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  <- 1234567891011 -> <- 1 ... 10, 11 ->
--
--
List results:
Search options:
Use \ before commas in usernames
HELLO!
I mean links to *collections* of data that aren't in the speeddemosarchive.com family of sites.  Those should not be prominent on the front page, tops of SDA pages, or SDA sidebars.

Links to individual videos or streams?  Great.  But if someone comes to speeddemosarchive.com, try to keep them there a while.
All the things
This probably deserves a bump. There have been a lot of ideas expressed here, and I'm sure some of it has taken hold with the staff in the background, but one of my larger concerns is still about the number of staff working on changes or even the traditional operations. It's great news that UA has set up automated pages, but how many people are working alongside him to develop backend improvements? Front-page updates have been a long-time staple of the site, but the authors of those posts has narrowed significantly. Vulajin currently mans the community posts, but are there plans to expand the writing staff for more consistent and varied updates?

I know that the staff has aspirations for ways to improve the site and content and that work is being made towards those improvements. At the same time, I know that staff are volunteers and time is split between many other endeavors. I am worried that just keeping the status quo is not sustainable. I still would like to see a wider invitation for developers, graphic artists, writers, and otherwise to build up the technical and creative staff to keep updates humming along. There is danger in diluting quality for teams that get too big, but as far as I can tell as a regular user the current teams of just a couple people are not going to be able to keep up with the demand for changes.
Exoray
You are correct in a lot of points.
So, when can you start? Smiley
HELLO!
What's the backend of the homepage look like? I could start this weekend.
All the things
Quote from moooh:
You are correct in a lot of points.
So, when can you start? Smiley


Whether in jest or not, I'm happy to help with what I can, when I can.

But a big part of my post is that if you want to expand the staff, it's probably good to get word out more widely, whether by way of announcement or front page post or otherwise.
Exoray
Far from a jest. What we could really use right now is someone to act as project lead and help hold things together and drive it forward. I feel like you have the exact qualities for taking a role like that. If you'd be up for something like that I can fill you in on some details around next week (out of town for a couple of days)
I'm available to help with whatever. I have a lot of experience with php, *sql databases, javascript, html, css, python, etc.
Dapper as fuck.
I have started helping the verification process doing PRC, but would be willing to do more in that area if there is need.  Unfortunately I can't help with server stuff as I'm not a programmer/DBA/whatever, just a network/voice guy.  Smiley
Formerly known as Skullboy
I don't know anything about coding, web-site design, etc but I can write up game blurbs, updates, etc. Writing and research are my specialties. I just need to know how to get the information to the site after writing something up.
Wake up and be awesome
I can also write up the blurbs. It was fun doing it for those few times.
Clear as a crisp spring morning!
I can do simple graphic design. I could also imagine writing for the site, I've done amateur journalism before, but the rest that offered that would probably do just fine Tongue
I'm addicted to games
We need someone to help with "community news". Vulajin was doing them, but now says he's too busy.

More importantly, I think, we need someone to help with "public relations". There are too many games on SDA with very old runs, and nobody submits improvements because "SDA is outdated". The contradiction is apparently lost on people. We need to get people to submit better runs, and realize that the days of "6 months to post" are gone. I don't know how to make that happen.
HELLO!
Respectfully, *are* the days of long publication gaps gone?

Looking at the most recent post:

Uninvited was ruled verified October 15 2014.

Zero the Kamikaze Squirrel was ruled verified October 5.

Fatal Frame III was ruled verified on November 11.

It's still taking 4-6 months to publish.
I'm addicted to games
I can't comment on any reasons for those particular run taking ~6 months, but I can tell you that the average age of all runs in submission is "only" 97 days. That's not great, but it's not six months. It will never be less than a week. These things will take time. So, how do we make people realize there is value in submitting SDA and waiting a few months? What's the point, when they can just post it in youtube immediately! Maybe there isn't one, or I'm too old and stuck in my ways to figure it out.
HELLO!
Far be it from me to suggest that standards be reduced (really), but I wonder if some of the steps involved here really are integral to the SDA quality level.

