On Thu, Jun 16, 2011 at 8:37 AM, Walter Bender <[email protected]> wrote: > On Wed, Jun 15, 2011 at 2:56 PM, <[email protected]> wrote: >>>Walter Bender wrote: >>>The eventual transition to GNOME 3.0 and PYGI will make a big difference in >>> our ability to support more interoperability as well. >> ------------------------------------------------------------------------------------------------- >> >> >> This is a very good news, Thanks, Walter! >> >> >> --------------------------------------------------------------------------------------- >> >>>Walter Bender wrote: >>>In any case, concrete feedback and criticism is welcome. >> ------------------------------------------------------------------------------------------------- >> >> Concrete Feedback: >> >> The journal is always full of garbage, hundreds of empty entries: is there >> a possibility to "CLEAN" the JOurnal?, or much better, that the programs >> won't write empty entries? > > Quip: Thank goodness the only problem we have is Journal spam. > > Seriously, this has been an issue we have gone back and forth about > for quite some time and I am afraid that our attempts to find a > one-size-fits-all solution have not been successful. > > In the early days of Sugar, the default behavior upon launching an > activity from the Home View was to launch a new instance, which ended > up with a Journal entry associated with that activity. This was > seemingly fine for Write, but pretty annoying for Terminal or Tetris. > > We turned it around about two-years ago such that the default behavior > is now launch the most recent instance by default. This has in my > experience almost entirely eliminated Journal spam, but we have had > complaints that it is too difficult to launch a new instance (you need > to use the hover menu and scroll past the list of most recent > actions). I had proposed (but never wrote the patch) to make "new > instance" be at the top of the list rather than the bottom, which I > think help quite a bit. Still, the debate continues. > > We have discussed adding to the activity.info file an indication as to > whether or not an activity should launch a new instance by default, > but this idea has generally been disparaged as being too confusing: > different launch behaviors for different activities. > > [I am curious which launch behavior you are working with, as I don't > understand how launch-most-recent by default would result in Journal > spam.] > > ---- > > In a separate thread, Ana just proposed a few other ideas, which I > will discuss as part of that thread: > > http://lists.sugarlabs.org/archive/iaep/2011-June/013521.html > > thanks. > > -walter > >
If it seems I deliberately by-passed your suggestion below, that was not my intention. I do think it would be good to allow an activity to "abort" its Journal entry in some way, but I think that the bulk of the Journal spam problem is associated with launching issue. regards. -walter >> >> Empty entry = when a kid opens "Browse", and he closes 10 seconds later, it >> is useless that entry. >> >> If the kid uses "Write" for more than 15 minutes, probably he wrote >> something useful. >> >> For example , If the kid opens "Write" but he don't write anything, that >> entry would not be save on the journal. >> >> ANother example: Programs that are not well "sugarized", they save lots and >> lots of useless entries on the journal. >> >> The problem is not the space, those entries don't occupy nothing. The >> problem is the amount of entries. >> >> TOO MUCH INFORMATION = NO INFORMATION >> ------------------------------------------------------------------------------------ >> >> >> Paolo Benini >> Montevideo >> >> >> >> >> _______________________________________________ >> IAEP -- It's An Education Project (not a laptop project!) >> [email protected] >> http://lists.sugarlabs.org/listinfo/iaep >> > > > > -- > Walter Bender > Sugar Labs > http://www.sugarlabs.org > -- Walter Bender Sugar Labs http://www.sugarlabs.org _______________________________________________ IAEP -- It's An Education Project (not a laptop project!) [email protected] http://lists.sugarlabs.org/listinfo/iaep
