This needs full consensus: at least 3 plusses and no negatives in 7 days. Ideally, we get unanimity before then.

From the meeting, jwb, decause, mattdm, robyduck, and langdon are +1, so we have the three +1s already. For unanimity, we need cwickert and tatica.

Oh, although Langdon is going to check my math. :)

Replying to [ticket:58 mattdm]:

{{{
Adjusted Percent of Percent of
FY17 Asked for Recorded FY16
Allocation Amount Spending

APAC: $11500 34.8% 151.3%
EMEA: $21650 65.1% 144.3%
LATAM: $10950 47.2% 99.5%
NA: $29400 49.0% 121.5%

}}}

Works for me and should work for EMEA. I cannot speak for the other regions though. I have some concerns about LATAM, they get less than they spent last year and due to bad tracking, this is already a subset of what they actually spent.

1. We approve this top level allocation
2. Back to the regions, who will come up with quarterly splits

What's the timeframe for this?

Doing quarterly splits means looking up what events fall into that quarter, look up how much budget was allocated for them, then talk to event owners where they can make cuts and get all of this in line with the total yearly budget. As this requires discussions among the event owners and the regional communities, I expect this to take some time.

Additionally, after each quarterly budget report (see https://budget.fedoraproject.org/), any unspent funds will revert to the council and be reallocated. (To be used for centralized projects like FADs, or to cover special requests by the regions.)

-1. Unlike the global/council budget, the regions already need to make cuts. If they can save money on event A in Q1, I see no reason why that should not use this money to spend the saved money on event B in Q2. In fact, they might only spend what that had initially planned, so I think it's unfair to take this money from them.

Replying to [comment:5 cwickert]:

Works for me and should work for EMEA. I cannot speak for the other regions though. I have some concerns about LATAM, they get less than they spent last year and due to bad tracking, this is already a subset of what they actually spent.

That is true; LATAM definitely came out the lowest here by any measure. That's largely because their ''request'' was significantly lower than that from the other regions, which doesn't really seem like a good reason to penalize. Do you have a concrete alternate proposal?

1. Back to the regions, who will come up with quarterly splits
   What's the timeframe for this?

There was talk about it being four weeks. If we consider that spending in the last two weeks of a quarter needs to be budgeted and accounted for in the next quarter (see meeting log), that could mean two weeks after quarter start.

Can you help me understand why this can't just be ''continuously'' kept up to date, though, making it just a matter of taking a snapshot?

Additionally, after each quarterly budget report (see https://budget.fedoraproject.org/), any unspent funds will revert to the council and be reallocated. (To be used for centralized projects like FADs, or to cover special requests by the regions.)

-1. Unlike the global/council budget, the regions already need to make cuts. If they can save money on event A in Q1, I see no reason why that should not use this money to spend the saved money on event B in Q2. In fact, they might only spend what that had initially planned, so I think it's unfair to take this money from them.

The goal is to make sure that the budgeted funds are actually all spent (in meaningful ways, of course). It isn't to penalize the region for underspending. I'm okay with modifying the above to say that regions can do their own internal quarter-to-quarter reallocation of unspent money as long as it is reported up (and as long as it really does end up being made use of). But if the region ''doesn't'' have a solid plan, we do still have plenty of central needs (or possibly there's a strong unplanned need in another region).

Replying to [comment:6 mattdm]:

LATAM definitely came out the lowest here by any measure. That's largely because their ''request'' was significantly lower than that from the other regions, which doesn't really seem like a good reason to penalize.

Their LATAM request was lower because they have learned their lesson of overbudgeting and underspending. APAC has overbudgeted way more, so I really think it's unfair to give LATAM less just because they are more realistic now.

Do you have a concrete alternate proposal?

AFAIR we approved the [https://pagure.io/fedora-budget/blob/master/f/FY17/adjustedannualbudgets/council/scenarios.txt#_25 hF/hDI scenario], which leaves USD 74,500 for regional support. The numbers above add up to USD 73,500, so there is a 1k margin. We could either
* give it to LATAM
* divide it by all four regions
* split among LATAM and APAC
* give APAC and LATAM the same amount, then it would match the [https://pagure.io/fedora-budget/blob/master/f/FY17/adjustedannualbudgets/council/scenarios.txt#_119 historic percentages].

Can you help me understand why this can't just be ''continuously'' kept up to date, though, making it just a matter of taking a snapshot?

I'm afraid I'm lost. You said the regions report quarterly splits after we approve the top-level allocation. This is a yearly thing, and so is the event planning in the regions. Therefor I assumed the quarterly splits were a yearly thing, too. Once a year, right tell the regions how much money they get and in return, they tell us how they plan to spend it throughout the quarters. How would one make this continuously?

