1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
|
Summary
-------
This paper presents and studies data spaces, which can be thought of a specific
kind of data exchange/market. The defining feature of a data space, compared to
generic data marketplaces, appears to be a greater focus on interoperability,
allowing value to be extracted from data more easily, typically within
a specific application domain.
A (fictional?) use case that is used as a running example throughout the paper
is “Green twin”: a data exchange pooling together data about a city's
buildings, vehicles, inhabitants and network infrastructure with the goal of
improving energy efficiency and quality of life. A survey of existing efforts
in the realm of data spaces (led by two related non-profit organizations, IDSA
and Gaia-X) is presented, highlighting the challenges of data interoperability
(with possible solutions involving the creation of data ontologies and the use
of standardized protocols) and generating value from data (with possible
solutions involving the use of automated machine learning techniques with
transfer of knowledge).
Comments
--------
While the concept of data spaces seems promising and an interesting object of
study, this paper suffers from a somewhat ambiguous scope and its contributions
are not immediately apparent:
1. As an overview of existing techniques and solutions, many of the
explanations were lacking in precision. For example, I was not able to find
a clear definition of data spaces and how they differ from data
marketplaces, and had to rely instead on slowly discovering the concept over
the course of the entire paper. The closest to this were the following
sentences in the introduction:
This concept serves as an abstraction for data management in case where
many stakeholders are involved and exchange data with each other. The
easy data exchange between the stakeholders will generate value,
especially in combination with data analytics. New trading mechanisms
can allow stakeholders to cooperate with each other based on the value
of the exchanged data and the analytics services.
2. As a position paper, I found that the description of future challenges lacks
in concreteness. What are the concrete open questions that researchers in
the data economy community should focus on? Do we need new algorithms? new
data structures? new machine learning techniques? What are specific ways in
which currently existing techniques unable to solve this challenges?
Overall, I think the paper would benefit from focusing on at most two of the
following three kind of contributions:
1. A systematic description and documentation of currently existing
technologies, protocols, and standards that can be used to build a data
space.
2. A concrete proposal for the “Green twin” use case following the standards of
the systems community: a clear description of a system that would solve this
use case, the different pieces it will contain and how they interact with
other.
3. Concrete open questions and conjectures and an invitation to the research
community to work on them.
|
