Tuesday, 16 October 2012

T 1741/08 – A Broken Chain

This is an appeal against the refusal of an application concerning a method of entering data into a data processing system. The Examining Division (ED) had found the claims on file to lack inventive step.

Claim 1 of the main request before the Board read (additions with respect to claim 1 as refused are underlined):
1. A method of entering of data in a data processing system comprising:
a) displaying a first horizontally aligned linear sequence (402; 702) of top-level icons (i),
b) displaying a second horizontally aligned linear sequence (420; 902; 1320) of second-level icons for a selected one of the top-level icons, the position of the leading icon of the second linear sequence being vertically aligned with the selected one of the top-level icons of the first linear sequence,
c) displaying at least one data entry field (422; 922; 1022; 1122; 1322; 1422; 1522) for a selected one of the second-level icons for entering the data, wherein the at least one data entry field (422; 922; 1022; 1122; 1322; 1422; 1522) is displayed spatially separated from the first and second linear sequence while the first and second linear sequence is displayed,
d) entering the data into the at least one data entry field,
e) going back to step c) for a consecutive one of the second-level icons until all data entry steps of the second linear sequence have been completed,
f) going back to step b) for a consecutive one of the top-level icons.
The Board confirmed the verdict of the ED:

[2.1.1] Claim 1 relates to a method of entering data in a data processing system using a particular graphical user interface (GUI). The layout of this GUI comprises two horizontally aligned linear sequences of icons. The second sequence is displayed for a selected icon of the first sequence. The leading icon of the second sequence is vertically aligned with the selected icon of the first sequence. In addition, one or more data entry fields are separately displayed for a selected icon of the second sequence.

[2.1.2] In the appealed decision, claim 1 was refused for lack of inventive step without citing a document. In section 6.1, it was stated:
“In the present case, it cannot be seen how data entry is technically made more efficient. In particular, the method of claim 1 still requires the user to enter data for every field for which data is needed.”
[2.1.3] In the grounds, the appellant proposes to take the notorious GUI type “wizards” as the closest prior art […]. They are already mentioned as prior art in the description […]. It is further stated […] that the technical problem of the invention is “making data entry more efficient”. In essence the argument, further elaborated in the oral proceedings (OPs), is that the display of the two rows of icons makes it easier, particularly for an inexperienced user, to identify the stage reached in a process of data input requiring a number of steps and sub-steps. The user will, as a consequence, grasp more quickly the nature of the data to be input at the present stage, and therefore respond more quickly than if the rows of icons were not displayed. The appellant further asserts, and the board accepts at least for the sake of argument, that less time-consuming input transactions have the technical effect that less computer resources are used.

[2.1.4] Claim 1 differs from wizards in what is displayed on the screen, i.e. in the particular GUI layout. Since the difference between the claim and D2 or the Microsoft Windows 95 start menu would also consist solely of particularities of the GUI layout, the board accepts that wizards are as good a starting point as either of these others for discussing the question of inventive step.

[2.1.5] It is established case law of the Boards of Appeal, and the appellant has not disputed this, that the subject-matter of a claim cannot be inventive if there is no technical contribution to the art, i.e. if there is no technical problem solved by the claimed subject-matter vis-à-vis the closest prior art, or equivalently if there is no additional technical effect over and above any technical effects present in the closest prior art. The appellant has argued that the reduction in the use of computer resources explained above is the required additional technical effect in the present case. The appellant further argues that any layout improvement which has the effect of “lowering the cognitive burden of the user”, at least in the context of an input operation, should therefore be at least potentially the subject of a patent. In other words a particular layout of information, not specifying the nature of the content, should not be considered to be a “presentation of information” in the sense of A 52(2)(d). The appellant seemed however to accept in the OPs that such an effect relying on a particular content of information, rather than its layout, would not be a patentable contribution. The board notes that this fact provides something of a “reality check” on the appellant’s argument, since the argument would apply equally well to amendments of content, such as replacing an instruction to move a cursor “vertically or horizontally” by “up or down”, as to amendments of layout.

[2.1.6] The board considers that a particular GUI layout could indeed shorten the search of an inexperienced user for where or what data to enter. As a result, less computer resources may be used. However, this reduction in use of resources would be caused by the way the brain of the user perceives and processes the visual information given by a particular way of presenting information.

The appellant is effectively arguing that there is a chain of effects: the improved layout (which is indeed a “presentation of information” according to case law - see below) “lowers the cognitive burden” for the user; the user therefore responds more quickly; and the computer therefore requires less resources. But in terms of technical effects this is a broken chain: the layout has an effect on the mind of the user; a mental transition takes place more quickly than in the prior art; the user responds more quickly, so that the computer uses less resources. Only the third of these links can be called a technical effect, in that the user leaving the computer idle for a shorter time than in the prior art reduces resource consumption.

The board does not accept that such a broken chain can be used as evidence of the required technical effect overall. It would seem that each of the links must be technical in nature for such a chain argument to be persuasive. Thus, the appellant has not established that there is an additional technical effect caused by the improved layout: the layout produces a psychological effect on the user; the user produces a technical effect on the computer. This is not the same as saying that the layout produces a technical effect on the computer.

