Marathon Donation App - Call for testers

2011-01-20 01:28:21 am

Fucking Weeaboo

One thing I thought of (don't know if it's been brought up) that might help with naming suggestions (or other things where a game has multiple items) would be to add in another column, so it would be like this:

GAME TITLE COLUMN - ITEM IN GAME - ITEM WANTED
Chrono Trigger - Lucca - (Insert Chosen Name Here)

From what little I worked with the thing, I think it would make finding something for a particular game easier, since you don't have to scroll though a giant list of character names from one game to get to another, for instance.

Mystery

2011-01-20 01:33:50 am

I have been thinking the same thing. Each entry should probably be associated with a game, a category and perhaps a subcategory.
Example:
Chrono Trigger, Find an item, <Insert Item Name Here>.
Final Fantasy 6, Rename a character, <Insert Character Name Here>, <Insert New Name Here>.

Though some things might not need a subcategory, so some flexibility might be needed.

Sir VG

2011-01-20 01:36:57 am

Fucking Weeaboo

Well, everything will technically have one category/items, so while it may seem obnoxious to have it for, say, a $40 goal to play X Cutscene, I think for consistancy and simplicity, it would be easier to have it all work that way. So for instance...

Mega Man X4 - Iris Cutscene - CUTSCENE ON!

It would make things easier for anybody who has to jump in to deal with the donation program, IMO.

UraniumAnchor

2011-01-20 01:55:15 am

Not a walrus

Having actually sat down and used the thing, I think most of the improvements need to be on our side of things in a behind the scenes way, as opposed to viewer experience. Having a place where people can check on the status of things such as name or challenge bids is the only thing I felt was missing from a viewer's perspective, and that mostly so that we don't have to spend so much time reading it off on the commentary stream.

Missing comments because of ChipIn weirdness is something we can solve from our end with simple html parsing, without having to present donators with a confusing and intimidating web form to fill out. And would be far easier to write in the first place.

Mystery

2011-01-20 02:04:25 am

Quote from UraniumAnchor:

Having actually sat down and used the thing, I think most of the improvements need to be on our side of things in a behind the scenes way, as opposed to viewer experience. Having a place where people can check on the status of things such as name or challenge bids is the only thing I felt was missing from a viewer's perspective, and that mostly so that we don't have to spend so much time reading it off on the commentary stream.

If we automate name bidding (which should be easy), then there would be a place to see the status of those.
Challenge bids, however, would be another thing since you're going to read them off PMs. So how do you suggest we create a page where viewers can see the status of those bids? Are we simply going to announce them in the stream when they're accepted/rejected, or manually update some page?

Quote:

Missing comments because of ChipIn weirdness is something we can solve from our end with simple html parsing, without having to present donators with a confusing and intimidating web form to fill out. And would be far easier to write in the first place.

You'll have to forgive me if I don't understand what missing comments refer to.
The current idea is to simply have the donors select a game, then a category (such as name a character), then what character they would want to rename and what to rename them to, then redirect to paypal so they can donate. That wouldn't be intimidating to me, at least, but I don't know about others.

[quote="UraniumAnchor"]Having actually sat down and used the thing, I think most of the improvements need to be on our side of things in a behind the scenes way, as opposed to viewer experience. Having a place where people can check on the status of things such as name or challenge bids is the only thing I felt was missing from a viewer's perspective, and that mostly so that we don't have to spend so much time reading it off on the commentary stream.[/quote]
If we automate name bidding (which should be easy), then there would be a place to see the status of those.
Challenge bids, however, would be another thing since you're going to read them off PMs. So how do you suggest we create a page where viewers can see the status of those bids? Are we simply going to announce them in the stream when they're accepted/rejected, or manually update some page?

[quote]Missing comments because of ChipIn weirdness is something we can solve from our end with simple html parsing, without having to present donators with a confusing and intimidating web form to fill out. And would be far easier to write in the first place.[/quote]
You'll have to forgive me if I don't understand what missing comments refer to.
The current idea is to simply have the donors select a game, then a category (such as name a character), then what character they would want to rename and what to rename them to, then redirect to paypal so they can donate. That wouldn't be intimidating to me, at least, but I don't know about others.

