Search This Blog

Labels

T 1273/11 - Provoking appeal

Of interest in this opposition appeal is the request of the respondent that his costs are to be apportioned to the appellant as, according to the respondent, the costs related to the present appeal have been deliberately provoked by the appellant because he withheld relevant material from the first instance proceedings, and therefore prevented the opposition division from taking a complete decision encompassing also this material.

Of further interest is the Board deciding on admissibility of late filed requests, which according to the respondent should be admitted since they are a reaction to the communication of the Board and have been submitted timely before the oral proceedings, such that the appellant had enough time to do a search and prepare for discussing them.

Summary of Facts and Submissions

I. The appellant (opponent) lodged an appeal against the decision rejecting the opposition against European patent EP-B-1 474 347, and requested that the decision under appeal be set aside and that the patent be revoked.

He also submitted further written evidence and offered two new witnesses.

II. The respondent (patent proprietor) requested that the appeal be dismissed, alternatively, that the decision under appeal be set aside and the patent be maintained on the basis of any of the auxiliary requests 1 to 4, filed with the statement setting out the grounds of appeal, or on the basis of any of the auxiliary requests 5 or 6, filed with the letter dated 23 June 2015.

Additionally the respondent requested not to admit the new evidence submitted with the statement of grounds of appeal, and that his costs be apportioned to the appellant.

T 1580/13 - Catch-up growth without risk of obesity



Here, the Board sets aside the decision of the opposition division by indicating that no inventive step could be recognized because the prior art that showed that 'catch-up growth' could be achieved using the compounds of the claimed invention. The term 'catch-up growth' was well defined in the specification and could therefore not be taken from other documents. Catch-up growth was in fact defined as a sudden spurt of growth that would result in catching up with subjects having a normal growth, although it may be that the subject would never attain the physical state that it would have reached had the stress not been suffered. Despite the arguments from the proprietor, the Board held that the problem of catch-up growth without an increase in caloric intake, with equilibrated lean and fat body mass and without promoting obesity had not been solved, because such could not be derived from the data in the specification.
Summary of Facts and Submissions
I. This decision concerns the appeal filed by opponent 1 (The IAMS Company), opponent 2 (Abbott Laboratories) and opponent 3 (N.V. Nutricia) against the interlocutory decision of the opposition division that European patent No. 1 940 248 as amended meets the requirements of the EPC.
II. The opponents had requested revocation of the patent in its entirety on the grounds under Article 100(a) EPC (lack of novelty and inventive step), Article 100(b) EPC and Article 100(c) EPC. [...]

T 557/13 - Referral: Mother poisonous for child?


This case refers five questions to the Enlarged Board of Appeal with respect to what is often referred to as partial priority, poisonous divisionals and poisonous priorities. This case has been discussed earlier in this blog, see our post "T 557/13 - Poisonous or nothing wrong?". 
The opposition decision under appeal relates to a patent that is a divisional from  D1 (which is an EP patent enjoying priority from D16). The divisional also claims priority from D16. The Opposition Division decided that claim 1 of the patent of this appeal case does not enjoy the priority from D16 and, thus, the filing date of D1 is the effective date of claim 1. It may be that the opinion of the Opposition Division is now summarized too much, but, according to the Opposition Division is D1 is an Art. 54(3) EPC document because the effective date of D1 is the priority date of D16, and D1 is published after the effective date of claim 1. 
An additional complication in this case is that claim 1 of the divisional application is a so-termed "generic 'OR'-claim" (see point 8.2.2 of this decision for a definition). Subsequently one could argue that a first portion of claim 1 enjoys priority from D16 and D1 is not an Art. 54(3) EPC document for this first portion, while another second portion of claim 1 does not enjoy priority from D1 and, consequently, the novelty of this second portion must be examined over D1 as an Art. 54(3) EPC document. 
Furthermore, the parties to the appeal proceedings had a diverging opinion about "Can a parent application of a divisional application be a novelty destroying Art. 54(3) EPC document?".
The Board of Appeal could not decide on these subjects because, as extensively discussed in the decision, the case law diverges in different directions. Therefore the Board formulated five questions for the Enlarged Board of Appeal. 

T 1279/10 - Deletion of reference to non-public DVD standard: problematic or OK?


