pull down to refresh

119 sats \ 25 replies \ @grayruby 18h \ on: Spoiler Alert: Stacker News Will Hit +1.5 Million Items in 2025 oracle
Are you suggesting @ek is putting his finger on the scale?
write
preview
40 sats \ 24 replies \ @Coinsreporter OP 18h
I can't say who's putting his finger where but the item count has definitely increased in the last two days. And I know that some items are missing.. For example: #1033158 #1033159
reply
93 sats \ 23 replies \ @ek 18h
You can create items but fail to pay for them. Then they will not show up for anyone else.
reply
20 sats \ 18 replies \ @grayruby 18h
I think that's a bug. Should probably be fixed asap.
reply
40 sats \ 13 replies \ @ek 18h
It's not a bug, it always worked like that.
The ids are sequential, and you can always still pay for an item.
Once an id was assigned, it's gone.
reply
120 sats \ 1 reply \ @grayruby 18h
I still think it should be fixed. Haha
reply
20 sats \ 0 replies \ @ek 18h
I respect that opinion haha
reply on another page
60 sats \ 6 replies \ @k00b 17h
It's more complicated than just unpaid items or deletions. Rollbacks after inserts also leave "holes" in sequences.
Because smallserial, serial and bigserial are implemented using sequences, there may be "holes" or gaps in the sequence of values which appears in the column, even if no rows are ever deleted. A value allocated from the sequence is still "used up" even if a row containing that value is never successfully inserted into the table column. This may happen, for example, if the inserting transaction rolls back.
Afaik the best we can say about sequences is that they're monotonically increasing (if NO CYCLE is used (which is the default)).
The id is not the item count. It's a rough overestimate of the item count.
reply
40 sats \ 5 replies \ @0xbitcoiner 16h
You could just make a temporary buffer (new table) until the item paid.
view all 5 replies
20 sats \ 3 replies \ @Coinsreporter OP 17h
What about a sudden spike in total daily items? Is SN going parabolic all of a sudden?
reply
20 sats \ 2 replies \ @denlillaapan 17h
THE RETAIL IS COMING!
lol, cope
view all 2 replies
90 sats \ 2 replies \ @k00b 17h
There's a bug in your bet. Our ids are working as intended. :P
reply
0 sats \ 1 reply \ @grayruby 17h
Haha there is only a bug if I lose.
reply
40 sats \ 0 replies \ @ek 15h
not too late to join the bright side
reply on another page
20 sats \ 0 replies \ @Coinsreporter OP 18h
Yes, I also think so. The way daily item numbers have increased in last two days, it's more likely a bug.
Even then I think @ek has suggested how we can manipulate that market and make profit.
Should we try it?
reply
91 sats \ 2 replies \ @0xbitcoiner 18h
Makes perfect sense. @mega_dreamer has gotta have access to the items that are paid up!
4.31?
reply
40 sats \ 1 reply \ @mega_dreamer 16h
I have no insider information - I tired to bribe @ek but he refused to divulge any intel 😈.
reply
40 sats \ 0 replies \ @0xbitcoiner 16h
After we found out that unpaid items are getting queued up, there's gotta be a way to figure out the exact numbers. Otherwise, the whole market's screwed. This is trickier than it looks, because you just assumed there were "X" items when you opened the market, but turns out it was "X minus Y." I figure the right way to fix the market is to do: items at the start of the market + new items - unpaid items from the start of the market.
reply
40 sats \ 0 replies \ @Coinsreporter OP 18h
Ahh! Didn't know that. Is there any chance of improving this to 'no item creation' for 'failing to pay'?
reply