UraniumAnchor

Edit history:

UraniumAnchor: 2011-01-20 02:13:23 am

2011-01-20 02:12:18 am

Not a walrus

Challenge bids are not something that come from PMs, I think you're conflating two different things here. We have challenge bids that are decided upon ahead of time, such as "$100 of total donations and I'll play Contra with no weapons". And then we have challenges that are more like pledges, such as "$10 for every minute on the game timer under 1:45". Viewers only really need to know about the former, and they won't change much, if ever, other than how much has gone towards them. The latter doesn't have to be kept track of at all, because presumably the person making the pledge is actually paying attention.

And missing comments comes from ChipIn assigning a timestamp to each comment and sorting them that way, but they don't necessarily show up in that order if somebody takes a while to write a comment. It's hard to explain without actually showing you the ChipIn page.

Making donors fill in anything more complex than a comment and a donation amount is shifting the workload in the wrong direction for not much benefit.

Challenge bids are not something that come from PMs, I think you're conflating two different things here. We have challenge bids that are decided upon ahead of time, such as "$100 of total donations and I'll play Contra with no weapons". And then we have challenges that are more like pledges, such as "$10 for every minute on the game timer under 1:45". Viewers only really need to know about the former, and they won't change much, if ever, other than how much has gone towards them. The latter doesn't have to be kept track of at all, because presumably the person making the pledge is actually paying attention.

And missing comments comes from ChipIn assigning a timestamp to each comment and sorting them that way, but they don't necessarily show up in that order if somebody takes a while to write a comment. It's hard to explain without actually showing you the ChipIn page.

Making donors fill in anything more complex than a comment and a donation amount is shifting the workload in the wrong direction for not much benefit.

Mystery

2011-01-20 02:19:54 am

Quote from UraniumAnchor:

Challenge bids are not something that come from PMs, I think you're conflating two different things here. We have challenge bids that are decided upon ahead of time, such as "$100 of total donations and I'll play Contra with no weapons". And then we have challenges that are more like pledges, such as "$10 for every minute on the game timer under 1:45". Viewers only really need to know about the former, and they won't change much, if ever, other than how much has gone towards them. The latter doesn't have to be kept track of at all, because presumably the person making the pledge is actually paying attention.

Ah, I get you now. So we should basically split the challenge bids into two categories.
So the predetermined would show up on the show donations or whatever page, so you can see status, the highest bid, etc.
And the pledges don't show up at all. They're handled manually by reading and announcing to the chat.
I got that right, I hope?

Quote:

Making donors fill in anything more complex than a comment and a donation amount is shifting the workload in the wrong direction for not much benefit.

It seems like you're proposing we go back to the app donation model. People donate, we read the comments and fill in the app.
But basically that's exactly what we wanted to avoid this time around, isn't it?

[quote="UraniumAnchor"]Challenge bids are not something that come from PMs, I think you're conflating two different things here. We have challenge bids that are decided upon ahead of time, such as "$100 of total donations and I'll play Contra with no weapons". And then we have challenges that are more like pledges, such as "$10 for every minute on the game timer under 1:45". Viewers only really need to know about the former, and they won't change much, if ever, other than how much has gone towards them. The latter doesn't have to be kept track of at all, because presumably the person making the pledge is actually paying attention.[/quote]
Ah, I get you now. So we should basically split the challenge bids into two categories.
So the predetermined would show up on the show donations or whatever page, so you can see status, the highest bid, etc.
And the pledges don't show up at all. They're handled manually by reading and announcing to the chat.
I got that right, I hope?

[quote]Making donors fill in anything more complex than a comment and a donation amount is shifting the workload in the wrong direction for not much benefit.[/quote]
It seems like you're proposing we go back to the app donation model. People donate, we read the comments and fill in the app.
But basically that's exactly what we wanted to avoid this time around, isn't it?

UraniumAnchor

2011-01-20 02:25:48 am

Not a walrus

The two biggest problems I ran across where a) keeping people apprised of bid wars, mostly because having to read off 30 different names gets old, and b) missing comments because chipin's timestamps are weird.

A web form from the donor's perspective is not the best solution to either of those problems. A) can be solved by having the donation app shoving a data dump onto the SDA server where the page parses it and displays it, and B) is solved by having the donation app parse ChipIn's page in a more robust fashion. It wouldn't be difficult at all for the donation app to attach a particular donation to a particular comment and store that too. 99% of the donations were uniquely identifiable by some combination of amount, donor, and timestamp, and the comment timestamps would be equal to the donation they were attached to, so it would be trivial to connect them in the program.

Mystery

2011-01-20 02:33:44 am

Well now, you've certainly given a unique perspective on what to do. I think I'll just let everyone in charge of the marathon make a decision about what exactly it is they need/want, to avoid further confusion and pursue a single path.
You'll forgive me, UraniumAnchor. I have no idea who are in charge except Mike and perhaps Flip and nate.
This are getting confusing. I'd rather see an "official" statement from all those who are in charge before discussing further details about how to go about stuff and all.
I'm over and out for the moment.

Sir VG

2011-01-20 02:39:17 am

Fucking Weeaboo

Something I'm trying to remember that may be of benefit for those donations that 'sneak' in - is there an export function on the chip in page? What could be done is a comparison check between the current export and the last one, then something outputs all the entries that were added, not just on top, but ones that appeared a ways in for whatever reason. It could avoid one of the biggest problems of the entry - human error. It'll be especially helpful when we start getting up to several hundred entries in.

UraniumAnchor

2011-01-20 02:42:26 am

Not a walrus

I'm mostly speaking as both a user of the donation app and somebody who knows what goes into writing this kind of program, so that's my perspective. From my discussions with others who were manning the station, they seemed to be mostly of the same thought, but if somebody wants to speak up and correct me feel free.

There's an export for the donation list, but not comments. HTML is trivial to parse, though. I could probably have something written in afternoon that would do pretty much exactly as you're describing, and was what I meant above in my solution to B).

Mystery

2011-01-20 04:23:02 am

Quote from UraniumAnchor:

I'm mostly speaking as both a user of the donation app and somebody who knows what goes into writing this kind of program, so that's my perspective. From my discussions with others who were manning the station, they seemed to be mostly of the same thought, but if somebody wants to speak up and correct me feel free.

I also know full well what goes into writing this kind of thing, as I have done it before. It isn't difficult. It is rather trivial, I should say. I already have lots of code written long before that could be integrated into something like this.
The main problem is that there are some (at least for certain, me) who believes that pushing more responsibility to the donors is a better idea, and some who believe that relieving donors of responsibility and handling this stuff manually is a better idea. These two views clash, so it would be nice to know what the ones who are actually going to be manning this the entire marathon thinks.

Sir VG

2011-01-20 04:33:21 am

Fucking Weeaboo

Well, I didn't man it much, but the program from what I used it was pretty simple - which is rule #1 - follow the KISS method. Honestly, the biggest concerns to me were simply that some items did get lost due to how Chip In stuffs entries in, and that there should be a method of output to a webpage so viewers can see how donations for X item are going without us having to constantly rattle off items, which may not be possible (aka SERIOUS TIME) or just long in general (Chrono Trigger/FF IV cast list, for instance). Overall, I thought the program worked really well.

(And BTW, I am a programmer, just not of JAVA. So I do understand programming concepts, so I can look at the program from a programmer standpoint and a user standpoint.)

Mystery

Edit history:

Mystery: 2011-01-20 04:38:52 am

2011-01-20 04:38:38 am

Quote from Sir VG:

Well, I didn't man it much, but the program from what I used it was pretty simple - which is rule #1 - follow the KISS method.

But simple for who? The viewer, or the ones manning the backend, so to speak? Or try to reduce the work overall?

Quote:

Honestly, the biggest concerns to me were simply that some items did get lost due to how Chip In stuffs entries in, and that there should be a method of output to a webpage so viewers can see how donations for X item are going without us having to constantly rattle off items, which may not be possible (aka SERIOUS TIME) or just long in general (Chrono Trigger/FF IV cast list, for instance).

This should easily be possible by simple storing the donations in a database and a php script polling the database. It's what we're aiming at doing.

Quote:

(And BTW, I am a programmer, just not of JAVA. So I do understand programming concepts, so I can look at the program from a programmer standpoint and a user standpoint.)

The serverside stuff is probably going to be php/perl (though admittedly, I don't know perl very well; only php).

[quote="Sir VG"]Well, I didn't man it much, but the program from what I used it was pretty simple - which is rule #1 - follow the KISS method.[/quote]
But simple for who? The viewer, or the ones manning the backend, so to speak? Or try to reduce the work overall?

[quote]Honestly, the biggest concerns to me were simply that some items did get lost due to how Chip In stuffs entries in, and that there should be a method of output to a webpage so viewers can see how donations for X item are going without us having to constantly rattle off items, which may not be possible (aka SERIOUS TIME) or just long in general (Chrono Trigger/FF IV cast list, for instance).[/quote]
This should easily be possible by simple storing the donations in a database and a php script polling the database. It's what we're aiming at doing.

