Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Edit: Making this thread public due to some concerns of certain users. Hopefully this will provide some transparency.
----------------------------------------------------
It has been brought to my attention that we lost a significant amount of viewers over the past week.
Now obviously people come and go, however it seems something happened on the 16th that led a number of subscribers to flee, and a bunch of people to stop watching.
I was not around at the time, so I don't have first hand knowledge of anything weird going on with the channel at that point in time, such as possible defacement, strange messages sent out, or something of that nature. However, checking some logs, it appears we've had over 40 re-encodes uploaded that day. I think that's some kind of record for us. While I'm all for breaking records, based on our loss in viewership we should probably avoid flooding the channel with re-encodes.
For the future, let's probably keep the amount of re-encodes uploaded in a day to under 10. Thank you.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Some users feel that this flooding of re-encodes has negatively affected the viewership on new movies published in close proximity to this flood. Is there some way we can make amends to the authors of said movies?
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
There's many factors involved in viewership. New, repeat, or re-encode. Whether it's finals season or not. Whether other sites also advertise it. Length of the movie too.
However here we're talking a loss of 100,000+ views, which affects more than just movies published close to the big drop off days.
I'm not really sure what "amends" can be made to the authors of all the movies that lost views here. Let's learn from this and improve for the future.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I'm not really sure what "amends" can be made to the authors of all the movies that lost views here. Let's learn from this and improve for the future.
I think acknowledgment of the issue and apologizing for it and reassuring people that it will not happen again in the future.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
In the interest of reducing these sorts of issues for movie authors, I'd also suggest limitations on publishing re-encodes directly after a new movie publication, so the new publication gets its fair time on top of subscription feeds before it may get hidden behind re-uploads of older movies. This has also been a noted issue recently.
For example, I'd suggest not to publish re-encodes for at least one or two hours after a new movie publication. This can be finetuned more, it is just to get the general point across.
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
I'm not really sure what "amends" can be made to the authors of all the movies that lost views here. Let's learn from this and improve for the future.
I think acknowledgment of the issue and apologizing for it and reassuring people that it will not happen again in the future.
It's a nice idea, but I think that words are meaningless without going with actual actions first. Since the movie that has been damaged the most is HL's warpless run (re-encode flood started 8 minutes after publication), what about re-uploading its encode to the YouTube channel? And maybe replacing the logo with something less menacing.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Why was that movie damaged the most?
We lost over 100,000 views, which affects many movies. The recently published Contra is also looking a bit low.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Because it was published 8 minutes before the re-encode flood started.
Nach wrote:
We lost over 100,000 views, which affects many movies. The recently published Contra is also looking a bit low.
The Contra movie has been published 4 days before that. Indeed, it got damaged too, and it could get something for improving its visibility too.
In any case, did anyone consider the idea of making the obsolete movies as unlisted on the YT channel? After all, most people are not interested in obsoleted movies, and the few people that are, are usually forced to find them on the main site anyway.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
ThunderAxe31 wrote:
Nach wrote:
Why was that movie damaged the most?
Because it was published 8 minutes before the re-encode flood started.
I'd like actual proof, not conjecture.
Nach wrote:
In any case, did anyone consider the idea of making the obsolete movies as unlisted on the YT channel? After all, most people are not interested in obsoleted movies, and the few people that are, are usually forced to find them on the main site anyway.
Maybe that's an idea.
Pros? Cons?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Joined: 10/12/2011
Posts: 6438
Location: The land down under.
Pro: People don't see old movies on the YouTube.
Con: People don't see old movies on the YouTube.
Con: People don't see the History of a movie on YouTube if they so desire to wanting to see it.
Con: People won't be able to search on YouTube for a specific movie if it's on TASVideosChannel.
Con: People go to WebNations.
Pro: Forces people to search on TASVideos Website.
Con: Forces people to search on TASVideos Website.
Con: SystemTVCControl isn't automatic for a long time. So manually making a Video Unlisted takes more time than just hitting Update and making it have [Obsoleted] next to its name.
I'm obviously missing a lot more Cons, and some few more Pros.
But that's what I have on the top of my head.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
I'd rather suggest publishing reencodes with some code-wise delay/timeout. For example, TVC logic could track the reencodes recently published, and only allow one per day for example. You can send it a command to publish 100, and it'd proceed them one at a time instead.
Pro: can't be abused even if we wanted.
Con: requires logic to be written, which might be tricky.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Pro: People don't see old movies on the YouTube.