I'm not saying they aren't. I'm writing this in genuine ignorance.  Timing?  Re-verifying the video quality?  Why does SDA, and only SDA, take 3 months to go from having a confirmed run to putting it on the site?

I like SDA. I'm a fan. I'm not asking these questions out of of hostility.  Just the other day on stream I snapped at a guy in my chat who came in trying to tell me SDA is pointless, blah blah blah blah.

I guess I'm just trying to ask questions to search for ways to improve, to make SDA easier to sell.
Weegee Time
Quote from Radix:
I can't comment on any reasons for those particular run taking ~6 months, but I can tell you that the average age of all runs in submission is "only" 97 days.

Er... if the average in the submission stack is 97 days, doesn't that mean there's probably a bunch of runs that are near 5-6 months, and a bunch of runs that recently entered the queue?
Quote from presjpolk:
Timing?

Potentially interesting thought: SDA is not a world record site, so why do we care so much about the timing?  I understand that it's an easy comparison metric.  However, consider some of the other, more amorphous metrics.  Is the run using modern strats?  Is the execution clean?  We pick these things up in verification.  I think that in many situations we can tell that a freshly submitted run has better quality than the currently posted run, or is of sufficient quality for a brand new game, without looking too hard at the time.  I don't know that Israeli's process takes too much time for him to perform, but it would still alleviate a potential bottleneck.  It's also possible that deemphasizing the time might motivate some people to submit, because it makes it clear they don't have to have the best time.
I originally saw this topic a long time ago and had a long post written up, but decided against it. Since some of this is related to what I wrote, I guess I'll write something up.

Quote from Radix:
More importantly, I think, we need someone to help with "public relations".

Yes, you need better PR. SDA has failed to define itself. Without a statement of purpose, without a well defined goal, people will not go on this website and find out SDA's goal, but try to find one itself. And the "goal" that people end up with is that "SDA is a WR site." The often used "SDA never claims to be a world record site" is a horrible excuse, when nowhere on SDA's website does it state what SDA claims to do. SDA is not a WR site? Fine, then what is it?

I know this may come off as aggressive, but here is the situation. Suppose I am honestly trying to find out what SDA is. I go on all of the websites. I cannot find the answer. I seriously cannot. The first question on the FAQ page is "Who runs SDA?" when honestly, I think it should be something like "What is a speedrun?" or "What is SpeedDemosArchive?" How is anyone supposed to find out what SDA is then? Is it actually really easy to find, but just slipping by me?

Here is a possible solution to the long delay between submission and publication. Do what TASVideos does. On the home page, there is a submissions section where I can see all submissions even before the final decision. In fact, there are many things that TASVideos does a good job. The issue I pointed out above does not exist on TASVideos. They have a page dedicated to explaining to someone who honestly wants to learn more about TASes and what TASVideos is. They give answers to what a TAS is, why make TASes, and what TASVideos does.

And honestly from here, I don't have much to say, not because I don't have ideas, but because I don't know what SDA is trying to do, nor can I really say why a person should submit to SDA.
umad
I said some of this on the first page of this thread, but it kind of got buried under the initial explosion of activity, so I think it's worth repeating, clarifying, and expanding, especially in light of what Rakuen and ahuynh just said.

Speedrunning has undergone a massive transformation over the past few years. In the past, speed running was about creating a final product. The goal of a "serious" speedrunner, in those days, was to get a run recorded that was fast, used the best feasible strategies, had few mistakes, and was something you were happy with and proud of. The term "World Record" was rarely, if ever, used back then, because that wasn't the point - the point was to create a final product for the community; a run that would, ideally, go up on SDA and serve as a representation of what was possible in that game. Obviously there was a competitive factor to it as well, but the community was small enough that there was usually only one person (maybe two) working on each category for a given game at any given time; most competition was between you and the run that was already on the site, not between you and another person who was working on the same category at the same time. And usually, once a run was put up on SDA, it was considered "done" until enough new strategies were found to warrant trying to improve it. In a way, speedrunning was a lot more similar to TASing back then - there was a competitive aspect to it, but the emphasis was usually on demonstration and showcasing, not on "World Records".

