Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

We can be compassionate about the folks getting laid off while still realizing that the company was grossly inefficient and needed to restructure -- these are not contradictory feelings. For me personally, I have no idea what "technology and development pace" these layoffs even could effect, as I have seen no real change in the service in the past ~5 years I've used it.

Actually, I have noticed that in the last year or two they refuse to show me more than a tweet or two until trying to force me to log in.



> as I have seen no real change in the service in the past ~5 years I've used it.

You underestimate the effort it takes to keep a service as large as Twitter running. Constant software upgrades to maintain high standard of security, inevitable technological break-age, deprecated dependencies, operations, root-causing failures, migrations (you can't avoid hardware migrations; hardware will die out), planning fallacy, etc, etc.


Yes it's incredible to see so many people on _hacker news_ who don't understand how much work it is to keep even moderately large (low millions of users) services running.


Yes, and repeatedly people point out that high dozens or low hundreds should be enough but they have thousands (allegedly).

I think the point you are missing that especially tech people might have been using twitter for 10 years like a public IRC client. Open App/Browser, read messages, write messages, complain about lack of linear timeline, close app.

Yes, it has grown, but they added a million features that probably none of the people here asked for. We don't necessarily follow Justin Bieber where the system needs to be able to handle 2m users connected, we follow a kernel developer with 2000 followers.

I guess it's a marvel of engineering that's mostly needed for whatever method they found to make money off, I'm still asking myself this after these years.


> Constant software upgrades to maintain high standard of security

https://www.washingtonpost.com/technology/interactive/2022/t...

"Zatko’s complaint alleges he had warned colleagues that half the company’s servers were running out-of-date and vulnerable software and that executives withheld dire facts about the number of breaches and lack of protection for user data, instead presenting directors with rosy charts measuring unimportant changes."

Firing incompetent people might improve things, not make them worse.


One of the reasons I was looking forward to the trial was to hear what (if any) communication between Mudge and Elon('s team) had been dredged up in discovery.


Here's story for you

A team of 30 works at zerodha for serving 6.2 million users. They are largest trading broker in India.

There story is worth a read https://zerodha.tech/blog/hello-world/

Edit: Apparently they have 1200 employees for operations and everything else. Thanks to child comment for pointing that out.


So what are the other 1170 people doing? 10 managers to each person in tech? Sounds like bloat /s


No they dont have managers or a separte QA team. 30 is their entire tech team which handles everything.

They are largest broker so others might be sales, legal, marketting, hr guys


> They are largest trading broker in India.

Trading brokers don't need to make spam-filtering tools, CSAM filters, ads tools, handling huge video/image data, or handling a global audience.

I suspect most brokers's tech is way simpler than Twitter's.


30 member tech team.


People also underestimate what it takes to keep advertisers happy. It might take a whole team just to manage one big account, and that account will cover the cost of the staff. With Elon threatening to fire half the company, it's no wonder brands are pulling out/pausing with no way to know who will be handling their accounts. A General Mills or Audi wants to keep working with the people it's come to trust over the years, not wonder if this new CEO is going to hand them off to some intern who just graduated with a major in advertising.


That's why WhatsApp had 50 employees before the acquisition - it's very difficult to do these things if you have only 3-5 people.


I thought Twitter was mostly on AWS?


According to whistleblower Peiter Zatko they had two datacentres.

He reports that they were worried that if a DC went offline they might not be able to bring it back up again!

Their servers in their DCs were running unpatched operating systems with no disk encryption!

I highly recommend reading his full report. The level of disfunction he reports is insane.


I don't think most of the issues mentioned are mitigated by AWS, sadly.


> Actually, I have noticed that in the last year or two they refuse to show me more than a tweet or two until trying to force me to log in.

Things like that are usually UX driven, developers often don't have any recourse beyond implementing what's demanded in such a setting.


Ironically, I think Twitter started actually building product for the first time in a long time recently.

One frame through which you could view Twitter is it was the place where famous / popular / important (athletes, movie stars, government officials) talk to the public. Spaces, enabling audio conversations for those people, makes tons of sense.

So also Twitter Blue letting you read certain news w/o a paywall. If you're a serious consumer of news on Twitter, wsj/nyt/wapo/latimes/ft probably cost $1500/year. If the economics worked for the news orgs, there was huge value there.

Of course, they really sat on the product for 5+ years, so oof...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: