Marathon Donation App - Call for testers

puwexil

Edit history:

puwexil: 2011-12-24 05:51:06 pm

2011-12-24 04:33:58 pm

Professional Second Banana

Odd - I didn't have any problems creating prizes earlier in the week.

Edit: Just tried again with tracker app v. 12/21 and got the above error.

SMK

Edit history:

SMK: 2011-12-24 04:43:38 pm

2011-12-24 04:43:13 pm

You and your lies UA! Smiley

Okay, I'm one of those strange people who is actually busy on Christmas Eve, but I'll get to this as soon as I can, probably boxing day.

Also, better searching is something I've considered for awhile now, I have some ideas, that I'm going to be trying over the next little while, but again, no promises that it'll be ready for the marathon. The multi-select thing is a bug (or rather, its an oversight, the list widget allows multi-select by default for some odd reason), but that's actually not a bad idea at all now that you mention it, and it should be super easy to implement.

When I have time.

Which is not now.

UraniumAnchor

2011-12-24 05:05:33 pm

Not a walrus

Well, I meant it more in the sense that I'd have to do my editing directly in production, which is not a good idea.

Now I need to go finish putting stuff under the tree. Have a Merry Christmas, those who are so inclined.

SMK

Edit history:

SMK: 2011-12-25 10:58:30 pm

2011-12-25 10:41:13 pm

Prize thing is fixed (edit: and uploaded), I'll get started trying to make searching better soon.

mikwuyma

2011-12-27 09:13:28 am

My feelings on The Demon Rush

Question, is it possible to add multiple entries for a donation draw? I'm thinking for the Master Sword we could allow people to donate for up to 3 "raffle tickets" for the sword.

Is this idea feasible, or is it too hard/late to implement it?

UraniumAnchor

Edit history:

UraniumAnchor: 2011-12-27 10:47:09 am

2011-12-27 10:45:07 am

Not a walrus

I don't want to speak for SMK too much, but if the raffle is already implemented (and I think it is) it shouldn't be too hard in theory to limit how high somebody's chance can go. I can't say for sure without looking at the code.

Besides this, at this point we should probably not try to implement any major new features, and just make sure everything works.

SMK

Edit history:

SMK: 2011-12-27 08:34:41 pm

SMK: 2011-12-27 08:30:11 pm

SMK: 2011-12-27 08:29:16 pm

SMK: 2011-12-27 04:20:48 pm

2011-12-27 04:18:44 pm

Quote from UraniumAnchor:

I don't want to speak for SMK too much, but if the raffle is already implemented (and I think it is) it shouldn't be too hard in theory to limit how high somebody's chance can go. I can't say for sure without looking at the code.

Yes, the raffle is implemented; at the moment, it allows unlimited 'tickets' (i.e. its just the floor of the total amount donated by a given donor divided by the minimum bid amount for that prize). It wouldn't be too difficult to have it cap it at some amount; the difficulty is actually in getting that information from the user somehow (like, there'd probably be an extra text box floating around, or I could just hard-code the amount for now).

Edit: This is now implemented. Hopefully it isn't too confusing.

Quote from UraniumAnchor:

Besides this, at this point we should probably not try to implement any major new features, and just make sure everything works.

Indeed. I looked over implementing some of the ideas for improving the searching features, and its turning out to require a lot more work that I initially thought (mostly due to the really, REALLY sloppy UI code that has accumulated over time (yes, nobody's fault but my own)). Hence I'm really hesitant to include that stuff into the main branch, so better searching will probably have to wait till next marathon.

Anyways, I'll finish the prize upgrades and the multi-searching thing (since that part is easy to implement) as soon as I can, just so that everything we need is in place.

Edit: Done, I'm uploading the latest version now. Unfortunately I'm at my parents' place, which has absolutely _awful_ internet, so, it'll be up in like, 10 minutes.

Edit 2: Good to go. I changed the master sword drawing method to the Random Weighted Draw method. I didn't set up a way to save the maximum number of tickets, so that'll be the one thing you need to set up manually for that draw.

