Rasmus Fogh wrote:
>Dear Gary,
>
>Going through the process, I can see one source of confusion:
>
>When the peak list is small you are asked to select a target and then you
>select it. When it is large, you are asked if you really mean it before
>you select the target list - that means that you have time to forget what
>you were asked to do ;-).
>
well thats some of the problem but quite all of the problem. One of
the problems was that the clone peaks button changes to 'stop cloning'
before the cloning started which is quite confusing as you aren't
cloning at that point... However, any other message I can think of is
too long
> It might be better if the 'do you really mean
>it' popup came up either before you were asked for the target, or after
>you had selected the target. Do you think that might help, and would you
>have a preference?
>
>
I think the 'do you really mean it' popup should come up as the last
thing before the actual clone operation starts...
>It is true that you will get problems if you clone a few thousand peaks
>into the wrong target list. But then, any command that makes large changes
>to a data set will cause trouble if you do it wrong (just think of
>'\rm -rf *'). It is hard to see how the program can protect the users from
>this eventuality.
>
>
Indeed without undo (which is a pianful problem to solve) the whole
things will always be dangerrous. One of the problem is that most
dangerous commands that are started by a mouse click typically work
along the following lines in other guis: select something (eg a cell in
a list and then press a button). Now this has some quite subtle human
interface twist in it as it is both an example of fitts law
(http://en.wikipedia.org/wiki/Fitts'_law) and what might be called the
'draggable forgiveness' of buttons.
1. the ease and accuracy of acquiring the list item with the mose is
quite low as it is quite a small area
2. however the ease and accuracy of acquring the button is much higher
as it is generally bigger
3. also the button is a more forgiving device thane the peak list table
as the peak list table selects immediately you click it and then starts
cloning whereas you can click on a button think abouyt the choise and
slide the mouse off it again if you have second thoughts
my ideal design (it may not be everyones favourite design and requires
time to impliment which you mayn not have) would be the following
1. select a peak list in the table
2. click clone
3. a modal dialog comes up with:
a. some text 'clone xxxx peaks from list xxx'
b. a second peak list table with suitable entries to clone to selected
c. two butons 'start clone' cancel
one other thought about this is that the progress dialog box would
really benefit from a cancel button...
regards
gary
>Yours,
>
>Rasmus
>
>---------------------------------------------------------------------------
>Dr. Rasmus H. Fogh Email: [log in to unmask]
>Dept. of Biochemistry, University of Cambridge,
>80 Tennis Court Road, Cambridge CB2 1GA, UK. FAX (01223)766002
>
>On Thu, 19 Jul 2007, gary thompson wrote:
>
>
>
>>Dear All
>>
>>
>>>Dear All
>>>I am using the current peak lists to clone peaks from one list into
>>>
>>>
>>another. However, the clone process appears to be hanging (the whole
>>python system is not hung but cancel >cloning comes up on the button and
>>then nothing seems to happen) help!
>>
>>
>>>another couple of comments / questions
>>>
>>>0. the ui is a bit painful it took several attempts to work out what was
>>>
>>>
>>going on, what with semi modal dialog boxes appearing saying that I had not
>>selected anything etc a small >dialog box with two lists and start
>>cloning/cancel button would seem more obvious to me at least...
>>
>>
>>>1. as the clone operation happens in the background do I have to be careful
>>>
>>>
>>not to change the peak list i am cloning from e.g. what would happen if I
>>deleted it....
>>
>>
>>>regards
>>>gary
>>>
>>>
>>Now I have found out what is going on and it was a human interface problem
>>
>>for the record you have to do the following
>>
>>1. click on a list to clone
>>2. click clone peaks
>>3. a dialog comes up saying 'please select a destination' [OK]
>>4. you click ok
>>5. if you have lots of peaks a second dialog comes up saying copy will
>>duplicate xxxx peaks. this may take some time. [OK] [cancel]
>>6. clone peaks changes to a blue button with 'cancel cloning' (note you
>>haven't started cloning yet though ...)
>>7. then you have to pick a peak list from original peak list [woe betide you
>>if you click badly and get the wrong peak list... you will trash it and have
>>to start again] or pick 'cancel cloning' button
>>
>>sorry the developers may shoot me, but this design is just not very
>>unintuitive
>>
>>
>>regards
>>gary
>>
>>(n.b. it has taken me about 3 - 4 hours to work out where I was going wrong
>>here)
>>
>>
>>
>
>.
>
>
>
--
-------------------------------------------------------------------
Dr Gary Thompson
Astbury Centre for Structural Molecular Biology,
University of Leeds, Astbury Building,
Leeds, LS2 9JT, West-Yorkshire, UK Tel. +44-113-3433024
email: [log in to unmask] Fax +44-113-2331407
-------------------------------------------------------------------
|