stan-dev / stan-dev.github.io Goto Github PK

Under each of the entries for the various stan https://github.com/stan-dev/stan-dev.github.io/issues/27warnings are some bullet point list entries.

The bullet points in the Rmd file need to be indented by some whitespace. I have made these edits but I am not sure what the convention is for pull requests. Let me know and I'm happy to submit it since it is a minor edit.

Hopefully the image shows.

Need a new page for stan math, that includes

• pointer to GitHub tarball (.tgz and .zip) this will be complete unlike other packages because there are no subrepos
• ? pointer to standalone manual
• ? pointer to development arXiv paper
• ? pointer to getting started, doxygen, etc. [this should go to Wiki]

https://mc-stan.org/users/documentation/ could be a much more transparent link to mc-stan.org/docs/ now that that page exists

Right now with every release we need to update version numbers for documentation -
cf., this commit:

6ff2353

instead, index pages should link unversioned page, which redirects to latest version. updating this link can be scripted as part of the release process or similar.

In documentation page, the link to the first written tutorial returns 404.

Michael Clark. (2015) Bayesian Basics: A Conceptual Introduction with Application in R and Stan. Center for Statistical Consultation and Research, University of Michigan.

@bnicenboim suggested here: #37 (comment)

you should think on adding DOIs to the case studies and the Stancon papers. If I want to refer to them in a paper, it's much better than to have a link (and journals usually ask for the DOIs).

We generate DOIs for the software, but we don't know how to generate DOIs for case studies. So any suggestions would be appreciated (feel free to just edit the issue rather than adding comments if there's a concrete workable suggestion).

stan-dev.github.io/development/index.md
says
"If you’d like to contribute code, please submit it through a Git pull request. The developer process for creating a Git pull request is described on the Stan Wiki."
and has link to "Developer Process Wiki" https://github.com/stan-dev/stan/wiki/Developer-Process
which produces a page saying "Create new page"
I couldn't find where the link should point to.

@betanalpha @akucukelbir, put up a page for NIPS.

Could you toss on a Stan at NIPS schedule on there too?

I have some suggestions for the existing Stan website from my experience of using the site and introducing new users to Stan via the site.

Discoverability

While the current site does a decent job of establishing Stan's presence online, I think it's limited in the ability for new user's to discover information/interfaces/case studies/etc. regarding Stan. I'm not a huge fan of having long sidebars on sites, but it might help with discoverability (e.g. see the JTD jekyll theme).

I've lightly surveyed some popular open source libraries and it looks like they mix the documentation-like sidebar theme with a aesthetically pleasing landing page:

• Pyro
• Pytorch
• TensorFlow

I'm not saying that we should emulate these sites. They have a lot more bells-and-whistles than we need. But their exposition of information might be worth considering.

One limiting factor for Stan (that is also a huge feature of the language) is how many interfaces we have. This makes it a little challenging to make the site a one-stop-shop for all users (the way the sites mentioned above are, more or less).

Flow for New Users

Another thing to note is that the purpose of the site is not clear for new users to Stan. One example of flow for new users could be,

1. Landing page
2. Interface/Installation
3. Quick Start
   • For popular interfaces RStan/PyStan/etc
   • Using a super basic example
4. Documentation
   • For users who are ready to go beyond the basic example presented in Quick Start

I'm not saying that these should be the only buttons on the nav bar, but rather that the site should be setup in a way that once a new(ish) user arrives at the landing page, there should be a clear path for them to get their hands dirty and get more information if they're interested.

I've suggested an example layout below,

• Landing Page
• Installation
  • Stan Interfaces
  • Higher-Level Interfaces
  • Lower-Level Interfaces
• Quick Start
  • RStan
  • PyStan
  • Others?
• Documentation
  • Stan User’s Guide
  • Stan Language Reference Manual
  • Stan Language Functions Reference
• Tools
  • Useful Tools
  • Syntax Aware Editors
• Forum (goes straight to Discourse)
• Case Studies
• Develop
  • Contribute to Stan
  • Submit an Issue
• About Us
  • Stan Development Team
  • Stan Governing Body
  • Support Stan
  • Stan Swag
• Search

The new rstan 2.19.2 provides warnings about bulk ess and tail ess, and those warnings include the following links:

http://mc-stan.org/misc/warnings.html#bulk-ess
http://mc-stan.org/misc/warnings.html#tail-ess

But those sections of the page do not currently exist. It looks like the sections do exist in misc/warnings.Rmd, so maybe that Rmd just needs to be compiled to html?

With Stan 2.33+ several old language syntax features produce errors. All the case studies would be good to update to use the latest syntax. Many case studies are in external repos and the authors have submitted only the rendered html and short md-part for the case study contents page. Only the html needs to be updated in users/documentation/case-studies/.

It would be good o contact the original authors and ask them if they are willing to update their repos and submit a new html. If the authors disagree or don't respond, we may consider updating just the syntax on html.

