Proposal:Boundary=forestry( compartment) relations/Review of second vote results

Opinions expressed in the Vote section

Here are the opinions expressed by voters in the wiki Vote section. Other opinions have been expressed, but, in the sake of improving the proposal, this table will only tell about what made voters reject the proposal, or the points voters asked to be improved, not what voters agreed with in the proposal. Such opinions do matter, of course, but I'll only take them into account when I'll investigate the possible improvements, in order to reject those who would simply increase opposition to an improvement of forest tagging.

During the count, I assumed that the voters telling "Same as Foo" agree with all the objections of the user Foo. I also had to interpret some comments, to tell if they match the hereunder opinions or if they were expressing a slightly different opinion; consequently, the count may be somewhat biased by my interpretation. That being said, the relative weight of the opinions should still match the reality of the vote comments.

Contradictory opinions per frequency

Opinion	Expressed n times	Notes
Do not use boundary for a landuse	27	These voters highlight that boundary should not be used to map how mankind use the land. This is a different opinion from those who want landuse=forest to be kept.
Do not deprecate landuse=forest	26	Such calls did not explain how to get rid of the current landuse=forest/natural=wood confusion without deprecating at least one of them. Some of these voters expressed doubts about the pace of the migration and fears about landuse=forest staying in the database despite deprecation. Others simply said that the tag is used for too many entities to allow deprecation. Others saw the proposal as a seventh approach to forests, confusing it further.
Feature is unverifiable	19	Given the mandated verificability of boundaries, I assume that these voters are saying that what is unverifiable is that the area is effectively under forestry.
Whether the area is under forestry is a niche interest which do not justify such a fundamental change	17	-
Forestry use and forest boundaries are different features and should be subject of different tags	13	-
Do not enforce relations	11	-
Unexplained opposal	4	-
admin_level tags should not be used on forests	3	This point seems to have especially troubled the US community
Do not promote mass retagging	2	Fears are about iD mass replacing a deprecated tag without explaining it to the user
Do not merge protected areas	1	-
Finer forestry-related tagging required	1	-
Legal problems	1	Unexplained issue, probably related to data import
Don’t import data from legal sources	1	-
Brand new tag needed for wooded areas	1	The purpose of this comment is likely to get rid of the landuse=forest/natural=wood confusion by deprecating both and having a one-tag-to-rule-them-all.
Hidden deprecation of landuse=forest	1	This issue may have led to numerous no votes, even when voters did not explicitly told it on the Wiki.
landcover=trees should be deprecated	1	This deprecation was proposed during the first vote and provoked a serious opposition from landcover proponents.
The proposal must also deprecate leaf_cycle/leaf_type	1	This comment said that theses tags are "completely unscientific and commonly mistaken".

