Open Questions. Our results leave open several interesting questions for followup work. Our constructions of SRDS offer a trade-off between cryptographic assumptions and setup as- sumptions (indeed, our lower bound indicates that some form of private-coin setup is needed). Is it possible to get the best of both, i.e., construct SRDS with bulletin-board PKI under standard, falsifiable assumptions? This in turn would imply O˜(1)-balanced BA from the corresponding com- putational assumption and setup. Alternatively, does SRDS in a weak setup model require strong computational assumptions: For example, do SRDS with bulletin-board PKI imply some kind of succinct non-interactive arguments (SNARGs)? Taking a step back: Is it possible to achieve O˜(1)-balanced BA unconditionally? While our SRDS-based approach inherently makes use of computational assumptions (and our lower bound implies this necessity for a one-shot boost from almost-everywhere to everywhere agreement in the PKI model), this leaves open the possibility of removing cryptography via an alternative approach. Can one further extend the lower bound in this work, identifying a minimal required round com- plexity for generically converting from almost-everywhere to everywhere agreement within various setup models? In the O˜(1)-amortized BA setting, known constructions consider stronger security models. Namely, the protocol in Xxxxx-Xxxxxxx et al. [19] is secure against static corruptions (similarly to our protocols); however, no trusted setup assumptions are required. The protocol of Xxxxxxx et al. [1] guarantees security against adaptive corruptions; however, it requires a trusted PKI as- sumption. On the contrary, the protocol of King and Saia [64] does not require setup assumptions and is resilient to adaptive corruptions, but it provides suboptimal total communication O˜(n√n). It is very interesting to explore if O˜(1)-balanced BA can be achieved without setup or in the adaptive setting. Regarding the communication model, the vast majority of sub-quadratic BA protocol are defined in the synchronous model. In the asynchronous setting unbalanced sub-quadratic BA in the trusted PKI model was recently proposed [35]. We note that balanced sub-quadratic BA is not known even in the partially synchronous model. An interesting question is to expand our techniques beyond the synchronous realm. Finally, all known BA protocols with o(n2) total communication follow either the approach of Xxxx et al. [65] or of Xxxx and Xxxxxx [29]...
Open Questions. Author Manuscript The MIIV approach to SEMs is far less studied than are the system wide estimators such as ML. As a result there are many open research questions. For instance, the field needs to learn more about the optimal selection of scaling indicators for each latent variable. I mentioned this issue earlier in the paper. A scaling indicator with little association with its latent variable seems certain to lead to poorer estimator properties than would a strong indicator. Weak instrument statistics and the R-squared for scaling indicators seem promising diagnostics that could help spot poor scaling indicators. Another scaling issue not explored is what happens when the scaling indicator actually loads on two but is assumed to load on one latent variable.8 What are the consequences? Would this be detectable with the equation overidentification test due to incorrect selection of MIIVs? Author Manuscript The weak instrument problem was mentioned several times in this paper and in the original presentation of MIIV-2SLS (Bollen, 1996). Although the econometric literature has examined this problem for some time now (e.g.,Xxxxxx & Xxxxxxxxxx, 1984; Xxxxx, Xxxxxx, & Xxxxx, 1995), it has been in the context of single equation or simultaneous equation models without latent variables. Because the L2O transformation makes latent variable models into a type of simultaneous equation model with a more complicated error structure, it would seem that much of the findings from the econometric literature should carry over to the MIIV-2SLS estimator. On the other hand, when indicators of the same latent variable are the MIIVs that predict the scaling indicator as is common in multiple indicator models, these relations should be moderate to strong. This might mean that weak instruments are less common than they are in the econometric literature where auxiliary variables external to the model are common. This too would be a fruitful area of research. Author Manuscript The overidentification tests also demand more attention. Based on the results of Xxxxx and Xxxxxx (2009), I have emphasized the Sargan test in conjunction with the MIIV-2SLS estimator. But other overidentification tests should be further considered. Furthermore, if the Sargan test statistic is significant, researchers need more guidance on the best way to locate the offending MIIVs. This would be helpful in determining how to respecify the model. Another question is whether it is possible for an equation that...
Open Questions. An essential part of the anaylsis of GD as a partial pro-drop language is connected to the fact that the second person pronouns can also be ommited in non-subject initial root clauses39. Those are characterised by the V2-property, i.
