The Journal of Peer Production - New perspectives on the implications of peer production for social change New perspectives on the implications of peer production for social change
Ten patterns developed by the Oekonux project image

by Stefan Meretz.

Read also the responses by Maurizio Teli and Toni Prug.


The Oekonux project seeks to establish a new basis for analyzing a new historical phenomenon: the emergence of peer production, starting with the creation of Free Software. If the initial hypotheses of Free Software being the germ form of a new mode of production beyond capitalism is valid, it would be necessary to develop new epistemological patterns to be able to analyze it adequately. This requires understanding and criticizing old analytical notions as historical products of the outlived capitalist way of producing our livelihood, including those which aim to be in opposition to capitalism. In this paper I present ten patterns which have emerged from the debates of the Oekonux Project. They demonstrate what it means to go beyond traditional affirmative and traditional oppositional or “leftist” patterns of analysis. Although taken from the debates in the Oekonux Project, these have never yet been presented in such a condensed way. Obviously not all patterns will be shared by all the participants of these debates, because in the end these are my personal conclusions drawn from over ten years of discussion.


In this text I will try to give some introduction to the main ideas which have been developed since the foundation of the Oekonux project in 1999. There is no fixed set of thoughts and personally I have my own perspective on Oekonux ideas.

Why is the Oekonux project so relevant for debates around commons-based peer production? There are two reasons. First, Oekonux developed many of the ideas many researchers are so familiar with many years before they reached a wider audience. Oekonux was founded as a project of reflection around Free Software, but from the beginning the question of generalizing observations about Free Software to other realms of immaterial as well as material goods was present. When Yochai Benkler (2006) coined the term commons-based peer production it only condensed a year-old debate into a catchy notion, but the insights itself were not very new and sound very familiar to Oekonux participants. Consequently the term has been adopted by the Oekonux project.

Second, Oekonux participants have gone much further than others in questioning the accepted way of thinking. New theses have been developed which did not only reject traditional discourse patterns in computer sciences, sociology, and economics, but also in emancipatory political and theoretical approaches. Stefan Merten, the founder of Oekonux who comes from an anarchist-marxist background, provocatively rejects “leftist and other capitalist ideologies” (Merten 2011) for the analysis of peer production. This sounds quite post-modern, but was meant differently: All means of emancipation are going to be developed right in front of our eyes, but we also have to grasp them theoretically. Traditional leftist patterns are not able to do that, because they adhere to the given mode of production for whose analysis they are made.

This was an enormous provocation to many people, traditionalists on all sides. And there have been many cultural and political clashes within the project. But there also have been a core of people, who continuously drove the Oekonux approach further. In the following I try to describe some Oekonux patterns, which of course represent my interpretation of the Oekonux debate. When I use the past when talking about Oekonux, it is not because the project does no longer exists. It still exists, and the Critical Studies in Peer Production journal is not the only spin-off of the project, there have been many others, so that the focus decentralizes to diverse projects inspired by Oekonux.

In an interview with Joanne Richardson Stefan Merten (2001) described Oekonux as a project to evaluate Free Software with respect to its “potential for a different society beyond labor, money, exchange”. Here, he gives the keywords Oekonux thinking was built around. I will take and extend them to illustrate why and how the main ideas contradict traditional leftist thinking so much, especially when Oekonux started in 1999 (Merten 1999).

Pattern 1: Beyond Exchange

Free Software, or more generally, commons-based peer production is not about exchange. Giving and taking are not coupled with each other. From today’s perspective this might not be surprising, but at the beginning of the Oekonux project it was. Still today traditional Leftist approaches are based on the assumption that someone is only allowed to get something, if s/he is willing and able to give something back, because if everybody is only taking then society would perish. This position could reference to a painful Socialist (and Christian) tradition saying that the one who does not want to work, should not eat. However, Free Software clearly showed that developers do not need to be forced to do what they love to do (cf. pattern 5).