Possibly relevant opinions, ideas and comments

• For the record, it seems that this vote break the record of the number of votes for a given proposal.
• landuse=forest is described by the Wiki as a "counterintuitive key name", explaining that "Commonly used for any area of trees, and does not imply a particular landuse."
• The proposal name did not explicit that the second vote was also about solving the landuse=forest/natural=wood by deprecating the former; numerous contributors, for instance in the German community, complained about what they felt a hidden deprecation, and that likely lead to numerous no votes; some users recognized that they learned of the proposal after the vote began and that, consequently, they did not read it and voted no because they weren't involved as early as they should have been. A clear, consistent proposal name could have prevented some no votes, by allowing the contributors to not feel betrayed, and could even have incited them to contribute in the proposal before the vote.
• The proposal should not enforce one forestry area per owner/manager or one name per forestry area; it is not currently the case but the proposal should be crafted to explicitly tell it without letting people think otherwise as it was the case during the vote. This led some German contributors to think that this enforcement about mapping owner/operator would lead to transform OSM in some sort of forestry cadastre, which would be impractical and could lead to conflict about personal data protection.
• Paulatthehug highlighted that the current 6 approaches to forest tagging are in no way equal; some of them are barely used and way easier to deprecate.
• Lectrician1 highlighted some difficulties of redefining a tag as used as landuse=forest.
• Fghj753 proposed to use the subtags operator:admin_level=* and owner:admin_level=*. This user also suggest to mention addr:* keys in the allowed forest-related tags, as in countries such as Estonia, addresses are about land lots and not about buildings. Finally, he recalled that rural areas are often mapped from armchair.
• Hauke-stieler advocated for keeping landuse=forest, but is noticeable because, contrary to other users, he gave some ideas about how to do that while at the same time removing the current confusion with natural=wood. His willingness may be useful, should this idea be kept.
• Many third-world populations have an influence on wooded lands (pasture, firewood gathering…) without long-term planification; is that management?
• Some people confuse forestry and wood extraction: forestry, as a discipline, is about human management of wooded and related lands; it may be about non-wooded lands and pursue other goals than wood extraction. Should OSM see management as an economical notion (i.e. wood extraction), or as a scientific-technic notion (i.e. management of wooded and related lands, whatever the resource extraction occuring)? Anyway, how to enforce the chosen definition on all mappers? Is it even possible? What if the management has concurrent goals, both economic and non-economic? If the economical notion is chosen, how do we prove that the purpose of the management is economical?
  • Because workers are paid? What if they are simply there for reforestation?
  • Because the extracted wood is sold? What if the money is spent into maintaining the wooded state of the land?
  • Because the extracted wood is used by humans? What if it is mere sustenance firewood extraction? It's sustenance, not economics.
• How do you tell for sure that a piece of wooded in managed (Western definition of management)? Short answer: you can't:
  • I asked some fellow French public forest rangers (at least 50 years of accumulated professional experience): their answer was "You can say that the land was managed through forestry with the traces of forest works, but you can't tell that it is currently managed."
  • A contributor in the German OSM forum asked a German forest ranger if you can tell the difference, and the answer was "Yes, but only by looking at the forest cadastre."
• Hungerburg made numerous contributions to this debate; among them, many highlighted the fact that, even in German countries where landuse=forest is assumed to effectively represent managed wooded lands, and natural=wood to represent unmanaged wooded lands, the current OSM entities also reflect the confusion around these tags:
  • in Tyrol, the offical forest cadastre tracks 5200 km² of wooded land, from which 3400 km² are subject to wood extraction, whereas OSM tracks 4570 km² of wooded lands, and announces 4515 km² as landuse=forest
  • around Gschnitztal and, worse, around Terfens, OSM records far more managed forests than the official Austrian forest cadastre; he assumes that the "95% of forests are managed in Germany/Austria/Switzerland" statistic is more or less a round number without real calculus, that this 95% number is a rule of thumb which was already assumed to be true when these wooded areas were managed and that, consequently, they were modelled with landuse=forest because "95% of wooded lands are managed and this is to be mapped with landuse=forest"
• Some mappers are inclined to use natural=scrub for wooded lands when the wooded plants on the ground looks more like scrubs than real trees, for instance after a clearcut or a final cut in even-aged management. Some other mappers contest this habit, saying that:
  • even after a clearcut with no trees or shrubs remaining, the land is still managed as a wooded land and should be mapped with landuse=forest or natural=wood (this argument may be partly raised because of the current tagging confusion between the landuse and the landcover, and may disappear if the tagging was unambiguously only about landcover)
  • such mapping would need steady maintenance, to switch back the natural=scrub or natural=heath entities to natural=wood when applicable (Hungerburg noted that such areas would keep their non-wooded cover for years, which is worth mapping, and that this would increase map quality and help orienteering on the ground, by reflecting the effective landcover; streckenkundler also noted that this may be solved with some open data sets, and gave the example of Brandebourg, which published aerial imagery with time frames allowing to do such maintenance from armchair)
