I pushed changes for figure 11 too

I’m in the office, stop by when you get in

I am teaching a lab at 10, I will come after.

only 4 lines to go

My version has 14 lines to go?

I just pushed a change

Indeed. So it’s now 2 lines to go.

I pushed some updates as well

@rjnw what font did you use in the plots?

because I don’t have it and thus I can’t usefully edit the pdfs

I don’t have any specific setting, just the default that racket chooses

what are you trying to do?

squash the legend a bit

it’s probably ok

if you want to try, you can open it in inkscape

hmm let me see

the paper on my machine is officially 12 pages

woo

me too (with two cites to spare)

shall I increase the size of plots? :smile:

lol

what about the appendix?

i think that’s the next thing to get working

and also we need to refer to it

I moved the stuff in a different file appendix.tex as jacques suggested

woohoo!

but it is just raw tex

Unfortunately, I may have to quit soon. I’ve been fighting a nasty headache all day. It was at bay for a while, but it seems to be coming back.

for the appendix, one possibility would be to \input it in the file itself, then use pdftools to cut it out again. We submit pdf anyways, right/

?

yes

that’s now pushed

make pipeline-split.pdf will do the right thing

@rjnw can you improve the caption on fig 12 in the appendix to say it’s the source of the code in fig ?

yes

Wow @samth that was fast.

i had the code ready from another pldi submission

and was already doing it when you said that

i’ve officially submitted

everyone should check the hotcrp site for their conflicts etc

and more editing is always good

@ccshan do we expect to hear more from you tonight?

wonderful, thanks. [And my illusion of your coding speed were much nicer than the truth… :wink: ]

I’ll do conflicts now, and then sign off.

Hey yall, I’m in the middle of Section 3.2, doing well with shortening and editing.

No new conflicts from me.

@ccshan excellent

@ccshan you should make sure to pull asap, so as to minimize conflicts from edits (I did small things all over). But I’m done, so after that you’ll be safe.

i now have only 36 pages of PLDI papers :slightly_smiling_face:

@samth can you check figure captions, especially figure 8

Sorry, I had not quite pushed everything I had editing. Now pushed.

Now 4 cites to spare.

Make that 12 cites to spare.

I guess you could bring some stuff back if you felt that keen…

In any case, thanks all. I really am off now.

I haven’t looked at any section other than Section 3. So if you feel like bring stuff back, do it now, or I might be tempted to bring intuition back to Section 3.3 when I get there.

i considered bringing back the longer title, but it didn’t fit then

I have not looked at sections 5, 6, and the appendix at all.

Can someone please fix the multiply-defined \label{e:distrib} in the appendix

@samth I scaled the legend for figure 10 so it doesn’t overshadow augur’s line

@ccshan what’s the natbib1 pkg?

I can’t build the paper because I need it, but I can’t find it on the internet

It’s mine and it should be in the ppaml repo

Doesn’t git pull give it to you already?

It defines \citearound and \citets

it’s not linked into the pipeline dir

Oops. Fixed in the repo (probably you fixed it yourself too already)

We have a half page free. Nobody wants to add their favorite dad joke or anything/

I think we need to add text along the lines we discussed to sections 5 & 6

agreed

I just did that in 6

let me know what you think

Ok, I’m almost done with section 3

I did that in 5 as well

I pushed Section 3 and I’ll read Section 6 (then 5) now. We still have almost a column to fill.

I restored the full definition of unproduct. Now 10 lines to fill

Table 1 (startup times) are in seconds, right?

yes

Would it help to put our testcode/hksrc in an appendix?

I doubt it

we could upload it as additional supplementary material

I changed the first evaluation goal to “compared to existing systems that execute different inference algorithms” - right?

Let me encourage you to spell-check and then submit

well, i wanted to suggest that say, jags is doing something similar to us but not the same

but this is an issue of connotation not denotation

I think it’s fair to say that collapsed Gibbs and uncollapsed Gibbs are not the same inference algorithm

sure, i agree with your sentence

(and that’s good for us — we handle this algorithm that no one else does)

ok

just saying why i wrote it that way

Got it

ok will spell-check

So, for me the remaining task is to edit Sections 4–7. And that’s an anytime task, which is why I encourage you to submit. I want you to go to sleep if you want, so I want to make sure I can submit any minor improvements I make. How do you deal with the appendix currently?

you run make pipeline-appendix.pdf and then you submit that as the appendix

you run make pipeline-submit.pdf to generate the pdf to submit

Got it, thanks!

spelling changes pushed

i resubmitted

done for the night

Thanks. Good night! (pats back)

(By the way, the page breaks you submitted are different from the ones I get, and I suspect it’s the version of newtxmath (in pipeline.log))