and when you try to rein in expenses to show less ads, the community begins to propose $10M buyouts and every refrain from reasonable optimizations of their API calls.
1. They are not "reining in" expenses, they are purposely pricing out 3rd-party clients (which probably has the goal to show more ads, not less)
2. The $10M buyout comments was a reasonable comment countering the statement that reddit loses $20M each year by users using Apollo. A comment which was a half-joke, as the audio recording shows clearly. In that case, buying Appolo for $10M would have been a steal and Reddit would have jumped on it. They didn't, their prior statement was probably a lie.
3. There were no suggestions for reasonable optimizations of API calls. Apollo claims they already optimized whenever possible and whenever they were asked for it, and that client was specifically not listed in an earlier discussion about ineffective API usage. Reported measurements confirm that, the official API uses more calls. Regardless, later Reddit made this accusation, showing that they are not truthful in this discussion.
Your pro-reddit position is not reasonable.
If something loses me 20M$ per year and I have an option to buy it for 10M$ or kill it for 0$, I would choose the latter without much thinking. And that seems to be exactly what Reddit is doing.
And that is something that any client using a third-party platform should expect that will happen eventually.
Sounds to me that shutting it down is the one sure way to lose money, but I also don't trust that the reddit team had made that calculation properly. This all seems like a short term profit or like a power play to me.
Weren’t the $20M reddit claimed they are losing, missed opportunity costs? I.e. if Apollo users wouldn‘t use Apollo but the official app, they would make that amount of money off of them. So buying that amount of revenue for $10M sounds definitely reasonable to me. If what reddit claims would be truthful that would be a guaranteed 100% ROI.
> A comment which was a half-joke, as the audio recording shows clearly.
…at some point people need to understand that jokes are about timing, about audience, about place, and the subjects of the joke.
again, i have no dog in the fight, but the number of times i’ve heard “it was clearly a joke” is wild. we’ve all seen people from every political corner from every walk of life proclaim “clearly that was a joke” when repeatedly it wasn’t received this way by those involved.
jokes are about timing.
all of that said, i do wish the founder of apollo well, if old.reddit wasn’t your thing apollo truly was one of the better alternatives.
slight tangent, but this is another lesson on how we inevitably screw ourselves when relying on non-open source.
if it isn’t ours, the thing is someone else’s. we have to face this. we repeatedly see the repercussions of ignoring this fact.
There is no reason reddit can't be profitable if they did these things. They don't need 100,000,000 users watching videos to bring in ad revenue. They need 10,000,000 users generating good content that ends up as top google results and they can find clever ways to monetize product tie-ins, annual meetups (those were a thing in a lot of cities a while ago), events, merch, and subscriptions. It isn't rocket science. The top stockholders just have to be satisfied to be millionaires instead of billionaires.
It'd pretty easy to be a profitable company with a $10 million turnover, but you're not going to be a unicorn this way. Founders and VC wants high return on their equity, and today being enormous without profitability is valued more than being reasonable and profitable.
There was once an age where investors would buy in simply on the belief of unicornedness, but that ship sailed several months ago. Investors have become much more critical of tech businesses, showing substantially less interest in tech companies that aren't displaying strong fundamentals. A recent Fidelity report indicates Reddit's own valuation has dropped by 41% on that change in sentiment.
In fact, if profitability was the goal, then he should have been ousted at this point, because he's been back for 8 years now without succeeding to make the company profitable.
I don't think he is that inapt, and that his board is stupid to let him in charge despite such a failure, profitability was just never the goal. Getting the biggest possible IPO is the goal, but the timing is actually very bad for them for the reason you mention and now they're trying as hard as they can, including by creating fake subreddits with translated old content for non-English speakers (at least French and German speakers).
Just fyi, reddit rose around the same time as Facebook so it’s likely they share a similar governance model, where founders control an outsized part of the board/voting shares. Since Uber and wework semi-imploded investors are less forgiving of that sort of thing but legacy companies like Reddit likely still are run that way.
Reddit needs 300 people max to run, and they got like 1500.
But did they try to be?
The feeds are definitely not cacheable (except for maybe on a per user basis with a low ttl). Popular and all are never the same upon reload. And every user is going to have a different combination of subreddits. Those feeds have to be generated on the fly. Most of the data is too dynamic and ever-changing to cache effectively.
Also a lot of their content is not "hot" and "commentable" anymore.
- Cache of most popular topics (again, even a 10s invalidation time) would help
For the logged users you could do something more dynamic, for example, cache the front-page snippets of the main subs + add the ones for the user. And again cache that
(unless the 'new reddit' makes this harder, to which I would say - well deserved! ;) )
Reddit already bought a third party app with Alien Blue, and they ruined it so badly that they had to close it down and start again. Nobody wants Reddit in charge of the third party clients, they have awful UI designers and a user hostile UX team. The whole point of a third party client is that it has to actually design a UI that it's users like, or it dies.
[citation needed]
(and not from someone known for lying and editing comments made by other people using their root access)