  • the different available aerial imageries will show different lancover during the years after the cut, depending of their updates, and that acknowledging the limits of such areas, especially when the area is irregular, is beyond expected mapper capabilities. According to my profesional experience and corresponding comments, this problem will occur more and more often with the global warming: some areas I know were subject to three different clear cuts with increasing size and irregular shape; the project can hardly expect the mappers to accurately map such areas.
• Nop, a German OSM forum moderator:
  • explained that the use of landuse=forest is somewhat extreme in Germany, as it has been used for wooded lands to represent the fact that they were forested centuries ago, even if it is no longer the case
  • expressed sympathy for the proposal gathering no votes for some details.
  • explained that the idea "landuse=forest means managed and natural=wood means unmanaged" was spread in German-speaking countries after a unilateral change enforcing that on the relevant wiki page; this edit went unnoticed at the time, and that this division was adopted by more and more mappers and effectively became law for some without any real debate
  • he reminded that most mappers simply want to map wooded lands and don't know/care about their management status. Some users expressed agreement with this opinion.
• tracker51 remarked that, even in countries with a management of virtually all wooded lands, the determination of the actually unmanaged wooded land is very difficult, and consequently that is would be reasonable to remove the difference between managed and unmanaged wooded lands. Other users advocated the same opinion.
• Woodpeck, a German contributor whose opinion motivated numerous no votes, explained that, though he does not reject the removal of the distinction between landuse=forest and natural=wood, the problem lies with the replacement of landuse=forest with natural=wood AND a boundary=forestry relation: though the proposal mandates that the relation must only be used when its limits are verifiable on the ground, it is likely that contributors will start to replace all landuse=forest entities with natural=wood and a boundary=forestry relation without checking that the land is effectively managed, in the end increasing the work needed to check data and displacing the mess.. Some users have a similar opinion: OK with the merging of landuse=forest and natural=wood for the land, but do not enforce a relation, especially a boundary which are not designed for landuse.
• SpaLeo showed understanding about the idea of mapping a unique forest with a relation: it allows to map it with its variations in leaf types, landcover and all without enforcing the use of a single landuse=forest/natural=wood entity.
• streckenkundler, in a comment about the proposal mixing different issues, explained the issues in terms that indicate a rather deep insight about forestry customs and match my professional knowledge; he may have relevant opinions and knowledge about German forestry customs. He too explained that, though he would agree with the merge of landuse=forest and natural=wood, he thinks that the relation part and the verificability check are immature and need more discussion. He also is one of the stricter contributor about the two weeks delay.
• GerdP warned against disconnecting the entity mapping the forest and the entity mapping its management status: mappers willing to replace a wooded land with an industrial or residential zone may simply remove the forest without removing the entity mapping its management, creating inconsistency in OSM data.
• Pfad-Finder emphasized that a Wiki change would never be enough for the deprecation of such widely used tags, and expressed doubts about the practicability, even if he also acknowledges that, worldwide, people use indifferently natural=wood and landuse=forest to say "here are trees". He also guessed that the binary vision between unmanaged and managed wooded lands is effectively mostly unapplicable in real world.
• cepesko also displayed insight on German forests and their history: he reminds that, since the XIX century and the use of coal, the main goal of wood extraction switched from firewood to timbers and that this switch led to the replacement of coppice by high forests, and that this operation also extended the number of years between forestry works, effectively rendering managed forest less distinguishable from unmanaged forests. He also recalled that non-foresters lack forestry knowledge, and, consequently, that they make mistakes in their interpretation of the landcover, especially during armchair mapping. cepesko also seams to have a deep insight of forestry customs in Germany.
• Comments showing that a proposal split could allow at least some parts to be more easily approved
• Shaun das Schaf interpreted that the proposal was more suitable for big forest and not for small chunks
• Some users advocated that enforcing a relation for forestry areas would be overly complicated and would effectively discourage most mappers to map forestry areas. On the other hand, some expressed that using a simple tag on natural=wood to map its management status or the goal of this management could meet the KISS principle and easily allow mappers to map forestry-related data, if they want to.
• Martin Koppenhoefer reminded that timber or wood production is not necessarily the goal of forestry.
• Mateusz Konieczny reminded some disadvantages, should landuse=forestry be pushed further; his arguments may be relevant to any landuse=* tag used for managed wooded lands
• Peter Elderson advised to give up the deprecation part in order to have the proposal finally approved

African ("Third World") point of view

Bert -Araali- Van Opstal made numerous contributions with a so-called Third World view.

He reminded that management is not limited to a Westerner harvesting notion and may perfectly be done by local communities with different customs which a Westerner may not acknowledge as management; that led to numerous wooded areas being mapped with natural=wood because they were not managed in a Westerner definition and that, consequently, natural=wood was the right tag because it was interpreted as "unmanaged wooded land"

Also, what is considered managed by authorities, and the assumed means of management by these authorities, may greatly differ of how it is effectively used by local communities; consequently, authoritative data about managed forests, especially their limits, should not be taken as is

When African contributors started to use landuse=forest because HOTOSM started to create base maps with landuse=* tags and, when they started to map forests, they simply sticked with landuse=* because they were accustomed to it. Later, there was debates with HOTOSM which referenced to the Western notion of management, which the locals do not use, though they effectively manage the forests at some level. This issue was further confused by:

• the confusion between what authorities mandates and what is effectively on the ground: official, legal limits are often not gazetted or materialized, and locals continue to manage/use the land as before (charcoal production, artisanal timber production) what is de jure a protected/national forest but is de facto a wooded land whose management does not match Western criteria;
• reforestation programs that used the official limits and led to reforestation of residential/agricultural areas, leading to conflicts with inhabitants;
• forest authorities managing some "borderline" areas, such as papyrus fields or bamboo stands, which can be understood as "forests", even if neither bamboo nor papyrus are trees.

All these areas (not tree-covered, with conflicting land use, unclear/unmaterialized boundaries, with plantation of exotic species…) are often mapped with landuse=forest. He feels that what is needed to allow improvement of mapping is:

• a way to model the species of the stand, especially for plantation of exotic species like Eucalyptus;
• a clear definition of what is considered managed and what is considered unmanaged;
• a clear tagging distinction between the land cover and its use by humans (landuse=forest is currently not of use for that).

What is forestry?

• English Wikipedia: "Forestry is the science and craft of creating, managing, playing, using, conserving and repairing forests, woodlands, and associated resources for human and environmental benefits. Forestry is practiced in plantations and natural stands. The science of forestry has elements that belong to the biological, physical, social, political and managerial sciences."
• FAO: "The science of establishing, cultivating and managing forests and their attendant resources."
• French consortium of the state forest manager (Office National des Forêts), public forestry engineering school (AgroParisTech) and various private forest managers and foresters: "Set of disciplines about forest management"; forest management is widely understood as the legally defined "sustainable forest management" which "guarantees their biological diversity, their productivity, their regeneration capacity, their vitality and their capacity to meet, presently and in the future, the relevant economical, ecological and social functions, at a local, national and international level, without causing harm to other ecosystems."
• "Forestry in Bavaria follows the Bavarian way of an integrative forest management. Its leitmotiv is: Protect and use the area as much as possible. With this approach, we take into account the various services provided by forests in a way which is balanced and adapted to local customs. Besides ecological and economical aspects, the numerous social benefits as creation of jobs, recreation and leisure possibilities find their place in the multi-functional management."
• An African definition, preferably official and with reference, would be greatly appreciated.

Consequent proposals

Forest boundaries

This proposal will be about mostly wooded lands with materialized boundaries, whatever their management/resource extraction status.

Managed wooded lands

There are several possibilities, each with advantages and problems.

Redefinition of landuse=forest to explicitly state that the area is under forestry

How to tell, upon seeing such entity, that it uses the new, explicit definition? Besides, what for natural=scrub and other non-wooded landcovers which are still subject to forestry? Personal example: a even-aged managed forest, whose natural regeneration failed after the removal of the last seed trees; a plantation that happened after also failed, leaving a natural=grassland which was legally a managed forest and was destined to be wooded.

May be more acceptable by the contributors who do not want to get rid of this tag, though. This could be done by stating that landuse=forest no longer implies anything about the land cover, and that it must be combined with natural=wood to say "the land is wooded (natural=wood) and subject to forestry (landuse=forest)."

natural=wood+managed=*

Main problem: what is management? Besides, we can only be sure of past management (e.g. by its traces on the trees), not of current management (very irregular in forestry)

That could be something like:

• managed=no: unmanaged, no voluntary human influence;
• managed=yes: unspecified management/voluntary human influence;
• managed=preservation: management for environmental protection;
• managed=production: management in order to produce goods; use with produce=timber/wood/latex/sap/charcoal/fruits… to explicit;
• managed=protection_forest: management in order to mitigate or prevent the impact of a natural hazard;
• managed=landscape: management in order to preserve the landscape;
• managed=leisure: management in order to provide accommodation;
• managed=CO2: management in order to optimize its carbon pit effect;
• managed=reforestation: management in order to restore the wooded state of the land or prevent its disappearance;
• managed=hunt: management in order to maintain game for hunting;
• managed=authoritative: the land is considered by local authorities under forestry even if it is technically not (i.e. bamboo plantation managed by the local forest administration and considered a forest by local laws/customs);
• list to be continued; values other than yes/no may be combined with a semicolon.

Some users advocated for such a move.

This could also come with a mean to tag the species composing the stand; something like composing_species=* with a list of species in descending order of magnitude.

Refinement of management tagging

Have a tagging scheme to describe the products of the area (wood, earth, sap, mushrooms, pasture…) and the methods of management (pruning, stand improvement, plantation, species selection…) Even in this case, how can you prove that the land is still managed? One will only see traces of past management, which do not prove that the land is still managed. Some plantations are never subject to forestry works before the final clearcut. Between the plantation and the clearcut, no forestry happens; is the land still managed during this time? If so, how do you distinguish for sure with reforestation plantation, which may not be logged at all? There were some advocates of such a solution

The human vs. nature dichotomy; or Should OSM enforce Western worldview?