The application contained several references to a (non-public) DVD standardization document including references with respect to certain terminology used, such as the term "video cell". The Examining Division required the applicant to delete the references to the (non-public) DVD standardization document. The Examining Division had subsequently refused the application after objections under Articles 83, 84 and 123(2) EPC, as the Examining Divison considered the skilled person to need non-public information to be able to understand the invention, the claims to be not supported by the description, and to extent the meaning of the term "video cell" beyond what was disclosed in the application as originally filed by its meaning no longer being defined by the standard document. The Board of Appeal assessed each of the objections and the assumed necessity of the standard document in detail in view of what is clear to the the skilled person from his common general knowledge and to what extent the invention is applicable to DVD only or broader.

T 217/10 Unsubstantiated auxiliary requests




The Board refuses to examine auxiliary requests I-V. The Board extensively discusses the Rules of procedure of the Boards of appeal and the interpretation thereof. Although the auxiliary requests were filed in time, they were not substantiated such that the Board could not see how prima facie these requests could overcome the objections against the main request. If auxiliary requests are submitted, it usually requires a justification to what extent the objections against the main request are overcome. If the auxiliary requests are not substantiated at filing but at a very late stage, a justification for the late submission is required.

T 2049/10 - No clarity or lack of support?





This Examination appeal concerns a decision to refuse a claim solely on the basis of clarity. The refused claim read:

1. A portable computer, comprising:

a display assembly coupled to a base assembly, the base assembly having palm rest areas positioned to support a user's palms;
a touchpad disposed on the base assembly , wherein the palm rest areas are formed by the touchpad,  wherein the touchpad includes a sensor that can detect an object on a surface of the touchpad, and wherein the portable computer estimates a probability that the object detected on the surface of the touchpad is an intentional contact.


Four different clarity objections were raised against this claim: The claim has palm rest areas, but no keyboard. The claim has a 'sensor', instead of a 'hand location sensor'. The claim introduces the surface of the touchpad with 'a surface of the touchpad', which should be 'the'. The claim does not specify the type of input provided by the touchpad that allows calculation of the probability.

The board considers clarity satisfied but considers support instead. Some observations are made regarding the claim that "the invention opens up a whole new field and is entitled to more generality in the claims".


In the end the Applicant settles and includes a keyboard and a hand location sensor in the claim. The application is remitted to allow the Examining division to decide on the inventiveness. 

The following claim is the main request that was eventually remitted:



1. A portable computer , comprising:
a display assembly coupled to a base assembly , the base assembly having palm rest areas positioned to support a user's palms; a keyboard disposed on the base assembly;
a touchpad disposed on the base assembly , wherein the palm rest areas are formed by the touchpad ; and a hand location sensor being able to detect a hand location when a user's hand is positioned on the keyboard ,
 wherein the portable computer filters each contact sensed by the touchpad to either accept the contact as an intentional input command, or reject the contact as unintentional, based on the hand location detected by the hand location sensor."