[quote="UraniumAnchor"]I don't want to speak for SMK too much, but if the raffle is already implemented (and I think it is) it shouldn't be too hard in theory to limit how high somebody's chance can go. I can't say for sure without looking at the code.[/quote]
Yes, the raffle is implemented; at the moment, it allows unlimited 'tickets' (i.e. its just the floor of the total amount donated by a given donor divided by the minimum bid amount for that prize).  It wouldn't be too difficult to have it cap it at some amount; the difficulty is actually in getting that information from the user somehow (like, there'd probably be an extra text box floating around, or I could just hard-code the amount for now).

Edit: This is now implemented.  Hopefully it isn't too confusing.

[quote="UraniumAnchor"]
Besides this, at this point we should probably not try to implement any major new features, and just make sure everything works.[/quote] Indeed.  I looked over implementing some of the ideas for improving the searching features, and its turning out to require a lot more work that I initially thought (mostly due to the really, REALLY sloppy UI code that has accumulated over time (yes, nobody's fault but my own)).  Hence I'm really hesitant to include that stuff into the main branch, so better searching will probably have to wait till next marathon.

Anyways, I'll finish the prize upgrades and the multi-searching thing (since that part is easy to implement) as soon as I can, just so that everything we need is in place.

Edit: Done, I'm uploading the latest version now. Unfortunately I'm at my parents' place, which has absolutely _awful_ internet, so, it'll be up in like, 10 minutes.

Edit 2: Good to go.  I changed the master sword drawing method to the Random Weighted Draw method.  I didn't set up a way to save the maximum number of tickets, so that'll be the one thing you need to set up manually for that draw.

UraniumAnchor

2011-12-27 08:45:42 pm

Not a walrus

I assume you mean it's an option in the prize draw window?

SMK

2011-12-27 09:25:05 pm

Quote from UraniumAnchor:

I assume you mean it's an option in the prize draw window?

Yep. I don't think its worth adding an extra column to the prize table for just that.

UraniumAnchor

2011-12-27 09:36:32 pm

Not a walrus

Not unless we end up with a lot more raffle prizes next time, yeah.

mikwuyma

2011-12-28 04:34:35 pm

My feelings on The Demon Rush

Okay so about prizes should we do that with any other prizes? I mean most prizes should be whatever donation hits the minimum or above should just be one chance, but maybe we should allow multiple chances with the super meat boy tenderizer too? Those are the only two I can think of.

puwexil

2011-12-28 07:07:17 pm

Professional Second Banana

Minor bug I just noticed with the tracker app: if I create a new prize record, populate any of the fields, and then set the start or end game, the other fields will rest back to the initial random key values.

SMK

Edit history:

SMK: 2011-12-30 01:43:20 pm

2011-12-28 07:11:04 pm

Yep, that's actually what I'm working on right now. Its part of the whole 'too much refreshing of the data' deal that I'm trying to address. I don't think it'll be a show stopper either way, but I'm hoping to at least have it partially fixed soon.

Edit: Partially fixed. I also fixed tab control (like, tabbing back and forth between fields) to work a little better. Uploaded.

UraniumAnchor

Edit history:

UraniumAnchor: 2011-12-31 01:42:22 pm

2011-12-31 01:38:29 pm

Not a walrus

I think the date in the first post is a lie, I'm pretty sure this has been updated since November 19th. Cheesy

Edit: I just noticed that the grand prize defaults to "donations made this minute", instead of having those boxes unchecked since there's no start/end games assigned. Tiny thing, but probably worth fixing if you can.

mikwuyma

2011-12-31 02:01:33 pm

My feelings on The Demon Rush

Oh yeah, about the grab bag, if I haven't mentioned this already, anyone who wins a prize from that bag should still be eligible for other prizes.

SMK

2011-12-31 02:39:08 pm

Quote from mikwuyma:

Oh yeah, about the grab bag, if I haven't mentioned this already, anyone who wins a prize from that bag should still be eligible for other prizes.

First I've heard of it. What is this grab bag you are referring to? I'm not sure if I can write a work-around, but give me some details and I'll see.

UraniumAnchor

Edit history:

UraniumAnchor: 2011-12-31 02:46:16 pm

2011-12-31 02:42:50 pm

Not a walrus

I think any game without a specific prize listed will have a drawing for a game from the "grab bag" at the end of it.