To start the process, I'm listing here all the case studies, and we can start tracking which have been fixed. Tagging also some authors that were easily found by github id autocomplete @mitzimorris, @WardBrian, @bob-carpenter, @charlesm93, @bbbales2, @imadmali

• Bayesian Structural Equation Modeling using blavaan: Feng Ji, Xingyao Xiao, Aybolek Amanmyradova, Sophia Rabe-Hesketh
• Multilevel regression modeling with CmdStanPy and plotnine: Mitzi Morris
• HoloML in Stan: Low-photon Image Reconstruction: Brian Ward, Bob Carpenter, and David Barmherzig
• Bayesian Latent Class Models and Handling of Label Switching: Feng Ji, Aybolek Amanmyradova, Sophia Rabe-Hesketh
• Bayesian model of planetary motion: exploring ideas for a modeling workflow: Charles Margossian and Andrew Gelman
• HMM Interface Example: Ben Bales
• Spatial models for plant neighborhood dynamics in Stan: Cristina Barber, Andrii Zaiats, Cara Applestein and T.Trevor Caughlin
• Predicting Engine Failure with Hierarchical Gaussian Process: Hyunji Moon, Jungin Choi
• Upgrading to the new ODE interface: Ben Bales, Sebastian Weber
• Bayesian Workflow for disease transmission modeling in Stan: Leo Grinsztajn, Elizaveta Semenova, Charles C. Margossian, and Julien Riou
• Reduce Sum Example: parallelization of a single chain across multiple cores: Ben Bales
• Stan Notebooks in the Cloud: Mitzi Morris
• Model-based Inference for Causal Effects in Completely Randomized Experimen: JoonHo Lee, Avi Feller and Sophia Rabe-Hesketh
• Tagging Basketball Events with HMM in Stan: Imad Ali
• Model building and expansion for golf putting: Andrew Gelman
• A Dyadic Item Response Theory Model: Stan Case Study: Nicholas Sim, Brian Gin, Anders Skrondal and Sophia Rabe-Hesketh (note: source link points to fork of example-models)
• Multilevel Linear Models using Rstanarm: JoonHo Lee, Nicholas Sim, Feng Ji, and Sophia Rabe-Hesketh
• Predator-Prey Population Dynamics: the Lotka-Volterra model in Stan: Bob Carpenter
• Nearest neighbor Gaussian process (NNGP) models in Stan: Lu Zhang
• Extreme value analysis and user defined probability functions in Stan: Aki Vehtari
• Modelling Loss Curves in Insurance with RStan: Mick Cooney
• Splines in Stan: Milad Kharratzadeh
• Spatial Models in Stan: Intrinsic Auto-Regressive Models for Areal Data: Mitzi Morris
• The QR Decomposition for Regression Models: Michael Betancourt
• Robust RStan Workflow: Michael Betancourt
• Robust PyStan Workflow: Michael Betancourt (also uses PyStan 2 which is no longer supported)
• Typical Sets and the Curse of Dimensionality: Bob Carpenter
• Diagnosing Biased Inference with Divergences: Michael Betancourt
• Identifying Bayesian Mixture Models: Michael Betancourt
• How the Shape of a Weakly Informative Prior Affects Inferences: Michael Betancourt
• Exact Sparse CAR Models in Stan: Max Joseph
• A Primer on Bayesian Multilevel Modeling using PyStan: Chris Fonnesbeck (also: rendered HTML was deleted?)
• The Impact of Reparameterization on Point Estimates: Bob Carpenter
• Hierarchical Two-Parameter Logistic Item Response Model: Daniel C. Furr
• Rating Scale and Generalized Rating Scale Models with Latent Regression: Daniel C. Furr
• Partial Credit and Generalized Partial Credit Models with Latent Regression: Daniel C. Furr
• Rasch and Two-Parameter Logistic Item Response Models with Latent Regression: Daniel C. Furr
• Two-Parameter Logistic Item Response Model: Daniel C. Furr, Seung Yeon Lee, Joon-Ho Lee, and Sophia Rabe-Hesketh
• Cognitive Diagnosis Model: DINA model with independent attributes: Seung Yeon Lee
• Pooling with Hierarchical Models for Repeated Binary Trials: Bob Carpenter
• Multiple Species-Site Occupancy Model: Bob Carpenter
• Soil Carbon Modeling with RStan: Bob Carpenter

critical landing page real estate is taken up by the logo (which bounces down to land in the middle of the loops which represent the ideal HMC trajectory) - can we simplify this to just the logo and allow more content to be visible?

The subject says it all.

https://mc-stan.org/events/stancon2018Helsinki/#videos

Clicking on the Thursday link goes to a youtube 404.

Now that we added a ScalaStan link from our users/interfaces page, we can add the rest of the high-profile interfaces to Stan.

The big qustion is how to marke these interfaces as ones that the Stan project itself doesn't maintain. I tried to do that with the ScalaStan section in the current (as of this feature request) interfaces page.

I think we should include at least these two packages:

• brms
• rethinking

http://mc-stan.org/about/team/

Wishlist item: having the citations page available in BibTeX format.