[2.1.7] Nonetheless, the board must consider whether perhaps the first step alone can be considered as causing an additional technical effect / solving a technical problem. In this context, it is necessary to consider the appeal cases cited during these proceedings. None of them relied on such a “chain” argument as employed by the appellant here. All however concerned effects on the perception of a user, one way or another. However in some cases, there was considered to be a technical effect and in others not.

[2.1.8] The appealed decision states […] that “human perception phenomena cannot be qualified as being of a technical nature”. The decision cites T 619/02 [2.3.2] as a basis for this general statement. The board agrees with this as a general statement.

[2.1.9] The board does agree with the appellant […] that the cited decision (T 619/02) is rather different to the present case. It relates to odour selection testing, and not to GUI design. The appellant argued during the OPs that the claim of the main request in T 619/02 contains “monitoring implicit odour memory by monitoring speed of response and subject confidence of accuracy of response in the subsequent step of odour recognition”. This meant that the claim monitored the response of the user, which was not the case here. The board agrees with the appellant that the claimed invention does not monitor the user’s behaviour.

However, the board nonetheless endorses the statements in the cited decision about human perception phenomena at least usually depending on “personal factors (cultural background, gender, age, past experiences, capacity to evoke dormant meanings and emotions, perception subjectivity, etc.)”. This observation also applies to the effects of GUIs in general. Merely the kind of perception is different.

[2.1.10] There is well-established case law which considers a GUI layout as such to be non-technical, being a “presentation of information” (A 52(2)(d)). In support of this position, the board cited T 1143/06. It concerns the representation of files satisfying some selection criteria as objects moving on a display screen, the speed of movement conveying information relating to the selection. On the issue of whether this representation is to be regarded as a presentation of information in the sense of A 52(2)(d), the decision argues as follows […]:
“Returning now to the present invention, the movement of the elements symbolising the data files over the screen is intended to convey information. This is clear from the patent application itself: ‘Patterns in the data are readily recognisable since each element moves in accordance with the relevance of the sort statements to the data file it represents’ (p.3, 2nd sentence). Regarded in isolation this feature must be held to be ‘presentation of information’ in the sense of A 52(2)(d). In its claim context the feature can therefore only contribute to an inventive step if it additionally produces a technical effect.”
The present board agrees with this view, which is equally applicable to the present case; the lines of icons displayed are also intended to convey information, namely at which step in the input process the user finds him- or herself.

[2.1.11] Decision T 1143/06 discusses a number of previous cases, some of which directly concerned the effects of particular layouts of information, others of which discussed the more general context of technical and non-technical effects. Amongst the cases discussed are T 49/04 and T 643/00, both of which have also been cited in the course of the present case. Decision T 1143/06 explicitly states that it does not follow T 49/04 and that,
“a feature which relates to the manner how cognitive content is conveyed to the user on a screen normally does not contribute to a technical solution to a technical problem. An exception would be if the manner of presentation can be shown to have a credible technical effect”. […]
The present board finds the reasoning in T 1143/06 to be convincing and consistent with the case law discussed therein. With respect to T 49/04, therefore, the board does not follow this decision for the reasons given in T 1143/06. The board notes that T 49/04 itself explicitly did not follow a previous decision, T 125/04, which this board would rather endorse.

[2.1.12] As to the other decision T 643/00, discussed in T 1143/06, the board agrees with the appellant […] that T 643/00 comes closer to the present case than T 619/02, since it deals with arrangement of menu items on a screen, and not with odour selection. In T 643/00 [catchword, 16], searching and retrieving images are considered as technical tasks. However, not everything that supports a technical task has itself a technical character. For example, the advice to have a good night’s rest in order to make searching images more “efficient” would not have technical character.

[2.1.13] In T 643/00 [17], displaying several images side-by-side in low resolution and allowing selection and display of an image at higher resolution is said to provide information to the user in the form of a technical tool.

The board noted during OPs that T 643/00 [10,16-17] always mentions the technical feature of a low resolution of the image in addition to the side-by-side arrangement of the images, whereas the present application merely defines the way in which the icons and the data entry field are presented (including their alignments and their spatial relationship). Even the appellant of T 643/00 included the reduced resolution in its argumentation […].

Moreover, the low resolution, permitting the simultaneous display of a number of images, is a distinctive technical feature over the closest prior art document […]. In the OPs, the appellant contradicted this and argued that the closest prior art referred to in the decision already displayed low resolution versions of high resolution images. The board considers that on this point the appellant misreads the decision. The board cannot find any reference in the decision to the prior art disclosing the display of low resolution versions of high resolution images. In section 9 the decision states, “The prior art of document EP-A-0 392 753 seems to aim at a high speed search of images by using the lowest resolution image data of the images hierarchically encoded and stored in the database 5 (...). The details how the search could be done, however, remain in the dark.” This lack of disclosure of display of the low resolution images in the prior art appears in fact to have played a significant role in the assessment of the inventive step in the decision.

