Re: Features from other IDEs and general usability (new features, borrowed features, and general changes)

Wade Chandler <[email protected]> Fri, 2 Nov 2007 09:05:28 -0700 (PDT)
Newsgroups gmane.comp.java.netbeans.user-interface
Message-ID <[email protected]>
Roman Strobl sent me this great blog post he wrote in 2005:
http://blogs.sun.com/roumen/entry/how_we_do_usability_studies

It is exactly what I envisioned, and hopefully it is still being employed, so it made me re-evaluate my statement a bit, and I replied to him ( have added more here though ):

I suppose the other answer the question and IZ issue
 I used as an example is real world usage over the course of some days
 then. I found in the issue I was working that it was very painful to
 keep going back and forth with the mouse. But, if I hadn't had a project I
 needed to do such a thing (running multiple configurations often) I would not have noticed how annoying the
 feature really was. The scenarios are probably what needs changed and not
 the method. 

I wouldn't even say changed, but instead would say more
 are needed, though the methodology used is probably not giving the user any input what so ever as to get their true opinion and usability issues. So, I suppose re-evaluating my last statement is also needed. It seems a two tiered user testing is needed in the lab. One with the normal and proven method of user testing. Another with scenarios where users are asked to perform certain tasks and to see which tasks they repeat most often and what comments they come up with.

Wade


----- Original Message ----
From: Anuradha G <[email protected]>
To: [email protected]
Sent: Friday, November 2, 2007 9:05:00 AM
Subject: Re: [nbui] Features from other IDEs and general usability (new features, borrowed features, and general changes)


Agree with you I have remove configuration  combo from tool bar  coz  I'm using tool bar as  quick actions that I regular  use.Changing configuration  is not a regular tack for me so I'm happy without combo BUT I prefer split button to change run  configuration  


On 11/2/07, Wade Chandler <[email protected]> wrote:
This isn't a rant, but just as I see some things which need to be talked about. I have seen others in the community mention such things. I don't want to single anything out, but I have picked an issue as an illustration.


See:
http://www.netbeans.org/issues/show_bug.cgi?id=105664

for some context.

I want to touch on usability as we have already mentioned Eclipse and IDEA in the issue above. No matter what any other IDE or project does we need to have our own identity and it needs to be designed, IMO of course, for ease of use and practicality. We can adopt features from other IDEs, but we should always try to make sure the implementation works into our framework and is very user friendly. Provide features yet innovate; we should strive to make it better if it can be.


I don't think we need to make UI choices in a vacuum, meaning just among a UI or dev group, and I'm not saying this is 100% what happens either, so we need to make sure we have community feedback and user studies if we could pull them off would be good. There is a book called, "Don't make me think", which is more web interface directed, but the premise applies, and it contains cost effective ways of implementing user studies.


It would be nice if someone could implement such things at NB days. We could have a booth with the IDE setup and a few sample projects for people wanted to volunteer to work through.  We would have to ask them if they are a long time NB user, from another IDE or not, or new to NB as the different contexts will matter (maybe a change was made to the IDE based on some other IDE...maybe it is a new feature..and if they never have used an IDE any feature goes). There could be a tutorial setup for which ever sample they're asked to walk through. Then have a camera so their reactions can be recorded, and there is software, as mentioned in the book, which can capture the screen activity. Then the users activity, facial expressions, and comments are all captured in place. Then everyone knows what is working and what isn't and has a better idea of the usability of different parts of the system.


Thanks,

Wade





-- 
Anuradha G