[quote](And BTW, I am a programmer, just not of JAVA.  So I do understand programming concepts, so I can look at the program from a programmer standpoint and a user standpoint.)[/quote]
The serverside stuff is probably going to be php/perl (though admittedly, I don't know perl very well; only php).

Sir VG

Edit history:

Sir VG: 2011-01-20 04:44:11 am

2011-01-20 04:42:48 am

Fucking Weeaboo

Frankly, I don't like the idea of forcing too much on the viewer, unless you can make it ridiculously simple. While I do like the idea of keeping some items more simplified by tying it into the donation at the time of running it though Paypal, that also at the same time creates a problem - we can't anticipate everything we're gonna donate for at the time of the marathon, and frankly putting in a "get in all your ideas before the marathon or we're not doing it!" idea is stupid. The way we ran things this year worked pretty well, for the most part.

And end users are often times sheep. The more you keep on the back end, the easier it would be to fix if something goes completely haywire. If you put more of it on the front end, the harder it will be to correct, I think.

Mystery

2011-01-20 04:49:45 am

Quote from Sir VG:

Frankly, I don't like the idea of forcing too much on the viewer, unless you can make it ridiculously simple. While I do like the idea of keeping some items more simplified by tying it into the donation at the time of running it though Paypal, that also at the same time creates a problem - we can't anticipate everything we're gonna donate for at the time of the marathon, and frankly putting in a "get in all your ideas before the marathon or we're not doing it!" idea is stupid. The way we ran things this year worked pretty well, for the most part.

The suggestion was basically to PM someone in the chat or basically select "other / what I want to donate for isn't listed here", which would then be processed manually, as before. But once something has been processed manually, it could be added to the list, which would make it faster for someone to donate, and easier for those manning the backend. They also wouldn't have to type as much in the comments, not that that is much of an argument.

Quote:

And end users are often times sheep. The more you keep on the back end, the easier it would be to fix if something goes completely haywire. If you put more of it on the front end, the harder it will be to correct, I think.

Completely agree, and you can't completely fix it. But I think (what others think is up for discussion naturally) that filling out some forms such as

"What do you want to donate for?" -> Renaming a character
"What character do you want to rename?" -> Insert character name here
"What would you like to rename the character to?" -> Insert new name here

...isn't too difficult, and it would save a lot of time, especially when there are a lot of donations.
So the question is, is this too much to put donors through?
*shrug* That's what we're trying to determine.

[quote="Sir VG"]Frankly, I don't like the idea of forcing too much on the viewer, unless you can make it ridiculously simple.  While I do like the idea of keeping some items more simplified by tying it into the donation at the time of running it though Paypal, that also at the same time creates a problem - we can't anticipate everything we're gonna donate for at the time of the marathon, and frankly putting in a "get in all your ideas before the marathon or we're not doing it!" idea is stupid.  The way we ran things this year worked pretty well, for the most part.[/quote]
The suggestion was basically to PM someone in the chat or basically select "other / what I want to donate for isn't listed here", which would then be processed manually, as before. But once something has been processed manually, it could be added to the list, which would make it faster for someone to donate, and easier for those manning the backend. They also wouldn't have to type as much in the comments, not that that is much of an argument.

[quote]And end users are often times sheep.  The more you keep on the back end, the easier it would be to fix if something goes completely haywire.  If you put more of it on the front end, the harder it will be to correct, I think.[/quote]
Completely agree, and you can't completely fix it. But I think (what others think is up for discussion naturally) that filling out some forms such as

"What do you want to donate for?" -> Renaming a character
"What character do you want to rename?" -> Insert character name here
"What would you like to rename the character to?" -> Insert new name here

...isn't too difficult, and it would save a lot of time, especially when there are a lot of donations.
So the question is, is this too much to put donors through?
*shrug* That's what we're trying to determine.

UraniumAnchor

2011-01-20 05:24:28 am

Not a walrus

It's far easier for donors to just type 'towards naming Terra Oglop' or 'weaponless Contra' instead of dealing with a giant, possibly intimidating wall of radio buttons, not to mention making sure it all displays properly on everybody's browser.