Therefore, in the board’s view the technicality in T 643/00 comes from the technical character of images and their resolution, and the technical effect is the ability to display several images simultaneously if low resolution versions of the image are used. In the present case, there is no analogy to the technical feature of an image resolution.

[2.1.14] Decision T 1143/06 [6.5, 1st paragraph] also cites a passage from T 244/00 [12, 3rd paragraph]:
“... the arrangement of the menu items on the screen, if it is not exceptionally determined by technical considerations, is not a technical aspect of a menu-driven control system.”
The board which decided T 643/00 (which also cites T 244/00) apparently considered that their case was such an exception.

[2.1.15] In its letter dated 15 June 2012, the appellant cited T 928/03 which was said to confirm “that visual information on a GUI which enables a user to perform a task on the GUI in interaction with a data processing system more efficiently is of technical character” […]. During OPs, the appellant pointed to […] its letter where the principle of “conflicting technical requirements” of T 928/03 [4.3] was transferred to the present application. The latter was said to resolve the technical conflict to present the user as much information as possible, but not too much in order to avoid confusion of the user about which data entry step he is actually performing.

The board cannot follow this argumentation, since choosing what to display is part of the presentation of information. If the choice of what to display were technical, then every GUI layout would be technical. The application contains several such choices, e.g. in figure 8 contains additional selection possibilities […].

Decision T 928/03 concerns a video game involving two teams of players and a ball (e.g. football). An image of the virtual game is presented to the user. In addition to presenting a “realistic” view of (part of) the playing field including players and ball, there are markers displayed, showing for example which player has possession of the ball and which is the nearest to pass to. Given the requirement to present a realistic view of the playing field in T 928/03, one object in a scene may obscure another one because of the geometrical constraints of viewing the scene from a particular point of view (thus giving rise to the conflict which the decision refers to). The invention in that case solved this technical problem by changing the size and positioning of the markers. On this basis, the board accepted that there was a technical effect. But in the present case, there is no scene. There is also no technical conflict solved by separating items and data entry fields spatially. The only effect (e.g. of avoiding the user being confused in order to improve inputting) emerges simply from the psychological reaction of the user.

[2.1.16] Applicant’s aforementioned letter alleges that T 333/95 is the base of T 928/03 [...]. This is not precisely correct; the latter mentions the former but does not base itself upon it. Be that as it may, the appellant’s letter further states that T 333/95 holds that a technical contribution can be achieved by a GUI which decreases the mental and physical effort of the user. This is said to be the situation of the present application.

Decision T 333/95 concerns using a computer to produce animations. In order to produce a scene in which a particular object moves, the selected object takes the place of the cursor, so that the user can move it around with the mouse as desired, while the system records the movements of the mouse and translates them into a script of commands for moving the object in the later display of the animated scene. However, the board in that case clearly considered the feature of “making said graphics object the current cursor” in T 333/95 as a technical feature per se. This “graphics object cursor” replaces the normal cursor (also technical) and its movements are recorded and translated into a kind of programming language […]. There was at least a technical effect in allowing the user to move the selected object around under control of the mouse. The decision does not say that every GUI design that makes user’s inputting more efficient has a technical character. Therefore, this decision does not apply to the present case.

[2.1.17] Thus, in all of the cases T 643/00, T 928/03 and T 333/95 the relevant board identified a specific technical effect, which made these cases exceptional. There was something other than the simple choice of what information to display and with what layout to display it, which meant that in these exceptional cases the displayed information might play a part in the assessment of inventive step of the claimed invention. In the present case the appellant has not put forward any convincing argument that there is such a technical effect. Neither is any such effect evident to the board as a result of its own analysis. In the absence of any technical effect arising from the claimed layout features, it is confirmed that claim 1 of the main request is not inventive vis-à-vis the well-known “wizards” for computer input.

Should you wish to download the whole decision, just click here.

The file wrapper can be found here.


Samuel Adams said...

In section 3.2 of the reasons the Board states that T 49/04 is a "single case deviating from the general case law".

This view appears to be incorrect. In particular, T 336/07 (reasons 4.3.3), T 1023/06 (reasons 3.7.3) T 1793/07 (reasons 3.5.1), and T 509/07 (reasons 5.2.2) all cite and seem to follow T 49/04.

Myshkin said...

I would say Board 3.5.04 in T 509/07 did not need to decide whether they agreed with T 49/04 or not. This is similar to point 4.1.2 of Board 3.5.01's T 928/03 (Konami): "even if that interpretation is followed ..." the shape of the guide mark only has an aesthetic effect. These decisions neither follow nor contradict T 49/04.

I agree however that T 336/07, T 1023/06 and T 1793/07 (all from Board 3.2.04) quite explicitly follow T 49/04 (from Board 3.4.03). The line taken by T 49/04 certainly seems determinative of the outcome of those decisions.

On the other side we have at least T 1143/06 and T 1575/07 (Board 3.5.01) and now T 1741/08 (Board 3.5.05), which reject T 49/04.

A referral to the EBA does not seem inappropriate to me. But what will happen if the EBA decides in favour of T 49/04? Will applicants start to submit experimental evidence gathered through the use of test panels to make alleged reductions in cognitive burden for certain categories of users credible?