Edit: At this point it might be better to just keep a manual list instead of trying to implement it in the tracker app.

SMK

Edit history:

SMK: 2011-12-31 04:42:13 pm

2011-12-31 04:40:11 pm

Quote from UraniumAnchor:

I think any game without a specific prize listed will have a drawing for a game from the "grab bag" at the end of it.

Edit: At this point it might be better to just keep a manual list instead of trying to implement it in the tracker app.

Well, an option is to add an extra boolean column to the prize table (that seems to be the solution for everything nowadays) that indicates that the prize drawing method is to always ignore the prize multiplicity check during draws. It's not impossible, it just means adding more data. The hardest part would be going through and flagging all such prizes. If you give me a list of all the prizes that this applies to, I'll make the change.

Edit: oh wait, I understand. Yeah, I guess we can just do the draw with 'exclude if already won' unchecked, and then record it separately, and then remove them. Gogo low-tech solutions. I can still implement the other thing too, if you want.

Quote from UraniumAnchor:

I think the date in the first post is a lie, I'm pretty sure this has been updated since November 19th. Cheesy

Edit: I just noticed that the grand prize defaults to "donations made this minute", instead of having those boxes unchecked since there's no start/end games assigned. Tiny thing, but probably worth fixing if you can.

Sorry, I accidentally ignored this. I fixed it in my local copy, and I'll push an update once I decide what to do about the grab-bag prizes.

[quote="UraniumAnchor"]I think any game without a specific prize listed will have a drawing for a game from the "grab bag" at the end of it.

Edit: At this point it might be better to just keep a manual list instead of trying to implement it in the tracker app.[/quote]

Well, an option is to add an extra boolean column to the prize table (that seems to be the solution for everything nowadays) that indicates that the prize drawing method is to always ignore the prize multiplicity check during draws.  It's not impossible, it just means adding more data.  The hardest part would be going through and flagging all such prizes.  If you give me a list of all the prizes that this applies to, I'll make the change.

Edit: oh wait, I understand.  Yeah, I guess we can just do the draw with 'exclude if already won' unchecked, and then record it separately, and then remove them.  Gogo low-tech solutions.  I can still implement the other thing too, if you want.

[quote="UraniumAnchor"]I think the date in the first post is a lie, I'm pretty sure this has been updated since November 19th. :D

Edit: I just noticed that the grand prize defaults to "donations made this minute", instead of having those boxes unchecked since there's no start/end games assigned. Tiny thing, but probably worth fixing if you can.[/quote]
Sorry, I accidentally ignored this.  I fixed it in my local copy, and I'll push an update once I decide what to do about the grab-bag prizes.

UraniumAnchor

2012-01-02 12:18:53 am

Not a walrus

I can't believe I just now remembered this, and it's probably not going to matter for the marathon since activity will be near-constant, but is there a way to make the tracker send keep-alives on the mysql connection?

SMK

Edit history:

SMK: 2012-01-06 09:55:10 pm

2012-01-02 08:10:37 am

The actual connection is being handled by a 3rd party connection pooling library, so I'm not sure, but I haven't had any problem with ~30 minutes of inactivity any time I had been connected, so I think we should be okay. Worst-case it errors out and you log in again.

Edit: it auto-reconnects, so that's fine, the issue is the ssh tunnel closes after about a few hours no matter what, and I don't know how to keep that alive.

Also, there will be a more hearty post-mortem after the marathon.

Rakuen

2012-01-07 08:26:34 am

Weegee Time

Posting this now so I don't forget. On UA's site when you click a challenge's page to see the donation detail, it has the total at the top, but it doesn't have the goal. Not a big deal, but it would be nice for that to be changed for the next marathon. :]

SMK

Edit history:

SMK: 2012-01-11 05:12:56 pm

SMK: 2012-01-11 03:39:17 pm

2012-01-11 03:37:23 pm

I'm hesitant to posting this because I already have absolutely no plans to touch this thing until a few weeks from now when I (hopefully) can get the rest of my life in order, but I figure it would be best to get this out of the way while everything is fresh in my mind.

