking they need to click Apply before clicking Ok to close the dialog, which is just a waste of clicks.

So, I'm thinking I should rename the Apply button back to Preview. This would take care of actual meaning.

However, the ideal button order according to ChatGPT then becomes "Exit, Help, Ok, Cancel, Preview" so I should change the order of the buttons too (then again, CoPilot does not agree lol). I hesitate to do this, however, because of muscle memory. For decades existing users have gotten used to click the last button to cancel changes (and the button before it to apply changes and close the dialog). Clicking Preview by mistake when you intended to click Cancel is not a biggie, but clicking Cancel by mistake when you intended to click Ok, is.

So, I'm facing three choices:

1. Keep everything as is, i.e. keep the Apply name
2. Rename Apply to Preview but keep the button in the same place
3. Rename Apply to Preview and move it to after the Cancel button.

What do you guys think? Need some opinions ASAP please.

PNGs are perfect for online use - and more besides - but WebP don't really add anything, as far as I can see.

Yeah, that makes sense in that context Jorge - absurd though the whole "extend full screen width" thing is for a dock.

Chat GPT and Copilot are both imbecilic programmes and only interested in harvesting your data/chats. Changing the order is just urinating in the wind.

I'd suggest rename to Preview and keep it where it is, simple.

Ahahhah I agree...

By the way, you are not going to believe how good the User Guide the AI is making for WorShelf is (with my help, of course, we are already at revision 17).

Take a look at THIS (still work in progress, of course).

Guys, please take a quick look at the updated User Guide in the link above and let me know if you spot anything wrong or out of place. Opinions are also very much welcome.

EDIT: Never mind, going through it myself already found (and fixed) a few problematic sections. Still at less than 1/3 though, this is going to be a lot of work.

On the other hand, perhaps v26.6 will be released finally featuring a complete and up-to-date User Guide.

Ladies and gentleman, I give you....

The Winstep User Guide!!!

Just kidding, still working on it. Despite the AI doing a huge amount of leg work, revising it, fixing mistakes and adding missing info has been an equally monumental task.

Also, don't know which screenshot I like more for the User Guide section in the v26.6 News announcement page.

I prefer the first one.

Argh Jorge, you're selling your soul - and all your data - to those AIs! You just don't know where all that could end up, or rather, will end up. Why not give the free Proton Lumo a try? Totally private, all E2EE zero access, no logging, nothing. The paid version of course is even better.

Sincerely, for this type of data I do not care one bit. It is pretty much public data already, anyway.

I too. Cleaner, more modern-ish look. 2nd one seems to have almost quasi-religious overtones.

I did start on checking over the text too but as you said you'd revised it already any way Jorge, I wasn't really sure if my comments would still have been accurate/relevant so gave up. There was a heck of a lot where the AI had screwed up, adding in redundant words etc., so would have made for a very long post!

Ouch! Surely user-guide should be copyright material. But apart from that, there are all your other data that the damn thing harvests!

Ahah, yes, you do have a point and I too prefer the first one.

You guys can already take a look at what the announcement page will look like if you want:

https://winstep.net/news/winstep_xtreme ... leased.asp



Yes, that's why I edited the post and why I am STILL knee deep in revising and adding to the User Guide. The AI did the best it could with the knowledge it had and sincerely I am surprised it actually managed to do as much as it did.

When I asked you guys to take a look, I was already exhausted after twenty‑plus revisions. I hadn’t yet gathered the strength to go through it again, paragraph by paragraph, and could only hope the AI had finally finished the job. Unfortunately, that wasn’t the case - and it became obvious the moment I woke up and started reviewing it.



I am aware of that and this is the type of data I truly do not care it harvests.

In my case and the way I use it, the benefits of using AI FAR, FAR, OUTWEIGH the costs, or any privacy concerns I might have (and if one has privacy concerns, we already sold our soul to the devil - Microsoft - just by running Windows).

Surely you must have noticed I have never been this productive before. This is because I use the AI like any other tool, indeed much like the "idiot savant" junior assistant I never had.