T 1602/10 - Those generalisation days are truly over


Here, the Board comes to the same conclusion as the ED albeit with another reasoning. Where the ED stated that the patent could not be granted because the disclosure showed that EAE mice treated with anti-TIM-3 antibodies exhibited exacerbated disease, which was not considered a therapeutic benefit and which therefore did not enable the skilled person to carry out the claimed invention (Art 83 EPC) [because an anti-cancer effect was claimed]. The Board on the other hand also denies patentability under Art 83 EPC but because cancer is not a single disease, not all cancers exhibit a TIM-3 mediated immune response and antibodies can have agonistic as well as antagonistic effects. It was concluded that in view of the potentially opposite actions of anti-TIM-3 antibodies, the skilled person would not have considered it plausible that substantially all embodiments of the invention defined in claim 1 were capable of being realised. The treatment of a single antibody in a single type of cancer does in this case not provide evidence in support of the entire scope.
Summary of Facts and Submissions
I. An appeal was filed by the patent applicant (appellant) against the decision of the examining division to refuse European patent application No. 03 723 627. The application, entitled "Compositions and methods related to TIM-3, a Th1-specific cell surface molecule" was filed as an international patent application and published as WO 03/063792.
II. The examining division dealt with a main and an auxiliary request. Claims 5 and 6 of the main request were held to concern an invention which was not disclosed in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art, contrary to Article 83 EPC. The same objection was held to apply to the subject-matter of claim 5 of the auxiliary request. These were the sole objections leading to the refusal of the application. The reason given for the refusal was that, although the application disclosed that mice with experimental autoimmune encephalomyelitis (EAE) treated with anti-TIM-3 antibodies exhibited exacerbated disease, this was not a therapeutic benefit and did not enable the skilled person to carry out the claimed invention.
It was held that post-published evidence of reduction of tumour growth in a tumour model by treatment with anti-TIM-3 antibody could not remedy the deficiency because sufficiency of disclosure had to be established at the filing or priority date of the application, as the case may be. Post-published documents could only be used to back up disclosures already made in the patent application but could not establish sufficiency of disclosure on their own.
III. With the statement of grounds of appeal, the appellant submitted a main and three auxiliary claim requests.
IV. With a letter dated 16 February 2015, in response to a communication of the board, the appellant filed a new main claim request replacing all previous claim requests.
V. Oral proceedings before the board were held on 17 March 2015. At the end of these proceedings, the chairwoman announced the decision of the board.
VI. The final request of the appellant was that the decision of the examining division be set aside and that a patent be granted on the basis of the set of claims filed as main request together with the letter dated 16 February 2015.
VII. Claims 1 and 2 of the sole request read:
"1. A TIM-3 binding molecule wherein the TIM-3 binding molecule is an antibody specific for TIM-3 or is a fragment of an antibody specific for TIM-3, for use in the treatment of cancer in a subject.
2. The TIM-3 binding molecule for use as claimed in claim 1, wherein said TIM-3 binding molecule binds to an extracellular region of TIM-3."
[...]
IX. The arguments of the appellant can be summarised as follows:
The experimental results disclosed in the application related to administering an antibody against TIM-3 in vivo in a mouse model of experimental autoimmune encephalitis (EAE) which is a model of the human autoimmune disorder multiple sclerosis. In an autoimmune disease a subject's own antibodies react with host tissues or the immune effector T cells are autoreactive to endogenous self-peptides and cause destruction of tissue, i.e. an anti-self response. The inventors discovered that administration of an antibody against TIM-3 resulted in more severe clinical disease and increased mortality in the EAE model. The inventors further found that these animals had increased inflammation in the central nervous system (CNS), and that the demyelinating lesions in the mice treated with antibody to TIM-3 were filled with activated macrophages (Example 6). Further examination of immune cell populations taken from the EAE model mice led to the inventors' discovery that macrophages from mice administered TIM-3 antibody showed increased proliferation and expressed increased levels of MHC Class II antigens, these being indicative of increased ability to present antigen. Both these parameters were measurements of the activation status of a macrophage, which is a type of antigen-presenting immune cell. Together, these data would have been understood by a person skilled in the art to be clearly indicative of an enhanced or increased immune response in animals treated with an anti-TIM-3 antibody. The application therefore demonstrated the suitability of a TIM-3 specific antibody for enhancing an immune response which includes an immune response mounted against a cancer antigen. The enhancement of the immune response by administration of anti-TIM-3 antibodies was of a general and non-specific nature. In the case of cancer, the skilled person knew from the prior art, for example from document D1, that macrophage activation was critical for the induction of