First off: BIG thanks to everyone who took the time to learn the program and helped out assigning bids and reading comments. I'd name names, but I'm sure I'll forget half of you (especially since chances are if you were using the program, its because I was asleep or otherwise not in the room). You know who you all are in any case; I know that it can be frustrating to use unpolished software, and I'm glad that you were able to make things work in spite of all the small bugs and stuff.

Okay.

What went well:
- Distribution/Remote MySQL server: the access latency was, in my opinion, a small price to pay to be able to have 2+ people managing the DB at once, especially during peak hours. It helped ensure that delays for challenges/naming/etc were kept to a minimum and that up to date information was available to pretty much everyone
- UA's website: I can take no real credit for this, as this is all him. I'm pretty sure he was a shoe-in for marathon tech-MVP even before you consider this, but yeah, huge props.
- Chipin merge: okay, this is a bit of a two-sided thing, but the fact that the only two problems with it were a) it getting progressively slower over the entire marathon (I tried fixing it, I really did), and b) problems with chipin itself making mistakes, I'd say that on the whole its a big win.
- Google Spreadsheet refresh: doing this in general was/is a bit of a pain, but considering it was a last minute feature, I'm astounded how stable it was.
- Not crashing: Unless I'm mistaken, the program lasted the entire marathon without giving up the ghost entirely, and that was with me doing last-minute hot-fixes.
- Task views: seriously, these were an afterthought when I first put them in, but they are probably the second most important feature of the program overall, next to the chipin-merge thing.

What was bad but I was able to fix:
- Change in google time format: for some reason, the date-time format on the google spreadsheet changed between when I lasted tested it before the marathon, and the end of the first game. No idea why, but its unsurprising given this was the least tested feature going in. Will need to investigate further.
- The back-and-forth donation state this: This was a caching bug in the client, and I fixed it by removing caching. This will require more in-depth work when I get back at it, but at least it got addressed before the real heavy donation blocks (meat boy/oot in particular)
- A bunch of stale state errors: these came up mostly when I fixed the first part, but they're mostly addressed, though a few remain (particularly when deleting entities). Again, will be addressed more fully when I get back into the development groove, but I don't think it caused too many issues.
- Slow prize drawing: I reversed the prize drawing algorithm to reduce communication overhead, so it works much faster now. Not much else to say.

What went meh:
- Assigning bids: Slightly better than last year, and also nice since it was possible to double-team it at peak hours, but it still could/should be automated even more. I think I had in-depth conversations with at least 3 different people about different solutions to make this process even easier; rest assured, it will be addressed.
- Grab bag prizes: I'd say it was my fault, and it kind of was, but it came up a little late to be addressed in any real way. I'll be better prepared next time. In the mean time, the last prize I assigned was for 8:11 pm - 11:11 pm on january 9th. I f someone can point me to the list of remaining grab bag prizes to be assigned, PM me the list so I can finish it off, or I can also include you in my 1-man private thread with all the current winners.

What went bad and was my fault:
- Slow chipin merge: Okay, I seriously didn't think it would get _that_ bad at the end. I have a few ideas in mind (either keep a cached copy of the page's state in the client, and only update the diff, or (preferably) have it automatically run as a server daemon every 'x' minutes (which would remove the communications bottleneck)), but it still needs to be mentioned.
- Google time zones: At one point, I discovered that logging in as somebody in a specific, non-EST time zone would change the google times to match that time zone for some reason. No idea how or why that happened, but the only fix was to login using my dummy account that had no location specified. I don't think it caused too many issues, but its worth mentioning.
- Donation read task latency: If you had the (mis)fortune of reading donations, especially during peak hours, or during games with very small, specific donation-reading-time windows, you'll know what I'm talking about. I'll probably looking at solutions to lessen that bottleneck one way or another for next time.
- Inability to edit comments: I technically fixed this right at the end, but it definitely was one of my less-inspired ideas to make chipin comments non-editable in the tracker. My bad, this is fixed now.
- The user interface: this is my umbrella term for all the other small pieces of crap in the program itself.

