Million sorries, Kristen, it was just a mistyping...:-)

All the Best,
Kristof

--- In reconstruct_users@yahoogroups.com, "Kristen M. Harris" <kharris@...> wrote:
>
> Hi Kristof,
>
>
>
> Kristen Harris is a 'she'.. J
>
> Thanks
>
> Kristen
>
>
>
>
>
>
>
> From: reconstruct_users@yahoogroups.com
> [mailto:reconstruct_users@yahoogroups.com] On Behalf Of kristofgiber
> Sent: Thursday, July 09, 2009 3:34 AM
> To: reconstruct_users@yahoogroups.com
> Subject: [Reconstruct] Re: Reconstruct Features and Bugs
>
>
>
>
>
>
>
>
> Hi Larry,
>
> Although Kristen told me he would forward my problem and suggestion to you
> regarding Boissonnat surface generation, I repeate it to you by copying my
> mail below. Your help in this issue would be more than useful.
>
> > The visible size of the objects are reduced when Boissonnat surface is
> applied: The same endings (ie., the endings facing towards the first section
> of the serious) of all objects is shrinked by the width of one section. This
> results in 1-section-wide spaces between objects which otherwise touch each
> other, and leads to the inproportional reduction of small objects (such as
> synapses which can become even half their original size) in comparison with
> large objects (such as axonal terminals or dendritic segments). To fix this
> problem, Reconstruct TM should have a user option (a checkbox or whatever)
> to decide whether
> 1) I would like the Boissonnat surface to be generated between the same side
> (or the central line) of all sections (as it is now), or
> 2) I would like the surfacing procedure eg. as follows (my suggestion): the
> central lines of each sections which have mutually both neighbouring
> sections containing traces of the object are matched with each other, while
> the side of the first and last section to be matched with the midline of
> their only neighbouring section containing a trace of the object is the side
> that is further from the rest of the object.
> The latter surfacing option would lead to a symmetric and faithful 3D
> generation of the reconstructed objects. Currently, there is not any options
> to fix the mentioned inproportional size reduction.
>
> Thanks
> Kristof
>
> --- In reconstruct_users@yahoogroups.com
> <mailto:reconstruct_users%40yahoogroups.com> , "larryflindsey"
> <larryflindsey@> wrote:
> >
> > Hello Reconstruct Users,
> >
> > My name is Larry Lindsey, and I have recently started in the Kristen
> Harris Lab as the main developer for Reconstruct.
> >
> > Soon, I intend to begin addressing some bugs and to introduce some new
> features. As I see it, my job is to make using Reconstruct as smooth and
> easy as possible for all of you. With that in mind, I would like to hear
> from you - have you ever thought that a feature of Reconstruct could be
> easier to use, had an idea for a new feature, or encountered a bug? If so,
> please let me know.
> >
> > Here are some examples of new feature ideas:
> > * A bigger memory cache for image files, to reduce the amount of time that
> it takes to flip between more than two images.
> > * A more user-configurable trace palette, for instance, to allow
> excel-like automatic naming for traces (say, from D01 to D99), or to allow
> more or fewer traces in the active trace palette.
> > * Save time, date, user name and system name with traces, and improve the
> trace list to sort and automatically group by this info.
> > * Associate each trace with a unique identifier that is not associated
> with its name, to prevent trace merge collisions in the case that two
> different objects have traces with identical names.
> >
> > I look forward to hearing from you all.
> >
> > Best,
> >
> > Larry
> >
>