immune responses to microbes as well to certain tumor cells. From document D2 (page 207, "Introduction") the skilled person knew that "the monocyte/macrophage system exhibits a wide array of powerful effector mechanisms that may be harnessed for therapeutic effect against infection and malignancy." The specification would have been considered in the light of the knowledge that monocyte-mediated immunity was active against malignancy. Post-published evidence also supported an inherent effect of the TIM-3 inhibitor in cancer models (WT3 sarcoma, TRAP-C1 prostate sarcoma) and some PD-1 blockade synergy (document D3) and EL4 lymphoma (document D4, see page 1386, Fig. 2B). The skilled person would have therefore had no doubt that the invention could be carried out as claimed. On the subject of whether the generation of a de novo immune response by anti-TIM-3 treatment was sufficiently disclosed, the appellant argued that, in fact, no generation of a de novo immune response was needed. The immune system carried out constant immune-surveillance, nascent cancers being naturally disposed of by activated macrophages. Disease was said to arise only when this surveillance failed. Administration of anti-TIM-3 antibodies acted to release a kind of "brake" on the immune system. Finally, it was pointed out that similar claims to second medical uses of compounds for the treatment of cancer in general had been accepted as meeting the requirements of Article 83 EPC, for instance in decisions T 1616/09 of 27 August 2014 and T 1492/09 of 9 January 2014 and T 1918/06 of 10 March 2010.
Reasons for the Decision
Main request [...]
Claim 1
4. The subject-matter of claim 1 is an antibody specific for TIM-3 or a TIM-3 specific fragment thereof, for the specific use of "treatment of cancer in a subject". The claim is therefore directed to a second medical use as foreseen by Article 54(5) EPC.
5. TIM-3 is a transmembrane protein which is preferentially expressed on differentiated Th1 cells and is termed "T cell Immunoglobulin and Mucin domain-containing molecule-3" (see the description of the application, page 1, paragraph 1 and page 2, paragraph 3).
6. The antibody of claim 1, specific for TIM-3 or a TIM-3 specific fragment thereof, may bind different epitopes of the target protein. For instance, claim 2 specifies that the antibody binds to an extracellular region of TIM-3.
7. Antibodies may have different biological effects, depending on their particular binding specificity, for instance, they may be agonistic or antagonistic. The description of the application on page 59 provides several possible explanations of the effect of anti-TIM-3 treatment on a subject: "[...] a cognate interaction between non-T cells and TIM-3-expressing Thl cells is affected by anti-TIM-3 treatment, resulting in the expansion and activation of CD11b+/F4/80+ macrophages. Several possible mechanisms may explain this finding: a) Anti-TIM-3 may cross-link TIM-3 protein on the surface of differentiated Th1 cells in vivo and amplify the production of pro-inflammatory cytokines (e. g., IFN-gamma and TNF), which in turn may induce activation of macrophages ; b) anti-TIM-3 antibody could enhance migration of differentiated Thl cells into the brain where these cells may increase the cellular influx of macrophages from the circulation; c) anti-TIM-3 could block a cognate interaction of TIM-3 with its potential inhibitory ligand on macrophages, thus leading to enhanced macrophage activation in the presence of pro-inflammatory cytokines produced by Thl cells".
8. The mechanism set out under (a) may be seen as an agonistic type mechanism, in which anti-TIM-3 binding triggers a signalling cascade, while that given under (b) is an antagonistic mechanism with the effects due to the blocking of the interaction of a ligand with its receptor.
9. The antibodies of the claim are for the treatment of cancer in a subject. Cancer is not a single disease with a single underlying mechanistic cause but is a collective term for a group of diseases involving abnormal cell growth with the potential to invade or spread to other parts of the body, see the description of the application, page 20, last paragraph: "Cancers include, but are not limited to, basal cell carcinoma, biliary tract cancer; bladder cancer; bone cancer; brain and CNS cancer; breast cancer; cervical cancer; choriocarcinoma; colon and rectum cancer; connective tissue cancer; cancer of the digestive system; endometrial cancer; esophageal cancer; eye cancer; cancer of the head and neck; gastric cancer; intra-epithelial neoplasm; kidney cancer; larynx cancer; leukemia; liver cancer; lung cancer (e.g. small cell and non-small cell); lymphoma including Hodgkin's and non-Hodgkin's lymphoma; melanoma; myeloma; neuroblastoma; oral cavity cancer (e.g. lip, tongue, mouth, and pharynx); ovarian cancer; pancreatic cancer; prostate cancer; retinoblastoma; rhabdomyosarcoma; rectal cancer; cancer of the respiratory system; sarcoma; skin cancer; stomach cancer; testicular cancer; thyroid cancer; uterine cancer; cancer of the urinary system, as well as other carcinomas and sarcomas."
10. In summary, claim 1 is directed to an anti-TIM-3 antibody or a TIM-3 specific fragment thereof, with any type of specificity for treatment of any type of cancer.
Sufficiency of disclosure - Article 83 EPC
11. Article 83 EPC requires that the claimed subject-matter is disclosed in the application "in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art".
