The story behind New Year’s Programme 2023 for Double Life World

This post described my side of story about 2023 DLW NYP. For more information about this project, please:

A never-intended project

After the semi-failure of 2022’s New Year’s Programme, as well as the no-update notice published by the officials, I was planning to conclude the GCGDS altogether – that include the archiving of many websites, wrapping up of some unfinished ends, as well as the disposal of the New Year’s Programme for 2023. This would be very sensible – after the final update, it is expected that only a marginal number of players would stick around, and few would like to spare their time watching a potentially 20-munute-plus video about a ‘dead game’. Actually, uploading any video longer then 10 minutes on bilibili would be a very ‘courageous’ move – and that is not a good thing.

However, some of the core members of the 2022’s programme, as well as some of the counsellors of GCGDS, suggested that it might be very possible for all of us to produce one last thing – to mark the end of all of these, and function as one of our last and best memories about this fantastic community, and the time and stories we have all experienced.

And – I went with it. It is, without doubt, not a decision would be made by a rational man. It is bound that this thing would not be a grand (restricted by the subscriber count and lack of support), and it is highly anticipated that the overall revenue the whole team could get would be quite marginal compared to the time and effort we must put into it.

Yet still, we went into the preparation phase.

Not enough people, not enough time

The very first thing stroke us is the lack of people. After the final update notice, we have witnessed a significant drop of active users in our accounts and inaccurately estimated a similar trend for the game itself though several reliable sources. The number of content creators willing to confirm to a future video production is very lacking, and if we do not do something, it is sure that we could not ensure a minimal length of 20 minutes.

In light of the shortage of people, a public recruitment notice was issued. The recruitment process lasted for almost two months while I did everything I can to get more creators on board, but in the end, we were only able to secure five productions, with another three having an uncertain or questionable statements, and only one of them were recruited from the public notice. Well… in the end, all of the three were dropped at various stages.

Although we manage to ensure the programme will run at least 20 minutes, it is still uncertain what kind of difficulties we might encounter during our period of production. The original plan was to make this possible within the year of 2022 so we thought it might be a good time for us to have this thing aired on the 3rd anniversary of the game. By general measurement of productivity, it should be fine… right?

No. The pandemic has something to argue. With all kinds of special and changing coronavirus regulations being implemented in China, along with the fact that many are already on their way into the fierce working class, almost all of the creators are being affected one way or another in their offline lives and makes them hard to spare time to produce the videos before the anniversary date. Understandable, but a bit too realistic as well.

One month before the production deadline, we made the decision to aim for a New Year’s Programme instead of an anniversary one, giving us two more months for all to do their part when they are available.

Not official, less restraints

Because bilibili do arrange and support a series of New Year’s Programme videos every year, and we got that in the 2022 round, we applied again in October 2022 when we could nail down a programme list.

However, for me it is more like a useless attempt: running this for an already-dead game is without doubt not promising in all aspects, and it would be highly irrational if they decide to accept us. I also asked several liaisons I have, and they all replied like – it is a good thing that you guys are planning and making this, but we have zero resources on hand to spend on anything about this ‘past’ game. Without many surprises, in November bilibili’s editors expressed their gratitude and ‘politely’ turned us down. To be honest, when they are not being able to decide in October is already telling something.

But that is not a dreadful thing to the creators. Last year during the preparation processes we encountered many barriers that are ‘official’ – like more strict control and prevention on out-of-character lines and contents throughout the programme: we only managed to get an exceedingly small proportion of lines and text that involve game characters approved in the last year. Also, lack of official support means we do not have to plan and coordinate a pre-defined hard release date, giving us ability to postpone any published date, including the publishing date of the publishing date.

Affected by these freedoms, as coordinator I also gave all these freedom to the creators to the maximum. I did not hard push for things to happen on-time, I do not actively monitor their progress unless they are willingly to asking for my opinion, and finally, I do not gather and provide resources for them unless they ask me to (and do my best if they asked). This is also an experiment for me – to see if this kind of supportive leadership without management may work out – and it did!

Although we all have things to tackle in our real life, in November and December I have witnessed substantial progress across all projects, with the earliest project being completed in late December, and most have entered the final stages in January. Considering that in January most have activities related to the Spring Festival, it is understandable that not much was made during that period.

And in the first full week in February, we suddenly got all the programmes ready – people had more time after the planned activities and is not working / studying, so the making of videos gained more priority, while I also found my timetable much flexible after the conclusion of the January assessments.

Before we move on…

It probably deserves a separate section to shout out something quite worthy to point out – bilibili’s editorial board really need polish skills on how to write this kind of messages in a more encouraging and acceptable way while preserving the dignity of both parties – it is exceptionally inappropriate to implicate any applying team is lacking or inadequate in either: experience in producing new year’s programmes; overall ability to cause influence (subscriber count); or, planning of the whole or singular programme. Although it sounds hard, but it is not. In comparison to their response, I wish to display a message from Cambridge during my failed attempt there, where they used just one paragraph and preserved the dignity of all applicants:

We also face the same challenge when replying to some of the candidates we have. Some productions are hardly acceptable, and we need to craft our words carefully to explain why we decide to turn that down while not being either too authoritative or having a possibility to make applicants feel what they are doing is a failure. To achieve this, we usually just give bullet points of reasons we feel might be of concern and tell them we are still amazed by what they have achieved and the unlimited potential of theirs and apologise for not being able to take that in at this moment while encouraging them to submit again when they feel confident to address the concerns. At times, we would actively encourage them to publish it themselves when we do not certainly believe the likely outcome is not promising, even though it would make us humble.

