Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
page  <- 123456789101112 -> <- 1 ... 11, 12 ->
--
--
List results:
Search options:
Use \ before commas in usernames
Waiting hurts my soul...
I watched both runs, saw others say what I already thought and speak up since there wasn't much dissent. I agree with Rakuen's assessment, including shortening the time frame.
I think you can cut it to 30 days even. Verifiers are supposed to be active here, or am I wrong? I mean you only have to click a few times to sign-up for a run with the new system.

Well it sounds good but I would like to see a faster process for IL runs in the future, since they're much shorter and improved more often than regular runs. Currently you're kinda forced to wait until you're finished with all the ILs you want to submit instead of submitting just a couple that you're really happy with now and then.
Magical. Flying. Bathtub
I think the time to private verify should be mainly based on the length of a run.  30 days isn't unreasonable of an IL or a run that is sub-10 mins, but games like FFXII are going to take a lot longer.
Caution: This user contains Kana ^_^
Quote from Qxybxjjllp:
Sounds good to me. The two trials were great examples of the types of runs public verification can most easily and quickly handle: runs that are well done, pretty easy accepts for well known games, and those that have been stuck in verification limbo because they're obscure.

My only worry would be if there was some dispute over the play quality of the run, as in these cases anonymity of opinion is more important. However, in this system, where 120 days should have passed before the run qualifies for public verification, this doesn't seem like a problem.

I think that this would have been an interesting further trial: A run, whose play quality is just bad enough to reject it (and Flip only just reaches that conclusion in private verification). It would be interesting to know how our community reacts to such a debatable case, which verdict the open discussion reaches (and maybe even: Would the inferiour play quality be detected at all, if the fanbase of a game is small?). However, I also don't think it's possible to test this in a controlled environment, so I would urge everyone in charge to closely monitor the first public verification that could (and does, in the end) fall into this group.

On the question of the time to wait for a verification to go public: I don't know how people so far signed up to verify; whether it all happened within the first 48 hours, or if there is an equally high chance for one application to drop in every week. If the first is the case, i.e. you can easily tell if a run will hang in verification hell after two weeks then shorten the time. If just as many runs find enough verifiers after four weeks as after two, then extend it. In any case, I hope the current system keeps timestamps of something (verifier applications or 'we have found enough verifiers' stati) that can reliably be used to determine which way to go. I also believe that Flip and UA came up with this already, and that they will make the correct decision anyway.
I think that giving the obscure games a better chance of having their runs be verified is excellent. Time frames and what not can always be changed in the future so I'd say play it safe and give it whatever time the staff think it needs. Optimizations and further discussion can be held after the open verification have been active for a couple of months.
I'm not sure if it's anymore relevant, but I'm posting here my idea of a better system.

First of all, there should be really a leaderboard, because then we could see the improvement of the runs over time and some newer runners could measure their abilities in running at a certain game like a "highscore".

Second, "waits for verfication" runs should be watchable, but should not be shown in the leaderboard. People could then approve the run, like at TASvideos.

For verification I would go for a "network of trust", like Linus does with his Linux Repo (in git). (e.g. if he trust someone he will just merge his branch with that person) (PS: I'm making this reference because Linux must have high quality)
If some trustworthy members did approve it, that we can garantuee that it's not getting approved by bot spamming or something like that. So that verification system is heavy community based.

Basically what I think is that, in the beginning there are the normal SDA validators, who approve the verification, and people can approve the run in advance.
By doing that you gain status points/reputation. You gain more if verdict of the run equals yours.
So by contributing a lot you will be more trusted. At a certain point you gain certain priviliges and become a trusted member or validator. So basically after some time more and more work will be done by the community which will decrease the gap of submitting and publication. Although it requieres alot of overthinking the concept and it flaws, that it could still provide an certain degree of quality.

Also people who are in the leaderboard (top 10 and often submit a run) of a certain game, gain a special status, because it is more likely they know if the run is legit or not (e.g. Cosmo can verify a OoT/TWW run). But should be done in oligopoly, like if two runners in the leaderboard approve it, then it will be verified, because mistakes can happen.