The goal is to make sure that the budgeted funds are actually all spent (in meaningful ways, of course). It isn't to penalize the region for underspending. I'm okay with modifying the above to say that regions can do their own internal quarter-to-quarter reallocation of unspent money as long as it is reported up (and as long as it really does end up being made use of). But if the region ''doesn't'' have a solid plan, we do still have plenty of central needs (or possibly there's a strong unplanned need in another region).

The easiest thing to do when you have money left at the end of the FY is to buy swag and save it for next year. This makes planning a lot easier: You don't have to worry about swag or about budget cuts. I'm not saying we should burn money, we should only buy swag we really need. As swag production happens on a regional level, the budget would remain there.

Only if a region has no other use for their budget, the council should claim it for other activities or regions. It is on the budget owner to have an eye on the spending and get touch with the regions if necessary. I would rather ask if we can use a region's unused budget than automatically reallocate it. In fact, quarterly reallocation can be counter-productive as it encourages regions to spend money on less meaningful stuff not only on a yearly but quarterly base.

Replying to [comment:7 cwickert]:

Their LATAM request was lower because they have learned their lesson of overbudgeting and underspending. APAC has overbudgeted way more, so I really think it's unfair to give LATAM less just because they are more realistic now.

That's hard to disagree with. :)

AFAIR we approved the [https://pagure.io/fedora-budget/blob/master/f/FY17/adjustedannualbudgets/council/scenarios.txt#_25 hF/hDI scenario], which leaves USD 74,500 for regional support. The numbers above add up to USD 73,500, so there is a 1k margin. We could either

Oh. Langdon, you were supposed to check my math. :)

That does make this easier, though. I guess if we are looking at $15k for APAC FUDCon but holding LATAM to $10k, my inclination is to just add the overlooked $1k to LATAM.

Can you help me understand why this can't just be ''continuously'' kept up to date, though, making it just a matter of taking a snapshot?

I'm afraid I'm lost. You said the regions report quarterly splits after we approve the top-level allocation. This is a yearly thing, and so is the event planning in the regions. Therefor I assumed the quarterly splits were a yearly thing, too. Once a year, right tell the regions how much money they get and in return, they tell us how they plan to spend it throughout the quarters. How would one make this continuously?

This is new, but we're also asking for quarterly updates and re-balancing from the regions. That way, we a) realize if we're getting out of sync before we get to the end of the year and b) can re-prioritize if needed.

The easiest thing to do when you have money left at the end of the FY is to buy swag and save it for next year. This makes planning a lot easier: You don't have to worry about swag or about budget cuts. I'm not saying we should burn money, we should only buy swag we really need. As swag production happens on a regional level, the budget would remain there.

I'm not sold on this. There are a lot of other things which might be more valuable overall to the project than swag, and it'd be nice to at least be able to consider them.

Only if a region has no other use for their budget, the council should claim it for other activities or regions. It is on the budget owner to have an eye on the spending and get touch with the regions if necessary. I would rather ask if we can use a region's unused budget than automatically reallocate it. In fact, quarterly reallocation can be counter-productive as it encourages regions to spend money on less meaningful stuff not only on a yearly but quarterly base.

I don't think "no other use" is right at all. It should be good use (both at quarter end and always).

Replying to [comment:8 mattdm]:

I guess if we are looking at $15k for APAC FUDCon but holding LATAM to $10k, my inclination is to just add the overlooked $1k to LATAM.

+1

If we do that–or at least make sure, LATAM does not get less than they spent last year- I'm +1 for the regional allocation. I suggest we consider this part done and move on in order to not delay the process further. '''Everybody ok with that?'''

This is new, but we're also asking for quarterly updates and re-balancing from the regions. That way, we a) realize if we're getting out of sync before we get to the end of the year and b) can re-prioritize if needed.

Sounds fair, but that was not my question. Do we want a quarterly split ''at the beginning'' or not? If so, this initial split takes some time, probably more than 2-4 weeks you estimated.

Last but not least I still think we should not only rebalance from the regions but also give them the chance to rebalance quarters. This is the regional support budget and it is called 'regional' for a reason.

I'm not sold on this. There are a lot of other things which might be more valuable overall to the project than swag, and it'd be nice to at least be able to consider them.

I fully agree, that's exactly why we used to handle it this way. By using (mostly) left-over money from one year, we make sure swag does not block more important things the next year. If we are out of swag, so be it, we can still carry on with events etc. But if we are out of money for events...

Anyway, this is not a must for me, I just wanted to share some background information for those who are not familiar with how the budget used to be managed.

I don't think "no other use" is right at all. It should be good use (both at quarter end and always).

Ok, "no other meaningful use" then. :-) The question is: Who is to decide what is good use and what is not? Who is to judge if something in region A is more useful for Fedora as a whole than else in region B? I think people in the regions know best what is good for them.

Over the past years, FAmSCo has been working hard to delegate responsibilities down to the regions. Ultimately, we want an open project with a flat hierarchy. We want to motivate people to be responsible for their regions. Therefor I'm very concerned about the centralist tendencies I see in the council and the budget.next process.

