My question though is STRICTLY about the running and engineering practices of Twitter/X. We have seen Twitter's revenue plummet, and a lot of negative press.
But what we haven't seen is twitter the actual website/platform implode like many engineers predicted.
So I want to ask for the opinions of y'all. Are most companies hiring too many engineers? How were they able to maintain this massive platform despite cutting off so many engineers. Have we over-estimated the importance of engineers?
Why have all of these take a plunge?
As engineers, sometimes, we tend to think of all problems as just software and forget about the large machinery that works behind us to make the product successful. A good analogy is the military. When you think of armed forces, you think of soldiers, the fighters. But, for each fighting person, there is a 5 - 10x the number of support personnel - from logistics, to chefs to camp maintenance to chaplains to everyone in the middle.
Similarly, for twitter/x to be successful, it needed marketing, sales, support, content moderation, legal, etc. If you get rid of all of those, then you maybe running a software, but not a business.
At the same time, it is true that most large companies run a bit fat. But this is also the first chapter in Mythical man-month, aptly named "The tar pit" and Frederick P. Brooks, Jr does a better job than me in explaining why it is so.
[https://web.eecs.umich.edu/~weimerw/2018-481/readings/mythic...]
Oh, right, then.
Middle management, project/product empires mostly increase overheads with sublinear increase in output.
I don't know what all those people at Twitter where doing, but I have worked at large companies and many engineers where working on developing possible new products and revenue streams, many of which, for different reasons, never see the light of day. If you're just looking at it from the outside it looks like there are a lot of people literally producing nothing.
With all that being said, I'm sure there are lots of companies out there (including pre Musk Twitter) who could lose a bunch of people without it affecting their bottom line or long term productivity.
1. Paying for blue checks. Which for me just means that blue checks are the hallmark of the lowest quality comments and posts now.
2. Renaming from Twitter to X. This one highlights how Twitter is basically unable to even do this correctly without a proper engineering team. Normally you'd implement this so that navigating to twitter.com would redirect to x.com, but Twitter had to do the opposite because changing the primary domain name likely would have touched a ton of underlying assumptions throughout the code, and Twitter was no longer equipped to be able to handle a change of that magnitude.
Why should it be any different for an IT system? Once the system is built, stable and does not require changes (Twitter's feature set hasn't meaningfully changed until recently, ironically post-Musk), it shouldn't require anywhere near the amount of engineers that it needed when it was being developed.
Anyone saying otherwise likely has their salary depend on it, which is a common thing around here thus all the noise that was made around the time the firings were announced.
Whoever said that Twitter would collapse has egg on their face now. But I wouldn't say we've "over-estimated the importance of engineers" but rather these big tech companies are mismanaged.
Twitter had plenty of issues post-firing. There were outages, features not working correctly, etc.
Let's not forget that the firing was a giant f-up, poorly planned, and done at the behest of a mentally ill billionaire.
But then it stabilised. While launching new features. Twitter's value destruction has been entirely a function of Musk's personal antics. Not their engineering quality.
I used to search within lists (was like the best feature of Twitter, for my usage) quite a bit -- after Musk gutted the team after taking over, it no longer works, and has been that way for many months.
But no. Now, after paying many millions in severance, he is hiring an entirely new team to do an LLM model.
* if the platform implodes, clearly the people who were fired didn't build a good platform
* if the platform doesn't implode, clearly the people who were fired weren't needed
