Liam wrote:

The single window mode has tabs, with a tab for each open image.

well, temporarily. there are no tabs in the design. it will not ship like this.

but it is good to find out that the trivial "just put some tabs like firefox" solution shows its weaknesses.

--ps

founder + principal interaction architect man + machine interface works

http://mmiworks.net/blog : on interaction architecture

On Fri, Apr 30, 2010 at 12:26 AM, Liam R E Quin wrote:

(1) there should be a View->show/hide image tabs option

This applies to the final image strip design aswell.

(2) probably the tabs should not be shown in full-screed mode, although I suppose one could have a preference for it (sigh)

I thing tabs/image bar should autohide and a have a tiny control you can click to show / hide it. It makes more sense with bar style control.

(3) it feels very strange to close a tab by trying to quit the application (pressing the [x] on the title bar). I think a pop-up context menu on the tab with Switch to

This is handled by single clic, no need to duplicate

Close

This is really missing. Firefox has close buttons on tabs. I think gimp should have the same.

Detach (or, Move to Window, with "(new window)" as an option)

This is not an option by design. There are problems we can not be solved sanely with multiple swm windows, so there can be just one swm main window.

On Fri, Apr 30, 2010 at 2:15 AM, peter sikking wrote:

well, temporarily. there are no tabs in the design.

No, theres an image bar, with pretty much identical function.

it will not ship like this.

I have problem with this attitude. Its not how open-source works. If its stable, you release it and then keep adding design and features in the next cycle. 2.8 has already taken too long. People who shoudn't be building Gimp from git are doing it. Heaping on "does not match design goals exactly" style roadblocks does more harm than helps. If it fills the basic requirements and is stable, its not a release roadblock. Id like to hear other developers opinions on this.

but it is good to find out that the trivial "just put some tabs like firefox" solution shows its weaknesses.

Firefox tabs have close buttons for some time. This is just something gimp lacks at the moment.

Hi Alexia

it will not ship like this.

I have problem with this attitude. Its not how open-source works. If its stable, you release it and then keep adding design and features in the next cycle. 2.8 has already taken too long. People who shoudn't be building Gimp from git are doing it. Heaping on "does not match design goals exactly" style roadblocks does more harm than helps. If it fills the basic requirements and is stable, its not a release roadblock. Id like to hear other developers opinions on this.

I'm a big fan of open source, but I am also a big fan of interaction design and usability. For me it is not only software stability that is important, I want things to be pleasant to use as well.

Once 2.8 is out I want to focus on GEGL integration. I won't be able to do that if we have a half-baked implementation of the single-window mode UI overhaul, so I support guiguru's wish to do what it takes to fix e.g. the problems the current tabs has before we release 2.8.

Yes, development of 2.8 has taken a long time, but it has also turned into a remarkable release. I expect 2.8 to be the stable version for a long time, so we should give it the development time it deserves.

Regards, Martin

On Fri, Apr 30, 2010 at 10:38 AM, Martin Nordholts wrote:

I'm a big fan of open source, but I am also a big fan of interaction design and usability. For me it is not only software stability that is important, I want things to be pleasant to use as well.

This is understandable and I share the view.

Once 2.8 is out I want to focus on GEGL integration. I won't be able to do that if we have a half-baked implementation of the single-window mode UI overhaul, so I support guiguru's wish to do what it takes to fix e.g. the problems the current tabs has before we release 2.8.

I don't have an issue with fixing any problems, ie the lack of an obvious way to close tabs. What I'm opposed to is saying that as long as we don't have a full image parade we aren't shipping. On the other hand, since you are the developer for those features, it is your call do decide when its good for 2.8 and if you do have resources to implement full image parade.

On Fri, 30 Apr 2010 09:25:05 +0300, Alexia Death wrote:

well, temporarily. there are no tabs in the design.

No, theres an image bar, with pretty much identical function.

Just a look at other's ideas – for example Opera browser (Windows .avi file, 21,4M):
http://www.secarica.ro/misc/opera_tabs_behaviour_ro.avi

Cristi

On 4/30/10, peter sikking wrote:

but it is good to find out that the trivial "just put some tabs like firefox" solution shows its weaknesses.

It can't show that because tabs in GIMP are not like tabs in Firefox :)

- images vs text, therefore too much space wasted - no "close" buttons per tab
- when too many images are opened, previews in tab captions disappear beyond horizon

I don't know what the final design is, but if those points were fixed, Liam probably wouldn't have reasons to write that mail :)

Alexandre

Martin wrote,

it will not ship like this.

I have problem with this attitude. Its not how open-source works. If its stable, you release it and then keep adding design and features in
the next cycle. 2.8 has already taken too long. People who shoudn't be
building Gimp from git are doing it. Heaping on "does not match design
goals exactly" style roadblocks does more harm than helps. If it fills
the basic requirements and is stable, its not a release roadblock. Id like to hear other developers opinions on this.

I'm a big fan of open source, but I am also a big fan of interaction design and usability. For me it is not only software stability that is important, I want things to be pleasant to use as well.

well I am happy with Martin's support, because without that there is no way to achieve results.

what Alexia describes is not typical open source, it is how at the moment the whole (commercial) software world works:

"when the going gets tough, usability becomes a 'nice to have'"

what is counterproductive is that right at that point the management and development people stop working with the interaction design and usability people and start 'going it alone.'

every time a developer stops working with an interaction designer on some project, a little bit this developer stops working with this interaction designer forever. this can only be repeated so many time (a cat has 7 lives) before this interaction designer stops working with this developer forever.

so the solution is: "when the going gets tough, you keep working with your interaction designer to achieve good usability"

I pride myself on being an interaction architect who design software with good usability and who works with engineers on getting it built in a practical way. I would prefer to do both of these (including the final call on what is usable) up to the last minute before releasing.

--ps

founder + principal interaction architect man + machine interface works

http://mmiworks.net/blog : on interaction architecture