What went bad and wasn't my fault:
- Donation reading backlog: Okay, well, at least partly not my fault; I'll accept responsibility for the delay between being able to read one comment to the next, but I don't think that alone would have prevented the kind of back-logging that came up during peak hours. The fact is that our 'best problem to have' is still a problem, and hence IMO still needs to be addressed at some point. This is less a technical issue that anything else here, but its still something that I hope I can help resolve. UA suggested that we create a 'marquee' program that we dynamically add to our xsplit display and stick up on-stream, and that would be able to display donations in a scrolling fashion (CNN headlines style I guess), during back-logged periods. At such a time, we would limit on-air reading to especially important/significant comments (such as when people like radix, bisquit, kari fry, theyetee, notch, notable forum members, et. al. donate, or when we receive especially large donations, or especially interesting comments, etc...).
- Chipin: if you have the misfortune of using firefox, you'll note that the chipin page is barely loadable at this point (protip, use chrome). Mercifully, its not the page size so much as the mess of jquery that they uses, but the point is that we're kind of stretching chipin beyond what it was designed for. I'm not saying we need to migrate away any time soon, and I'll continue to support the chipin merge functionality either way since I think that its a good feature to have/fall back onto, but we might want to at least start investigating our own paypal reception system, especially since development on chipin has basically ended, so its unlikely they'll be adding a paging system (or better unicode support for our international donors) to their donations pages any time soon.

Things that really need to be added:
- search by option name: instead of choice -> then option name
- 'run & bid name' search listing: only bid name was listed in the search window, which meant that 'character choice' ended up being the worst thing ever to search for.
- search by comment contents: it came up far more often than I thought it would
- fixing the bid tasks' 'done' predicates (sorry, I was too tired to work out the boolean logic on-site)
- better profanity/negativity filtering: this will be addressed with the attempt to automate bids on donations
- More 'task' views: for things like prize drawing
- A bunch of other stuff that I can't remember right now.

So, there. That's my take. Any and all non-redundant input (from both sides of the stream too) would be greatly appreciated, especially since if our marathons are just going to be getting bigger and bigger (experience shows they will) then having good technology to support this end of things is a must.

Edit: I, unlike uyama, will not be doing any dick-punching if you post feedback on this right now, just be aware that I probably won't be too fast to respond since I need to sleep too.

Edit 2: forgot also to thank the people who's computers I totally commandeered during the marathon: moooh, RaneofSOTN, tri-hex, and puwexel. Seriously, thanks.

I'm hesitant to posting this because I already have absolutely no plans to touch this thing until a few weeks from now when I (hopefully) can get the rest of my life in order, but I figure it would be best to get this out of the way while everything is fresh in my mind.

First off: BIG thanks to everyone who took the time to learn the program and helped out assigning bids and reading comments.  I'd name names, but I'm sure I'll forget half of you (especially since chances are if you were using the program, its because I was asleep or otherwise not in the room).  You know who you all are in any case; I know that it can be frustrating to use unpolished software, and I'm glad that you were able to make things work in spite of all the small bugs and stuff.

Okay.

What went well:
- Distribution/Remote MySQL server: the access latency was, in my opinion, a small price to pay to be able to have 2+ people managing the DB at once, especially during peak hours.  It helped ensure that delays for challenges/naming/etc were kept to a minimum and that up to date information was available to pretty much everyone
- UA's website: I can take no real credit for this, as this is all him.  I'm pretty sure he was a shoe-in for marathon tech-MVP even before you consider this, but yeah, huge props.
- Chipin merge: okay, this is a bit of a two-sided thing, but the fact that the only two problems with it were a) it getting progressively slower over the entire marathon (I tried fixing it, I really did), and b) problems with chipin itself making mistakes, I'd say that on the whole its a big win.
- Google Spreadsheet refresh: doing this in general was/is a bit of a pain, but considering it was a last minute feature, I'm astounded how stable it was.
- Not crashing: Unless I'm mistaken, the program lasted the entire marathon without giving up the ghost entirely, and that was with me doing last-minute hot-fixes.
- Task views: seriously, these were an afterthought when I first put them in, but they are probably the second most important feature of the program overall, next to the chipin-merge thing.