• Source of the division: XVII century Europe:
  • Francis Bacon (Novuus Organum, 1620; according to EN.WP: Bacon starts the work saying that man is "the minister and interpreter of nature", that "knowledge and human power are synonymous", that "effects are produced by the means of instruments and helps", and that "man while operating can only apply or withdraw natural bodies; nature internally performs the rest", and later that "nature can only be commanded by obeying her". Here is an abstract of the philosophy of this work, that by the knowledge of nature and the using of instruments, man can govern or direct the natural work of nature to produce definite results. Therefore, that man, by seeking knowledge of nature, can reach power over it – and thus reestablish the "Empire of Man over creation", which had been lost by the Fall together with man's original purity. In this way, he believed, would mankind be raised above conditions of helplessness, poverty, and mystery, while coming into a condition of peace, prosperity, and security.)
  • John Locke (Two Treatises to Government, 1689: To properly understand political power and trace its origins, we must consider the state that all people are in naturally. That is a state of perfect freedom of acting and disposing of their own possessions and persons as they think fit within the bounds of the law of nature.)
  • René Descartes (Discourse on the Method, part VI, 1637: For by them I perceived it to be possible to arrive at knowledge highly useful in life; and in room of the speculative philosophy usually taught in the schools, to discover a practical, by means of which, knowing the force and action of fire, water, air the stars, the heavens, and all the other bodies that surround us, as distinctly as we know the various crafts of our artisans, we might also apply them in the same way to all the uses to which they are adapted, and thus render ourselves the lords and possessors of nature.)
  • This was the start of the naturalist ontology: [2]
• Anthropologists: this dichotomy is purely westerner; these notions of nature and culture/managed, if effective, are at least deeply intertwined and not easily separated
  • Philippe Descola:
    • study of the Achuar (a Jivaro people)
    • interview "There is no such thing as nature."
    • "Mais en dépit de cette utilité tactique que je reconnais à l’idée de nature, il me semble nécessaire de répéter que cette notion a fait son temps et qu’il faut maintenant penser sans elle."
  • Alessandro Pignocchi, cognitive science researcher and art philosopher:
    • "Un des drames de l’humanité et des questions d’identité, c’est que les connaissances et les valeurs avec lesquelles on organise notre monde nous semblent transparentes, on a l’impression qu’on voit le monde objectivement, que nos valeurs sont universelles, c’est la cause de tous les problèmes. Si nous étions conscients que nos valeurs sont relatives et qu’elles ne sont pas universelles, tout irait beaucoup mieux."
    • About the "colonisation" of "wild lands" such as the Amazonian forest, and their representation as modernity progressing through nature: "Les sciences cognitives, dont on pourrait croire qu’elles progressaient sur la nature qui est en nous comme des tronçonneuses dans la forêt, voient cette nature ressurgir partout, y compris du mauvais côté de la frontière, du côté de ce qui était supposé être pleinement humain. Les facultés de « haut niveau » — le langage, le raisonnement logique, tout ce qui sous-tend les phénomènes les plus éminemment culturels — apparaissent pétries de « nature ». Et ce surgissement est général ; partout prolifèrent ce que le sociologue et philosophe des sciences Bruno Latour appelle des « hybrides », des objets dans lesquels nature et culture sont si entremêlées que leur seule contemplation suffit à faire ressentir l’inadéquation de cette distinction. Le changement climatique en est le paradigme, à la fois produit de la technologie et phénomène naturel, enjeu politique et monstre incontrôlable.
      Et alors ? Alors un concept n’existe pleinement que lorsqu’on est capable de se représenter son contraire. Sans « nature » bien identifiable, plus de « culture », sans sauvagerie à conquérir, plus d’humanité à faire progresser."
    • "Parmi les outils de composition du monde partagés par la plupart des Occidentaux contemporains, l’un des plus fondamentaux est la séparation bien nette que nous érigeons entre la nature et la culture. Cette distinction unifie la « nature », la met à distance du monde des hommes et confère aux êtres qui la peuplent, à peu de chose près, un statut d’objets. Associée à d’autres outils presque aussi fondamentaux, comme les notions de travail et de progrès, elle conduit à l’exploitation sans limites des créatures que nous avons rejetées dans la sphère autonome de la nature. Cet outil nous est si familier qu’il est bien difficile d’identifier le rôle organisateur qu’il joue dans la composition de notre monde. […] [L]es Indiens emploient des outils de composition du monde qui diffèrent des nôtres dans certaines de leurs dimensions fondamentales et, en particulier, […] ils font l’économie de la distinction entre nature et culture. En Amazonie, pas de concept de nature ; aucun mot dans les langues indiennes ne la désigne. Les plantes et les animaux sont considérés comme des personnes et les relations que l’on entretient avec eux sont apparentées à des relations sociales."
    • "Pour les Jivaros, par exemple, la nature n’existe pas. Aucun mot, dans aucune langue amazonienne, ne se rapproche même de loin de notre concept occidental de nature. L’anthropologie permet de s’apercevoir qu’un concept qu’on jugeait comme universel — ou « naturel », justement — ne l’est pas."
  • Pablo Servigne and Gauthier Chapelle, L’entraide, l’autre loi de la jungle:
    • Le second mythe est assez complémentaire du premier [la nature repose sur la compétition, qui est son état normal]: il nous dit que nous devrions nous séparer et nous extraire de la nature. D’ailleurs, c’est cette vision du monde — que l’Europe adopte à partir du XVII^e siècle — qui crée le concept même de nature pour décrire précisément ce qui n’est pas humain. Séparation nature/culture, séparation corps/esprit : l’être humain de l’époque des Lumières acquiert la conviction qu’il se distingue des « autres qu’humains » par sa subjectivité, sa réflexivité, son langage symbolique, son esprit.
      Pour expliquer la persistance et la puissance de cette croyance tenace, le philosophe Jean-Claude Michéa nous invite à remonter aux interminables guerres de religion que l’Europe a connues au Moyen Âge. Les philosophes politiques du XVII^e siècle (donc Locke et Hobbes), las de ces conflits et désespérés des comportements humains, ont inventé un cadre politique à l’éthique minimale. Un cadre le plus neutre possible, qui ne demande l’intervention d’aucune religion ou morale et qui rende possible de cohabiter sans nous entretuer : le libéralisme était né.
      Selon Michéa, cette doctrine politique s’est formée à partir de l’imaginaire d’une nature humaine sombre et égoïste, et, par extension, agressive et sanglante. La conclusion — logique — a été de croire que seule une organisation humaine aussi puissante que l’État pouvait nous permettre de sortir collectivement de ce monde « sauvage », et que seul le marché (neutre et protégé par l’État) pouvait nous permettre de satisfaire les besoins de tous en laissant libre cours à nos pulsions égoïstes. Ces croyances ont évidemment façonné toute l’idéologie occidentale, dominée par ce que le sociologue Alain Caillé appelle l’axiomatique de l’intérêt.
    • Le mythe d’une séparation nature/culture est aujourd’hui bien ébranlé, comme en témoignent les travaux d’anthropologues tels que Tim Ingold ou Philippe Descola. Ce dernier, qui a étudié les Achuar en Amazonie, montre bien que notre conception du monde séparée des êtres vivants leur est totalement incompréhensible. À la question: « Laquelle, de la nature ou de la culture, contribue davantage à la personnalité ? », le psychologue canadien Donald Hebb est connu pour avoir répondu par une autre question : « Qu’est-ce qui selon vous contribue davantage à la surface d’un rectangle : sa longueur ou sa largeur ? ».
• Ellis et al., People have shaped most of terrestrial nature for at least 12,000 years, Proceedings of the (US) National Academy of Sciences: Even 12,000y ago, nearly three quarters of Earth’s land was inhabited and therefore shaped by human societies, including more than 95% of temperate and 90% of tropical woodlands. Lands now characterized as “natural,” “intact,” and “wild” generally exhibit long histories of use, as do protected areas and Indigenous lands […].

Consequence: it may simply be impossible to tell which wooded lands are managed (nature) and which are not (culture), because these notions are semantically not opposed, but form a continuum. In that case, it is likely impossible to draw a line and tell "here stops nature and begins culture"; such drawing would be highly dubious and lead to constant turmoil, effectively maintaining the current confusion about forests in OSM.

OR we may simply acknowledge that the issue is too difficult, that human influence over wooded lands and their stands cannot be accurately modelled, and we can simply drop the issue. This point of view likely influenced people who described forestry as a niche interest: they simply want to know if the land is wooded or not, period.