One important approach which tried to grasp the new developments of Free Software, although sticking with old thinking, was the “gift economy” approach. However it is not coincidental that the correct term should be “gift exchange economy”: The giver can expect to get something back, because it is a moral duty in societies based on the exchange of gifts. This kind of personal reciprocal duty does not exist in Free Software. Even if a developer says that s/he wants to “give something back”, then this giving is not a precondition to receive something. In general, commons-based peer production is based on unconditional voluntary contributions.
From a Leftist perspective, uncoupled giving and taking could only be possible in a mythical land in a distant future called Communism – if at all. But never today, because before communism is possible, an unfriendly interphase called Socialism sticking with the exchange dogma is necessary (cf. pattern 8). Historically, “real existing Socialism” trying to implement this necessity failed, which will happen with all Socialist approaches accepting the exchange dogma.

If one does not want to give up exchange, then capitalism is the only option.

Pattern 2: Beyond Scarcity

It is a common misconception that material things are scarce while immaterial things are not. It seems justified to keep material goods as commodities while immaterial goods are required to be free. However, this assumption turns a social property into a natural one. No produced good is scarce by nature. Scarcity is a result of goods being produced as commodities, thus scarcity is a social aspect of a commodity created for a market. In the digital era this is obvious for immaterial goods, as we can clearly see the measures to artificially make the good scarce. Such measures include laws (based on so-called “intellectual property”) and technical barriers to prevent free access to the good. It seems to be less obvious for material goods, because we are used to the non-accessibility of material goods unless we have paid for them. But the measures are the same: law and technical barriers, accompanied by continuous destruction of goods to keep the commodities rare enough to obtain a suitable price on markets.

Furthermore it seems obvious that we all depend on material goods which may not be available in sufficient amount. Even immaterial goods depend on a material infrastructure, at least our brains (in the case of knowledge), which also need to be fed. This is definitely true, however, it has nothing to do with a “natural scarcity”. Since all goods we need are to be produced, the only question is, how they are to be produced in a societal sense. The commodity form is one option, the commons form another. Commodities must be produced in a scarce manner to realize their price on the market. The commons good can be produced according to the needs of the people using the given productive capacity. There might be current limitations, but limits always have been subject to human creativity to overcome them.

Maybe some limitations may never be overcome, but this again is no reason to make goods artificially scarce. In these rare cases social agreements can be used to organize responsible use of the limited resource or good. The commons movement learned that both rival as well as non-rival goods can be produced as commons, but they require different social treatment. While non-rival goods are agreed to be freely accessible to prevent under-use, it makes sense to avoid over-use for rival goods by finding appropriate rules or measures either to organize sustainable use or to extend collective production and thus availability of the rival good.

Scarcity is a social phenomenon which is unavoidable if goods are produced as commodities. Often scarcity is confused with limitations which can be overcome by human efforts and creativity.

Pattern 3: Beyond Commodity

In her studies Elinor Ostrom found, that “neither the state nor the market” is a successful means for commons management (1990). Based on traditional economics she analyzed the practices of natural commons and finally simply proved liberal dogmatics wrong. Markets are not a good way to allocate resources, and the State is not a good way to re-distribute wealth and manage the destructive results of markets. Best results occur if the people organize themselves according to their needs, experiences and creativity and treat resources and goods not as commodities, but as common pool resources.

This is exactly what happens in Free Software. Interestingly it took many years to understand that Free Software is a commons and that it is basically identical to what Elinor Ostrom and others were talking about much earlier. One weak aspect of the traditional commons research and the early phase of Free Software was that a clear notion of a commodity and a non-commodity did not exist. It was the Oekonux Project which clearly said: Free Software is not a commodity. This dictum is closely related to the insight that Free Software is not exchanged (cf. pattern 1).

Critics from the left argued that being a non-commodity is limited to the realm of immaterial goods like software. From their viewpoint Free Software is only an “anomaly” (Nuss, Heinrich 2002), while “normal” goods in capitalism have to be commodities. This assumption, however, is closely linked to the acceptance of the scarcity dogma (cf. pattern 2). Moreover, it treats capitalism as a kind of normal or natural mode of production under conditions of “natural scarcity” (as they think). This view completely turns real relations upside down. Capitalism could only establish itself by enclosing the commons, by depriving the people from their traditional access to resources in order to transform them into workers. This enclosure of the commons is an ongoing process. Capitalism can only exist if it continuously separates people from resources by making them artificially scarce. A commodity – as nice as it may appear in the shopping malls – is a result of an ongoing violent process of enclosure and dispossession.
The same process occurs in software. Proprietary software is a way of dispossessing the scientific and development community from their knowledge, experiences, and creativity. Free Software was first a defensive act of maintaining common goods common. However, since software is at the forefront of the development of productive forces it quickly turned into a creative process of overcoming the limitations and alienations of proprietary software. In a special field Free Software established a new mode of production which is going to spread into other realms (cf. pattern 10).

