[OSGeo-Discuss] Re: Will there be an OSGeo Desktop shootout atFOSS4G 2010?

[Brian Russo]

I do not think a simple feature comparison is very useful. Seeing workflows
that happen to use XYZ software or.. how we transitioned from ABC
proprietary software to XYZ open source and improved performance 10% while
reducing costs 20% etc.. that's useful and convincing. Knowing that ABC
proprietary supports 3 methods of kriging while XYZ open source supports 2
may be earthshattering or completely irrelevant. The real answer is an
unexciting "It depends".

You can tweak feature comparisons to make yourself look good, the
competition look bad.. etc..  It's just like statistics. I see this in
camera reviews all the time, The Pentax K200D has a 96% viewfinder.
Comparable models from Nikon and Canon offer 95% viewfinders. Call me
cynical, but I find it hard to believe that someone at Pentax didn't say
"Let's make our number bigger". Of course, in many reviews of those models,
the Pentax scores higher on that feature because 96 > 95 [1]. Does that make
it a better camera? Well gee I guess if you only cared about that 1 thing; I
don't know anyone that does (or should).

What you don't see in feature comparisons are solid, no-B$ analyses of how
they let you do your job better. Usability for example is something that you
cannot easily quantify. You can have the best product/software in the world
but if I can't get the results due to UI/UX failure, or an unnecessarily
steep learning curve, etc; then for me the user - your software is 100%
useless (actually it's worse because now I have to find a tool that does
work). Handtools are a classic example of this; anyone that works with wood
or mechanical parts will understand how some tools just don't "feel" right.
Do they feel 20% less right? Doesn't work that way.

Not to say that feature comparisons are completely useless, especially for
new people they can be good; but overall they're coarse, imprecise, and not
very knowledge-rich IMO. Case studies of transition are much more powerful;
speaking both as a user and a decisionmaker. I think moving towards active
real-world presentations is far more powerful than lifeless comparisons.

Another example is people that love SSDs (solid state drives) and rave about
their Windows boot times. Yeah SSDs are great but.. do you just sit around
and reboot your computer all day? A 2000% improvement on something I do once
a month is probably not that big of a deal.

 - bri

p.s. I shoot nikon but I really don't care what you shoot and have 0 vested
interest; just an example.

1. http://www.dpreview.com/reviews/pentaxK200D/page20.asp

On Mon, Dec 21, 2009 at 11:01 PM, Miguel Montesinos <
mmontesinos at prodevelop.es> wrote:

>  Hello,
>
>
>
> I think that a simple comparison to what ArcGIS does is limitating. Several
> issues arises:
>
>
>
> -          Why compare to ArcGIS 9.3 and not Geomedia, MapInfo,…?
>
> -          What about features that OS GIS desktops provides not present
> in ArcGIS 9.3?
>
>
>
> I’d rather have a comparison among all of them under equal conditions, for
> instance a feature comparison based on the maximum features all products
> offer, as well as a perfomance analysis.
>
>
>
> For this, a common dataset of both file and service based data should be
> available. In Spain there are “a lot” of public official geodata which could
> be used as test datasets.
>
>
>
> I also like very much Paul Ramsey’s approach about what I like and what I
> don’t made by people belonging to different projects.
>
>
>
> Regards,
>
>
>
>
>
> ---------------------------------------------
>
> Miguel Montesinos
>
> CTO
>
> PRODEVELOP, S.L.
>
> mmontesinos [at] prodevelop [dot] es
>
> www.prodevelop.es
>
>
>
> Miguel Montesinos
>
>
>
> *De:* discuss-bounces at lists.osgeo.org [mailto:
> discuss-bounces at lists.osgeo.org] *En nombre de *Daniel Ames
> *Enviado el:* lunes, 21 de diciembre de 2009 19:25
> *Para:* Maxim Dubinin; OSGeo Discussions
> *Asunto:* Re: [OSGeo-Discuss] Re: Will there be an OSGeo Desktop shootout
> atFOSS4G 2010?
>
>
>
> Folks, I like the structured comparison approach that Cameron outlined.
> Also equally (or perhaps more useful) would be to put together a wiki page
> with goals and benchmarks based on ArcGIS 9.3. And then indicate where the
> os packages compare. This would provide us with the ability to answer the
> most important question which is "can this do what the proprietary software
> does."  For example, we could post a couple of maps made in AG and then
> challenge each desktop team to create and upload the same maps. Etc.  I have
> a line shapefile with 200 shapes. We could upload it and have everyone do
> some timing to show how fast to load,pan, etc on the data. This could also
> serve as a way for some of the teams to see their own deficiencies and find
> critical tasks to work on (they could then update their reporting on the
> wiki and indicate the version number)... - Dan
>
> _______________________________________________
> Discuss mailing list
> Discuss at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/discuss
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/discuss/attachments/20091221/59af3787/attachment-0002.html>