[GRASSLIST:9040] Re: ArcView vs GRASS

[M S]

INTRO:
oh I could have a field day with this thread, and I think i will. I just
cant pass on this opportunity. I didnt want to start a thread like this, but
Its rather lengthy, so grab a comfortable seat, a cup of joe, and be glad
you use GRASS. Since there was some digression from the ArcView vs GRASS
topic, i too shall digress. If you are a blind fan of ESRI, dont read this,
you will just make silly responses. These are not petty issues. My criticism
is harsh at times, but please read through this to see where i am coming
from and why i have the opinions that i do. These are enormous amounts of
money being paid for a now poor product. Dont perceive this as "bashing" of
ESRI, but rather just a taste of the sincere and genuine experiences I have
personally experienced with ArcMap/ESRI in a professional, deadline driven
environment. This is why I was driven to find alternatives such as GRASS.
And I am so thankful that I did.

There is one word that describes ESRI, and that word is GREED. There are
many words that describe GRASS: excellence, refined code, dedication to
producing a quality product, pride in the product, multi-platform
compatibility, quick bug fixes, stability, innovation, free, intrigue to new
users, relief for users of ESRI, and best of all: an embracing and
supportive community and author(s). (I salute you Markus)

HISTORY:
ESRI had a good thing with workstation arc/info and the extra modules. the
command line was king. i learned GIS on A/I workstation on UNIX. THose were
the days when, if there was an erroneous result in the analysis or data
creation, it was the user/operator's fault. On the ArcMap platform, if the
result is erroneous, its advised to first check that the software algorithm
isnt bugged, because I have too often found this to be the case. (I could
cite dozens of examples, a recent one i will say is the bugged HPGN/HARN
projection in ArcMap's on the fly projection, and physical projection of
data in ArcCatalog)

Around the turn of the millennium, with the advent of ArcMap, ESRI
completely fell to pieces in the quality of their product and the quality
and availability of their tech support. They sold out to windows, they
turned their back on the very operating system that helped their software be
stable and crunch on analytical data for hours on end without bombing out.
They ticked off a large portion of their existing GIS user base when doing
this. They price gouged the community. They wove their software in to
government agencies, large corporations, and the trickle down effect was
felt as many other businesses had to now cater to the proprietary "datamodel
specific" ESRI way of things. Now even if other companies and people want to
use, better, open source, alternative software, they cant without major
hassle, because final products are being asked to be delivered in a
"geodatabase" fomat which as far as i know, only ESRI software can write. I
beg of ANYONE in the open source community to create a program that can
read/write to geodatabases (microsoft access databases with spatial
component).

BROKEN COVERAGE DATA MODEL ON WINDOWS NETWORKS:
How in the world do they justify the SAME cost of their software that ran on
UNIX on Windows? in the UNIX days, everything was expensive, the UNIX OS,
the UNIX Hardware, the UNIX Administrators... a $9,000 Arc/Info fell right
into the price scheme... and it WORKED. I personally experienced tons and
tons of software bombouts/errors when building or doing overlays across the
network. When they ported arc/info workstation to Windows with WINNT around
1996, it was BROKEN right off the bat, yet the prices were the same. A
technician later told me, "You cant build, overlay or do anything that
rebuilds topology across a pure windows network... because a windows server
does not return the data specifically like a UNIX server does. it messes up
workstation, and causes it to bomb out. our solution: COPY THE DATA
LOCALLY." i said this is unacceptable. networks have advantages. especially
centralization and backing up. in a department of many, local copying of
data is a night mare. this was hardly a "solution", this was BROKEN. I said
"well how can this be? how can this product be sold as it is, if the windows
networks dont work with the coverage data model???" the response, "It never
worked. We never tested it in a pure windows networked environment. this
came to light after it was in production by other companies". I was
astonished and disappointed. this was the beginning of the end. due to this
and the open nature of reading/writing shapefiles, ESRI needed a new,
proprietary data model that only THEY could write/read. Enter the
Geodatabase.