What was bad but I was able to fix:
- Change in google time format: for some reason, the date-time format on the google spreadsheet changed between when I lasted tested it before the marathon, and the end of the first game.  No idea why, but its unsurprising given this was the least tested feature going in.  Will need to investigate further.
- The back-and-forth donation state this: This was a caching bug in the client, and I fixed it by removing caching.  This will require more in-depth work when I get back at it, but at least it got addressed before the real heavy donation blocks (meat boy/oot in particular)
- A bunch of stale state errors: these came up mostly when I fixed the first part, but they're mostly addressed, though a few remain (particularly when deleting entities).  Again, will be addressed more fully when I get back into the development groove, but I don't think it caused too many issues.
- Slow prize drawing: I reversed the prize drawing algorithm to reduce communication overhead, so it works much faster now.  Not much else to say.

What went meh:
- Assigning bids: Slightly better than last year, and also nice since it was possible to double-team it at peak hours, but it still could/should be automated even more.  I think I had in-depth conversations with at least 3 different people about different solutions to make this process even easier; rest assured, it will be addressed.
- Grab bag prizes: I'd say it was my fault, and it kind of was, but it came up a little late to be addressed in any real way.  I'll be better prepared next time.  In the mean time, the last prize I assigned was for 8:11 pm - 11:11 pm on january 9th.  I f someone can point me to the list of remaining grab bag prizes to be assigned, PM me the list so I can finish it off, or I can also include you in my 1-man private thread with all the current winners.

What went bad and was my fault:
- Slow chipin merge: Okay, I seriously didn't think it would get _that_ bad at the end.  I have a few ideas in mind (either keep a cached copy of the page's state in the client, and only update the diff, or (preferably) have it automatically run as a server daemon every 'x' minutes (which would remove the communications bottleneck)), but it still needs to be mentioned.  
- Google time zones: At one point, I discovered that logging in as somebody in a specific, non-EST time zone would change the google times to match that time zone for some reason.  No idea how or why that happened, but the only fix was to login using my dummy account that had no location specified.  I don't think it caused too many issues, but its worth mentioning.
- Donation read task latency: If you had the (mis)fortune of reading donations, especially during peak hours, or during games with very small, specific donation-reading-time windows, you'll know what I'm talking about.  I'll probably looking at solutions to lessen that bottleneck one way or another for next time.
- Inability to edit comments: I technically fixed this right at the end, but it definitely was one of my less-inspired ideas to make chipin comments non-editable in the tracker.  My bad, this is fixed now.
- The user interface: this is my umbrella term for all the other small pieces of crap in the program itself.

What went bad and wasn't my fault:
- Donation reading backlog: Okay, well, at least partly not my fault; I'll accept responsibility for the delay between being able to read one comment to the next, but I don't think that alone would have prevented the kind of back-logging that came up during peak hours.  The fact is that our 'best problem to have' is still a problem, and hence IMO still needs to be addressed at some point.  This is less a technical issue that anything else here, but its still something that I hope I can help resolve.  UA suggested that we create a 'marquee' program that we dynamically add to our xsplit display and stick up on-stream, and that would be able to display donations in a scrolling fashion (CNN headlines style I guess), during back-logged periods.  At such a time, we would limit on-air reading to especially important/significant comments (such as when people like radix, bisquit, kari fry, theyetee, notch, notable forum members, et. al. donate, or when we receive especially large donations, or especially interesting comments, etc...).
- Chipin: if you have the misfortune of using firefox, you'll note that the chipin page is barely loadable at this point (protip, use chrome).  Mercifully, its not the page size so much as the mess of jquery that they uses, but the point is that we're kind of stretching chipin beyond what it was designed for.  I'm not saying we need to migrate away any time soon, and I'll continue to support the chipin merge functionality either way since I think that its a good feature to have/fall back onto, but we might want to at least start investigating our own paypal reception system, especially since development on chipin has basically ended, so its unlikely they'll be adding a paging system (or better unicode support for our international donors) to their donations pages any time soon.

