THE LOG← all posts

OPINION · POST 001

Why there are no quality games anymore, only slop

·3 MIN READ·KODO
tl;dr
  1. 01Volume went up. The floor for what counts as a game went down with it.
  2. 02Hiring now selects for a portfolio and a posting habit, not for the ability to finish.
  3. 03Long projects are the only thing left that can't be printed overnight.
contents

I posted a thread at 5:28 in the afternoon and went back to work. It had three views. It says most of what I think, so I am putting it here first, unedited, and then taking it apart slowly.1

POST EMBEDARCHIVED COPY
kodo@tarik06784901

Why there's no quality games and we only get SLOP? Game creators and amazing people with ideas like @TizzzyRBLX struggle to find proper people for the job. Real devs you actually need for the job don't get the right opportunities. Everyone seems to be a dev these days. #roblox

01 · Everyone is a dev now

The community I joined was small and slow and mostly made of people who had shipped something badly at least once. That is gone. What replaced it is a very large group of people with a portfolio site, a content-rich account, and no finished project behind either.

The problem isn't that they exist. The problem is that the signal a studio uses to pick a developer is now the same signal anyone can fake in a weekend. If you have never shipped, the fastest route to looking hireable is not building — it's posting.

“The arrival of AI changed the paradigm of development everywhere, roblox development is no different.”

— from the thread, 8 minutes in

02 · What AI actually changed

It did not make anyone a better developer. It made unfinished work look finished, which is worse, because the audience for this platform is mostly kids, and kids do not audit anything. We are the ones setting the standard they grow up thinking is normal.

03 · The listing vs the job

Four listings I was sent this month, and what each one turned out to be after one call.

THE LISTINGTHE JOBTERMS
“Experienced scripter, long term”Finish an abandoned prototype in a week% of revenue
“Systems developer, portfolio required”Rebuild a copy of a front-page gameflat, low
“Looking for a technical co-founder”Unpaid, indefinite, idea already publicequity, verbal
“Small fix, quick job”A save system with no schema and no backupshourly
TABLE 1Paraphrased, names removed.

04 · What good looks like

Here is what the careful version looks like: a round loop where the client only ever asks and the server decides. It is longer than the version that trusts the client, and nobody claps for it.2

.LUAUsrc/server/RoundService.luau
-- authoritative: the client only ever asks
function RoundService:RequestStart(player)
  if not self:IsHost(player) then
    return false, "not_host"
  end
  if self.state ~= "lobby" then
    return false, "busy"
  end

  self:SetState("starting")
  self.deadline = os.clock() + 3
  return true
end

05 · What I'd change

Nothing structural, because I can't. Three things I would ask for:

  1. Ask for the game, not the portfolio. Play it. Ten minutes tells you everything a grid of renders won't.
  2. Offer a paid trial task instead of a revenue share on a game that does not exist yet.
  3. Write the postmortem in public, including the parts that make you look slow.

That's it. I'll be in Studio.

NOTES

  1. The thread has four replies; the rest of it is mostly me arguing in the same direction. ↩

  2. This is simplified. The real one carries a per-player intent buffer, which is the boring part nobody screenshots. ↩

FILED UNDER