But as speedrunning got more popular, this paradigm started to break down. With more and more people interested in doing runs, records were getting beaten much more frequently, to the point where the idea of a "finished product" didn't even make sense anymore. Runs weren't really considered to ever be "finished" anymore; there was always pressure to keep going, to shave off those few extra seconds by optimizing just a tiny bit more. Each category now had dozens of people running it and going for the World Record™, not just one or two people "working on" it and trying to get a finished product. Speedrunning went from being almost like an art form (again: think TASVideos) to being like a competitive sport (like high scores in arcade games), and because of this huge paradigm shift, SDA started to fade into irrelevance.

The problem is that SDA, in general, has always been about the "demonstrative" or "artistic" side of speedrunning, where doing a speedrun is about creating a final product to showcase the game pushed to its limits. Submitting a run to SDA meant that it was a finished product that would serve to represent the game on the site for the foreseeable future until enough advancements in the route, tricks, glitches, strategy, etc. were developed to warrant doing a new run to obsolete it. In this new, ultra-competitive climate, why would someone submit a run to SDA if it's just going to get beaten within the next month by a few seconds using the exact same strategies? If I spend months working on a Majora's Mask run (for example) for SDA, submit a finished product, and then someone beats it by 30 seconds a month later and then submits that, what significance did my contribution even have? It's one thing for someone to beat my time on a leaderboard shortly after I get it (which is expected); it's another thing completely when my "finished product" that was supposed to represent the game on SDA gets replaced on the site by a new one with barely any differences a week after I submit it.

The upshot is that SDA doesn't do well with the competitive, WR-driven style of speedrunning that has come to dominate the community within the past few years. That style of speedrunning is more suited to leaderboards, where the emphasis is just on ranking times rather than showcasing clean runs of the game. It's not just that SDA takes too long to update, or that the submission process is slow - it's that SDA serves a completely different purpose than leaderboards do. As has been repeated over and over again: SDA is not a world records site. It's not supposed to be. It's supposed to be a site that hosts videos to showcase the current state of speedrunning in a particular game. If someone wants to submit to SDA for that purpose - to contribute a run that will serve to represent the game for the foreseeable future - the fear that someone is going to submit another run 2 weeks later that's a few seconds faster is going to be a HUGE disincentive. I can't speak for others, but I know that in my own case, the biggest thing discouraging me from submitting to SDA is that someone else is going to submit a slightly faster run 2 weeks later, and then it'll just turn into a constant submission war to the point where we might as well just be fighting for WRs on a leaderboard. And at that point, what's the point of even using SDA in the first place, as opposed to already-established leaderboards like ZSR or Speedrun.com?

My proposition is this: SDA needs to revamp its image to distinguish itself from a leaderboard or World Record site. I think Rakuen's idea about ditching the timing metric is a good starting point for discussion, although I don't know if ditching timing completely is the right way to go about it. I also think that it might be a good idea to think about adding interesting "arbitrary" categories that don't get a lot of exposure, as a way of emphasizing the "showcasing" aspect of what makes SDA unique from your typical leaderboard site. (A good example of what I mean is the 2-pause Majora's Mask run that Kaztalek made, which is a rather arbitrary goal but results in a REALLY interesting route that most people have never seen or even considered before - although the 2-pause run was done on an emulator with savestates since it's rather unfeasible to do in real time.)