An alternative and pure players’ creation

After we have confirmed the final rundown and nailed down a targeted release date, everyone got much intense to prepare for the big event. For the spaces between individual programmes, this year I went with a minimal approach and applied the module borrowed from the 2nd Anniversary’s official video and use a clicking effect to make viewers think they are ‘unlocking’ all the sections one by one, partly due to stories were not in place in time. The whole team managed to handle all the technical details and finished any refinements needed in the final week and made it possible for the final output 72 hours before the livestream.

When I finished the output of a low-resolution preview and projected it onto the wall of my room, it deeply stroke me that how long can I focus on watching this whole rundown without checking phone or other things – it is extremely addictive and enjoyable to watch compared to last year’s programme where the overall quality and ability to adhesive was lacking despite best effort being poured into every possible detail. I do not have a good explanation for this – and probably never will – but because everyone participating are doing this out of pure and deep affection.

Without the constrains from the official, we can plan things more closely when we are almost ready. We released the announcement only 6 days ahead compared to last year’s 31 days, and this time all the programmes only went through two lines of check (by creator and me respectively) for any issues that may impact the release, without the time-consuming higher-level checks by the officials. This made everything more harmonious – we can discuss all the concerns, instead of taking some hard orders.

For 2023, I decided to do both a livestream and a ‘playback’ – as lengthy videos would be very ‘courageous’ to release.^[1] But this is almost the opposite for a livestream, where longer the programme the better. I usually use a combination of techniques to ensure the maximum number of viewers may be accumulated by the time the programme have started, including creating appointments and start streaming early by arranging countdowns. The final videos were completed in early hours of Friday and luckily successfully passed the platform screening of bilibili.

Time flies, and we came to…

The Release Day

There’s no way to cover it up, I woke up late. It is not just one or two hours after when I should be, it is that if I did not check my phone and realised the catastrophic potential, I could entirely miss the airing period by spending it in dreamland.

Luckily, my bowel movement was gentle, and there are some quick ways to ejaculate urine (which I certainly do not recommend), and I was able to fully boot up my workstation with OBS loaded up and proxy promptly connected, just less than four minutes before the designated streaming start moment, which is 16 minutes before the appointment time.^[2] I rushed to the streaming control panel and pressed the ‘start streaming’ button on time.

But, although it looks everything did not go off the rails compared to the worst outcome, but a) my heart is pounding like an overloaded engine and my smart watch was ready to call 999 at some point, b) I skipped the 5-minute delay calculation and connection quality checking process, c) danmuku monitoring was initiated 3 minutes after the streaming has started instead of the planned 5 mins ahead, and d) playback with danmuku (which is a screen recording on a separate device) was cancelled.

It is quite weird that I am much more concerned about anything going sideways compared to the same time of last year – although the statistics are much lower, the consequence of any hiccup will be much gentle, and my general workload is marginal compared to what I had last year (setting giveaways in the process).

There is, again, only one good explanation: I cannot let down this great creation of pure and precious affection.

Luckily, the streaming concluded with remarkable success technical wise.

(Almost) a Statistical Nightmare

We gained a total revenue of a bit short of CN￥200 (about £24) through livestream and platform’s share. This is a much lower revenue compared to the CN￥2000 revenue last year. Since this year I did not use the advanced commercial-grade proxy software, and we do not have to pay for the shipping costs of all the giveaways, all the revenue we received was evenly split between all creators. I did not take a dime and gave a little to make the sum to exactly CN￥200.

Viewer wise, the livestream had a total viewer count of around 1000, but it is expected to be about 1/10 of the amount we gained last year, by estimating using the popularity value given by the platform. However, the playback video gained over 9000 plays in just two weeks, which is just 10% short compared to the last year – a proof that the overall quality is indeed higher.

Gain of subscriber count was not impressive, but it successfully pushed the total number to 4700+, which was the hundred mark the account has struggled with for months. So, not entirely a terrible thing, I guess?

But the numbers also implied that it would be extremely hard for us or me to ever arrange a third programme ever. But why this statement sounds a bit familiar?

[1]: Just a few days before the publishing of this post (and almost a month after the stream), there came a message that bilibili might be planning to change its algorithms in a way that could potentially give significant advantages to medium to long contents. Such rumours do not even exist when we started the project, and we were witnessing the direct opposite until the date of publishing for this post.

[2]: A proxy software is needed if streaming from where I am operating (United Kingdom) and the software I am using is known to not prune to be used with procedures other than it has planned, although being the most stable one among all the choices.