Friday Facts #141 - Mod Portal

[LotA]

The new loco kicks ass, can't wait to see how you managed to make it stretch while turning. Anyway, with these new graphics, will you update the trailer ? xD

[Smarty]

Dat new train tho

[Nova]

Just because a system is not really much used, doesn't mean it should not be available. What's the problem with a five star system? People who would use "like" / "dislike" can just rate it 5 or 1 star - but the other people can use 2, 3 or 4 stars. To be honest, I first look at every not-5 an not-1 star rating for products - that reviews are in most cases of better quality, because the reviewers have mostly considered more aspects about the product to come to the conclusion of 2, 3 or 4 stars.
It's not a good way to look into products by focusing at things many people use, because if you remove that you will anger the people which did use it, and that are sometimes the biggest fans.

[Erigalus]

Although there isn't any 0.13 in sight, I just have discovered that the achievements are already there. But considering that the global achievement stats of them all is at 0%, I assume that they aren't really implemented yet. Which is good, since I discovered a typo.
The achievement 'Streamrolled' says in the description: 'Destroy 10 Spanwers by impact.'
Just wanted to point that out. Really hyped for the full version, the train looks epic.

[jorgenRe]

How much of an impact would it be to perfomance to use the global variables compared to normal ones?

[vaderciya]

That is a sexy ass locomotive, I love it!!!!
Even more reasons to use trains

[bNarFProfCrazy]

I don't know whether you actually read the responses on this thread because there are so many here.

But could you please make longer clips of the train aka a full turn (or at least a "full" half turn), because the current one is too fast for me to look at and the virtually jumping train distracts my eyes.
Or I'm just getting old.

You will be able to rate mods, probably using a 5-star scale and filter/sort mods by rating

I don't like a 5 star system.
Either use a 3 star system: Don't like - Neither like nor dislike - Like
Or a 2 star system: Don't like - Like (With a hidden third option aka do not vote)
Reason: The "Awesomeness" of the mod is quite opionionated. Someone might vote a good mod 4 stars because it still have some bugs, while others still rate it 5 stars.
However if they have rated a mod once they won't update their rating even if the bug gets fixed at some point.

Another thing is:
Use timespan based votes. Mods develop so it would be cool to get the current and overall rating (Similar to Steam).
(It must be possible to update his ranking vote)

How much of an impact would it be to perfomance to use the global variables compared to normal ones?

Almost none, but its
a) Harder to keep clean/remove outdated/unneeded data.
b) Harder to implement proper garbage collection.

[Kane]

What is your secret to keeping the exhaust so clean?

[Neemys]

Nice locomotive, really factorio-style

Although there isn't any 0.13 in sight, I just have discovered that the achievements are already there. But considering that the global achievement stats of them all is at 0%, I assume that they aren't really implemented yet. Which is good, since I discovered a typo.
The achievement 'Streamrolled' says in the description: 'Destroy 10 Spanwers by impact.'
Just wanted to point that out. Really hyped for the full version, the train looks epic.

Typos already reported, and there is another typo you haven't see : "Streamrolled" should be "Steamrolled" (this one was reported too)

[AlexTheNotsogreat]

That train will defiantly take Bentham by surprise, that is just amazing!

[MiiNiPaa]

Actually, in niche groups where signal/noise voting ratio is higher, Bayesian ranking works pretty well. Vote weight decaying over time is something worth considering too. Requiring to write a review before voting, rating reviews by other users (binary: useful/not useful) and changing vote weight based on that would prevent compulsive voting.

[againey]

I'm mostly fine with whatever rating system is used, but I suspect much displeasure with rating systems comes from the fact that a single value is a severe oversimplification of just about any sort of judgment. Trying to break it up into multiple categories is tricky, though; don't wanna go overboard with too many categories, or an unhelpful selection of categories.

As a suggestion, I'd recommend two categories: "Concept" and "Execution" (or quality?). Many mods might have a great concept but poor execution. Or high quality execution but most people kinda think "meh" about the concept. The best mods of course will be highly rated on both concept and execution.

I'd expect the Concept category to have at least three ratings, so that the ideas which are neither terrible nor phenomenal can sit in the middle, instead of getting forced into the exceptionally bad or good categories. Execution might be okay with just two, more or less indicating if the mod either functions as intended, or needs more work. Whether it functions perfectly or only nearly perfectly or whether it is mostly broken or completely broken are details that don't really need to be reflected in the rating.

And regarding the time based ratings bNarFProfCrazy suggested, they would make perfect sense for the Execution category, though it probably makes more sense for the Concept ratings to persist indefinitely.

[KatherineOfSky]

Oh, wowee! What a super sexy stempunky loco! Can't wait! ...meow...

[Adil]

New train seems weird. Both the organic design and the stretching that can be seen on gif.

[xxninjaxx25]

Looks like they just have to finish the mod website and multiplayer connection issues

[Kewlhotrod]

the actual problem with 5star systems

[Ojelle]

O-god, that train!
Nice!

And a mod portal will be lovely too ^^

[pr0n]

Just because a system is not really much used, doesn't mean it should not be available. What's the problem with a five star system? People who would use "like" / "dislike" can just rate it 5 or 1 star - but the other people can use 2, 3 or 4 stars. To be honest, I first look at every not-5 an not-1 star rating for products - that reviews are in most cases of better quality, because the reviewers have mostly considered more aspects about the product to come to the conclusion of 2, 3 or 4 stars.
It's not a good way to look into products by focusing at things many people use, because if you remove that you will anger the people which did use it, and that are sometimes the biggest fans.

Absolutely it does mean it shouldn't be available. It's a ranking system, thus it doesn't work on pure principle if not everyone is using it the same. Which in a 5 star system is as sure to happen as gravity existing. For further reading it's the same point that is argued in the link on my previous post.

[Erigalus]

Nice locomotive, really factorio-style

Although there isn't any 0.13 in sight, I just have discovered that the achievements are already there. But considering that the global achievement stats of them all is at 0%, I assume that they aren't really implemented yet. Which is good, since I discovered a typo.
The achievement 'Streamrolled' says in the description: 'Destroy 10 Spanwers by impact.'
Just wanted to point that out. Really hyped for the full version, the train looks epic.

Typos already reported, and there is another typo you haven't see : "Streamrolled" should be "Steamrolled" (this one was reported too)

Alright, then my post is from now on flagged as 'Feel free to ignore'.
I just noticed it after reading the fff, so I thought it might've been appropriate here, and didn't look on the other forums for it. Yes, I'm lazy and I apologize for it. ^^

[Bulisek]

The locomotive looks really awesome.
You are doing a great job.