e. the finite verb moves to C and any other maximal projection of the clause has to be fronted in Spec-CP40. Pro-drop can only be observed in cases where a projection other than the subject precedes the verb, consequently leading to what is generally refered to as subject-verb inversion, i.e. the subject follows the verb. In the most simplest cases the subject will then be in Spec-TP directly dominated by the verb in C◦. This immediately raises the question of what happens in main clauses under the proposed analysis. Recall that in this approach GM is viewed as part of the subject DP and is being attracted to C due to its finiteness-feature. We need to account for the fact that in main clauses the verb ends up in C and not GM – whereas the exact opposite happens in embedded clauses. There are at least two 39 As pointed out to me by Xxxxxx Xxxxxxxxx, this is not only true for varieties displaying CA, but also for a number of other German dialects. However, given the diachronic link between the development of second person verbal agreement and the rise of pro this is not surprising. The question is then much rather why not all German varietes display CA – at least in the singular, where the reanalysis took place much earlier than in the plural and affected all German varieties including the Standard. In line with Poletto (2006) the difference would lie in the option of raising the lexical DP to an internal specifier. However, at the current stage of research, I do not have an answer as to why one variety would develop this option whereas others would not. 40 As for subject-inital clauses it has also been argued that no CP is projected at all (cf. section 2.3). However, this point is not of importance for the issue at hand. straightforward options: one is to assume different lexical entries for the pronoun, i.e. [[GM] [pronoun]] and [pronoun], and the other involves some sort of deletion of GM. As for the first option, I consider it highly unlikely due to a number of reasons. First and foremost, if GM is indeed the marker of [+addressee], I do not see any reason whatsoever why the lexicon should contain two versions of the pronoun. Even more, this would mean that the pronoun without GM would not include an addressee-feature41 – s...
Open Questions. Are you using the state-‐of-‐art tools, methods and knowledge to assess the impact of air pollution on human health? . What do you see as future research needs? . What are the largest uncertainties in HIA?
Open Questions. Explain to what extent it was possible to achieve the objectives of your source apportionment study and what were the limitations of the used methodology? . Are the source apportionment capacities in institutions of your country/region/city enough to achieve reliable source identification and support air quality management? . Is data collection in your country/region/city appropriate for source identification?
Open Questions. In general, to the question `Can you describe your experience performing the task with the robot?`, most participants tended to be more positive towards their own training. Most participants considered their own training as interesting and easier to use. They used words such as "better understandable", "more logical", "exciting" or "fun". However, a few participants answered that the experience was "not good"or that the programming could use some tweaking". This was due to some incidents with the motions and the joints angle. But one of those participants also indicated that with "some more trial and error", it could be better. Finally, two participants said their training was "slow".
Open Questions. In this paper we have explored many characterizations of conformal mappings and M¨obius transformations and we have seen that quasiconformal maps and quasicircles are natural extensions of conformal maps and circles. Furthermore, many of the properties of Mobius transformations and circles also extend naturally to quasiconformal maps and quasicircles simply by multiplying by a finite constant greater than or equal to 1. This led me to question whether the same was true with our two key results about M¨obius transformations from Section 2.2. We answered the first of these with the introduction of quasim¨obius maps, and although not every quasiconformal map is necessarily quasim¨obius there do exist sufficient conditions under which this is true. However, whether or not Theorem 2.11 can be extended to quasiconformal maps is still unknown.
Question 1. Suppose that f : C∞ → C∞ is a sense preserving homeomorphism. Is f quasiconformal if and only if f takes circles to quasicircles? We do know that one direction of this question is true: if f is quasiconformal then it takes circles to quasicircles by the definition of a quasicircle. However, the converse of this is unknown. I would hypothesize that the converse is true globally because of the similar result with quasim¨obius maps, but a rigorous proof still needs to be written. But if this is not true then like the quasim¨obius maps there are probably conditions we could place upon f to make it true, and this results in the following question.
Question 2. If a quasicircle preserving map is not quasiconformal in general, then what conditions can we place upon either f or the domain to make it quasiconformal?
Open Questions. The Ethics and Society Opinion on Protection and Privacy makes several recommendations [6, p15]. The following open questions ask how best HBP researchers can be aided in meeting the spirit of the recommendations, in the light of their specific research tasks. Following this are recommendations for follow-up that should contribute to the meeting of the challenges contained in these questions. The table indicates the SP or SPs primarily connected with the potential issues, a label for the issue, and a comment. The comment reflects something of prior discussion on the topic area that has taken place. Table 3: Specific issues, SPs concerned, Comments and Recommendations SP Topic Comment Relevant recommendation 2, 5, 11 Data transfers Explore Memoranda of understanding for transfers among projects. Issues of international transfers of data will require SP11 actions, in coordination with ethics management. 2 2, 3, 5 Incidental findings, and informed consent Contribute to standard operating procedures in these areas. 4, 5 3, 5 Data registration and audit Follow developments in DGWG. 3, 4 5 Data flow diagrams These are a useful, if not essential , part of demonstrating reflection upon, and coherent system-design for, data protection as between e.g., Data Registration -> Repository -> Active data repository -> User Retrieval. 1c 2, 4, 5 Data provenance Contribute to DGWG, and to Platform user guides. Broad consent, and de/re-identification are relevant here. The status of published data. 2, 7, 8 8 Privacy Impact Assessment This is of special importance in the light of SP12 Opinion recommendations, and for the forthcoming GDPR. 1c 11 Potential for commercial use of research data This connects with the work of the DGWG, as well as the topic of broad consent, and the field of data provenance. 3, 6, 8
Open Questions. Were synergies among policies at different scales important in your assessment? .Did you identify conflicts or inconsistencies among scales? If yes can you list them in order of priority (from more to less crucial aspects)? .How did you combine the results at different scales? .What do you see as remaining issues?
Open Questions. What does the patient understand about their treatment, and which medicines do they actually take?