Pretty much in the same way we have been using Google and other search engines for years to search for the information we want and filter out the rest. This is where the AI truly excels, and is a thousand times better than any search engine.

It cannot do what I do, but it can certainly point me in the right direction most of the time. And, like a real human assistant, I can burden it with tasks that would take me AGES to do - if I ever even got the courage to tackle them in the first place.

How long do you think it would have taken me to get latitude and longitude coordinates for over 70,000 cities, match them all one by one and verify that they are indeed correct, for instance, even if I had made a script for that? I would never even have STARTED to do such a thing, as my time is valuable and best spent on other things.

Likewise for the User Guide. It is already such a tedious and mammoth task going over what the AI did (and it took 20 supervised revisions to get something I could finally begin to go over) that I probably would have already given up otherwise.

With the AI I am no longer afraid to tackle issues I know nothing or very little about - I know it will help me zero in on the stuff that really matters, and on how to get the job done. Adding Dark Mode to the forums is another example - not only do I not know PHP, as I would have been too afraid to break the forums to even attempt to tackle it on my own.

And graphics? Don’t even get me started. ‘AI slop’ or not, it still beats anything I could have produced on my own - I’m a developer, not a graphics artist - especially at the speed it does it, and it’s the kind of work I would have absolutely dreaded doing. That’s assuming I would, or even could, have done it at all.

Properly directed and used, the AI is like any other tool, or more like an assistant who works for me and allows me to unburden the brunt work onto him, brainstorm ideas, do research and graphics, etc. In pretty much the same way computers revolutionized the speed calculations are made by several order of magnitudes, properly used AI can do the same for productivity in certain aspects.

The actual creativity and final implementation, however, still comes from me. The AI by itself can do NOTHING.

Of course, besides the obvious privacy issues, there are some VERY REAL CONCERNS about the risks of using AI for everything (for instance, humans becoming so reliant in AI that they stop making decisions for themselves, becoming the perfect slave so many tyrants can only dream of).

However, this is neither the place nor the time to discuss those concerns. I will just finish by saying that like any other tool, AI can be used for good or for evil. The use I am making of it is benefiting me and every Winstep user in a way that was simply IMPOSSIBLE a handful of years ago.

Announcement page looks good Jorge.

Re: the UG text, there were an awfull lot of "fill words" and the like the AI had put in, also things like qualifying non-qualifiables e.g. unique - something either is or is not unique, it can't be qualified with totally, almost, most, etc. etc.

As for "the devil" MS, in the old days up till XP, there was comparatively little data harvesting by them, and they certainly didn't get much from me and certainly nothing that would have been personally identifiable. (I already had stuff that stopped a lot of things even then and never used real IDs etc.) Post-XP times I already was using Linux a fair lot also, and as soon as things like VPNs and the TOR network came along I was using those already. Never used Google search engine - there were others, that were more private and secure even back in XP days, late anyway.

I'm not disputing your use of AI to increase your productivity Jorge, far from it. I use AI quite a lot myself - e.g., in learning to code in new languages when I can find the time and brain power It's also helping me rediscover at least a bit of the classical Latin I learned at school and the Island or Church Latin I learned later - I might even get far enough eventually to re-read Dante Alighieri's Divine Comedy - better known for one of the chapters, 'The Inferno' - etc. in the original, what a joy that would be! And it's helping me with all kinds of other tasks - even the book I'm ghostwriting!

But here's the difference. The AI I'm using is entirely, 100% private, with no snooping, no logging, everything E2EE with zero access - only *I* have the keys to decrypt - and I could even use it anonymously if I want. Any tasks/conversations that I save are saved - encrypted of course - to my local or cloud drive (also E2EE etc.) It is AI as it should be - private and secure. That's Proton's Lumo, and it even will only remember what you want it to remember, but no-one else can access that except I and it when working. And of course it's open source and regularly independently audited, as are all Proton's privacy apps. Everything including Lumo has a free version, and even that is pretty good. That's also useful as a basic "test drive" if one needed features of the premium version.

You really ought to have a look/go at it, Jorge - I think you'd be pleasantly surprised.