Solution suggestion: I don't know how the build process of the site works, but perhaps jekyll-scholar could be useful.

the Stan documentation has been converted to RMarkdown format which is markdown plus latex's mathjax for the math and R chunks for producing charts and graphs and running programs.
we are using the R bookdown program to generate the HTML documentation.
the generated docs have mediocre in-document search.

it would be possible to provide better cross-document search via lunr.js which provides decent lucene/elasticsearch-like behavoir. with lunr.js search is client-side. the sticking point is that the client must first download the search index which can be quite large - once the search index is downloaded, search is lighting fast.

this would be a nice project for a javascript developer.

They're coming out as "Stan -" and "Documentation -" on Safari.

Right now, there's a link out.

That should be marked that it's going out to GitHub, or ideally, it should link to one of our internal pages with the same structure as the other interfaces.

Summary:

Stan Website's documentation page https://mc-stan.org/users/documentation/ does not link to the CmdStan User's Guide. The page does have links to RStan and PyStan documentation (although the PyStan link is 404) but no mention of CmdStanR or CmdStanPy, either.

Here's the Korea group we heard about in today's meeting:

https://www.facebook.com/groups/StanKorea

and there are more Stan meetup groups we can link to.

From issue stan-dev/stan#2743 originaly submitted by @lismartin72.

Hi, I am a new staff member at NumFOCUS. We have updated our donation provider from Flipcause to Salsa. The new link is:

https://numfocus.salsalabs.org/donate-to-stan/index.html

Please update your website accordingly. : )

Thanks,

Lisa Martin
NumFOCUS

All of the search on the site is broken from Google. And any link that existed is now broken. It's standard operating procedure in these cases to leave redirect links in place of the dead links. Google's smart enough to index the new things.

Is there an easy way to move the workshops material over? Or did you drop it on purpose?

This is from @andrewgelman, who wants to reduce the number of menu items on the top page in order to focus user search. The plan will require more clicks to get to some things but there will be fewer choices at each stage.

• What should we do about the workshop pages and other "hidden" content?

• Any URLs that go away should forward somewhere reasonable or be replaced with a page to manually follow.

Here's the new plan from @andrewgelman:

Home Page

As is.

Top-level Menu Items

These will be the menu choices on the home page and the targets of the menu items:

• Interfaces
• Documentation
• Issues
• Community

The first three will remain the same.

Community will get a landing page with links to all the previous pages:

• Users list and dev list (was what we called "Community" before)
• Bug Reports & Feature Requests
• Events
• How to Cite & Citations
• Team
• Developer Resources
• Contributions & Supporters

If we can figure it out, it might make sense to have a submenu of some kind. Otherwise, each of the pages will only have the top-level menu items and it'll be a challenge as to how to lay out that community page index, as it'll require a bit of explanation for each item and some scrolling.

What we have now

Thsi is just an outline of the current site (as of this issue submission):

CAPS are menu pages. Bold items have their own web page.

HOME

• what is Stan
• how to get started
• how to donate (remove NuMFOCUS logo and simplify link to contributions)

INTERFACES

• Language
  • RStan, PyStan, ...
• Higher Level
  • RStanArm
  • ShinyStan
• Lower Level
  • Math library
  • Stan library
• Useful tools
  • Loo
  • Bayesplot
  • rstantools

DOCS

• Manual
• Example Models
• Case Studies
• Tutorials
  • video
  • web tutorials
  • papers
  • books

ISSUES

• Bug Reports
• Feature Requests
• Submit fixes

EVENTS

• meetups
• talks
• courses
• cons

COMMUNITY

• users group
• dev group

CITE

• How to cite Stan
• Books about Stan
• Boosks with examples translated to Stan
• Papers about Stan
• Papers using Stan
• Theses using Stan
• Papers using NUTS
• Software using Stan
• Software using NUTS

TEAM

• Core Developers
• Alum developers

DEVELOP

• suggest or discuss code patches
• contribute code patches
• to do list and projects
• how to join the dev team
• source repos

SHOP

• US shop
• UK shop

SUPPORT

• Contribute to Stan
• Donors
• R&D grants
• Industrial consulting

We also have a hidden page that doesn't have a menu item or any links to it from the site:

Workshops

• birds
• icl_epi
• ...

we should indicate what the license is on all the material, including the workshop handouts

this page:
https://mc-stan.org/users/documentation/
should contain links to earlier versions of the docs.
online docs exist since 2.18 (functions ref added for 2.19)

Both mc-stan.spreadshirt.com and mc-stan.spreadshirt.co.uk are dead links on https://mc-stan.org/shop/.

I also could not find Stan at all on the Spreadshirt website to let me update the links on that page via a pull request, so perhaps we can no longer infer in style?

On the Team page under Join the Development Team the link where is says "read about how to contribute to Stan" seems to be broken (it points to a non-existent Contribute page). Is the intent to link to the github wiki or just to the "Development" page on the website?

Link is broken to CmdStanPy here: https://mc-stan.org/users/interfaces/

Add syllable for ISBA 2016 workshop along with the to be used data.

... also add the ASA event in October.

add new page and link from interfaces page

I went to https://mc-stan.org/documentation/case-studies/grsm_latent_reg.html and got the following message:

mc-stan.org uses an invalid security certificate. The certificate is only valid for the following names: *.github.com, github.com, *.github.io, github.io Error code: SSL_ERROR_BAD_CERT_DOMAIN