The volume and duration of the actual things we need done do not justify the setup and testing time. From the operator side of things it does not take that much time to read a comment and assign the donation to the proper option. We could even set up keywords that make comments get flagged for closer perusal. Most of the difficulty in assigning things to the proper place was because you had to scroll the ChipIn page so damn much in order to align comments/bids with donation amounts. Solving that and making the bid data available to viewers would go a long way towards cutting the workload and would be far easier to implement than trying to fully automate a system. The volume of donations we were receiving does not justify that headache. Though that would be a good problem to have.

It's far easier for donors to just type 'towards naming Terra Oglop' or 'weaponless Contra' instead of dealing with a giant, possibly intimidating wall of radio buttons, not to mention making sure it all displays properly on everybody's browser.

The volume and duration of the actual things we need done do not justify the setup and testing time.  From the operator side of things it does not take that much time to read a comment and assign the donation to the proper option. We could even set up keywords that make comments get flagged for closer perusal. Most of the difficulty in assigning things to the proper place was because you had to scroll the ChipIn page so damn much in order to align comments/bids with donation amounts. Solving that and making the bid data available to viewers would go a long way towards cutting the workload and would be far easier to implement than trying to fully automate a system. The volume of donations we were receiving does not justify that headache. Though that would be a good problem to have.

puwexil

2011-01-20 01:33:30 pm

Professional Second Banana

Having run the donation app during a pretty high-volume part of the marathon (the finale), the only change to that app that I think would have made it go more smoothly would have been some mechanism for publishing the bid competition totals (ie for character names, run challenges, etc) onto our website. The Chipin page was a bit difficult to work with (slow refresh times, donations not getting published in chonological order, no mechanism to flag individual donations as logged into the database); but since I highly doubt we have any control over Chipin's interface, I think we just need to live with it and find our own workarounds where possible (like maybe using dedicated SDACommentator accounts in the IRC chat and putting something in the FAQ below the streams that viewers should PM a commentator if their donation comment doesn't get read within 15 minutes or so).

Like I said in my post-marathon shoutouts post, I think SMK did an awesome job with the donation app. The bid and prize tracking functionality especially saved tons of time and energy.

Having run the donation app during a pretty high-volume part of the marathon (the finale), the only change to that app that I think would have made it go more smoothly would have been some mechanism for publishing the bid competition totals (ie for character names, run challenges, etc) onto our website.  The Chipin page was a bit difficult to work with (slow refresh times, donations not getting published in chonological order, no mechanism to flag individual donations as logged into the database); but since I highly doubt we have any control over Chipin's interface, I think we just need to live with it and find our own workarounds where possible (like maybe using dedicated SDACommentator accounts in the IRC chat and putting something in the FAQ below the streams that viewers should PM a commentator if their donation comment doesn't get read within 15 minutes or so).

Like I said in my post-marathon shoutouts post, I think SMK did an awesome job with the donation app.  The bid and prize tracking functionality especially saved tons of time and energy.

Mystery

2011-01-20 01:41:07 pm

Using a website system would allow us NOT to miss any donation comments! That is not a problem. We won't be using the chipin from the last time. We'll store the information about the donations ourselves, making for easier filtering, tracking, etc. Statistics could be tracked all the time.
Don't settle for good. Settle for best! We have the time to make a better system now. We should take advantage of this to its fullest instead of just some small changes to the interface.

Rakuen

2011-01-20 01:45:24 pm

Weegee Time

When you think about it, it's making a customized version of Chipin. Now, I don't know if SDA really needs that or not, but given time is available for people to develop it, I'd still be interested in what it would end up looking like.

You've also got to consider that if it gets developed, other people can use it too. There's no reason this marathon's donation app couldn't be used by any other group. If we develop an extension that's useful for someone else's purposes, I see that as a good thing.

UraniumAnchor

2011-01-20 04:42:39 pm

Not a walrus

As I said before, having the donation app parsing the comments would prevent us from missing any, because it's easy to check for uniqueness, preventing duplicate data but also making it so we don't miss anything. Mark a comment as 'unread' upon first import. There, bam, no lost comments. It would be more worthwhile to ask ChipIn to have an export option that includes comments so we don't have to parse the HTML and attach them ourselves.

SMK

Edit history:

SMK: 2011-01-20 06:03:10 pm

SMK: 2011-01-20 06:02:43 pm

SMK: 2011-01-20 06:00:17 pm

2011-01-20 05:59:26 pm

Wow, you guys move fast.

