Mailing List lswitcher-dev@2rosenthals.com Archived Message #96

From: "Lewis" <lswitcher-dev@2rosenthals.com> Full Headers
Undecoded message
Subject: Re: [lswitcher-dev] lSwitcher-2-92-0-RC_12.wpi
Date: Wed, 30 Sep 2020 21:44:30 -0400
To: lSwitcher Developers Mailing List <lswitcher-dev@2rosenthals.com>

Hi...

On 09/30/20 07:17 pm, Alfredo Fernández Díaz wrote:
Hi Gregg,

On 20/09/27 20:33, Gregg Young wrote:
On Sun, 27 Sep 2020 13:44:59 +0200 Alfredo Fernández Díaz wrote:
...
You have a plan for forcing the update of XWP on all systems and
preventing the use of Warp Center which we can't fix and you know
for certain these are the only programs that use this?

I always have a plan. Phase I is getting to know how all of this works
without looking at the code. Thanks for your help ;)

Hi Alfred

The real solution would be modifying  WinSetDesktopWorkArea in PMMERGE  to
check to see if the desktop was already reduced and return an error  on
any subsequent call except one to restore the desktop to its original
size. Of course this isn't going to happen.

I certainly don't count on it happening, but it sounds like a pretty well known thing. Hm.


Indeed. It's just another of those well known PMMERGE quirks (I say "quirks" vs "bugs," as there is little to no chance of ever seeing such things fixed). Of course, there's the DOT patch, so who knows?

We need to remember this is an undocumented call and IBM never intended
for others to use it. As such it was not designed for that possibility.
Just like it wasn't designed to be called with a less than full width
window over the reduced area.

I see.

For completeness of your planning I can
probably hide the stupid desktop titlebar.

Yes, that's probably better.


But I like to know what that "big window" is consuming my screen... :-)

I imagine the lack of a WarpCenter object in ArcaOS kind of prevents
its use. If the people at ArcaNoae were really determined to do that
I imagine they could also prevent the SmartCenter class from being
registered in the first place in future versions. That would be an even
safer way to be sure nothing will interfere with what's not even there.
And I imagine those people would want as many customers as possible to
be reasonably updated. Seriously, how many people are left outside that
lot (besides me ;)?

Probably a lot; OS/2 users aren't exactly a progressive bunch. I know one
that is still using a prehistoric version of lSwitcher;-)

You think that's prehistoric? You should see the WordPerfect macros I'm writing lately ;p


:-)

Point taken, though (listening to Lewis' input as well -- I used the warp 3 launcher on Warp 4 for ages, hehe).


As OS/2 users, we value our freedom of choice as opposed to whatever the software publishers *think* is best for us. The LaunchPad caught on quickly for many of us, and served us well for some time. Heck, it took me a long time to switch away from WarpCenter to XCenter.

Then it's all a matter of documenting things properly (do not do this
and that for such and such reason: just a twist on the current docs) and
wait for reports of the three people in the entire universe who still
see something break up in their systems.

Fortunately they never bother to report anything. Think my breaking of all
the non-English languages in 2.91 and never hearing anything about it.

Hey, *I* reported that. Maybe not in time for 2.91, but...

...
I'll have a look at that later. I've changed a couple of strings to stay
closer to the current English menu (lSwitcher settings -> Properties,
and help mnemonics).

Am I to read this that I need to change the hotkeys again. Remember the
mnemonics and hotkeys must match.

As you probably have realized by now, no; I meant the mnemonics for "Help" on the taskbar menu. So those two letters change on that menu alone, and the rest is untouched.

Yeah, I've written better pieces. Have a look at the screenshot.
'Programs' is a WPS folder: you can make it 'Close' but not 'Quit'.
Maybe the 'Close' menu should be replaced by its first child item, which
is the only one that works in that situation.

I think Quit just needs to be removed it has no advantage over close and mostly doesn't work.

It seems to work here for *applications* (not WPS folders). I am not clear on the (dis)advantages when closing programs, though, but if there are none, why was it there in the first place?


I think Gregg's point is that Quit doesn't really work any better or worse than Close. IOW, I've never had a running process refuse to close but was then able to use Quit. It works, surely, but when it does, Close should also work, so there's no real advantage to having the extra option.

Having seen r13b, I think I concur with Lewis: I'd prefer a single 'Close' lSwitcher menu entry for folders in the taskbar (since only that makes sense), but for programs, and if possible, I'd rather have a 'Close' submenu with the different methods. I think Dos/Sigkill look a bit daunting, not being normal words and that, and it would be better if at least non-tech-savvy people can tell right away they're just modes of closing a task.


Exactly.

-Packaging: any particular reason why we keep old English readmes,
i.e.
readme.v*?

I always package the historic readmes on my first release. They can
probably go away now. Thanks

Don't forget for the GA - they're still in the rc13 wpi, I don't know whether intentionally.

Thank you for all you do ^^

Your welcome. Thanks for your help and translations.

:)


Cheers

--
Lewis


Subscribe: Feed, Digest, Index.
Unsubscribe
Mail to ListMaster