12. The experimental evidence provided in the application relates inter alia to mice having experimental autoimmune encephalomyelitis (EAE), a Thl-dependent autoimmune disease being a widely accepted model for multiple sclerosis. The data provided in the examples show that the Th1-specific cell surface protein TIM-3 is involved in regulating the level of T-cell trafficking to target tissues and macrophage activation in these mice (description of the application, page 1, lines 5 to 11 and Examples 5 to 16). Example 6 shows that an existing autoimmune reaction triggered by immunisation with encephalitogenic proteolipid protein (PLP) is exacerbated by administration of an anti-TIM-3 antibody.
13. The board concludes that, from the evidence provided in the application, the skilled person would have considered it plausible that an extant immune reaction, such as that existing in the EAE mice, could be amplified by administration of anti-TIM-3 antibodies.
14. The evidence provided in the application does not concern establishment of a de novo immune response to a tumour antigen or to any other antigen.
15. The skilled person at the priority date knew that
i) cancer cells were "self" cells, ii) an immune response to self antigens was the exception to the rule (cf. description of the application, page 40, lines 7 to 29) and iii) there were a wide range of different cancers with diverse underlying causes (see point 9.9. above).
16. In view of the above it is considered that the skilled person would not have considered that the disclosure of the application makes it plausible that it would be possible to generate a de novo immune response to cancers where there was no established native response. Moreover, the skilled person would not have believed that substantially all cancer types inherently generate a Th1, TIM-3 mediated immune response and be treatable by administration of anti-TIM-3 antibodies. It is noted that the post-published documents submitted by the appellant support this conclusion, see point 23.23. below.
17. Thus, the subject-matter of claim 1 does not meet the requirements of Article 83 EPC with respect to the whole scope claimed in relation to the disease to be treated.
18. Secondly, anti-TIM-3 antibodies do not all have the same specificity or functionality (see points 6.6. and 7.7. above). The particular antibodies used in the examples of the application are termed "8B.2C12" and "25F.1D6" (see Example 1) and were not deposited with a recognised depositary institution according to Rule 31(a) EPC. It was not disclosed in the application whether the interaction of TIM-3 with its not yet identified ligand (see Example 17) was blocked or activated by the antibody to achieve the claimed therapeutic effect. In view of the potentially opposite actions of anti-TIM-3 antibodies, the skilled person would not have considered it plausible that substantially all embodiments of the invention defined in claim 1 were capable of being realised (c.f. Case Law of the Boards of Appeal of the EPO, 7th edition, II.C.6.1.2).
19. Thus, the subject-matter of claim 1 lacks sufficient disclosure to be carried out by the skilled person over the entire claimed scope with respect to the specificity of the anti-TIM-3 antibody to be used in the claimed medical use.
[...]
23. With respect to the evidence provided in the form of post-published documents, the board notes that document D3 provides evidence which confirms the board's earlier conclusion (see point 17.17. above) on the insufficiency of disclosure of the application with respect to de novo carcinogenesis. Document D3 reports "an extensive characterization of the therapeutic activity and mechanism of action of an anti-mouse TIM-3 mAb [monoclonal antibody] against experimental and carcinogen-induced tumors" (see the abstract). The anti-TIM-3 antibodies used were specifically antagonistic antibodies (see page 3541, "Tumor models"). Anti-TIM-3 was reported to display only modest prophylactic (page 3546, column 2, final paragraph) and therapeutic activity (page 3547, paragraph 1) against a small fraction of carcinogen-induced sarcomas (13%). In the discussion section (page 3550, column 1) it is stated "We have shown that dramatic therapeutic effects are not observed with monotherapies (including anti-PD1 or anti-TIM3) in this model [of de novo carcinogenesis]. This is despite the fact that anti-PD1 is an extremely promising therapeutic in some human cancers".
24. In summary, document D3 discloses that anti-TIM-3 antibodies fail to successfully treat the majority (87%) of induced sarcomas (see the sentence bridging pages 3546 and 3547).
25. Document D4 reports inter alia that "treatment of EL-4 tumor-bearing mice with anti-TIM-3 Ab resulted in delayed tumor progression coincident with lower frequency of CD11b**(+)Gr-1**(+) cells (Fig. 2B)" (see page 1386, column 2, penultimate sentence).
26. However, in view of the diversity of cancer types (see point 9.9. above), evidence of successful treatment of one tumour type with a specific antibody ("clone 5D12") does not provide evidence in support of the entire scope claimed. Moreover, even if the disclosure of document D4 were to be considered as evidence of the successful treatment of cancer, a particular post-published disclosure relating to a particular antibody and a specific type of cancer cannot remedy a problem of general lack of sufficient disclosure at the priority date (see also decision T 609/02 of 27 October 2004, Reasons 9).
[...]
29. In view of the above, the board concludes that the subject-matter of claim 1 is not disclosed in the application in a manner sufficiently clear and complete for it to be carried out by the skilled person and therefore does not meet the requirements of Article 83 EPC.
Order
For these reasons it is decided that:
The appeal is dismissed.
This decision has European Case Law Identifier: ECLI:EP:BA:2015:T160210.20150317 and can be found here. The file wrapper can be found here. Photo "Antibody" (link) by Jim obtained via Flickr under CC BY 2.0 license.