Replying to [comment:9 cwickert]:

If we do that–or at least make sure, LATAM does not get less than they spent last year- I'm +1 for the regional allocation. I suggest we consider this part done and move on in order to not delay the process further. '''Everybody ok with that?'''

+1

Sounds fair, but that was not my question. Do we want a quarterly split ''at the beginning'' or not? If so, this initial split takes some time, probably more than 2-4 weeks you estimated.

Ah, yes. I think we do want the split at the beginning, but I don't really specifically care about it from the point of view of overall allocation. And, yeah, 2-4 weeks seems reasonable, especially this time around. The various plans do have quarterly splits already, but they will need to be adjusted and I can see that taking some time. Although I guess also, looking at the calendar, that will basically coincide with Q1 being complete already.

Last but not least I still think we should not only rebalance from the regions but also give them the chance to rebalance quarters. This is the regional support budget and it is called 'regional' for a reason.

Okay, fair enough. I do hope that regions with under-spending and no immediate plans consider overall Fedora needs, too, though.

Ok, "no other meaningful use" then. :-) The question is: Who is to decide what is good use and what is not? Who is to judge if something in region A is more useful for Fedora as a whole than else in region B? I think people in the regions know best what is good for them.

Over the past years, FAmSCo has been working hard to delegate responsibilities down to the regions. Ultimately, we want an open project with a flat hierarchy. We want to motivate people to be responsible for their regions. Therefor I'm very concerned about the centralist tendencies I see in the council and the budget.next process.

I see the concern. But, we do need to spend money globally, and on things which may be regional but which aren't necessarily ambassador-focused (for example translation events). The FY14 budget that we've been continuing from did not allocate anything for FADs, when in retrospect it probably should have; that puts us in a hard spot for shifting funding back to FADs now.

We ultimately ''are'' responsible for deciding if money has more useful for Fedora as a whole in region A or region B. We need evidence that what we're doing is working, whether decentralized or not. Decentralized has a natural tendency to go in all sorts of different directions, which can be powerful, but it can also be inefficient — and can be hard to put together again into a coherent story. Right now, we don't have a way to show where the money went, why, and what we got out of it. I think Remy's plan (with the three different roles in each region) will help — and help keep things decentralized.

Replying to [comment:5 cwickert]:
I have some concerns about LATAM, they get less than they spent last year and due to bad tracking, this is already a subset of what they actually spent.

Lets please remember that LATAM member not always have an easy way to get reimbursements, and reimbursements usually take quite long.

That is true; LATAM definitely came out the lowest here by any measure. That's largely because their ''request'' was significantly lower than that from the other regions, which doesn't really seem like a good reason to penalize. Do you have a concrete alternate proposal?

LATAM region has always been the last on budget; we have learn to make smaller budgets more adjusted to reality, yet, when larger Regions ask for more, they ended up taking it from the LATAM budget. Our budget keeps getting smaller and smaller, despite part of those funds aren't granted due the difficulty of transfer the funds to our Fedora members, more than just a '''I don't want to use it'''.

Just imagine next year budget, when we end up using only 80% of our budget will look like just because we aren't '''the bigger spenders'''?

Our main event is FUDcon, which take a lot of money since travel airfares are quite more expensive in our region than in any other else. Would there be a way to increase (at least) only the FUDcon budget? We usually take part of our Regional budget to cover FUDcon, and if each year the cuts keep coming, to me, it seems that budget for LATAM is meant to disappear in order to provide more budget to everyone else.

Don't get me wrong, I understand that everyone is getting cuts, yet, seems that LATAM is getting always the worst.

Replying to [comment:11 tatica]:

Just imagine next year budget, when we end up using only 80% of our budget will look like just because we aren't '''the bigger spenders'''?

This is an important point, and continues on from the discussion Christoph and I were just having. When we get to FY18 and FY19, I don't think we should base our allocation on percent spent at all. We're just using that as a proxy for the relationship between plans and actual activities.

Instead, allocation should be based on the demonstrated results from whatever was spent. If LATAM spends 80% of what was planned but gets really great, measurable results, that's ''significantly'' better than spending 100% and getting question marks — and future allocation should reflect that.

Don't get me wrong, I understand that everyone is getting cuts, yet, seems that LATAM is getting always the worst.

Yeah, that wasn't the intention, and I hope adding in the previously-uncounted $1000 helps balance that a little.

Hey everyone. We need to move forward on this. Christoph, Maria — is the LATAM budget acceptable with the additional $1000 added? That would put the amount at an 8.6% increase from recorded-actual spending last year, and at 51.6% of the requested amount (the second-highest fraction). If that doesn't seem right to you, let's discuss alternatives rather than letting this sit....

The new budget (with the additional $1000) is acceptable, mattdm.

Okay, so, looks like discussion has wound down on this and we are basically living it -- so, marking approved. Thanks to everyone for your input and work on this.