Examples from Buckland et al. (2015)

[erex]

@dill

We discussed the matter of bringing the case studies from the 2015 book into the fold with the other examples. I said I would mock up something for us to discuss.

In the interim, the May introductory workshop got launched, so my efforts were delayed. But now...

I've created 3 new repos containing

• DSsim example of Laura's
• Cornelia's model based bobwhite analysis
• Steve's crossbill lure study

There are four other case studies referenced in the book (and the old website)

• ever popular amakihi
  • for which there already exists an entry in examples site, but it has the errors I noted in a critique I shared with you
• spotted dolphins
• Montrave robins
  • the robins from the book have morphed into winter wrens (lines and cues), maybe that needs attention
• cork oak simulation
  • I must have performed it in the dark past, but I can't resurrect it, may have to start from scratch

I don't think we need to duplicate amakihi, spotted dolphins or Montrave robins (I could be convinced otherwise). Cork oak simulation is an outstanding issue.

The three repos (Dssim, bobwhites, crossbills) have not yet been exposed, pending decisions about our next steps.

Including the 3 most recent additions, the examples page will have 18 examples. That's a healthy number, but time has come for some organisation beyond a bulleted list. I don't know what might be best. Feeble options from my brain:

• subheaders (don't know how many or what categories)
  • this would be simplest for re-directing people looking for the Buckland et al. (2015) materials, a link to the anchor formed by the "Buckland et al. (2015) case studies" subheading would be sufficient
• some kind of a tagging system
  • again I don't know how many tags, or their definition (points, lines, covariates, strata...). Seems the cool kids are happy using tags to organise things like their Gmail inbox. Don't know how that might work in this case. I'd like to think each example would have two/three tags that could lead folks to the desired destination.

How to proceed?

[dill]

Following on from the tags idea: How about a table with sortable headers, perhaps with search (similar to the bibliography page) be an option? This would require some javascript fiddling so might be a "stage 2"

…

On 13/04/2020 16:39, erex wrote: @dill <https://github.com/dill> We discussed the matter of bringing the case studies from the 2015 book into the fold with the other examples. I said I would mock up something for us to discuss. In the interim, the May introductory workshop got launched, so my efforts were delayed. But now... I've created 3 new repos containing * DSsim example of Laura's * Cornelia's model based bobwhite analysis * Steve's crossbill lure study There are four other case studies referenced in the book (and the old website) * ever popular amakihi o for which there already exists an entry in examples site, but it has the errors I noted in a critique I shared with you * spotted dolphins * Montrave robins o the robins from the book have morphed into winter wrens (lines and cues), maybe that needs attention * cork oak simulation o I must have performed it in the dark past, but I can't resurrect it, may have to start from scratch I don't think we need to duplicate amakihi, spotted dolphins or Montrave robins (I could be convinced otherwise). Cork oak simulation is an outstanding issue. The three repos (Dssim, bobwhites, crossbills) have not yet been exposed, pending decisions about our next steps. Including the 3 most recent additions, the examples page will have 16 examples. That's a healthy number, but time has come for some organisation beyond a bulleted list. I don't know what might be best. Feeble options from my brain: * subheaders (don't know how many or what categories) o this would be simplest for re-directing people looking for the Buckland et al. (2015) materials, a link to the anchor formed by the "Buckland et al. (2015) case studies" subheading would be sufficient * some kind of a tagging system o again I don't know how many tags, or their definition (points, lines, covariates, strata...). Seems the cool kids are happy using tags to organise things like their Gmail inbox. Don't know how that might work in this case. I'd like to think each example would have two/three tags that could lead folks to the desired destination. How to proceed? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAAPIJ6JOE7366VRSJDP3LRMMW3TANCNFSM4MHBD5VA>.

[erex]

perhaps...

seems first-order problem is identifying categories (strata for the examples), second-order problem is how to present the strata

[erex]

I'm not particularly thrilled with the table view (no place for brief intro text), also need to make special provision for Gui's auxillary files.

I'm leaving it for now.