Things that really need to be added:
- search by option name: instead of choice -> then option name
- 'run & bid name' search listing: only bid name was listed in the search window, which meant that 'character choice' ended up being the worst thing ever to search for.
- search by comment contents: it came up far more often than I thought it would
- fixing the bid tasks' 'done' predicates (sorry, I was too tired to work out the boolean logic on-site)
- better profanity/negativity filtering: this will be addressed with the attempt to automate bids on donations
- More 'task' views: for things like prize drawing
- A bunch of other stuff that I can't remember right now.

So, there.  That's my take.  Any and all non-redundant input (from both sides of the stream too) would be greatly appreciated, especially since if our marathons are just going to be getting bigger and bigger (experience shows they will) then having good technology to support this end of things is a must.

Edit: I, unlike uyama, will not be doing any dick-punching if you post feedback on this right now, just be aware that I probably won't be too fast to respond since I need to sleep too.

Edit 2: forgot also to thank the people who's computers I totally commandeered during the marathon: moooh, RaneofSOTN, tri-hex, and puwexel.  Seriously, thanks.

UraniumAnchor

Edit history:

UraniumAnchor: 2012-01-11 04:49:58 pm

2012-01-11 04:47:23 pm

Not a walrus

For the marquee: I'm not sure if we should juts have it show censored names (like my page when you aren't logged in) and donations, or actually show approved comments too. It would probably be better to do the former, since people can look up approved comments on my website. The stability of the website has me a bit worried, but CoolMatty offered to host it next time so I'm looking into moving it over there for the future. If not, I can temporarily upgrade my webhost account to something a bit more stable, probably.

Edit: Also, maybe I just missed it in your list, but comment approval could probably stand to be it's own tab, like read/bid tasks.

SMK

2012-01-11 05:01:00 pm

Quote from UraniumAnchor:

For the marquee: I'm not sure if we should juts have it show censored names (like my page when you aren't logged in) and donations, or actually show approved comments too. It would probably be better to do the former, since people can look up approved comments on my website.

Plus it would be kinda dumb to print out entire comments when like, 80% of them are like "20 for naming tifa esports!", and the other 20% are too long to read in a scrolling text thing anyways.

Quote from UraniumAnchor:

The stability of the website has me a bit worried, but CoolMatty offered to host it next time so I'm looking into moving it over there for the future. If not, I can temporarily upgrade my webhost account to something a bit more stable, probably.

For what its worth, it held up like a champ, though I'm still sad that it terminated the chipin merge because it was using too much memory (just loading the chipin page I think killed it).

Quote from UraniumAnchor:

Edit: Also, maybe I just missed it in your list, but comment approval could probably stand to be it's own tab, like read/bid tasks.

Sorry for the text wall, but no, that wasn't in there. That would make sense. However, I'm planning on making the task system a little more flexible too, such that you should be able to do all 3 in one, or split it across 3 depending on your preference. Again, more when I'm actually in implementation mode, rather than sleep mode.

[quote="UraniumAnchor"]For the marquee: I'm not sure if we should juts have it show censored names (like my page when you aren't logged in) and donations, or actually show approved comments too. It would probably be better to do the former, since people can look up approved comments on my website. [/quote]
Plus it would be kinda dumb to print out entire comments when like, 80% of them are like "20 for naming tifa esports!", and the other 20% are too long to read in a scrolling text thing anyways.

[quote="UraniumAnchor"]
The stability of the website has me a bit worried, but CoolMatty offered to host it next time so I'm looking into moving it over there for the future. If not, I can temporarily upgrade my webhost account to something a bit more stable, probably.[/quote]
For what its worth, it held up like a champ, though I'm still sad that it terminated the chipin merge because it was using too much memory (just loading the chipin page I think killed it).

[quote="UraniumAnchor"]
Edit: Also, maybe I just missed it in your list, but comment approval could probably stand to be it's own tab, like read/bid tasks.[/quote]
Sorry for the text wall, but no, that wasn't in there.  That would make sense.  However, I'm planning on making the task system a little more flexible too, such that you should be able to do all 3 in one, or split it across 3 depending on your preference.  Again, more when I'm actually in implementation mode, rather than sleep mode.

Emptyeye

2012-01-11 05:32:58 pm

Talk to the Hand

By censoring, I guess you mean comments and not names of donators?