In the last couple of years I have participated in building quite a few web applications and we always have put a lot of emphasis in the user experience, navigation, in having pretty styled widgets, visual effects, smoth transitions, drag & drop, etc. Sometimes we used some out of the box styling or guidelines such as twitter bootstrap or google's material UI, others just bare javascript and css.
I think it makes a lot of sense to be careful about this stuff if you want that your users feel comfortable and want to use your application. Many times a user *buys* your software because of it looking *modern *and often it is better to have less features that are comfortable to use instead of lots of features but that are difficult to find or understand. I have learnt the importance of having a *good* web designer (or even better a UX expert). That said, and even when Pharo has evolved a lot in the last few years, I think that there is still room to improve and much to gain both in Morphic as in Spec or Glamour (I do not have experience with Bric widgets). I think it would be wonderful to be able to combine the power and simplicity of Smalltalk with a UI that allows the flexibility that I can achieve in web applications. It is not enough to say "it is possible", it has to be easy, understandable, robust, well documented. A little example: drag & drop has been possible in javascript + css since a lot of time ago, but 10 o 15 years ago, when someone asked me for dnd, I always tried to avoid it. "Yes I can do it, but it will take a lot of time, and it will never *feel *like the dnd you are used to in your host OS." Now tools evolved and my perspective changed a lot about this kind of *goodies*, I even empower my designers to include that kind of stuff, we can do it fast, it is easy, think about the best UX and we will figure out how to build it. I think that it would be great to achieve the same level of confidence in Pharo. On Sat, Jun 25, 2016 at 9:54 PM, Tommaso Dal Sasso < tommaso.dalsa...@gmail.com> wrote: > Just a couple of thoughts from the last days coding GUI in Pharo: > > I don't think that the Spec vs. Morphic vs. Bric has really to do with > "ugliness". > True, the Glamour team is doing an amazing work with providing an > interface that is more pleasing and friendly, but the point for me is not > how Spec or Morphic look. I don't use a library because its widgets look > cool, I use it because the tools I build with it fit in the system i.e., > they look like the rest of the system and respect its settings (provided > that borders, spacing and alignment of the labels are aligned). How the > system and your application look like, is responsibility of a theme manager. > > The beauty that I am looking for as a developer for has to do with the > code and the API. I consider important how fast I am able to prototype a > working example --in this Spec is excellent-- and how much I can extend the > code to refine my application --in this Spec is really hard to use--. > > These last days I spent a lot of time with Morphic and later with Bric. > What I noticed is that Morphic is *really* powerful, but to define a > minimal working example you have to write a lot of code to take care of the > details. On the other hand, Bric abstracts many of these details, allowing > for a more pleasant looking result in less time. > > I now understand why I hear saying a lot that Morphic got out of hand: you > have so many options that one is not not able anymore to discern the > essential ones from the once-in-a-lifetime settings. Since our main source > of documentation comes from inspecting the code, and iterate over a list of > methods requires a human at least linear time (who knows if we can improve > this), a class like Morph becomes a nightmare. > Tommaso > > > > On 25/06/16 16:45, Peter Uhnák wrote: > > > > On Sat, Jun 25, 2016 at 2:12 PM, Johan Fabry <jfa...@dcc.uchile.cl> wrote: > >> >> >> On Jun 24, 2016, at 19:00, Yuriy Tymchuk < <yuriy.tymc...@me.com> >> yuriy.tymc...@me.com> wrote: >> >> >> […] >> >> I know that we don’t have designers to inverted something cool, but what >> GT teem did is they took Google material design guidelines and implemented >> widgets in Pharo, so if I use them it will be pleasant to the user’s eye. >> Otherwise we are saying how cool Pharo is and how good we are, but a random >> Android developer can build nicer interface because they have this sexy >> widgets. >> >> >> Just to understand: are you saying that a better set of underlying >> widgets for Spec would solve the ‘ugliness’ issue for you? >> > > Then this would not a Spec issue, but Morphic/Bloc one. > > >> >> And yes, in the end for me the big turn off with Spec is the direct >> example with Tommaso. You know that something is wrong, so you want to >> change the size of a font. And then you ask on the mailing list and it >> turns out that it’s not that easy and you have to either hack something, or >> submit patch to Spec… In the end I’ve decided to use plain morphic because >> at least there I can change whatever I want. >> >> >> So, again to understand: you want to be able to change the font size of >> any widget from within the application, ignoring the System preference >> settings? >> > > You should be able to do that even now with #whenBuiltDo: > > But it's true that Spec doesn't really have any API for styling, and if > something is customizable (some colors for some widgets, some other > options), it is quite haphazard, similarly for drag and drop (there was a > brief discussion about this about a month ago, but we don't have a solution) > > >