LICENSE CONFUSION, BAIT AND SWITCH:
Here's just one example of the mess they created. Now the young GIS useres
will not appreciate this if they have joined GIS post year 2000. In the
"good old days", a common software combination was workstation arc/info and
arcview. workstation for data creation and analysis, and arcview for
cartography. some people were paying maintenance on workstation and arcview.
if i recall right, thats 3000$ & 500$ respectively PER YEAR. When ESRI came
out with ARCmap, arcview was now just a subset of arcmap arc/info. there was
no need to maintain two separate licenses for the sofwares because one was
just a subset of another. Their licensing got majorly screwed up, so much to
the point that when you asked a sales representative to explain it, the
response was "we are confused about it too". what kind of a response if that
from sales reps who you are paying MASSIVE amounts of money EVERY single
year??? Anyhow, there was no longer a need for arcview if you owned arcmap
arc/info. would they credit you? would they allow the maintenance to be
applied elsewhere? no, I personally was laughed at for even ASKING such a
question by them, even though after the bait and switch a credit is the
least they could have done. but no, they wanted more money. I had such a
horrible experience at a local level with them that i took my issues to the
east-coast sales manager. we hashed things out, i expressed sincere and
genuine dissatisfaction with ArcMap, and all the bugs and plotting problems
and production inhibiting issues in arcmap. His big solution: "WAIT till the
next release(s) to fix the problems". i wondered "well why have i been
paying for this software for 4 years???" its really beta software, where the
users have to find the bugs and report them to ESRI. Not to mention that
with the immense complexity of ArcMap (i mean the context "menu" for the
main "menu" is so big, it scrolls off a 21" monitor!), the tech support
quality and availability diminished to the point where I would not get calls
back for DAYS. by then, my deadline was missed. i would have to find work
arounds. a VERY common statement from ESRI was, "well ask the community if
we cant get to you fast enough, and post to our forums". I"m thinking our
organization is not paying $10,000 PER YEAR to go seek out the answer on
line. we are paying this money to talk to techs who are knowledgeable and
can solve the problems quickly. thats how it was prior to arcmap. immediate
responses and problems solved for the user.

EMBEDDING ESRI INTO AGENCIES AND BIG BUSINESS:
there is no excuse for the complete garbage software that ESRI has out now
nearly 6 years into this software's life, and over a half dozen release of
software versions. The users are the beta testers, the users are the
de-buggers, the users are the victims. ESRI sits back collecting millions if
not billions in yearly maintenance, working on their own schedule, and fixes
stuff as they see fit. Try a lenghty editing session in ArcMap, and try to
keep the software from crashing. Good luck. The personal touch and care for
customers is gone. Unless you are a government agency, then I heard that
they get taken out to lunches frequently by ESRI reps and treated great.
THis makes sense because if they can weave their software in to government
agencies, then other companies working for them/with them have to be ESRI
compatible. not "open GIS" data exchange formats, but rather the proprietary
Geodatabase. An example is agencies asking for data in Geodatabase format
and maps in MXD format. this is unacceptable!

SYNONYMS:
Sadly, much like PC's are synonymous with MicroSoft, GIS is synonymous to
ESRI. THis is truly a shame. Monopolies are bad for the consumer. Monopolies
are bad for GIS. ESRI is bad for GIS.

ADMISSION OF MESS:
As of this year ESRI redid their license scheme (and as i read in this
thread, no you dont have to buy a dongle every year, you have to buy the
"maintenance package" every year). ESRI must realize that the liensing is
out of control. they simplified it by putting modules we used to pay for
into the main license just this year, 5 years later after the software
re-write and re-licensing. they made it sound like this stuff was free now.
however our maintenance every year is still the same price as they jacked up
the price on the main items.

BIND ESRI SUPPORTERS:
When this type of frustration is vented in ESRI forums the response is "stop
whining". There is plenty of justification for this disgruntled-ness. ESRI
pulled the "bait and switch" maneuver. They had a solid product for 30
years, then FORCED the community to move to ArcMap, dumped UNIX OS's, and
sold out to windows. This is my career they are messing with, this is my
livelyhood, this is my future. New users do not understand. They just live
with the horrible bugs, and keep on clicking their way through GIS without a
solid understanding of the data and analysis. They dont realize the enormous
price that one pays for the software initially AND every subsequent year
there after. They dont realize what quality GIS was. When deadlines and be
met, and bugs are the reason why, thre is plenty of justification to vent
this frustration. I have to use this cruddy software, or find a new job or
even career. As consultants, we have to be compatable with our clients.

RESPONSE TO PREVIOUS THREADS:
ArcView doesnt just "run better under windows"... it ONLY runs on windows,
provided we are talking about ArcMap ArcView. ArcView 3.X used to run on
UNIX and was so solid. People who can really speak to this issue are the
ones who were working with GIS prior to ArcMap's existance. The fact it runs
on windows is not a plus. This is the reason arcmap can not plot large
images because of the file capture function is relied upon by windows. where
the old arcview (3.X) used an ESRI file capture format and was nearly
flawless. You have to put your map plot quality to poor if you are working
with large images you want to plot on paper or the data gets dropped off the
map. As per ESRI, they now rely on the windows subsystem to handle the file
capture and print. Which is one reason why now they include their "arcpress"
module, so that the maps can be rasterized for plotting capabilities. before
this, their business partner would scour the forums looking for people with
plotting problems and solicit them for their $3000 plotting "solution". ESRI
probably owns the "business partner" as well. this is unacceptable. plotting
maps and large images should be an "out of the box" function for $9,000
software.