I'm also thinking about using tags like "live run", "glitched run", "controller cam" etc. Reason for that is, that a run where the controller is filmed is most likely legit and requires less knownledge members, when it comes to legitimate the run.

Also the system could be expanded, like if someone approves a lot of runs for a certain game or system, well probably because he knows a lot about running the game, but is bad at actually running it, that he is in favor for verifying it.

A little pro and con situation:
PRO:
-faster verification
-more verifications
-more workforce
-less stress/work for the current verifiers
-seeing "not verified" runs

CONS:
-might decrease the quality
-alot of planning
-could fail if the community does not contribute

So that is what I think is a better system. Hopefully you can understand my points. (And I should acquire better english skills TT_TT)
<(^_^)>
Quote from Sarrg:
By doing that you gain status points/reputation. You gain more if verdict of the run equals yours.


This I disagree with. There can be controversial runs that are very 50-50, which shouldn't really affect "reputation". People can make mistakes as well or missed something that almost nobody knows about aside from the runner, and so on.

Leaderboards are a great idea in general, but it'd be so hard to implement, and there are many games that almost nobody touches anyways. Then we'd probably have to go through the trouble of verifying every submission, and there's the question of "should this go on SDA or not? or just the leaderboards?"
The artist formerly known as Qxy
While this is a pretty good idea saarg, I'm not sure that it is actually necessary. It solves a lot of problems I just don't think exist.

For instance, having a "network of trust" begs the question of who is trustworthy on the site. The system as it stands now is: if Flip trusts (or has no reason not to trust) that you've beaten a game, and will help contribute to the verification process, he will add you as a verifier. Plus there are checks built in such that if untrustworthy people do suspicious things, their opinions get weighted less or ignored entirely.

Having the top ten of the leaderboard (Assuming it exists. I wouldn't want it to.) be trusted is fine, but these are the people who sign up for verification of a game popular enough to have a top ten. These people are also usually known in the sub-community around the game, and are recognized and respected when they show up in verification.

Aside from the the ways this system fails to change things, there is one real problem with the system: how trust is gained. You gain more trust and reputation if the verdict of the run equals yours. What if there was a dispute, and you were on the losing end but you provided compelling arguments and meaningful discussion. Surely that is worth more than an echo verification, but in this system the opposite is true. The trust gaining system also for this reason has a lot less weight. Its not game specific, so if I were to walk over and verify a game I knew less, but I had been here a long time and gained a lot of trust in games I ran or had a hand in planning, I would hope that the admins in charge of this system would not bias me because of the trust, but instead evaluate my opinion based on how much I know what I'm talking about. In that sense, the fact that you have trust means something (you're not malicious/a bot) but the amount of trust you have doesn't after a couple runs.

What you describe in terms of trust is something that seems to be implicitly done around here for the most part, but in my opinion should not be made explicit.
Edit history:
Flip: 2013-01-15 10:46:59 am
1-Up!
I have no interest in any form of points system or hierarchy for verifiers. I can see where you idea comes from, Sarrg, it seems to me to be lots of unnecessary work and headaches for a small payoff. When we go to public verification for obscure runs, I won't care who has the most "verifier points" or who is most well-known in the community-at-large. I just want to know what experience people have with the game and what they think about the run.
@kirbymasta well it was the only way I come up with to sort out the more knowledge/trustworthy member from the lesser ones.

@Qxybxjjllp Well "network of trust" has nothing to do with if I trust him. More like the system trust him because he contributed a lot, therefore he spends his time with helping the site. Like he is doing good for the system. The Verifier Status will be given by the admins I guess. Don't know if letting the system handle that will cause flaws, but it's discouraged to do so.

Well what i meant with gaining more is like, when the verdict equals then you get like some extra reputation. Nethertheless it equals or not you will gain some for efford. I didn't really think through that well. It was just an example. In the end it comes to if you contribute a lot you gain a lot of reputation like in any community.

What also is an opinion is then when you disapprove of the run that you can write your complains why it is like that.

Well, it was just an idea which came into my mind so I though I might share it. Never expected to really implement it.

