Widen Manual Recruitment Filter

[The Nation of the People of the Nation]

As of late, manual recruitment telegrams have taken longer to deliver than API and Stamp telegrams.

I've run a few experiments where I make a puppet and immediately send a tg to it, and in the majority of cases the api or stamp tg from Europeia comes before my manual tg sent seconds after the nation was founded.

Through anecdotal evidence this seems to be a game wide issue. So is it possible for the filter to be widened and for more manual recruitment tgs to be delivered per minute?

[Sedgistan]

When a recruitment telegram is sent, it goes into one of three queues: the Stamp queue, the API queue and the Manual queue. Delivery is rotated between these queues to give each an equal chance. Your anecdotal evidence is just that.

Because almost no-one send manual recruitment TGs (less than a hundred sent in the last week), they are much more likely to be amongst the first-received telegrams compared to Stamps/API-sent recruitment telegrams, because there likely isn't anyone ahead of you in that queue.

[The Nation of the People of the Nation]

Because almost no-one send manual recruitment TGs (less than a hundred sent in the last week), they are much more likely to be amongst the first-received telegrams compared to Stamps/API-sent recruitment telegrams, because there likely isn't anyone ahead of you in that queue.

What do you mean by "less than a hundred sent in the last week?" There are ~350 in the queue.

[Sedgistan]

I mean exactly what I say. Where's your information coming from?

[Galiantus III]

Going by the tgqueue shard of the API, this is what the telegram queue looks like right now:

Manual: 374
Stamps: 52422
API: 7274

So for someone currently recruiting, the chances of a manual telegram reaching its recipient is an order of magnitude better than API, and API is an order of magnitude better than stamps. From what I've seen there are fluctuations in the relative effectiveness of stamps and API, but manual is consistently superior to both. Of course, the reason it ends up not being used as much as the others is people don't like work, and dealing with the rate limit can get pretty annoying.

[The Nation of the People of the Nation]

I mean exactly what I say. Where's your information coming from?

That Europeia alone has sent nearly 2000 this week alone?

[WayNeacTia]

I mean exactly what I say. Where's your information coming from?

That Europeia alone has sent nearly 2000 this week alone?

How exactly do you know this?

[The Nation of the People of the Nation]

That Europeia alone has sent nearly 2000 this week alone?

How exactly do you know this?

We have a tool that keeps track of how many tgs our recruiters have sent.

Anyways I alone have sent about 300 in the past week, the 100 number makes no sense.

[Sedgistan]

How exactly do you know this?

We have a tool that keeps track of how many tgs our recruiters have sent.

Anyways I alone have sent about 300 in the past week, the 100 number makes no sense.

Thanks. Mods have access to some internal data on TGs, and the "less than 100" figure is what it's showing for Manual. It's also showing 0 manual recruitment TGs sent by Europeia in the last 7 days. I can't explain the discrepancy.

[The Nation of the People of the Nation]

We have a tool that keeps track of how many tgs our recruiters have sent.

Anyways I alone have sent about 300 in the past week, the 100 number makes no sense.

Thanks. Mods have access to some internal data on TGs, and the "less than 100" figure is what it's showing for Manual. It's also showing 0 manual recruitment TGs sent by Europeia in the last 7 days. I can't explain the discrepancy.

Interesting, could that perhaps be looked into?

[Bernie]

Is it possible tag:template TGs are not being treated as manual except on the tgqueue api?

[Refuge Isle]

Eyeballing Refugia's internal records, there should be around 600 this week just from us.

[Homyland]

We have a tool that keeps track of how many tgs our recruiters have sent.

Anyways I alone have sent about 300 in the past week, the 100 number makes no sense.

Thanks. Mods have access to some internal data on TGs, and the "less than 100" figure is what it's showing for Manual. It's also showing 0 manual recruitment TGs sent by Europeia in the last 7 days. I can't explain the discrepancy.

I think you're looking at the manual recruitment queue, but TNoTPoTN and Galliantus are looking at the whole manual queue, without any tags.

[The Nation of the People of the Nation]

Thanks. Mods have access to some internal data on TGs, and the "less than 100" figure is what it's showing for Manual. It's also showing 0 manual recruitment TGs sent by Europeia in the last 7 days. I can't explain the discrepancy.

I think you're looking at the manual recruitment queue, but TNoTPoTN and Galliantus are looking at the whole manual queue, without any tags.

All of our recruitment tgs (any any other UCRs, it's illegal otherwise) are tagged as such.

[The Nation of the People of the Nation]

Has a problem been identified?

[Sedgistan]

I'm not yet convinced there's a problem anywhere, aside from with the stats mods get presented with on manual recruitment.

[Roavin]

I'm not yet convinced there's a problem anywhere, aside from with the stats mods get presented with on manual recruitment.

Sedge, with all due respect, you can't be convinced one way or the other yet, nor can you say at this point that there is or is not a problem, because your initial conclusion was based on ostensibly false data.

EDIT: apologies, I didn't realize this was a gravedig, I clicked through this link from elsewhere. My point stands, however.

[Sedgistan]

No problem with the gravedig; that's fine if you believe a bug report is still outstanding.

This report is based on "anecdotal evidence" (The Nation of the People of the Nation's words) that manual recruitment TGs are taking longer to deliver than they should. No-one else appears to have backed this up, so there's not yet a case for flagging this to admin for investigation. Give me more information.

(There is a separate issue that the little-used internal stats on recruitment by type appear to be wrong, but that is irrelevant to the bug report.)

[Eluvatar]

I think there are two issues with the stats.

One is that it's only looking at the telegrams from a region that got delivered. If a nation has already received a recruitment message from a region, it will not receive another from the same region for 28 days. So if a manual recruitment TG from a region is beaten into a nation's inbox by an API or stamp recruitment TG from the same region, it won't show up in the stats shown to moderators. I'm not sure how to fix this.

Another is that the moderator-shown interface is assuming any TG addressed to "tag:template" is an API TG, which is not true. This should be easy to fix.

I haven't investigated the actual claim. For that, I think the stats we're talking about would be insufficient even if they were accurate. I imagine I'd want to look at the following for all TGs processed in the last little while:

1. Whether they are manual/api/stamp
2. When they were queued
3. When they were processed (delivered or dropped)
4. Whether they were delivered or dropped

I'd want to look at what the distribution of the difference between the two times would be for manual, for api, and for stamp TGs.