IMO one/week is more realistic than one/day. One of the stated reasons for doing this is:
It would give some of us the motivation to actually go through the Sequences finally
It seems like the above goal would be better realized with a slower pace; it would be pretty easy to miss a posting or two if they are posted one/day. Also, what is the rationale for this:
I recommend re-posting the entire article to LW, including any edits or additions that are new in the book
It seems unnecessary to post the entire article; why not just have a link to the original article? Are there a lot of edits in the book version? If so, and if copyright is not an issue, why not just edit the original post to match what is in the book (assuming this could be done easily). If copyright issues preclude this, then presumably the same copyright issues would preclude posting the book content as new postings. It seems like it will be confusing to have two potentially different versions of each post available on LW.
So I recommend the following:
One post a week
Each post would contain a link to the original post
If possible, the facilitator would provide a summary of the original post and a list of questions/points as a catalyst for discussion, like in the Superintelligence reading group (albeit this would require a fair bit of effort on the part of the facilitator)
Consider updating the original post to match the print version (if there are no copyright issues or other issues)
Seconded, this was my reason for suggesting one a day. I believe there are over 800 articles, so even at one a day that’ll take a few years! [Edit: Correction, it seems there will be about 300 articles—see link below.]
Once a day was pretty much the original pace as well.
I was originally for a pace of two per week, just knowing my own work schedule. But if there are truly going to be 800 articles represented in the book, then one a day is the only workable solution. Do we know that the book will be broken out into something like 800 articles?
I was wrong about the number—the number is approximately 300 articles. I’m basing this on the kickstarter page for the audio version, on the sidebar under where it says “Pledge $50 or more”.
Still, 300 articles would take almost a year at 1 per day and almost 6 years at once per week.
One additional thing to note is that most of the articles aren’t that long. If you can more or less keep up with Slate Star Codex then I’d guess you should have no problem with once a day Sequences. If you missed a day or two, or even a week or two, it wouldn’t take you all that long to catch up.
Maybe we should look at a compromise: Would every other day (so about two years total) maybe work better?
Looks like it’s time for yet another poll. My, this thread is getting rather full of those.
If you split up some of the longer articles, you might be able to get it to exactly 365 days of blog posts. :)
The eBook will be organized into 26 sequences, all of similar length; so if you want to start new discussion threads, perhaps you should do one thread per sequence rather than one per blog post.
26 sequences one every other week is precisely 52 weeks = the number of weeks in a year. Not bad.
Assuming there are about 300 articles total, that comes out to about 11 or 12 articles per sequence on average, which at one every other week is a little less than once per day.
I think I’d still prefer an article per day (or every other day, as per earlier poll), but I’ll let others weigh in on this. Should I do yet another poll?
I’d sooner go for one a week; I think that’s closer to the likely reading pace and it means we quickly find out whether it works. We could easily follow it with an article-a-day presentation if that’s what we think is best after learning from the sequence-a-week presentation.
It would facilitate new discussion. Assuming we want people to comment on the new rather than the old thread, it would be annoying to keep having to refer back to an old post when commenting on the new post.
Much simpler to just copy/paste the old post & add in some edits than to provide a summary.
Only Eliezer can update the original post, as far as I know. So if we want to put in the edits we’ll need a new copy.
But a link to the original post is obviously a good idea.
IMO one/week is more realistic than one/day. One of the stated reasons for doing this is:
It seems like the above goal would be better realized with a slower pace; it would be pretty easy to miss a posting or two if they are posted one/day. Also, what is the rationale for this:
It seems unnecessary to post the entire article; why not just have a link to the original article? Are there a lot of edits in the book version? If so, and if copyright is not an issue, why not just edit the original post to match what is in the book (assuming this could be done easily). If copyright issues preclude this, then presumably the same copyright issues would preclude posting the book content as new postings. It seems like it will be confusing to have two potentially different versions of each post available on LW.
So I recommend the following:
One post a week
Each post would contain a link to the original post
If possible, the facilitator would provide a summary of the original post and a list of questions/points as a catalyst for discussion, like in the Superintelligence reading group (albeit this would require a fair bit of effort on the part of the facilitator)
Consider updating the original post to match the print version (if there are no copyright issues or other issues)
At one a week I think it will take decades to complete.
Seconded, this was my reason for suggesting one a day. I believe there are over 800 articles, so even at one a day that’ll take a few years! [Edit: Correction, it seems there will be about 300 articles—see link below.]
Once a day was pretty much the original pace as well.
I was originally for a pace of two per week, just knowing my own work schedule. But if there are truly going to be 800 articles represented in the book, then one a day is the only workable solution. Do we know that the book will be broken out into something like 800 articles?
I was wrong about the number—the number is approximately 300 articles. I’m basing this on the kickstarter page for the audio version, on the sidebar under where it says “Pledge $50 or more”.
Still, 300 articles would take almost a year at 1 per day and almost 6 years at once per week.
One additional thing to note is that most of the articles aren’t that long. If you can more or less keep up with Slate Star Codex then I’d guess you should have no problem with once a day Sequences. If you missed a day or two, or even a week or two, it wouldn’t take you all that long to catch up.
Maybe we should look at a compromise: Would every other day (so about two years total) maybe work better?
Looks like it’s time for yet another poll. My, this thread is getting rather full of those.
What pace would you prefer?
[pollid:828]
If you split up some of the longer articles, you might be able to get it to exactly 365 days of blog posts. :)
The eBook will be organized into 26 sequences, all of similar length; so if you want to start new discussion threads, perhaps you should do one thread per sequence rather than one per blog post.
We should totally try this before doing sequence reruns—do a discussion group on, say, a sequence a week.
26 sequences one every other week is precisely 52 weeks = the number of weeks in a year. Not bad.
Assuming there are about 300 articles total, that comes out to about 11 or 12 articles per sequence on average, which at one every other week is a little less than once per day.
I think I’d still prefer an article per day (or every other day, as per earlier poll), but I’ll let others weigh in on this. Should I do yet another poll?
I’d sooner go for one a week; I think that’s closer to the likely reading pace and it means we quickly find out whether it works. We could easily follow it with an article-a-day presentation if that’s what we think is best after learning from the sequence-a-week presentation.
FYI, each sequence is (very roughly) 20,000 words.
Assuming it is slower to read than the standard 200 wpm, that’s still only a couple of hours each; seems doable!
Poll:
[pollid:837]
Reasons for re-posting the whole thing:
It would facilitate new discussion. Assuming we want people to comment on the new rather than the old thread, it would be annoying to keep having to refer back to an old post when commenting on the new post.
Much simpler to just copy/paste the old post & add in some edits than to provide a summary.
Only Eliezer can update the original post, as far as I know. So if we want to put in the edits we’ll need a new copy.
But a link to the original post is obviously a good idea.
[pollid:827]