Anyways, a system where community members can approve of a run like in tasvideo would be not that stupid tho. And I guess implementing a system like that which is basically just a forum, might not be that hard.
I trust that Flip can spot whether people know what they're talking about or not. I'm verifying for the first time at the moment, but I hope my detailed explanations will make it clear both that I've beaten the game, and that I know what I'm talking about when I review the run. If not, he can just ignore me and trust the other verifiers.
HELLO!
An automated system like Sarrg is talking about might have more value if we didn't have reviews of games after verification, before publication.  That check step means we don't have to overcomplicate it.
I myself would like to say I do agree with the idea of being able to watch runs that are 'waiting for verification'. The lag time between submission and verification is one of the biggest problems I've noticed, and this would definitely help to alleviate it a bit.
What about a kickstarter to get the money to pay a professional to finish SDA 2.0 with a bunch of cool features and ideas suggested here as stretch goals? I know the video said they felt it was closer than ever but for some reason I'm not feeling confident about that. A professional paid to dedicate time to it could do more and faster than the volunteer staff right?
Edit history:
UraniumAnchor: 2013-02-07 11:08:26 am
UraniumAnchor: 2013-02-07 11:04:56 am
Not a walrus
No.

To be less terse: The amount of custom features we'd need the backend to implement would mean that we'd have to pay a lot of money for a "professional" would be prohibitive, and then we'd still need to be able to modify it ourselves afterwards. The kind of quality worth paying for is money better spent on the fundraisers we do.

Even besides that, we'd have to shop around for somebody we could trust to do quality work, and considering that most of the groundwork is already laid it would mean another migration of all the data to whatever new system we'd end up with.

Now that the marathon is over and most of the video stuff is squared away (other than just letting the encoder churn), I'll be able to start working on the backend again.

To be terse again: Professional web development is better left for professional websites. This is a hobbyist website.
Exoray
But feel free to send us free money whenever
HELLO!
I'd be happy to send code starting next month when I have a life again. Smiley
SEGA Junkie
I haven't had a lot of time to do speedrun-related stuff lately owing to moving state and going to university again, but I've still been keeping my eye on community issues and thinking about improvements all the time. It's now been six months since the roundtable, so it's a good time to assess what's come out of it, and what is still to be done.

The main thing to come out of the roundtable specifically was that SDA had to change (read: speed up) or risk irrelevance. How has that progressed?

Verification
Pretty much immediately, public verification trials were conducted, and deadlines were set and enforced in all verifications. A quick look through the verification board shows that in the time period of the much-vaunted Big Push, as well as 126 runs being posted, about 70 were verified (these numbers have considerable overlap, suggesting the postverification process has improved significantly too... more on this in a bit), and the "runs needing verification" topic is at 3-year lows. This suggests verification is back under control.

Backend Improvements
UA has clearly put a huge amount of work into automating many of the time-consuming tasks that has made a lot of that improvement possible. Just as important though, has been the transparency of these updates: every update is explained with its impact on the submission process in mind. It won't be long until verification is the only process that isn't automated. You're all sacked! Wink

SDA as an Information Repository
This was the major recommendation from this very thread: start using SDA (and in particular, the Strategy Wiki and Knowledge Base) as a source of knowledge, allowing anyone to pick up their favourite game by looking at the best strategies. Of course the videos already kind of do this, but there's nuances of control and movement that aren't always obvious, and in my main game (Sonic 3 & Knuckles) videos are completely useless in explaining certain tricks. Now, by its very nature this isn't a job that can be left to a small group of people - it requires everybody to actively participate. A few of us have been throwing ideas around, but Mr K has taken the lead and developed a prototype for the kind of pages we want to populate the Strategy Wiki with. It's just about ready, so you'll all be hearing a lot more about this very shortly, and all be able to contribute to this goal.

Now, is there a rift between SDA and SRL? Not really, certainly not on an institutional level. Any stuff like that is generally pursued by individuals. But the perception still exists, and it's time we actively did something about it. In this sense it would be much more productive to think of the entire speedrunning community as one group rather than separate factions... but I'll talk more about that at the end.

