Hacker Newsnew | past | comments | ask | show | jobs | submit | flyingramen's commentslogin

> Online, you can create a private Slack channel or even a separate off-Slack group for the in-crowd where you exclude others, and no one can see it.

Since you say politics reasons wholly because of this reason, it’s not like Slack is not available if you are not fully remote.


It's available when you're not fully remote, but people prefer to have in-person meetings instead, because they have higher communication bandwidth.


Post with more votes and made a few minutes earlier to this one: https://news.ycombinator.com/item?id=14338328.


For my benefit, would you be willing to share your salary or range (and currency :)?


I'm not the grandparent post, but, my salary in Canada was $80K starting in 2000, and is up to $160K now.

I've been making more than double the median wage in my city for the last 20 years or so.


$80K CAD is $59K USD, which is not that much different than the original post. $160K CAD = $119K USD is fairly good, but most developers with 17 years experience would take home north of $200K in Silicon Valley (I'm not familiar with other US markets).

All the anecdotes still seem to point to software engineers making more than double in the US. I would love to return to Canada some day, but I'm not likely to do so until the numbers I see people post online start beginning with at least a 2 or 3.


Exactly. A USD 200K software eng simply cannot move to Canada. It is messed up. That is more than a director or higher's salary in a place like Toronto.


If you're living in Canada, you're spending money in CAD. I don't think its fair to convert Canadian salaries to USD to make the comparison. Still, $160K CAD is considerably less than $200K USD.


If you are living in Canada, everything retail is more expensive anyways, so just converting to USD isn't very accurate.


$160k CAD = $120k USD.

And you have 20 years experience? If you are good, you could be making 4x this in the Bay Area. Or more,


We hire a lot in Montreal. Mid-size startup. Range is more like CAD 85-140k for generalist frontend and backend developers.

In the larger cities, salaries have really gone up in the last few years.


I think that actually makes the other posters points. That's 63-104k in USD, which is low. Most of us in the US would be taking significant pay cuts at that level. I have no idea how that stacks up as far as PPP.


Name the company my man! Then don't mention Montreal again, we don't want everyone coming.


Might be a well-known streaming company that definitely has interesting technical issues. Problem for some is the content and reputation.


It sprang to mind.


French is the key barrier .. not naming the city :)


Eh not in Montreal. As an Ottawa native I've never had trouble in Montreal with my English and horribly broken French. Now Quebec city...


I disagree. I'm here, it's not ideal, I'm learning but it's not a blocker. Unlike say London when you cannot live in a decent area.


This is just bringing up the Nebraska center up to speed with the California and Vermont centers. This is done because Nebraska will restart processing H-1B application which it had not been. USCIS clarified there is no change in policy.


A few people lost all their emails on Gmail in 2006. https://techcrunch.com/2006/12/28/gmail-disaster-reports-of-...

There have been other Gmail outages in the past though. My two favorites are:

1. The multi-hour outage in 2009 http://www.cnn.com/2009/TECH/09/01/gmail.outage/index.html?e.... 2. And one where they had to restore data from magnetic tape backups https://gmail.googleblog.com/2011/02/gmail-back-soon-for-eve....

Of course, there are other email providers too and their own outages. But I like Gmail and that's what I follow.


What would an ideal way forward be for Uber? Recent incidents combined with a tainted past make public trust in Uber not the best.

Lot of people have suggest these problems stem from the culture. Culture of companies and people is similar to the culture of, say, bread. The starter really matters and sets the tone. I'd be interested in hearing from people who have seen a drastic change in culture at a big place or better yet, have been behind that change.

The other idea is it could just be a few rotten apples giving everyone a bad name. I don't know the answer.


>The other idea is it could just be a few rotten apples giving everyone a bad name.

There's no such thing as this.

Honestly, is everyone these days completely ignorant of what happens when you leave a few rotten apples in a big bunch of apples? In case you don't know, the correct answer is that very soon, ALL the apples become rotten. That's why we have the old saying about "bad apples": "one bad apple ruins the whole bunch". Somehow, these days, everyone seems to have forgotten the "ruins the whole bunch" part of the phrase which is so important to its meaning.

It applies outside of apples too: it applies to police, corporations, any human organization really. There's no such thing as "only a few bad apples". An organization that tolerates rotten people very quickly becomes thoroughly rotten.


Thanks for taking time to lay that out. I've felt like I was living in some alternate reality the last few years every time someone trots out that saying as an excuse to duck responsibility.


That may be a good analogy to make but it is not that simple. I am going to try find it, but there was at least one paper showing how it really depends on the placement of the kind of actors in a system. Enough (but an easy minority) of good actors in a system could "clean up" the system by ensuring everyone acts well. If you want a more relatable simple analogy is a benevolent dictator cleaning up the mess.