Dylan and Rich have extremely valid points in my. they capture the essence
of this whole issue. Very well stated by the both of you.

Rich captures the essence with this, " I found too many people who "knew"
GIS because they knew how to
operate the program controls. My response was that teaching someone how to
use a word processor did not make her a writer; teaching someone how to use
a
graphics program does not make him an artist. There's a huge difference
between knowing how to do things with a computer application and knowing
what
to do to get the proper results."

Dylan is right on as well... I couldnt agree more with what he wrote. The
dangers of point and clicking your way through a map or analysis, without
understanding what is really going on can be disastrous.

An example is someone i encountered who said their experience with GIS was
"spatial analyst". This was peculiar to me, because they were asking about
BASIC data creation procedures. This indicates that they can click their way
through the "analysis", but dont understand how the data was created or how
it works. You can run the "analysis" but have absolutley NO CLUE about the
data itself, or how to create it??? Eventually even a monkey couldl press
all the right buttons to get the analysis to work.

GRASS ROCKS!:
GRASS is awesome. Its capabilities amaze me and intrigue me to keep learning
about it. Quantum GIS is awesome. Mapserver is awesome. these three
softwares on a Linux platform, have the ability to replace Arc/Info, ArcView
and ArcIMS respectively. And its already happening. I saw an ArcView
3.Xscript that writes out a Mapserver File. people are catching on.
the
community is becoming aware.

I should write a book or large composition. I have had so much interaction
with ESRI tech support, and sales reps that i know more than they want me
to. I have so many stories and real life experiences with ESRI that it would
be enough to turn many people away. After much prying from tech support and
sales, I see the full picture. I would not be surprised if this privately
owned company is shoving out tons and tons of copies of their software to
continue to be the monopoly, and then suddenly go public and cash out and
leave the mess for the shareholders. this is a common trend with private
tech companies.

This is just some of my experiences. one good thing the horrible ESRI
software did for me, it drove me to GRASS, QGIS and Mapserver. i thank them
for that. I love Linux and the next logical step was GIS on linux. I know
most all of you on this thread are of the same feeling that GRASS is
awesome. I was so relieved to find GRASS. it revived my hope in GIS. It
brought me back in the game. I am so thankful that all the developers are
continuing to work on GRASS and make it the best out there.

Top Accolades go out to all the open source guys, developers especially but
also the users. this movement is too strong to die or be squashed by
inferior, proprietary software. Let ESRI feel the squeeze. Let the
revolution begin.


On 11/11/05, Rich Shepard <rshepard at appl-ecosys.com > wrote:
>
> On Fri, 11 Nov 2005, Trevor Wiens wrote:
>
> > The common reply to this is that ArcView runs better under Windows and
> is
> > easier to use, so it is worth it to buy a license and have no control
> over
> > your software. To this, I would suggest that the appropriate counter is
> > that if users want Windows integration and ease of use, they should buy
> > Manifold GIS, for 1/10 the cost of ArcView and is considerably more
> > powerful than the base ArcView product.
>
> If I may be permitted peripheral comments again, I have two to offer.
>
> I first learned of GIS in 1987 when we used it at the state agency where I
> was a technical program manager. I quickly learned that most people wanted
> pretty maps, not deep technical analyses. I think that's still the case in
>
> many situations: make the map sufficiently visually attractive and no one
> questions the quality of the data.
>
> One problem with "ease of use" is the emphasis on the wrong aspect. When I
> was more heavily involved with GIS than I am now (my current business
> focus
> being on the application of approximate reasoning models to environmental
> issues), I found too many people who "knew" GIS because they knew how to
> operate the program controls. My response was that teaching someone how to
>
> use a word processor did not make her a writer; teaching someone how to
> use a
> graphics program does not make him an artist. There's a huge difference
> between knowing how to do things with a computer application and knowing
> what
> to do to get the proper results. I've seen this mis-match with a lot of
> folks
> who do multiple T-tests rather than ANOVA on their data because they don't
> understand statistics, but they have a computer application that "does"
> statistics for them.
>
> Your analogy is very pertinent: understand what you're doing before you
> use
> a tool that takes away the tedium of calculation.
>
> Carpe weekend,
>
> Rich
>
> --
> Richard B. Shepard, Ph.D. | Author of "Quantifying Environmental
> Applied Ecosystem Services, Inc. (TM) | Impact Assessments Using Fuzzy
> Logic"
> < http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-8863
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20051113/3ec5dfe0/attachment.html