Skip to content

Commit

Permalink
modif slide : "Data storage : reminder of the context and the issues"
Browse files Browse the repository at this point in the history
  • Loading branch information
Olivier Norvez committed Nov 8, 2024
1 parent 763d45e commit c9cccf4
Showing 1 changed file with 9 additions and 67 deletions.
76 changes: 9 additions & 67 deletions sections/slides-data-storage.qmd
Original file line number Diff line number Diff line change
Expand Up @@ -4,78 +4,15 @@

## Data storage : reminder of the context and the issues

:::: {.columns}

::: {.column .center width=50%}
![](images/image-Page2016.png){width=80%}
:::

::: {.column .center width=50%}
![](images/image-Michener1997.jpg){width=80%}
:::

::::


:::: {.columns}

::: {.column .center width=50%}
**Heterogeneity (data types, origin, standards) &<br/>Diversity of "objects" to be linked together**[^2]
:::

::: {.column .center width=50%}
**Loss of information over time**[^3]
:::

::::


[^2]: [Page (2016)](https://riojournal.com/article/8767)
[^3]: [Michener _et al._ (1997)](https://doi.org/10.1890/1051-0761(1997)007[0330:NMFTES]2.0.CO;2)


<!-- page X -->

## Data storage : reminder of the context and the issues

:::: {.columns}

::: {.column width=50%}
![](images/image-Powers2018.jpg){width=100%}
:::

::: {.column width=50%}
![](images/image-JenkinsReproductibility.png){width=100%}
:::

::::

:::: {.columns}

::: {.column width=50%}
> Computational reproducibility frequently refers to the ability to generate equivalent analytical outcomes from the same data set using the same code and software[^4].
:::

::: {.column width=50%}
> [...] all raw data and metadata, code, programming scripts, and bespoke software necessary for fully replicating any analyses that lead to inferences made in a published study[^5].
:::

::::

[^4]: [Powers & Hampton (2018)](https://doi.org/10.1002/eap.1822)
[^5]: [Genkins _et al._ (2016)](https://doi.org/10.1002/ece3.9961)



<!-- page X -->

## Data storage : reminder of the context and the issues
- **Heterogeneity** (data types, origin, standards) & **diversity** of "objects" to be linked together[^2]
- **Loss of information** over time[^3]
- Toward a better **open science** and **reproducibility** [^4] [^5]

:::: {.columns}

::: {.column width=33%}

![](images/image-DIKW_Pyramid.svg){width=100%}
![](images/image-DIKW_Pyramid.svg){width=70%}
![](images/image_DataInformationknowledgeHumanities.jpg){width=100%}

:::
Expand All @@ -93,6 +30,11 @@

::::

[^2]: [Page (2016)](https://riojournal.com/article/8767)
[^3]: [Michener _et al._ (1997)](https://doi.org/10.1890/1051-0761(1997)007[0330:NMFTES]2.0.CO;2)
[^4]: [Powers & Hampton (2018)](https://doi.org/10.1002/eap.1822)
[^5]: [Genkins _et al._ (2016)](https://doi.org/10.1002/ece3.9961)

<!-- page X -->

## Data storage : Hot *versus* cold data storage
Expand Down

0 comments on commit c9cccf4

Please sign in to comment.