Taking the measure of the prior art (T 1943/15)

A picture may speak a thousand words, but how many of those words are clear and unambiguous? In a recently published Board of Appeal decision (T 1943/15), the Board considered the circumstances under which a disclosure of a component's dimension may be derived solely from the patent drawings, either by eye or by measurement. 

What do patent drawings disclose?

As with any disclosure, the disclosure of a drawing used in a novelty attack should be clearly and unambiguously derivable. In order to assess whether a measurement in a drawing is clear and unambiguous, it is necessary to assess what technical teaching a skilled person could be said to derive from the drawing (T 204/83). Consequently, dimensions derived solely by measuring a patent drawing (and not complemented by the description) are not considered part of the disclosure of a patent.

What about relative measurements? Are these disclosed by a patent drawing? In a previous decision of the Boards of Appeal, the relative size of two components deduced solely from the drawings in a patent were not considered part of the patent's disclosure (T 1664/06). Several other decisions, based on different facts, have reached the opposite conclusion. In one decision, the relative thickness of two drawing components were considered disclosed, as the relative thickness of the components was identified in the specification as an essential component of the invention (T 748/91). Practical technical teaching could thus be said to be derivable from measurement of the drawings. In other decisions, the drawing in a patent were considered “quite” precise (T 1200/05) or akin to construction drawings (T 422/95). Measurement of the drawings could therefore again be said to confer practical teaching.

Drawing

Of course, it is not always necessary to measure a drawing. It may also be possible to obtain information on the relative size of components just by looking closely at the drawing. The case law provides no general rules by which the disclosure of drawings deduced by eye should be assessed (T 204/83). This is instead determined on a case-by-case basis.

Measuring the threads

The patent in question (EP 2366075) in the most recent Board of Appeal decision on the issue of drawing measurements, related to oil extraction machinery. A novelty attack against the claims relied on dimensions derived from the drawings of the prior art. The granted claim related to a threaded connection between a first and second tube. The threads of the first tube increased in size towards the end of the tube, whilst the threads of the second tube correspondingly decreased in size towards the end of the tube. The relative max and min sizes of the threads on the two tubes were characterised as having a ratio of greater or equal to 0.2.

The Board of Appeal were satisfied that the lower limit of the range (0.2) could not be clearly and unambiguously derived merely by looking at the drawings of the prior art. Particularly, inferring by eye that one dimension was five times the size of a second dimension was said by the Board to verge on speculation. The Board contrasted the situation to one in which an open range was claimed (i.e. one dimension is greater than another). The Board acknowledged that it may sometimes be clear that one dimension is intended to be larger than another. However, this is different to determining the ratio of the dimensions, which would require mentally ascribing sizes to the dimensions. The skilled person would therefore have to measure the drawings of the prior art.

After reviewing the case law, the Board of Appeal (3.2.05) concluded that any patent drawing falling short of the standard of construction drawings could not considered part of the clear and unambiguous disclosure of the patent. Construction drawings, the Board reasoned, could generally be relied on to show dimensions and proportions of elements to scale. In the opinion of the Board, even patent drawings in the mechanical field could not be relied on to the same extent as construction drawings.

The drawings of the prior art in question were not considered by the Board to correspond to the standard of construction drawings. The Board was not even convinced that the drawings of the prior art were to scale. Multiple dimensions were also missing from the drawings which, in the Board’s opinion, meant that there was no way telling whether the figures actually showed the relevant part of the threads of the tubes. The claims were therefore found to be novel in view of the prior art.

Is this latest Board of Appeal decision in line with all those of the previous Boards? A 2007 decision cited by the Board itself considered a measurement from the drawings of a patent as prior art for the purposes of an inventive step attack. These drawings were considered "precise", but the quality of the drawings compared to construction drawings was not considered. However, as highlighted throughout the case law on this issue, establishing rules for assessing the disclosure of drawings is difficult. As a general principle however, relying on measurements from drawings either for basis or a novelty attack can be an uphill battle. A picture may speak a thousand words, but words are clearer.