Goods which are not made artificially scarce and are not subject to exchange are not commodities, but commons.

Pattern 4: Beyond Money

Since money only makes sense for commodities, a non-commodity (cf. pattern 3) implies that there is no money involved. Thus Free Software is beyond money. On the other hand, there is obviously a lot of money around Free Software: developers are paid, companies spend money, new companies are formed around Free Software. This has confused a lot of people, even on the left. They stick to an either-or thinking, being unable to think these observations as a contradictory process of parallel development in a societal period of transition (cf. pattern 10).

Money is not a neutral tool, money can occur in different social settings. It can be wage money, invested money (capital), profit, cash money etc. Different functions have to be analyzed differently. In Free Software there is no commodity form involved, so money in the narrow sense of selling a commodity for a price does not exist. However, Eric Raymond explained how to make money using a non-commodity: by combining it with a scarce good. In a capitalist society where only a few goods had broken out of the commodity realm, it is beyond question that all other goods continue to exist as commodities. They are kept scarce and they are combined with a priceless good. Using a perspective of valorization this is nothing new (e.g. spreading gifts to attract customers). Using a perspective of recognizing a germ form this way a new mode of production starts to develop within the still existing old model.
But why do companies give money if this money is not an investment in the traditional sense, but a kind of a donation, e.g. to pay Free Software developers? Why did IBM put one billion dollars into Free Software? Because they were forced to do so. Economically speaking they have to devalue one business area to save the other profit-making areas. They have to burn money to create a costly environment for their sales (e.g. server hardware). As the enclosure of the commons is a precondition for capitalism, the other way around is also true. Extending the commons in a field currently dominated by commodities means that this field is replaced by free goods.

However, the “four freedoms” of Free Software – use, study, change, redistribute – (Free Software Foundation, 1996) do not speak about “free” in the sense of “gratuitous”. The slogan “free as in freedom, not in free beer” is legion. This is completely fine and does not contradict the “beyond money” dictum, because the four freedoms do not say anything about money. The four freedoms are about free availability, are about abundance. Thus, the absence of money is an indirect effect. Abundant and thus non-scarce goods cannot be a commodity (cf. pattern 2) and cannot make any money. However, making money is not forbidden per se.

There have been a lot of attempts to integrate the non-exchange, non-commodity, commons-based free circulation of Free Software into the traditional economic paradigm, which is based on exchange and commodity. The most prominent one was the “attention economy” saying that the producers do not exchange goods, but attention (Goldhaber, 1997). They concluded that attention is the new currency. But this was only a desperate attempt do cling to old terms which neither worked properly nor delivered any new insights and thus was not relevant. Various other similar attempts are skipped here.

Being beyond money directly results from not being a commodity.

Pattern 5: Beyond Labor

Free Software and commons in general is beyond labor. This can only be understood if you grasp labor as a productive activity specific to a certain historical form of society. Selling labor power – i.e. the ability to work – to some capitalist who uses it to produce more value than the labor power is worth, is unique in history. This has two important consequences.

First, it turns productive activity – which has always been used by people to produce their livelihood – into alienated labor. This alienation is not imposed by personal domination, but by structural coercion. In capitalism humans can only survive if they pay for their livelihood, which compels people to make money. Making money can be either done by selling their own labor power or by buying and valorizing the labor power of others. The result is a distorted process where structural requirements prescribe what a person has to do (cf. pattern 6).

Second, it creates the homo economicus, the isolated individual seeking for maximization of his/her own utility – if necessary even at the expense of others. Traditional economists then assert that the homo economicus is the archetype of a human being, which confuses the specific historical result with a natural presupposition.

Instead of labor, Free Software is based on Selbstentfaltung. The German notion of Selbstentfaltung is not easy to translate. On the one hand it starts from “scratching an itch” (Eric Raymond), “doing what you really really want” (Fritjof Bergmann), and “having a lot of fun” (the Free Software developer). On the other hand it integrates other fellow developers to strive for the best solution possible. This also means high engagement, passion, and effort, not just picking the low hanging fruits. It includes a positive reciprocity with others striving for the same goal in a way, that the Selbstentfaltung of the one is the precondition of the Selbstentfaltung of the others. Not by chance this is reminiscent of the Communist Manifesto where the “the free development of each is the condition for the free development of all” (Marx, Engels 1848). However, in Free Software it is not a goal of a future society, but it is an inalienable feature of the beginning new mode of production on the way to that new free society.

