Page nesting about lost-stats.github.io

[NickCH-K]

I like this setup! I think GLS will likely get big enough at some point that it will need to be split further, but its not there yet.

Maybe take Linear Hypothesis Tests and put it in a section called "Statistical Inference" that would also include standard errors?

[grantmcdermott]

Hmm... I like the 'statistical inference' section header. I'll use 'research design' for the DiD stuff then to avoid confusion between section names.

[NickCH-K]

Sounds good to me

[grantmcdermott]

Okay @NickCH-K it's pretty much done (for model estimation at least).

https://lost-stats.github.io/Model_Estimation/Model_Estimation.html

Minor note: Since this is going to result in a bunch of broken internal hyperlinks -- still needs to be fixed -- I went ahead and also changed the underlying Estimation.md file to Model_Estimation.md to improve file naming <--> navigation consistency.

[grantmcdermott]

Oh, another question:

Shouldn't Time Series (and maybe even Machine Learning) be nested under Model Estimation?

[grantmcdermott]

If I'm being picky, we should probably split the 'Presentation' section into 'Figures' and 'Tables' subsections and move 'Summary Statistics' into the latter...

[NickCH-K]

Figures and Tables separate makes sense.

If Model Estimation gets all of Time Series and ML that's going to make it huge. Maybe Model Estimation gets a title that excludes those two somehow? I'm not sure what that would be.

[khwilson]

A note on the broken links: Would you like me to install htmlproofer? It can run on build and let you know what the broken links are.

[NickCH-K]

That would be awesome! It was a pain trawling through to find all the broken links the last time we did something like this, and the site was smaller then. Nick Huntington-Klein Assistant Professor of Economics Seattle University From: Kevin Wilson <[email protected]> Sent: Sunday, August 16, 2020 5:47 PM To: LOST-STATS/lost-stats.github.io <[email protected]> Cc: Huntington-Klein, Nicholas <[email protected]>; Mention <[email protected]> Subject: Re: [LOST-STATS/lost-stats.github.io] Page nesting (#55) A note on the broken links: Would you like me to install htmlproofer? It can run on build and let you know what the broken links are. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FLOST-STATS%2Flost-stats.github.io%2Fissues%2F55%23issuecomment-674603175&data=02%7C01%7Cnhuntington-klein%40seattleu.edu%7Cf3c950ce59df44ec628508d8424715e5%7Cbc10e052b01c48499967ee7ec74fc9d8%7C0%7C0%7C637332220365384490&sdata=O2nAB%2BrBf2PImN%2Bcf9VKAAB3EjnFp17hwSEDzNpgYpc%3D&reserved=0>, or unsubscribe<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAKV5TDOGZ2KGEX67MAPD5YTSBB4ZFANCNFSM4P74W75A&data=02%7C01%7Cnhuntington-klein%40seattleu.edu%7Cf3c950ce59df44ec628508d8424715e5%7Cbc10e052b01c48499967ee7ec74fc9d8%7C0%7C0%7C637332220365394448&sdata=SegDbu6EuL%2FJkXi%2Bj3LjiM1PSbDd0AcM8%2BSN3cKuCgk%3D&reserved=0>.

[khwilson]

Cross referencing #59 which adds in the proofer. Note that it requires switching the linking style as currently internal links are being implemented as absolute links (i.e., https://lost-stats.github.io/Time_Series versus just /Time_Series). There's a bit of a cleanup project to switch those, but afterwards, we can, if you'd like, make the GH Actions job fail if there are missing links.

(Note that this may not be desirable as, e.g., the Time Series section currently links to, say, ARIMA models even though the page doesn't exist. OTOH, perhaps the requirement for merge would all those pages get placeholders?)

[grantmcdermott]

Awesome @khwilson. On your follow-up points:

1. The use of absolute paths has been bothering me for a while, so I'm happy to see those go.
2. Getting the scaffolding up for the desired non-existent pages would make it easier for contributors to add code and ensure nesting stays correct. (Just edit the existing, blank template.) But I'll be honest and sayt hat I don't have time rn to dedicate to that!

[NickCH-K]

1. Agreed! 2. I don't have time for this either, but this would be a good solution (I've long wanted the default missing-page result to be a Wikipedia - style "write your own!" but I imagine this isn't possible. I don't have time for this now but could do it in September. On Aug 17, 2020 10:19 AM, Grant McDermott <[email protected]> wrote: Awesome @khwilson<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fkhwilson&data=02%7C01%7Cnhuntington-klein%40seattleu.edu%7C96a586250695404d829e08d842d1b386%7Cbc10e052b01c48499967ee7ec74fc9d8%7C0%7C0%7C637332815722918454&sdata=l1pB5N7mPD2FqJ2AYH8J%2B4%2BC9PCS7Yu5WbflDOE4oAI%3D&reserved=0>. On your follow-up points: 1. The use of absolute paths has been bothering me for a while, so I'm happy to see those go. 2. Getting the scaffolding up for the desired non-existent pages would make it easier for contributors to add code and ensure nesting stays correct. (Just edit the existing, blank template.) But I'll be honest and sayt hat I don't have time rn to dedicate to that! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FLOST-STATS%2Flost-stats.github.io%2Fissues%2F55%23issuecomment-675006200&data=02%7C01%7Cnhuntington-klein%40seattleu.edu%7C96a586250695404d829e08d842d1b386%7Cbc10e052b01c48499967ee7ec74fc9d8%7C0%7C0%7C637332815722928450&sdata=wFubfaIJfxyiChxE%2BHfaY0Rv6ls75wvoNmWGLOu%2FjSQ%3D&reserved=0>, or unsubscribe<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAKV5TDOXDKWWVXJNQ6BFX2LSBFRCDANCNFSM4P74W75A&data=02%7C01%7Cnhuntington-klein%40seattleu.edu%7C96a586250695404d829e08d842d1b386%7Cbc10e052b01c48499967ee7ec74fc9d8%7C0%7C0%7C637332815722928450&sdata=QrKd8JzACWB2KZqn%2Bd0g9ogARzfz6XXY%2F1F8NiuHe8M%3D&reserved=0>.

[khwilson]

1. Sounds good. We hopefully have a new season of interns arriving sometime in the fall, so there's some chance I can make this an intern project. Just have to haggle with the intern czar. :-)

2. You actually can create a custom 404 page in gh pages, GH serves the 404 from the address you linked to. See e.g. this nonexistent page. So a little JavaScript could probably do exactly what you're asking Nick.

2b. If you setup the TOC as a YAML, it's very easy to write a tiny script to just create all the blank pages and wouldn't take too long.

[grantmcdermott]

Okay, I've nested the Presentation section now too: https://lost-stats.github.io/Presentation/Presentation.html

FWIW I fixed the (absolute) paths here with a quick bit of sed work since otherwise none of the figures would render. I probably should do the same for the Model Estimation section... but maybe that should wait seeing as Kevin is setting up the HTML checker anyway? Maybe I'll see if I have time later.

I'm going to close this issue now since the main nesting cases have been taken care of.