You've flipped the situation. Arizhel isn't making an analogy, they're correcting your misuse of one ("a few rotten apples") to defend a conclusion that is the exact opposite of the one it implies.


Allow employees to sue as a class for cases of sexual harassment. Currently, Uber's employees are bound to confidential, bilateral and binding arbitration.

Next, deputise an ombudsman to whom every employee has a confidential and direct line. This ombudsman reports directly to the CEO and the Board. They should have a large amount of authority surrounding recommending dismissal of any employee to the CEO; whenever a recommendation is made, it should be copied to the Board. In return, the ombudsman should carry a significant amount of personal liability in connection with sexual harassment at the company.

This is how compliance works at many financial firms, and it's very effective. Compliance officers are deputized by the U.S. Treasury for anti-money laundering purposes. This makes them personally responsible if something goes wrong. It shows.


> What would an ideal way forward be for Uber?

How about firing sexual harassers? And any HR staff that systematically lied to cover it up. And the people who gave them those orders. And not taking forever to do it.


Simply firing people that have committed a certain offense may not fix the underlying problem—only treat the symptom. But I agree that sexual harassers should be separated from the company.


I'm not saying that's all they should do but, fundamentally, all the policies, training, press releases, investigations, CEO statements, etc. are secondary to the core test of whether bad behavior is encouraged, ignored, tolerated, slapped on the wrist or punished severely.


Not just that: cancel any stock options or other perks they have too.


New leadership is one option. Depends if investors feel a change would improve things or not


I think this open letter gives a very clear recommendation; have a truly independent investigation that leads to recommendations for change.


There's a grand tradition of founders being fired/resigning once their companies grew up, in order for the companies to grow up more. Kalanick's resignation is likely the ideal way forward for Uber as a company; it will be very difficult for him to solve this problem as it's tainted his reputation.


> What would an ideal way forward be for Uber? Recent incidents combined with a tainted past make public trust in Uber not the best.

Usually these are handled with a shake-up at the top, and other involved people being let go.


Your one machine's disk fails and you'll lose data if you have anything other than synchronous replication. And once you have synchronous replication let's hope it is on at least a different rack if in the same data-center. Preferably in a different geographical region actually. And now you no longer have "really fast" QPS.

All this assumes you have only so much data that can fit in one machine. If that's not the case you're in need of a database that spans more than one machine.


There is a better response to this up in this thread (https://news.ycombinator.com/item?id=13661735).

The test is testing the worst-case scenario of everything needing locking. CockroachDB uses an optimistic "locking" approach which makes this bad. But if you're use-case is strictly linearizable high-throughput reads good luck finding anything that is not a single-machine database.


That's actually the only good answer I received to my comment.


Eric Brewer who is the author of the CAP theorem works at Google now. He has a post here: https://cloudplatform.googleblog.com/2017/02/inside-Cloud-Sp....


It is fascinating that more and more people are using Cassandra. DataStax believes they have fixed problems with prior guarantees claims that were exposed by Jepsen. But there has been no official Jepsen testing since.

On the topic of looking at Scylla next, I wonder why did the team not just start out with it to begin with. Also, are they people with experience running both. How is the performance? And what is the state of reliability?


The problems that Jepsen found were centered around the "transactions" feature that Cassandra added. We don't use these and don't need them since we don't need 100% consistency and prefer availability (for example we read at quorum to trigger read repair, but downgrade to single node reads if we need to).

Also ScyllaDB is a new product and it would be crazy to start off with it. We plan to run a long-term double write experiment before we are comfortable with using it as a primary data store.


The Jepsen tests were not completely centered around transactions. It also had to do with data loss when replicas go down and pure "last-write-wins" approach. For those wanting more info around this the original post is here:

https://aphyr.com/posts/294-jepsen-cassandra


Last-write-wins is a semantic we are okay with for this data, and dealing with one of our conflicts is outlined the article.


I enjoyed your article and I do appreciate your transparency!


I find it fascinating that people still think Cassandra is some risky new tech - been running it in production since 2010, and the fact that people are still worried about it makes me snicker a bit.


Not everyone need strong consistency, that's why those Jepsen reports are not relevant for all cases.


The whole ideas behind Jepsen report is not that people need Strong Consistency. It is that products should tell you precisely what they guarantee or not.

This is why Aphyr always test claims.


We're running Scylla in production. Much easier to setup, no tuning necessary, great performance, no issues. It has a strong team behind it.

Missing some features that Cassandra has that should be fixed by v2. Read their site/blog for a good overview of the tech and progress.


Cassandra now runs jepsen regularly see. If you dont believe they are fixed you can check yourself: https://www.youtube.com/watch?v=OnG1FCr5WTI


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

Search: