Le portail francophone de la géomatique

Bonjour,

Thanks for copying me on your message. Regrettably, my limited command of
French prevents me from replying with any sort of sophistication. However,
I will venture a brief reaction to what I guess to be the gist of your
comments:


> 1/ la NIMA n'a pas compile une couverture complete du monde au 250k.
>
> VMap 1 est une co-production internationale qui associe 16 pays. La

Yes and no. If you look at the history of the data set known as VMAP1 it
is clear that it originated mainly using US military data sets prepared by
the US. It certainly existed as a coherent, world-wide data set before any
content was contributed by other countries.

The other issue is that as an activity of the US government, in the US VMAP1
is governed exclusively by US law. It may well be that other countries
contributed data to VMAP1; however, if those contributions and attendant
agreements did not follow *exactly* the requirements of US law to exclude
such contributions from the public domain in US jurisdictions, then... tough
luck - the data has entered the public domain. It comes as a surprise to
many foreigners the many ways in which data provided to US agencies can
enter the public domain and the numerous technicalities that must be
observed to prevent this. However, it is NIMA's and the foreign partner's
job to deal with US law.

Note that even if NIMA has managed to exclude certain data legally the US
origin data is still in the public domain. NIMA *must* separate out all the
US origin data for those regions and provide them. It is this effect, where
entering into foreign partnerships blackens US origin data that is the
illegal effect.

> NIMA n'est donc pas proprietaire des donnees et toutes les discussions
> aux U.S. qui s'appuient sur le fait que cela aurait ete cree avec
> l'argent du contribuable U.S. sont par ce fait vides de sens. Les gens
> qui s'agitent ici ou la sur leur chaise en repetant cette anerie
> feraient bien de se renseigner.
> Pour la France le cederom qui couvre le pays (l'essentiel du territoire
> des morceaux de pays voisins) a ete fabrique par l'etat-major des armees
> francais a partir de donnees preparees par l'IGN.
>
> 2/ La source n'est donc pas toujours (directement) la carte papier
>
> Pour la France les donnees sources ont ete elaborees a partir de la BD
> Carto.
> D'autres pays europeens ont aussi fourni le VMap 1 en partant de bases
> de donnees nationales (c'est le scenario evidemment preponderant en Europe
> occidentale).
>

Again, yes and no. The original projects according to NIMA were entered
from paper maps and from the sources used for those paper maps.

>
> 3/ Les instituts cartographiques et les critiques qui leur sont adressees
>
> La critique la plus courante est - a des variantes pres - celle de la
> Free World Map Foundation dont le l'URL est fournie par le message
> d'origine repris par Georges Bertrand, je cite :
> Government Cartographic Monopolies - Regrettably, many so-called free
> societies must still purchase the landscape of their democratic
> discussions
> from a state-run monopoly and must utilize the maps so purchased only as
> allowed by the licenses granted by the state. Outside the United States,
> this is the rule, not the exception.
>
> Il faudrait un jour que ces reproches s'adressent a ceux qui ils doivent
> s'adresser. Dans tous les pays europeens ou presque les gouvernements
> (representes par leurs ministeres du budget ou des finances) cherchent
> a limiter les depenses. Les instituts doivent donc obtenir une part des
> financements necessaires des ventes qu'ils realisent, exactement comme les
> chemins de fer ou le service de l'eau (sont-ils devenus gratuits depuis
> la ou ils ont ete privatises ou rendus concurrentiels ???
> sont-ils gratuits
> aux etats-unis ?).
> Si donc d'aucuns veulent que les donnees ou cartes soient gratuites pour
> l'usager, que n'ont-ils de cesse de harceler les ministeres des finances
> pour que ces derniers (avec nos impots a tous) financent a 100% la
> creation et l'entretien desdites donnees et cartes ?
>

I agree with the Free World statement. Anti-democratic governments come up
with lots of excuses why the context of democratic decisions should be
denied to the public. Charging for access is an effective way of denying
access. If you believe in democracy, you'll finance it. Democratic
governments spend money to educate their populations and to publish proposed
laws and other government activities so their citizens can participate.
It's part of the cost of freedom.

>From a purely economic perspective, charging for GIS data that supports
country-wide infrastructure is simply foolish. It is like charging to use
highways or to attend public school. The efficiencies of using GIS bring
many more benefits to the nation at large than the comparatively microscopic
cost of the GIS data held by the government.

Speaking as someone in the GIS business, by the way, it's no accident that
France and other European nations have effectively zero participation in
world GIS. It's a case of infrastructure driving industry. Germany had
autobahns and the US lots of highways... the result in both cases was growth
in the car industry. In GIS, the vast array of free public data in the US
has provided the fertile soil from which numerous, highly competitive, cool
GIS and data-analytic software tools have sprouted. The state cartographic
monopoly of GIS data in many European countries has had the opposite effect
and has discouraged growth of local GIS.

Look, there is no way something like Manifold (http://www.manifold.net)
could have sprouted in Europe. The sad part is that much of Manifold *is*
European but it could never have happened there. Only in the US was there
the USGS with lots and lots of data to provide the raw material, the fuel,
for something like Manifold. So, it is no accident that once again in what
emerges as a new class of world software Europe is left in the position of
buying its future from the USA.

The utterly rediculous part is while European bureaucrats invent reasons to
make it difficult for their citizens to get cool cartographic data of their
own countries, Internet makes it possible for those same European citizens
to bypass the dead hand of their local bureaucracies and buy the maps of
their own homes from... the USA. Is that good for Europe? What next? Put
the Americans in charge of collecting tolls on French roads? Try to
establish a ban on Internet so that no one in France can buy cool maps of
France over Internet?

>From the purely parochial interest of France, unless you want to face yet
another computing lexicon that is entirely English in origin [ cederom ?!
:-)] it would be good to develop some French capability in GIS before the
young people interested in GIS feel they need to move to the US. GIS in
French will not happen to a significant degree until the French government
puts GIS data in the public domain. Even simplistic things like US
companies deciding to translate their software into French depends upon the
availability of French data. No data => limited market => lower priority on
translation. This is especially important to attract French translations of
the young, new companies like Manifold who are creating the next generations
of innovations in GIS.

> Les gens de la Free World Map Foundation poursuivent en enoncant :
> Regrettably, the long arm of some state cartographic monopolies has
> extended into influencing agencies within countries such as the United
> States in an attempt to prevent freely usable data from emerging that
> may threaten the state monopoly. In the US, for example, the National
> Imagery and Mapping Agency (NIMA) has refused to release VMAP1 , the
> public domain, non-classified, highly detailed world map it possesses,
> because of influences from state monopolies outside the US.
> The sharing of data between governments has now introduced the repressive
> policies of some state monopolies into formerly free areas.
>
> Ce texte, veritable morceau d'antologie de propagande, enonce que
> la donnee
> est public domain , ce qui est faux, confere supra, non classified , ce
> qui
> est faux (beaucoup de CD sont Restricted diffusion ), que la NIMA possede
> ces

You may be surprised to read the actual texts of the US laws that regulate
the intellectual property status of VMAP1. NIMA can describe the data
however it likes, but that does not prevent the law from applying.

> donnees (faux encore, confere supra) et, d'une facon generale repose sur
> le mode plus on assene et plus on repete, plus ca s'impose ...
> je passe sur le fait que les societes europeennes ne sont pas
> libres, voire
> sont des repressed societies .. Du Pat Buchanan au bas mot. Ce jargon
> revient sans cesse dans la prose de la page de la Free World Map
> Foundation .
>

It's surprising to read a French person dragging in Pat Buchanan's name.
For all my tremendous respect for French people and French institutions, I
am struck by how Pat Buchanan has little in common with most Americans but
how his major theme, limiting immigration by certain classes of foreigners,
is almost identically the sentiment expressed by many French people. It is
an ironic situation. We all have our problems with immigration (even in
Russia) but it is still a surprising you would use his name.

> On peut noter que les memes s'en prennent aussi aux organisations a but
> non lucratif :
> In other cases, non-profit organizations often publish maps that
> are provided free of cost but which may not be used for
> commercial purposes
> or by commercial organizations.   We strongly believe that
> speech that is
> edited, even for a good purpose, is still not free speech. For
> someone to
> say you may use a map but not for commercial purposes as a
> practical matter
> is a very significant intrusion into your organizational life.
>
> En somme tout doit toujours premettre de faire du business et du profit,
> on est loin de l'idealisme angelique que certains supposent...
>

As far as I understood the above as well as the original Free World rant, it
seems to me that the comment was a discussion of the GNU approach to
copyleft publication. If one disagrees with that, it seems one also
disagrees with LINUX, GNU, and other free software items.

>
> 4/ Quelques precisions au sujet de Dimitri Rotow:
>
> Dimitri Rotow is the Founder and President of Bell Technologies,
> a hardware
> and software company that was purchased by Intel. He headed up all of
> Intel's
> marketing efforts in the former Soviet Union where he produced a cadre of
> software developers that are very highly regarded. Mr. Rotow is also the
> Founder and Chairman of Manifold.net, a leading company in the GPS
> marketplace

I'm not responsible for what other people write about me. In point of fact,
I'm merely a product manager for manifold.net, and neither the founder nor
the Chairman. I was a General Manager at Intel and launched several new
businesses for them, including their integrated workstation manufacturing
business. Intel now has numerous facilities and offices in the former
Soviet Union and employs over 1500 people in Russia, mostly software
developers. They also have 3000 developers in China.

It is interesting to note that in the vast land rush of technology
movement out of the United States that the Old World is, once more, being
passed by. It seems that just as manufacturing left the US and went straight
to Asia, that this time high technology design is leaving the US and jumping
straight to China, India, Russia and other places, but not to Europe. The
386 was the last microprocessor designed mainly by Americans. With the 486,
over 50% of the designers were foreigners, mostly Chinese, Indians,
Pakistanis, Russians and Israelis. With the 586 (Pentium) over 80% of the
designers were foreign and in recent chips I've heard the ratio is now well
over 90%. Almost none of these people are Europeans or Americans.

I think part of this is the reduced mathematical content of education in
Europe and in the US as compared to places like China, India and Russia. One
sees this all the time in GIS software design, where very few people from
Europe or America have the requisite skills in math to do good GIS
programming. I strongly suggest that freeing up GIS data and using
spatial/geometric/cartographic concepts in schools is very important for
developing European students with good math skills.

At manifold.net we all love Europe. It would be good to see Europe take a
leading position in GIS and database science. Why abandon the field to the
Americans, or to hired guns from India, China and Russia?

On a personal note, you may be amused to see the Release 5.00 news items at
http://www.manifold.net/news/news_set.html. This is the upcoming new
version of Manifold. One of the main images used in the Help topics for
Manifold 5.00 is a photograph of the famous statue of Henry IV at the Pont
Neuf. It appears in some of the sample thumbnails and Help topics at the
above URL.

Best regards,

Dimitri Rotow

Merci monsieur pour votre mise en garde et surtout pour le temps que vous
avez consacre a lire ce email et a en faire l'analyse. Je suis totalement
en accord avec vous et je tenais tout simplement a le dire.

Meilleures salutations

Olivier Louy

Maitre es science       
Sion (Valais)
Suisse

Bonsoir,

Il semble qu'il faille apporter quelques complements, ainsi
que quelques precisions car il doit y avoir un ou deux malentendus.

1. le cederom qui couvre l'essentiel de la France (au 1 : 250 000
en theorie, meme si la definition est pour la France plus compliquee,
et non au 1 : 100 000) est effectivement reserve a usage militaire.
Le produit civil comparable est Route500 (plus riche pour les besoins
civils en fait). Il est disponible a l'IGN (mais pour l'heure pas
a 55 j'en conviens). On peut noter quand meme que le Vmap a ete
cree en 1998 a partir de donnees 97, tandis que le produit civil
est mis a jour annuellement (ce qui est un facteur de cout, est-il
besoin de le rappeler).

2. Les donnees numeriques en France (et d'ailleurs dans les
autres pays de l'UE) sont bel et bien disponibles (!?!)
Je ne me susbtituerai pas ici a la Direction Commerciale
de l'IGN pour exposer les tarifs et conditions. Ajoutons
simplement qu'a compter de 2001 les conditions de diffusion
sont tres larges (application directe des decisions
interministerielles a venir qui font suite au rapport Lengagne).
Sauf a considerer que l'acces a la donnee veut dire acces
au cout marginal de reproduction, on ne peut dire que l'acces
soit interdit.

3. Toujours au titre des evolutions de ces dernieres annees : la
comptabilite IGN (fournie aux tutelles) distingue les differents
produits et permet de savoir ce que finance la dotation de l'etat.
Le financement public s'applique donc a des productions approuvees
et choisies par la tutelle. La tendance est de plus que le CNIG
joue un role dans l'approbation de la definition des produits
voire leur definition (de plus en plus). En revanche le financement
ainsi defini des productions de service public repose a la fois
sur la dotation et sur les revenus des ventes de licences. La
dotation dans son entierete couvre 70% des charges de personnel
(les autres charges restant a couvrir...)

4. Je n'ai jamais eu connaissance d'une seule intervention d'un
IGN europeen aupres de la NIMA pour interdire la diffusion de
CD VMap americains. En revanche les partenaires europeens du
programme (dont la France: EMA) ont bien sur indique qu'ils
n'autorisaient pas la diffusion des donnees relevant de leur
territoire hors des armees alliees. Si donc la NIMA ne diffuse
pas les donnnes correspondant au territoire americain c'est
qu'elle choisit de faire ainsi, soit par prudence pour ne pas
creer un precedent, soit pour une raison strictement nationale.
En tout etat de cause la version paranoiaque exposee sur le site
de M. Rotow n'a aucun fondement ni aucune realite.

5. Je n'ai jamais dit que je trouvais anormal de gagner de
l'argent en s'appuyant sur des infrastructures etablies avec
l'argent public ! Si le gouvernement francais voulait bien
considerer l'infratsruture geographique comme une autoroute...
je ne serais pas le dernier a esperer que cela puisse contri-
buer a la richesse et au bien etre nationaux.
Ce que j'ai releve, c'est que M. Rotow s'indigne du fait que
des organisations a but non lucratif (ou des chercheurs plus
haut dans le texte) ne veuillent pas que les donnees qu'ils
creent et fournissent a titre gracieux fasse l'objet d'un
emploi a but lucratif. J'aurais tendance a dire que c'est
leur affaire et que tant que les lois du copyright s'appliquent,
je ne vois pas au nom de quel arrogant totalitarimse on
leur denierait leurs droits !
je rappelle un morceua du texte :
In other cases, non-profit organizations often publish
maps that are provided free of cost but which may not be
used for commercial purposes or by commercial organizations....
For someone to say you may use a map but not for commercial
purposes as a practical matter is a very significant intrusion
into your organizational life.
Vous m'avez mal lu.

6. J'ai bien dit que je trouvais normale (et non simplement
logique ) l'attitude de Manifold. En effet, confere discussion
supra, ils font partie des acteurs qui peuvent esperer un
accroissement de leur activite et de leurs benefices si
des infrastructures adequates sont mises a disposition par les
gouvernements (apres que tel ou tel secteur soit plus ou moins
prioritaire a cet egard est un choix politique). Ce qui est
tres contestable en revanche c'est d'associer sa pub a cette
demarche, de facon a faire envelopper sa societe d'une aura
de defense des libertes, des citoyens etc. et d'associer son
image a celle de GNU. Que je sache les produits Manifold ne
sont pas distribues au prix de reproduction au nom de la
liberte des citoyens !

7. Je ne disconviendrais certainement pas du fait que
l'investissement dans les infrastructures peut generer des
richesses, y compris pour l'etat. Il peut etre utile a cet
egard de comparer les etats-unis et la France. Le budget
2001 de l'USGS est de 895 millions de dollars, avec une
dotation federale en augmentation de 82 M sur 2000 (celui
de la NIMA a aussi fortement augmente). Par comparaison le
budget 2001 de l'IGN sera d'environ 110 M avec une
dotation (de 53 M) en augmentation d'un peu moins de 2%.
Meme si le pays est plus grand et peuple, il est vrai
que le contribuable americain paye plus...

8. La question du dommage au prive par la disponibilite de
donnees publiques bon marche est a regler en amont, par un
choix politique. Si quelque chose est juge utile ou
necessaire a la collectivite nationale mais pas rentable
economiquement sur une base de mise a disposition pour
l'ensemble du territoir, alors il est juste que cela soit
cree, entretenu et mis a disposition par l'etat, plutot
que de ne le voir exister que dans les zones rentables.
Par ailleurs et comme je l'ai indique au point 2, les
conditions d'acces aux donnees numeriques de l'IGN sont
radicalement revisees (mais pas encore la dotation de
l'etat !)

Et tant pis si c'est trop long