Just to re-iterate, our plan is to replace, not use, chip-in. I understand that it might seem like overkill for our purposes, since, for the most part, we were able to keep up with donations, however I still would rather reduce the required manpower for it as much as possible (the person/people manning donations could ideally be able to double as IRC mods or commentators without going insane). In addition, even if it is a lot of development effort, I don't really see a problem with that; we have a lot of time and willingness to work on it, and quite frankly I think it will be a fun project to work on. Also, I think we can disable seller comments when we post to Paypal, so maybe we can even save some people missing their comments entirely.

As for whether to have the users enter bids as an html form vs in their comments, regardless of my personal opinion on the matter, I think it would be worth implementing both ends of it regardless. Having the ability to modify/update ANYTHING in the system from our end is a must, regardless of whether we can/should automate it from the user's end as well. Once we have a basic system up and running, we can discuss whether such additions would be a good thing or not.

*EDIT*

Another possibility with a custom system is allowing multiple people to make updates at a given time (i.e. not have to go through 1 person). For example, if somebody gets a request in an IRC PM, they could make the changes on their system without having to go through someone else.

Wow, you guys move fast.

Just to re-iterate, our plan is to replace, not use, chip-in.  I understand that it might seem like overkill for our purposes, since, for the most part, we were able to keep up with donations, however I still would rather reduce the required manpower for it as much as possible (the person/people manning donations could ideally be able to double as IRC mods or commentators without going insane).  In addition, even if it is a lot of development effort, I don't really see a problem with that; we have a lot of time and willingness to work on it, and quite frankly I think it will be a fun project to work on.  Also, I think we can disable seller comments when we post to Paypal, so maybe we can even save some people missing their comments entirely.

As for whether to have the users enter bids as an html form vs in their comments, regardless of my personal opinion on the matter, I think it would be worth implementing both ends of it regardless.  Having the ability to modify/update ANYTHING in the system from our end is a must, regardless of whether we can/should automate it from the user's end as well.  Once we have a basic system up and running, we can discuss whether such additions would be a good thing or not.

*EDIT*

Another possibility with a custom system is allowing multiple people to make updates at a given time (i.e. not have to go through 1 person).  For example, if somebody gets a request in an IRC PM, they could make the changes on their system without having to go through someone else.

UraniumAnchor

Edit history:

UraniumAnchor: 2011-01-20 06:22:17 pm

UraniumAnchor: 2011-01-20 06:21:42 pm

2011-01-20 06:21:15 pm

Not a walrus

The main issue I have is that getting that sort of thing working would be a lot of work for not that much return. When I was running the donation app the biggest headache for me was having to constantly scroll the chipin page to find new things. Getting a webform system from the donor's end working reliably would take a stress test that we probably can't realistically do until the system is live and that's a recipe for disaster.

I mean, hell, look at what the rotating banner did when 3000 people were trying to pile onto the front page at the same time.

arkarian

2011-01-20 07:00:33 pm

gamelogs.org

;-[

safe to say if i do the page again next year i'll just write that script myself.

SMK

2011-01-20 08:32:12 pm

I think the main reason we don't want to use chip-in is in fact because it didn't scale well to the volume of donations we received.

Getting a webform system working is not at all difficult, and it has roughly the same overhead as a page request plus one or two database lookups on the server side. I don't think that we're going to be hitting any major limits, especially since its not 3000 people donating at once, its like, maybe 30 max. Implementing the user view of the current bid status will probably require more server overhead (number of people refreshing that page versus number of people actually donating at once), and we've already decided that is an essential feature.

I guess what I'm saying is, once we have the current bid status implemented (i.e. have web-pages accessing a db containing the donation info and such), its a very small step towards just having the system do the donation insertions for us. I don't really know what else to say to alleviate your fears, other than why don't we at least try it and see where it goes.

I think the main reason we don't want to use chip-in is in fact because it didn't scale well to the volume of donations we received.

Getting a webform system working is not at all difficult, and it has roughly the same overhead as a page request plus one or two database lookups on the server side.  I don't think that we're going to be hitting any major limits, especially since its not 3000 people donating at once, its like, maybe 30 max.  Implementing the user view of the current bid status will probably require more server overhead (number of people refreshing that page versus number of people actually donating at once), and we've already decided that is an essential feature.

I guess what I'm saying is, once we have the current bid status implemented (i.e. have web-pages accessing a db containing the donation info and such), its a very small step towards just having the system do the donation insertions for us.  I don't really know what else to say to alleviate your fears, other than why don't we at least try it and see where it goes.