I obviously don't have all the answers, but I can at least throw ideas out there and hope that someone catches them and can find some way to use them.
The artist formerly known as Qxy
I don't think forgoing the time or even some of the competitive nature of speedrunning is a viable solution. While it may seem like allowing more arbitrary categories and focusing on other qualities will give SDA it's own niche, it's not a niche I think anyone wants. SDA is about speedrunning and everything that comes with it; there are only a couple reasons why they can't really claim to be a WR site either. They need permission from runners, and would prefer to host high quality unaltered gameplay videos, both of which cannot be said of every WR. But in an ideal world all the WR runs should go to SDA if the game has mellowed out enough. And while I agree that it isn't as often, games absolutely get into that state, where someone gets TEH URN, no one can really improve it without a route change, and the community moves on to something else. Suggestions for how to facilitate this:

1) Keep the hype of a new run fresh with a more proactive or informal verification process: The community for any decent sized game is pretty big these days, usually agrees when a run is good/done, and chills out on other places than SDA. If people affiliated with SDA happen to be involved in these communities watching the stream at the time of teh urn, or chilling in an IRC/teamspeak when a number of relevant people are talking about a recent submission, it can be pretty easy to gauge an opinion of quality. A system isn't foolproof, but hey, neither is actual verification. Odds are that these people would be the ones verifying anyway so it's subject to the same sort of bias pitfalls. Related to this, and helping to facilitate it:

2) Allow verifcation of gameplay quality from a livestream and video quality in a more compact form during the submission process that doesn't involve outside verifiers. Push how easy recording a high quality hard copy in addition to streaming it and make guides on the subject more widespread. If the process can be verification first, send the torrent, SDA checks the video quality by looking at ~30 seconds and extrapolating ala a video quality test in the tech support forums, then there's no feeling of slowness on the submitter's end. People submit runs that are going to be accepted anyway, may as well line up with intuition. What do we need to do this, and replace Vulajin?

3) A network of "community outreach" members of the site, each with not a lot of responsibility. People that watch a lot of streams and keep up with community news for the hell of it without having to be told. People who could greenlight a run as verified before it comes in using the informal process because they're trusted members of SDA. People who could rotate the responsibility of writing the community news posts on the front page with all the tips the site gets. We couldn't get complete coverage, and this wouldn't replace regular or even public verification for the medium to small games, but it would still keep the hype of the major aspects of speedrunning relevant to SDA. All the "must watch" runs could slowly gravitate here again, and that's pretty much SDA's only problem as far as reputation.

I don't know how much sense all of this makes but I think there would be a lot of people willing to do something like this. With a single dedicated outreach coordinator, as long as each person who volunteers can be trusted and isn't given too much work, I could see SDA as a more active force in the community, which would inspire the community to give back.
Edit history:
puwexil: 2015-04-04 11:38:37 am
Professional Second Banana
To add on to what ING-X said about SDA videos being more of a final product than leaderboard links, something unique I think SDA provides are the text comments and (post-run) audio commentary that most runs on the site have, which I feel make the runs more accessible to people outside the community for that particular game.  Usually watching a Twitch VOD with live commentary doesn't give me much insight into the route/strategies, since most runners (myself included) focus more on interacting with chat there and now rather than on making a video for future viewers in a wider audience.
Edit history:
ConHuevos: 2015-04-04 05:29:14 am
Ciento Dos Huevos
Quote from presjpolk:
Respectfully, *are* the days of long publication gaps gone?

Looking at the most recent post:

Uninvited was ruled verified October 15 2014.

Zero the Kamikaze Squirrel was ruled verified October 5.

Fatal Frame III was ruled verified on November 11.

It's still taking 4-6 months to publish.


And this still goes unanswered.  No surprise.  What exactly takes 4-6 months in time to update an archaic page of the game when a runner provides comments and the video is already verified?
All the things
While I can't speak universally, Zero's publication was delayed since it was waiting on commentary and final encodes from me. I can bet that a lot of the other older entries in the queue are waiting on some kind of submitter interaction as well.