T 0562/13 - Missing opponent in appeal



An opponent is allowed by the board to correct his appeal by including his joint opponent. 

The opposition had been filed by two joint opponents. Later, only one of the two files an appeal against the decision of the opposition division. Before the Oral Proceedings, the board brings this issue to the attention of the parties, informing them that 'the admissibility of the appeal (...) is at stake'.


The opponent-appellant submits a request for correction and to include the other opponent in the appeal under Rule 101(2). This request is allowed, even though the request is filed less than 1 month before the Oral Proceedings in the appeal case. 


Reasons for the Decision

1. Admissibility of the appeal

1.1 The notice of appeal against the decision of the Opposition Division, filed with letter of 19 February 2013 in Dutch (together with an English translation), was expressly "on behalf of the opponent, Unilever N.V.".

According to the Representative's statement in its letter dated 27 May 2015 (supra), this notice of appeal was however to be meant to have been filed on behalf of both opponents.

1.2 The Respondent did not comment on this issue.
 
1.3 It is undisputed that in opposition proceedings Unilever N.V. and Unilever PLC acted as joint opponents and were jointly represented by Mr Kan.

1.4 According to the reasoning given in decisions G 3/99 (OJ EPO 2002, 347) and R 18/09 of 27 September 2010, an opposition filed by several persons in common is to be dealt with as an opposition filed by only one party, and such a group of common opponents is to be considered as a single party represented by a common representative (G 3/99, Reasons, 14 and 15). Should such a group of common opponents intend to file an appeal, they can only do so jointly as a single party acting through their common representative (G 3/99, Reasons, 17; R 18/09, Reasons, 5).

1.5 G 3/99 furthermore states (Order, 3) that "[i]n order to safeguard the rights of the patent proprietor and in the interest of procedural efficiency, it has to be clear throughout the procedure who belongs to the group of common opponents or common appellants. If either a common opponent or appellant (including the common representative) intends to withdraw from the proceedings, the EPO shall be notified accordingly by the common representative or by a new common representative determined under Rule 100(1) EPC in order for the withdrawal to take effect" (emphasis added).

1.6 In the present case the EPO has not received any notification by the common Representative of Unilever N.V. and Unilever PLC that Unilever PLC intended to withdraw from the proceedings.

1.7 No arguments or any proof pointing to the contrary have been submitted by the Respondent either.

1.8 The facts and evidence at hand, including the statement by the joint Representative, convincingly lead the Board to the conclusion that it was the Appellants' true intention to file the appeal jointly by both Opponents, i.e. Unilever N.V. and Unilever PLC, thus as joint Appellants.

1.9 However, the notice of appeal does not reflect or express this true intention of the parties on whose behalf it was intended to be filed. In other words, the notice of appeal omits the designation of Unilever PLC. In this respect, a correction under Rule 101(2) EPC as referred to in G 1/12 (OJ EPO 2014, A114) has been requested by the Representative of the joint Opponents.

1.10 Hence, the Board considers it necessary, for both Opponents, Unilever N.V and Unilever PLC, to continue acting through the common Representative before the EPO as joint Appellants, that the file be corrected, in line with G 1/12, so as to mention both Opponents as joint Appellants, i.e. to clarify or introduce what was originally intended in the notice of appeal.

1.11 The appeal is consequently admissible.

 This decision T 0562/13 (pdf) has European Case Law Identifier: ECLI:EP:BA:2015:T056213.20150624. The file wrapper can be found here. Cartoon "Join, or Die" by Benjamin Franklin, obtained from DonkeyHotey via  Flickr under CC BY 2.0 license (no changes made).