I respect Ed's programming abilities and his decisions on what to put into
Vuescan. As a programmer for over 20 yrs I also instantly bristle whenever I
hear someone claim to be able to put together a UI in "just a couple weeks."
(note you did say prototype - the rule is the last 10% of the work takes
90% of the time applied to your estimate means at least an additional 9
weeks to get something working. My guess, and I make my living guessing how
long a software project will take, is that to finish a GUI that is
acceptable to a multi-national crowd of users would take at least 6 to 12
1. Do you realize that VS is multi-platform - Mac, Linux, Windows.
2. Have you ever seen Ed's code base and have any idea what the interface to
the GUI would look like.
3. Most UI suggestions include features like Histograms, Levels adjustments,
batch control with individual frame adjustments, job and/or work flow
management, etc. Are those included in your estimate.
4. If you can do this as a "wrapper" which drives VS external to the code
base, then do it and sell it.
5. Does your estimate include multi-lingual/multi-national support?
6. Apparently you have little respect for Ed's abilities since he has
already stated that there are complications with some of the GUI parts based
on idiosyncrasies of individual models/makes of scanners.
Keep up the good work. What most programmers/engineers miss is that the
question is not "can you" but "should you". The clean data and vast scanner
support is obviously more important to most of your registered users (me
included) and we thank you every time we use your program.
> -----Original Message-----
> From: email@example.com
> [mailto:firstname.lastname@example.org]On Behalf Of Johnny Deadman
> Sent: Monday, July 23, 2001 1:52 PM
> To: Filmscanners
> Subject: Re: filmscanners: Re: Vuescan gripes
> on 7/23/01 11:25 AM, Shough, Dean at email@example.com wrote:
> > I don't think we miss the point, but rather we have different
> priorities. I
> > would love it if VueScan had a better (and more Mac like) interface, but
> > given the choice between improving the guts of VueScan or the
> interface, I
> > will take the guts anytime. Especially since I can work around
> the portions
> > of the interface I don't like. If the raw scan is bad there is no work
> > around. Ed could hire 5 programmers to assist him, spend 6
> months getting
> > them up to speed before getting anything useful out of them,
> and raise the
> > price of VueScan from $40 to $400, but I think it would kill VueScan.
> no honestly this is nuts. If I had a week to spare I could prototype a GUI
> in RealBasic. There's nothing hard about it.
> John Brownlow
> ICQ: 109343205