Instead of selling one’s energy for alienated purposes, usually called labor, Free Software is based on Selbstentfaltung which is the free development of all the productive forces of the people.

Pattern 6: Beyond Classes

Capitalism is a society of separations. Buying vs. selling, producing vs. consuming, labor vs. capital, concrete vs. abstract labor, use value vs. exchange value, private production vs. social distribution etc. Capitalist development is driven by the contradictions between these separated parts. Among them, labor and capital is only one contradiction, but it seems to be the most relevant one. A person seems to be defined by being a labor seller or a labor buyer, a worker or capitalist. However, in fact labor and capital are not properties of individuals, but opposite societal functions like all other separations capitalism generates.

Therefore, it is not true that only one side of the various separations represents the general or progressive one. On the contrary, both parts of a separation depend on each other. Labor produces capital, and capital creates labor. It is an alienated cycle of a permanent reproduction of the capitalist forms. Thus, both sides of these separations, e.g. labor and capital, are necessary functions of capitalism. The so called antagonism of labor and capital is in fact a purely immanent mode of historical development of capitalism. The working class does not represent emancipation, by no means.

Free Software and peer production in general is not recreating classes, it is rather beyond that mode. It represents a germ form (cf. pattern 10) of a new mode of production which generally is not based on separations, but on integrating different personal needs, behavior and wishes as a powerful source of development. Exploitation does not exist, because selling and buying of labor does not exist and money can only play a role in retro games about antiquated societies called “capitalism”.

Selbstentfaltung as a free developing human being is the source of societal transition towards a free society, not the class adherence.

Pattern 7: Beyond Exclusion

One of the most basic separations capitalism generates is the separation of those who are inside and those who are not. This inside/outside pattern is not a class separation (cf. pattern 6) and it is not only one big separation. It is a structural mechanism of inclusion and exclusion along all possible lines of society: job-owner vs. jobless, rich vs. poor, men vs. women, people of color vs. white people, bosses vs. subordinated, owners of means of production vs. non-owners, members of social security vs. non-members etc. It has to be recognized as a basic structural principle of capitalism: An inclusion of the one side implies an exclusion of the other side. For the individual this means that any personal progress is realized at the expense of others who stagnate or regress.

In general the commons are beyond the mechanism of exclusion. In Free Software, for example, the more active people join a project the faster and the better a goal can be achieved. Here, the relationship between people is not structured by inclusion-exclusion mechanisms, but by an inclusive reciprocity (Meretz 2012). The maintainer of a project tries to include as many active people as possible, strives for a creative atmosphere, and tries to solve conflicts in a way, that as many people as possible can follow the “rough consensus” and the “running code”.

If a consensus is not possible the best solution is then a fork: a risky but valid option to test different directions of development. If you look at existing forks (e.g. between KDE and GNOME), then many of them are working closely together or maintain an atmosphere of cooperation. Yes, there are other examples of fights against one another. But these non-productive forks are mainly due to alienated interests playing an important role. Oracle tried to implement a command and control regime after having bought OpenOffice as part of the Sun package. The fork to LibreOffice by many important developers was an act of self-defense and self-determination to maintain their environment of Selbstentfaltung. They don’t want to go back into the old “labor mode” of development (cf. pattern 5).

While capitalism is structurally based on exclusion mechanisms, commons-based peer production generally creates and advances inclusion.

Pattern 8: Beyond Socialism

Socialism, as defined by Karl Marx in the “Critique of the Gotha Programme” (Marx, 1875) is a commodity-producing society ruled by the working class. Historically this was realized by the so called “real existing Socialism”. There have been many critiques of real socialist countries (lacking democracy, etc.) from within the left. Nevertheless, a good part of the left shares the assumption that an interphase between a free society (which may be called communism) and capitalism is unavoidable. The general concept is that the working class holding the power can reconstruct the whole economy according to their interests which represent the majority of the society. In short: power comes first, then a new mode of production will follow, in order to build a really free society. This concept has failed historically.

