Hacker News new | past | comments | ask | show | jobs | submit | sorval's comments login

Part of the idea was that it would be more authentic and kinder. When people are completely anonymized they are free to be their worst selves. Your voice is personal. Its slower though because its more 'single threaded' for lack of a better term, than a reddit sub.


> Part of the idea was that it would be more authentic and kinder.

Did that work out? All I know about clubhouse in this regard is stuff you you can find at google searches for e.g. "clubhouse misogynist", "clubhouse racist", "clubhouse anti-semitic"

I have never been there, and part of the reason is the reputation that it has from afar be being the opposite of "kind".

Wikipedia entry for "Clubhouse (app)" says as much in para 2.

Maybe, being not anonymous isn't a cure-all for bad behaviour. After all, there were racists before they could be anonymous online.


I've seen in leaderless orgs that getting new headcount for your part of the org is very hard, and everyone just does what they want. Theres multiple tools competing for each other. Getting promoted is hard. Ownership is difficult-- everyone wants to be an advisor and not an owner. "No managers" sounds good, but I havent seen it work well.


I'd be curious to know how common this is. I absolutely do not miss any part of the commute, or being constantly distracted by coworkers noise and needs.


Well, there goes the intercept.


I've found Adobe to be pretty handsy with their consumers' pocketbooks in the past...


I worked there. It was a greasy and miserable place run by people who should have retired long ago. Everything was waterfall development, and there was a yearly slave-rush to their miserable convention show, where only half the stuff really worked. No one really knew what they were doing. It was like a time machine back to 20 years ago. So glad I left.


Thank you for confirming the things I have suspected for a long time.

Maybe it's the path I took in the industry but I had the displeasure of having to automate the installation of Adobe security updates to tens of thousands of Windows boxes a decade ago right through having to (briefly) write software targeting a customer's Adobe Experience Manager product in the last couple of years. There's a long rant bubbling up about AEM, but I'll resist the temptation and just say a suitable 4-letter word, instead.

Their products cost a fortune and for the privilege of forking over your hard-earned money, the company does things like this. I still own a license for one of the older versions of Photoshop post-Activation. I corrected it in a manner that would probably qualify as "cracking it" before ever activating after having run into grief with the product at my day job (at the time). Anymore, if there's a reasonable alternative, that's the kind of stupid I won't spend my money on.


As a long time user of one of their tools (Illustrator, which I have a strong suspicion is not a prestigious team to work on) I have always suspected something like this. Sigh.


moronic right wing analysis. The article had personally identifying info so it got blocked. the end.


Most of the difference between a Sr engineer vs Staff/Sr staff/Mgr is all social skills. Its not just what you get done, but how you get it done and how you leverage other people that matters. Social skills are absolutely vital to climb the engineering career ladder.


Maybe those were older people? Threads came about in 96 and became default in the linux kernel in 2.6 (released in 2003).

Sharing memory between processes was a mess before that. Java green threads (emulating threads in user space, which often blocked badly on IO) were horrifying.

Python threading implementations were even worse of a joke till fairly recently.


Thats just utterly untrue and the reverse of whats actually true.

"states that were coded as blue based upon results from the 2004 presidential election were significantly higher in education funding than were states coded as red. Students in blue states scored significantly higher on outcome measures of math and reading in grades four and eight than did students in red states."

https://files.eric.ed.gov/fulltext/ED503486.pdf


Is there any correlation at all between increased school funding and better educational outcomes? For instance, Utah spends $5k per student per year while Baltimore spends $25k per student per year. Utahan children have higher graduation rates, better GPAs, and better standardized test scores. It's the same across the US from NYC to LA.

I'm growing more and more skeptical that throwing more and more money into the furnace that is the US education system is suddenly going to produce better outcomes. How much more must Baltimore spend? $30k a student? $50k a student? $100k a student? Where's the magic number?


According to [1] Utah spends $10,510 per student per year and according to [2] Baltimore spends $16,184. Where did you get your figures?

[1] https://auditor.utah.gov/audit_reports/public-education/

[2] https://www.census.gov/newsroom/press-releases/2019/school-s...


I got it from the Annual Survey of School System Finances, U.S. Census Bureau, though on checking it again it's between $6,000 and $7,000 (I was recalling from memory). So not what I put in my comment exactly, but doesn't change my main point in any way.


It could be four things:

  - Comparing two statistical flukes.
  - Socio-economic factor: Utah has better wealth distribution.
  - Socio-economic factor: Utah has affordable housing and Baltimore so the effective wealth gap is lower.
  - Baltimore has other issues.


That’s the most facile thing I’ve ever read. Did it control for cost of living? Did it control for demographics?

In 2019, and I’m using this because you can explore it at https://www.nationsreportcard.gov/mathematics/states/groups?... , Mississippi’s 8th grade math mean score was the same as New Hampshire’s, when comparing white students. Mississippi is just way less white than New Hampshire. California, also near the bottom overall, has better white performance than New Hampshire, but not as good as Massachusetts, which has always had a lot of smart people moving to Boston.

The ranking there is simply a diversity/brain drain ranking. Mississippi and California are way more diverse, so they rank low. States with nothing to attract or keep smart parents, like West Virginia, rank low too.


Yes, the whole puzzle becomes much simpler to think about when you consider the possibility that the school scores primarily reflect the quality of students admitted, and not the quality of the education administered.


I note that koolba's comment was evidence-lite and really rather partisan. So it is probably incorrect.

However, claiming the reverse is true is also evidence-lite. That paper looks like it was a fairly cursory piece of work and didn't account for population size or characteristics (eg, number of disadvantaged students) in the correlation work.

I suspect what it really shows is that political affiliation isn't the biggest driver of outcomes. Looking at the table at the end of the paper and thinking about population, I suspect the aim would be more of what Massachusetts does and less of what California does (both Democrat states) if possible.


Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: