home *** CD-ROM | disk | FTP | other *** search
- From: Mark.Baker@mettav.exnet.com (Mark Baker)
- Date: 29 Jul 94 10:36:10
- Message-Id: <UUCP.775504910@mettav>
- Subject: Re: Digest
- To: gem-list@world.std.com (gem-list@world.std.com)
- Precedence: bulk
-
- Ofir:
- >> Looks like Ofir started a trend :)
- >
- > What trend?
-
- Sending blank messages. There were two more after yours :)
-
- > Unless Atari release AES 4.1 to the public and continue development, MTOS
- > is a waste of time. I'd love to multi-task on the Falcon, but MTOS is not
- > reliable enough and is still very slow at disk access (I have the latest
-
- The main problem is that it still uses the bios routines for disc access, which
- are blocking. There is a driver for the falcon's SCSI interface, and someone
- got minixfs to work with it to give non-blocking disc i/o. Really we need more
- device drivers and versions of minixfs and tosfs that use them, but this is
- probably quite a long way off.
-
- Timothy:
- > Where/When can I get the new version of MultiTOS?
-
- You can't. Definitely not legally and I haven't seen illicit copies on any
- bulletin boards.
-
- Evan:
- > No, the code is not correct. The assembly binding for Frename pushes
- > a 0 on the stack instead of 86. I believe this will terminate your
- > program as opcode 0 is Pterm0() isn't it? Yep, it is.
-
- Lots of other bindings are wrong, particularly vdi ones. And nowhere does it
- mention that returns from gemdos, bios and xbios functions are in d0.
-
- > There are numerous others. Ahh .. wind_set does not have the iconify
- > options mentioned at all. wind_get has them listed, but not wind_set.
-
- They are mentioned in mine, which is the new version.
-
- > There are other problems. You'd think after all this time that the
- > idea of waiting for reverse mouse button event would be documented.
- > evnt_multi barely mentions anything, other than see evnt_button,
- > and evnt_button doesn't mention reverse events.
-
- It mentions it now, but it's wrong! It says that it detects them
- independently, without mentioning the reversing of the state. And it ors 0x100
- to the mask not the number of clicks.
-
- > I bought mine used. I better find out what in in this upgrade :-(
-
- Lots of errors are fixed but there are still a _lot_ in it. Extra material
- -programming .xdd and .xfs drivers for MiNT, the xbra protocol, and extended
- style guide and memory map.
-
- Warwick:
- > Not just keyboard shortcuts, but general application defaults.
-
- Yes.
-
- > Application-specific and `global' setting all in the one file.
-
- No. Look at the windows win.ini file, which has application specific stuff.
- Every time you install an application it screws around with it and deinstalling
- an application is difficult as you don't know what it has changed. As a result
- new programs use separate settings files.
-
- > Format is "attribute_pattern: value", ie:
-
- Fine.
-
- > An application, say `MyEdit', would search for matches to the attributes
- > it is interested in. For example, it would look for:
- >
- > WordProcessor.MyEdit.FontSize
-
- As I said I'm not happy with application specific stuff being in here.
-
- > A set of common attributes and application types needs to be decided.
- > From there, we can create a APP_DEFS.SYS file that conforms to the
- > shortcuts standard.
-
- Yes. I don't object to application types, as word processors do need different
- shortcuts to drawing programs for example.
-
- > Multiple formats of the `value' should be supported. For example, for
- > key shortcuts, a standard notation like "^O" should be used by default
- > app files, etc., but the user should be able to have a scancode if they
-
- Yes, I prefer this to what someone suggested putting the ascii code or the
- scancode here.
-
- > have some key on there keyboard that is not one of the one available
- > on all keyboards. They'll have to provide a name for it in that case -
- > so it can be described in a menu item. eg. shift-< and shift->, Alt-
- > number-pad-digit.
-
- You can't have alt-num pad anyway as it's used for entering characters by their
- ascii code.
-
- > Q: When windows are non-overlapping, how do you know where
- > your keypresses are going?
- > A: The top window is highlighted differently.
-
- No, because you topped it.
-
- > Q: So you can tell the focus by the window decorations?
- > A: Yes.
- > Q: So if I changed the appearance of the window with the focus,
- > you would have exactly that.
- >
- > This `change the appearance' could be changing the appearance of
- > the cursor (ala TOSWIN and XTERM), and/or it could be changing the
- > window colours. On AES >=4.1 and under WINX, this can be done
-
- AES >= 3.30 actually.