That's all for the SDA end, which as we all know is just one facet of the community these days. Let's now look at some of the other things that have gone on in the community at the same time:

New Innovations in Speedrunning
On a more positive note, the last few months have seen some really interesting innovations in speedrunning: SRL Seasons, the Speed Gaming League, and the weekly Showcase Race. I'd be very interested to hear the thoughts of people on how these have affected speedrunning. My initial thoughts are basically as follows: Seasons encourages people to pick up a narrow range of games (but in greater numbers), SGL's format has revealed a new class of quality speedrunners in the #blind crew, and Showcase Race will be interesting to monitor... I'm not sure how it'll go in the medium term, to be honest. If JRDQ taught us anything, being an intermediary with money like that is a rabbit hole that it's a great risk to pursue.

With all of that said, and the assertion that the community is more dynamic than ever, it's time to propose...

A Model for the Future
Despite all the derision they get for some of their attitudes, the fighting game community is a worthy case study here. When they put on big events, we don't talk of them as IPlayWinner, or LevelUpLive, or Team Spooky events. They're FGC events. And even though there's some divisive figures in the community, overall they're all working towards a common purpose. We have an even more obvious common purpose, particularly when we're at marathons, and we should be presenting a united front at them. I'd be very interested to hear what people think about the following proposal: when our main commentator introduces Awesome Games Done Quick or whatever marathon it happens to be, we speak of it as a "speed running community" event, not an SDA event with assistance from SRL or however we did it last time.

Relatedly, it's probably about time we decided as a community whether we're going to refer to it as "speed running" or "speed gaming". The difference might seem insignificant on the surface, but it appears to be one other way in which factions assert their difference.

One final thing... I'm presently putting together a survey designed to see how different players in the community see collaboration in the community manifesting itself now, and how they'd like to see it manifest in the future. I want to get views from speedrunners, both SDA and SRL regulars, and people who just watch streams on Twitch. If you think you can/want to help, shoot me a PM!

I look forward to some interesting and productive discussion!
Quote from mike89:
Now, is there a rift between SDA and SRL? Not really, certainly not on an institutional level. Any stuff like that is generally pursued by individuals. But the perception still exists, and it's time we actively did something about it. In this sense it would be much more productive to think of the entire speedrunning community as one group rather than separate factions... but I'll talk more about that at the end.


I'm not sure there's even a perception of it anymore. I haven't heard anything about this supposed rift in months. In any case, as long as we treat each other respectfully and stop talking about a rift that doesn't exist, it and any perception of it should go away on its own. I think the solution on this one is to stop treating it like a big deal if it isn't one (and an issue that doesn't exist is not a big deal, if you ask me).
INTJ
Quote:
Relatedly, it's probably about time we decided as a community whether we're going to refer to it as "speed running" or "speed gaming". The difference might seem insignificant on the surface, but it appears to be one other way in which factions assert their difference.


Might just be me, but I have not heard of "speed gaming" a great deal. Where does that come from?

I have nothing else of value to add, but I'm interested in seeing how this develops
Edit history:
moooh: 2013-05-07 11:10:15 am
moooh: 2013-05-07 11:08:37 am
Exoray
If people wanted to refer to the act of playing through a game quickly and skillfully as "speed gaming" then wouldn't that be their choice? I mean sure, they'd be in the minority but I don't see why that would be an issue. It's not like you could confuse the meaning of the two terms anyways.

Oh and I'm eagerly awaiting the new kb suggestions. It'll be good to have a coherent information resource once again.
sda loyalist
Despite speed running being the older term, I prefer 'speed gaming' because I'm fed up of having to explain to non-speedrunners that the running does not actually involve running. 'Gaming' is much clearer, and the norm these days is to assume video gaming rather than any other type of gaming.
1-Up!
I know that older is not always better, but really "Speed running" is pretty well ingrained into the speed running community (not just SDA, imo).

Just doesn't make sense to me. "What is the product of speed gaming?" "A speed run." "Why don't you just call it speed running?"
Caution: This user contains Kana ^_^
A carpenter doesn't make carpets. Sir, your argument is invalid Wink

On-topic: Both terms can be understood, and I think everyone should go with what they prefer. For me, that'll be speedrunning, because tradition.
.
I'm not a fan of this site using the term 'speedgaming', just because The Speed Gamers already exists as a collective community and I wouldn't want people to get confused.