The reason for this failure is not due to internal tactical differences and shortcomings. Instead it is due to the unrealistic concept of qualitative historical transformation. Never in history was the question of power placed first, it was always the new mode of production which emerged from the old way of producing which prepared the historical transition. Capitalism initially developed from craftsmanship in medieval towns, then integrated manufactures, finally leading to the system of big industry. The question of power was solved “on the way”. This does not diminish the role of revolutions, but revolutions only realize and enhance what was already developing. The revolutions of the Arab Spring do not create anything new, but try to realize the potentials of a normal democratic bourgeois society.
This analysis of historical developments (discussed in more detail in pattern 10) has to be applied to the current situation. Historical transition can not be realized by taking over political power – be it by parliament or by street actions – but by developing a new mode of production. The criteria for being “new” can be derived from the negation of the old mode of production: instead of commodities: commons production, instead of exchange and mediation by money: free distribution, instead of labor: Selbstentfaltung, instead of exclusion mechanisms: potential inclusion of all people. However, care needs to be taken since not all developments of capitalism are to be abolished. Rather some continue – though in a transcended form.

Commons-based peer production transcends capitalism as well as commodity-based socialism.

Pattern 9: Beyond Politics

Since commons-based peer production is mainly about constructing a new mode of production, it is basically a non-political movement. Here, politics is understood as addressing the state and its institutions to demand changes in some desired direction. Such politics are based on interests which in capitalism are generally positioned against each other. If a society is structured along inclusion-exclusion patterns (see pattern 7), then it is necessary to organize common but partial interests in order to realize them at the expense of the common partial interests of others. In this sense commons are beyond politics, because they basically do not operate in the realm of interests but of needs.

It is important to distinguish between needs and interests. Needs have to be organized in the form of interests, if the usual mode of realization is the exclusion of the interests of others. Commons on the other hand are based on the variety of needs of their participants, which act as a source of creativity. The mediation of these different needs is part of the process of peer production. Thus, it is not necessary that participants additionally organize their needs as interests and try to implement them politically. Instead, they achieve this directly.

One aspect which makes this clear is the question of hierarchies. Usually hierarchies are part of capitalist commodity production. Therefore, a common left topic was to reject any hierarchies to avoid domination. This ignores the fact that hierarchies as such do not generate domination, but rather the function hierarchies have in a given context. In a company hierarchies express different interests, for example the interests of workers and of the management (cf. pattern 5). However, in a peer production project a hierarchy may express different levels of expertise or different responsibilities, which are shared by those who accept someone in a leading position. Being a maintainer does not mean following different interests at the expense of project members. Such a project would not prosper. On the contrary, a maintainer is keen to integrate as many active and competent members as possible. This does not avoid conflicts, but conflicts are solved on the common base of the project’s goals.

Commons-based peer production does not require to articulate people’s needs in the form of opposing interests and thus is beyond politics.

Pattern 10: Germ Form

Last but not least, the most important pattern is the germ form or five-step-model (Holzkamp, 1983). It is a model to understand the concurrent existence of phenomena with different qualities. When discussing peer production the debate is often dominated by two groups: those who are in favor of peer production and who try to prove peer production is anti-capitalist and those who see peer production only as a modernization of capitalism. The challenge is to think it as both. The germ form model accomplishes this by viewing the emergence and development of commons-based peer production as a process of its own contradictory unfolding in time.

Normally applying the five-step-model is a retrospective procedure where the result of the analyzed development is well known. By mentally assuming the result of a transition towards a free society based on commons-based peer-production the emergence of this result can be reconstructed using the model. Here is a very rough sketch of the five steps applied to the case of peer production.

1. Germ form: A new function appears. In this phase the new function must not be understood as a rich germ or a seed enclosing all properties of the final entity which only has to grow. Rather in this phase the germ form shows only principles of the new, but it is not the new itself. Thus, commons-based peer production is not the new itself, but the qualitatively new aspect it shows is the need-oriented mediation between peers (based on Selbstentfaltung, see pattern 5). During this phase this is visible only on a local level.

2. Crisis: Only if the overall old system falls into a crisis can the germ form leave its niche. The capitalist way of societal production and mediation via commodities, markets, capital, and state has brought mankind into a deep crisis. It has entered a phase of successive degradation and exhaustion of historically accumulated system resources. The recurring financial crisis makes this obvious to everyone.