Con: People don't see old movies on the YouTube.
Con: People don't see the History of a movie on YouTube if they so desire to wanting to see it.
Con: People won't be able to search on YouTube for a specific movie if it's on TASVideosChannel.
Con: People go to WebNations.
Pro: Forces people to search on TASVideos Website.
Con: Forces people to search on TASVideos Website.
Finding a specific obsoleted video with the YouTube search engine is often impossible.
Also, it's really confusing to see an obsoleted movie without having the obsoletion chain in plain view, which can only be found on TASVideos.
Spikestuff wrote:
Con: SystemTVCControl isn't automatic for a long time. So manually making a Video Unlisted takes more time than just hitting Update and making it have [Obsoleted] next to its name.
Can't this be fixed?
feos wrote:
I'd rather suggest publishing reencodes with some code-wise delay/timeout. For example, TVC logic could track the reencodes recently published, and only allow one per day for example. You can send it a command to publish 100, and it'd proceed them one at a time instead.
But then, wouldn't the reencodes just get piled up in an ever-growing queue?
I insist about making the obsolete videos unlisted. I'm sure that no one would really mind, and if you're unsure, just ask directly to the community.
In any case, we could still make some exceptions for obsoleted movies that we deem worthy.
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Oh you know what. Linking to obsoleted videos from the ones that obsolete them, and having the obsoleted ones unlisted, would be identical to what we do on the site!
The TVC updater needs to be fixed so badly! I suspect that if the updater part is fixed, some part of the code used for publishing will also be fixed.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
feos wrote:
The TVC updater needs to be fixed so badly! I suspect that if the updater part is fixed, some part of the code used for publishing will also be fixed.
I was hoping to fix it last week, then I ended up in the hospital.
I'd fix it now, but I can only use one hand, which is super hard to develop with.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
This was all dealt with back in July-August over IRC, but I'll put a recap up:
The TVC updater issue was fixed, bug turned out to be a change in YT API:
[Fri Jun 15 2018] [19:08:18] <feos> Nach: will you have time on tvc updater bug this weekend?
[Fri Jun 15 2018] [19:09:25] <Nach> feos: talk to me about it on Sunday
-------------------------------------------
[Wed Jun 27 2018] [23:46:12] <feos> Nach: if rating discussion (and specifically my suggestions) look bothering, they can be completely ignored if it helps you find time on tvc updater :)
[Wed Jun 27 2018] [23:47:02] <Nach> feos: I didn't have a computer that was usable for stuff till yesterday
[Wed Jun 27 2018] [23:47:17] <Nach> due to a weird string of events with multiple hardware dying on me in the past few weeks
[Wed Jun 27 2018] [23:47:25] <feos> right
[Wed Jun 27 2018] [23:47:30] <Nach> so now I need to catch up on some stuff
[Wed Jun 27 2018] [23:47:34] <Nach> then I can look into that updater
[Wed Jun 27 2018] [23:48:01] * feos 's hope wakes up
[Wed Jun 27 2018] [23:49:04] <Nach> I have multiple computers in my lab, as well as tons of parts, but my 3 top machines all had major components stop working within weeks of each other, completely out of the blue
[Wed Jun 27 2018] [23:49:16] <Nach> and the other machines are so old, they're not exactly useful these days
[Wed Jun 27 2018] [23:50:34] <Nach> I mean they're useful for stuff other family members do, but not development stuff
-------------------------------------------
[Sat Aug 18 2018] [21:26:49] <feos> Nach: with this out of the way, I can start asking about tvc updater more often, until you finally have time for it. how often should I be asking?
[Sat Aug 18 2018] [22:28:32] <Nach> feos: I probably will have time in a few days
[Sat Aug 18 2018] [22:28:42] <Nach> and I'm able to type with both hands again!
[Sat Aug 18 2018] [22:28:59] <feos> \o/ for both news!
[Sat Aug 18 2018] [22:28:59] <ThunderAxe31> hurray!!