That said, it may still be time to think of a better way to go about submissions/verification overall. The question of "Why should I submit to SDA?" is still hanging open from earlier in the thread. Verification is the crux of the current mission statement, but either the mechanisms in place for it are no longer adequate or the "meaning" of being verified just doesn't hold enough water anymore. It's worth looking into new ways to set it up.
Edit history:
Serris: 2015-04-04 05:30:24 pm
I don't think the video archive side of SDA has to compete with Twitch and YouTube for SDA to be relevant. Striving for popularity over quality was never the mission, and the increase in exposure is much, much larger than the gain the community got from it, with a larger percentage being passive viewers.
The fact that segmented runs are basically dead speaks volumes about the change in culture. The average stream audience doesn't care about speedrunning, they want skill showcases and entertainment, and while speedruns just happen to be both most of the time, there's a significant difference in philosophy that doesn't fit with SDA's mission. The very fact they're a "stream audience" implies they don't care about a final product, since they're there to watch the process and chat.
If people don't care about submitting runs anymore, so be it. The community desperately needs a central hub that has links to game/series-specific communities and serves as a forum of discussion for everything else.
The decentralized nature of streaming actually seems to have made it harder to share things if you're not already part of some circle surrounding a specific game (that isn't Zelda or Metroid) who are hanging out in some Skype/IRC room, if you're not content with just slapping it on YouTube and hoping it reaches people eventually.
A search-by-game news page for new discoveries ("Someone found glitch x today! Here's a video") would also be easy to maintain if many people have the ability to contribute. And yes, the Knowledge Base should be a huge focus.
Not a walrus
Quote from ConHuevos:
Quote from presjpolk:
Respectfully, *are* the days of long publication gaps gone?

Looking at the most recent post:

Uninvited was ruled verified October 15 2014.

Zero the Kamikaze Squirrel was ruled verified October 5.

Fatal Frame III was ruled verified on November 11.

It's still taking 4-6 months to publish.


And this still goes unanswered.  No surprise.  What exactly takes 4-6 months in time to update an archaic page of the game when a runner provides comments and the video is already verified?


Short answer: The amount of staff power we currently have is too low for the amount of post-verification work that needs to be done on a run. It's gotten better because I've been able to automate more of these steps (game pages still have several things that need to be worked out before they can be fully automated, as does posting news items), so once I finish those we can a) bring in a couple more "staff" (ie news writers) and b) shift some of the manpower back into the stuff that can't be (fully) automated, like timing runs.
Borderlands 2 Glitch Hunter/ router.
Quote from UraniumAnchor:
Quote from ConHuevos:
Quote from presjpolk:
Respectfully, *are* the days of long publication gaps gone?

Looking at the most recent post:

Uninvited was ruled verified October 15 2014.

Zero the Kamikaze Squirrel was ruled verified October 5.

Fatal Frame III was ruled verified on November 11.

It's still taking 4-6 months to publish.


And this still goes unanswered.  No surprise.  What exactly takes 4-6 months in time to update an archaic page of the game when a runner provides comments and the video is already verified?


Short answer: The amount of staff power we currently have is too low for the amount of post-verification work that needs to be done on a run. It's gotten better because I've been able to automate more of these steps (game pages still have several things that need to be worked out before they can be fully automated, as does posting news items), so once I finish those we can a) bring in a couple more "staff" (ie news writers) and b) shift some of the manpower back into the stuff that can't be (fully) automated, like timing runs.


Also as a Pre-release checker you find that some delay is actually caused by the files given to SDA that are incorrect and which can only be resent by the runner / runners.

As an example; there is a recently checked run which had an error with one segment (excess footage [failed attempts]) when the runner was contacted it was found that he had not been on-line in 5 - 6 months and it took a further month to get a reply for that person and have the corrected file sent.

Some time is due to workload, some is the fact that we are volunteering our own free time to do it, some / many of us do OUR OWN speedrunning and some days you just say "I wanna do attempts today".

It's being worked on and our numbers are increasing but it's still done in our free time. im not saying that everything isn't our fault but some times it just isn't, stuff gets delayed for all kinds of reasons.

<3 <3