3. Function shift: The new function leaves its germ form status in the niche and gains relevance for the reproduction of the old system. The former germ form is now double-faced: On the one hand it can be used for the sake of the old system, on the other hand its own logic is and remains incompatible with the logic of the dominant old system. Peer production is usable for purposes of cost-saving and creating new environments for commercial activities, but it rests upon non-commodity development within its own activities (cf. pattern 3). Cooptation and absorption into normal commodity producing cycles are possible (De Angelis, 2007), and only if peer production is able to defend its own commons-based principles and abilities to create networks on this ground will the next step be reached. Free Software as one example of peer production quite clearly is at this stage.

4. Dominance shift: The new function becomes prevalent. The old function does not disappear immediately, but steps back as the previously dominant function to marginal domains. Commons-based peer production has reached a network density on a global level, so that input-output links are closed to self-contained loops. Separated private production with subsequent market mediation using money is no longer required. Need-based societal mediation organizes production and distribution. The entire system has now qualitatively changed its character.

5. Restructuring: The direction of development, the backbone structures, and the basic functional logics have changed. This process embraces more and more societal fields which refocus towards the new need-based mode of societal mediation. The state is stripped down, new institutions emerge, which no longer have a uniform State character, but are means of collective Selbstentfaltung (cf. pattern 5). New contradictions may come up, a new cycle of development may begin.

This is only an epistemological model, not a scheme for immediate action. The main advantage is the possibility to escape unfruitful either-or debates. It allows for thinking the emergence of a new mode of production being useful for the old system while maintaining its transcending function towards a free society as concurrent phenomena.

The germ form model adapted in the Oekonux context is a dialectical conceptualization of historical transition.


Far from being a consistent theory of historical transition towards a free society these patterns give a fairly good impression of why they don’t fit into any of the traditional approaches. There might be some accordances with one approach or the other, and most of the Oekonux participants will not agree with all of the patterns, but no single approach could answer to all challenges at once in a consistent way.

This is not coincidental. On the one hand the formation of a new society can not be entirely grasped in terms of the already fully developed society which is going to be made history. On the other hand, there are overarching aspects which continue to exist in all societies, but which undergo a reconfiguration. Other aspects dissolve completely. And finally some aspects are leveraged in a way that they hardly have anything in common with their origin. These three forms of transition – preservation, dissolution, leverage – describe the meaning of what G.W.F. Hegel called sublation (Aufhebung). Ten patterns of societal transition presented in this paper try to fulfill this requirement.


Special thanks to Stefan Merten and Mathieu O’Neil for editing support. Tomislav Knaffl gave valuable hints.

Works cited

Benkler, Y. (2006), The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven: Yale University Press, URL: (2011-10-10)

De Angelis, M. (2007), The Beginning of History. Value Struggles and Global Capital, London: Pluto Press.

Free Software Foundation (1996), The Free Software Definition, URL: (2011-10-10)

Goldhaber, M.H. (1997), The Attention Economy and the Net, in: First Monday, Vol. 2, No. 4, URL: (2011-10-10)
Holzkamp, K. (1983), Grundlegung der Psychologie, Frankfurt/Main, New York: Campus.

Marx, K., Engels, F. (1848), Manifesto of the Communist Party, URL: (2011-10-10)

Marx, K. (1875), Critique of the Gotha Programme, URL: (2011-10-10)

Meretz, S. (2012), The Structural Communality of the Commons, In: Bollier, D. et al. (2012), Self-Sustaining Abundance, in print.

Merten, S. (1999), Willkommen bei ‘oekonux’, URL: (2011-10-10)

Merten, S. (2011), Leftist and other capitalist ideologies and peer production, URL: (2011-10-10)

Merten, S., Richardson, J. (2001), Free Software & GPL Society. Stefan Merten of Oekonux interviewed by Joanne Richardson, URL: (2011-10-10)

Nuss, S., Heinrich, M. (2002), Freie Software und Kapitalismus, in: Streifzüge 1/2002, URL: (2011-10-10)

Ostrom, E. (1990), Governing the Commons. The Evolution of Institutions for Collective Action, Cambridge: Cambridge University Press.