[Sat Aug 18 2018] [22:29:05] <Nach> also, as of this past week, I can finally pick stuff up with both hands, which is fantastic
[Sat Aug 18 2018] [22:31:10] <Masterjun> that's pretty useful
[Sat Aug 18 2018] [22:31:23] <Nach> yes, it is
[Sat Aug 18 2018] [22:31:34] <Nach> and sometimes you take stuff like that for granted
[Sat Aug 18 2018] [22:31:49] <Nach> and realize how difficult things are with just one hand
[Sat Aug 18 2018] [22:32:28] <Masterjun> I sometimes try to type with only one hand and it takes like 4 times as long
[Sat Aug 18 2018] [22:32:48] <Nach> indeed
[Sat Aug 18 2018] [22:33:07] <Nach> I'm just glad I'm ambidextrous, otherwise it would have made things even more annoying
[Sat Aug 18 2018] [22:33:33] <Masterjun> neat
[Sat Aug 18 2018] [22:34:10] <Masterjun> is it possible to learn that or is that rooted in the brain somewhere?
[Sat Aug 18 2018] [22:36:46] <Nach> the answer I'm pretty sure is both
[Sat Aug 18 2018] [22:37:00] <Nach> there's a whole field of study how learning new motor controls enhances your brain
[Sat Aug 18 2018] [22:37:20] <Nach> and augments your brains handling of more complicated motor control in general
[Sat Aug 18 2018] [22:38:04] <Nach> techniques to start with: Try brushing your teeth with your other hand, eating your cereal in milk with your other hand, and using your mouse with your other hand
[Sat Aug 18 2018] [22:38:23] <Nach> keep it up for a week, it should start feeling comfortable
-------------------------------------------
[Wed Aug 22 2018] [22:07:44] <TASVideoAgent> root /var/svn/repos/tasvideosorg r6193 : Some cleanup, saner parsing, and fix YT updating, in the name of feos (by Nach).
[Wed Aug 22 2018] [22:08:17] <MemoryTAS> OH
[Wed Aug 22 2018] [22:08:44] <Mothrayas> oh
[Wed Aug 22 2018] [22:08:51] <fsvgm777> Oh?
[Wed Aug 22 2018] [22:10:01] <Nach> and yet feos is silent like a lump of tuna
[Wed Aug 22 2018] [22:10:17] <feos> I'm opening a ton of tabs to properly test it!
[Wed Aug 22 2018] [22:10:18] <Mothrayas> feos, please stop being a lump of tuna
[Wed Aug 22 2018] [22:12:14] <Nach> so the bug was hilarious
[Wed Aug 22 2018] [22:12:26] <Nach> we make sure to only update YT if its our channel
[Wed Aug 22 2018] [22:12:40] <fsvgm777> feos: Wake up, you lump of tuna!
[Wed Aug 22 2018] [22:12:45] <Nach> YT used to return the channel for the movie in their API as all lowercase, so that's what we checked for
[Wed Aug 22 2018] [22:12:57] <Nach> but now it returns the proper case, and was mismatching our case sensitive check
[Wed Aug 22 2018] [22:13:23] <feos> having to watch the supper on the pan doesn't speed me up
[Wed Aug 22 2018] [22:13:35] <Nach> is the supper in the pan a lump of tuna?
[Wed Aug 22 2018] [22:13:52] <Mothrayas> dang youtube constantly changing their api and breaking things all over
[Wed Aug 22 2018] [22:14:26] <fsvgm777> Dang YouTube breaking 60 FPS on anything lower than 1080p.
[Wed Aug 22 2018] [22:14:47] <Mothrayas> dang youtube just breaking things all over
I spoke to our publishers, and we agreed to take feos' suggestion about not flooding.
I don't have an exact log, since I wasn't on one of my normal computers at the time (see above quote as to why). IIRC, we agreed to keep amount of new stuff appearing on TVC to 10 a day. Although my exact memory of that time was fuzzy also due to loss of blood and stuff. If someone has more accurate info as to what we agreed to, please post.
Regarding some encoders/publishers allegedly trying to bury publications, I've spoke to all involved at the time. They gave me their word that they will not perform any activities which look like publication burying in the future. As these are private discussions, they will not be quoted.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
I think the cooldown for publishing is 3-4 hours. For pushing reencodes, I don't know if anyone wanted it to be shorter.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Now obviously people come and go, however it seems something happened on the 16th that led a number of subscribers to flee, and a bunch of people to stop watching.
I was not around at the time, so I don't have first hand knowledge of anything weird going on with the channel at that point in time, such as possible defacement, strange messages sent out, or something of that nature. However, checking some logs, it appears we've had over 40 re-encodes uploaded that day. I think that's some kind of record for us. While I'm all for breaking records, based on our loss in viewership we should probably avoid flooding the channel with re-encodes.
For the future, let's probably keep the amount of re-encodes uploaded in a day to under 10. Thank you.