-
- >> cut and paste from text fields
- >
- > This I really miss on my machine at home, after using X11 all day at
- > work.
-
- LTMF does it in modal dialgues, and also any non-modal dialogues that use
- form_keybd/objc_edit. Of course if you use your own editable fields and they
- don't support it you can't really complain.
-
- Timothy:
- > I see no point in absolutely abandoning the GEM style. It makes sense to
- > make some modifications, yet some of the things you people are talking
- > about like sending key-presses to a background window are not only hard
- > to implement and counter-intuitive, but possibly DANGEROUS. I will not
- > send keypresses to a background window, and unless it's absolutely
- > necessary, I don't see any point in sending mouse-clicks to a background
- > window either. It's just not GEM and it will only confuse and frustrate
- > people.
-
- I agree with you entirely here.
-
- > I am writing a window library. At the present it is about 20k. Already,
- > it cuts userinterface development for me into a small fraction of the
- > time it took before. It sets things up for me, does amodal dialogs, and
- > handles and directs window events automatically for me. It also makes
- > any application that runs under it a little more object oriented,
- > treating each window as a seperate object and giving the application
- > simple means to handle data independantly for each window.
- >
- > With that, I have to ask what is in some of these other libraries that
- > take up over 200k? Loads and loads of more features.... most of which
- > someone looking to get work done would never use.
-
- Firstly for particular types of windows such as images or text they handle it
- all for you. Secondly the dialogue handling includes progdefs to add more
- features such as CUA-style toggles and radio buttons.
-
- And of course if you don't use the features your program won't grow that big.
-
- Vincent:
- > In your button code (very useful, thanks), you set the high byte of
- > the second parameter. Is it a GEM feature or an accident? (The Atari
- > Compendium doesn't say anything about it, nor the other books I have.)
-
- That was the whole point of his code, to demonstrate an undocumented feature in
- the aes, namely that ORing 0x100 with the number of clicks will cause it to
- exit when the specified state is _not_ used, so you make the mask 0x03 and the
- state 0x00 then it will wait for the state not to be both up, ie either button
- down.
-
- This is not in the first edition of TAC, it is documented totally wrongly in
- the second.
-
- David:
- > For keys, I think we need to formalise as many keys as are used in a wide
- > number of applications. This means all of the keys that Atari/GemList
- > specify, plus a few more if we can think of them.
-
- All keys that will appear in more than one application doing the same thing,
- should be in it.
-
- > Other functions I think we need names for are:
- >
- > Dialogs in Windows (on/off)
- > Novice User Help (on/off)
- > Confirm on Exit/Save on Exit
- > Auto Save Prefs on Exit
- > Default Window Size (Lattice C really annoys me when every new window is
- > full screen!)
-
- I agree with all of these apart from the last, as this surely depends on what
- application it is.
-
- Warwick:
- > I know what you mean - I often double-click on a field, hoping it will
- > give me a file selector.
- >
- > It's not really a problem: older applications will still have the
- > hidden feature, and newer/updated ones could have an icon that clearly
- > indicates that double-clicking is not how to access the feature. A
- > standard icon could be chosen from the characters set. ('*' in BOXCHAR?)
-
- A problem with that is that if you are used to new applications which have an
- icon you will not expect the file selector to appear on older programs.
- Actually a lot of programs only use single clicks for the file selector ATM.
-
- Dan/Ken:
- >> I see no point in absolutely abandoning the GEM style. It makes sense
- >> to make some modifications, yet some of the things you people are
- >> talking about like sending key-presses to a background window are not only
- >> hard to implement and counter-intuitive, but possibly DANGEROUS. I will
- >> not send keypresses to a background window, and unless it's absolutely
- >
- > Nobody should *ever* send keypresses to anything but the window under
- > current focus. It is *incredibly* confusing to do otherwise. Just flip
- > this option on in some X11 window manager and you'll learn really damn
- > quick to hate it (as well as auto-window topping.. this is a totally
- > STUPID idea!)
-
- I never thought I'd say this, but I actually agree with you here :)
-
- > By the way, I'm not sure what you mean by 'abandoning the GEM style'.
- > There really is none! There is almost no consistency in GEM user interfaces,
- > the way things work under different versions of TOS, etc.
-
- But there is a standard for what windows look like and what the gadgets do. Yes
- it depends on TOS version but people only use one TOS version and with that all
- their applications _except_those_written_with_your_library_ will look and feel
- the same.
-
-
-
-