Really? Here in Norway children would only be vaccinated against influenza if they already have other severe conditions such as diabetes, chronic heart or lung disease, etc.
Flu vaccine reduces risk of death. Coronavirus has no vaccine yet. I do think we take the flu too lightly. Urgent care and hospitals definitely need to be more risk averse even after this pandemic is over, and I hope we learn that lesson in this pandemic.
I tried this in college. I was doing 6 hours sleep per day (3 hours and 2 90 min segments with 6 hours awake time between them) Worked great while taking 12 credit hours over summer semester. But you end up being up when everyone else is asleep and sleeping when others are awake. Not great for social life but I did get 4.0 that semester.
Not sure if I would choose to get irritated with X-rays if I think I have a cancerous tumor. CT scans are already scary IMHO if you already have cancer. MRI may be safer option.
Not exactly a good comparison if you don't generate the data the same way for the test setup. Your generated data is more compressible by clickhouse, that skews the comparison. Would have been better to not change the test data if you wanted to do a comparison.
I bet results would be roughly the same even for the exact same dataset - Scylla and other K/v data stores can’t compete with columnar databases that are purpose built for complex analytics queries. the many orders of magnitude query performance differences (not to count storage, compute overhead) show it enough.
It was kind of a crummy use case for Scylla anyway (it’s a transactional write store, not an analytics engine)
The important difference is that we used a more realistic temperature profile, which as you say does affect compression for that column. Schema design (including sort order, compression, and codecs) for the remaining columns is just good ClickHouse practice. Much of the storage and I/O savings is in the date, time, and sensor_id and columns.
It's also useful to note that the materialized view results would be essentially the same no matter how you generate and store data because the materialized view down-samples temperature max/min to daily aggregates. The data are vastly smaller no matter how you generate them.
The article illustrates that if you really had such an IoT app and designed it properly you could run analytics with surprisingly few resources. I think that's a significant point.
That's what you wanted to show, but what you ended up showing is that if you have different data, then the query performance can be quite good.
I get the desire to critique the temperature profile, but completely changing it makes the comparison worthless. From a data perspective it's like saying "if all the sensors just report 1 for temperature every reading, computing the min, max, and average is super fast". No shit, that wasn't the task though.
But they didn't set the temperature reading to anything that would advantage their tests. Without access to the original data they simply generated a dataset as close to the original dataset and volume as possible. The fact they took a few sentences talking about the temperature doesn't equate to invalidating the test.
Looking at this your way - Scylla used an INT, Altinity used a Decimal type with specialized compression (T64). I can tell you that this would have hampered ClickHouse and advantaged Scylla. It's the opposite of what you're saying. They actually performed this benchmark with one arm tied behind their back.
It's a funny benchmark anyway because the two systems have very different use cases but it doesn't invalidate the result.
Then you should provide results for both test datasets to make the point of using a more realistic approach. Materialized views are not news, nor is properly designed analytics applications. For me the importance is how click house is better and why.
A column-store will be magnitudes faster at analytical queries than any rowstore system. This is fundamental architecture and the data used makes little to no difference. You could use the exact ScyllaDB dataset duplicated to trillions of rows and still arrive at the same relative performance figures.
It doesn't matter. ScyllaDB is a Cassandra clone, an advanced nested key/value database that stores data per-row and requires slow iteration to scan through an entire table.
Column-oriented databases will always be much faster at analytical queries because of the difference in physical layout and vectorized processing. Scylla's has very impressive OLTP performance but really shouldn't be compared to OLAP databases at all. That original 1B rows/sec blog post by them is kind of a strange benchmark to begin with.
Problem is what use cases are strictly OLTP? At this point, I’d consider Scylla/C* to usable for a write-only workload with single-row lookups, or a single-column range lookup.
Same question has to be raised: do you have enough rows to justify a distributed Scylla/C* or could you have used MySQL or Postgres on a giant box?
Plenty of OLTP scenarios that need distributed scale and/or high availability of C* - we use it for user profiles/session storage, counters and some high-volume logging that needs access to individual events.
The compression is a property of the table and done on the fly, transparently to the user. If the difference was compressing/decompressing as part of the user task, I'd agree. But this is something that comes for free by a few extra characters in the schema.
Drivel. Nothing directly actionable. If a CEO has to be physically in his office to successfully run the company, he has done a poor job in building leadership under him/her. Jack is doing what he should be, finding new growth opportunities. Milking the users with more Ads will just piss off users for short term gains.
With respect, what you're saying doesn't address the main thrust of the letter. The nine hour time zone difference between Dorsey and Twitter HQ is only one component of the "drivel", as you put it. Frankly it's just the rhetorical opener to the main point.
Reasonable people can disagree about whether Dorsey should be CEO. But I think the author's concerns are more varied and more cogent than your rebuttal would make them seem.
What are your thoughts on Dorsey leading two tech companies at the same time, Twitter's share performance over the past several years, and the political waters Twitter has to navigate?
Are you kidding? Twitter is a tech company with a great brand and a monopoly platform that has a stock that performs like a mall chain. The rate of product change is so slow. There are so many opportunities for growth that are not advertising.
Seriously, if you can't push out a CEO for under-performance in this case, when can you.
The rate of product change at twitter is something every company should learn from!
Devs like new features a lot more than users do. Users feel disrupted by new features. Users don't open up their favorite app and say "gee whiz,I wonder what new feature there is today". Users don't even think about twitter, they think about tweets,replies,likes and other users. You focus on what matters to users, not what matters to devs and everyone else who does not get the user base.
Imagine feature-creep at their scale! Twitter is the most tolerable mainstream social media network I know of for many reasons -- all centered around their resistance to change and refusal to follow the trends at fb,instagram,snapchat,linkedin,etc...
Repeat with me: you don't need more growth if everyone is happy and you already have a steady and growing profit. You want more growth of course but you don't need it. You focus on what you have and need first then you consider your wants and features...very slowly.
One example of a feature that I (a user) would love to see get a little love is Lists.
I use them frequently, but Lists are not well-designed at the moment. I can't see if I've already added a user to a list, the "update" is weirdly different from that of the main timeline, there are no aggregations/stats/summaries and AFAIK, there is no interaction with push notifications. They do nothing on the platform to draw anyone's attention to the feature.
But Lists are great! I can pare down my feed to a curated set of users who are all talking about the same thing -- I have an NLP list, and ML list, a "politics" list, etc. I can follow other people's curated lists too. Twitter is huge, it's awesome to have some path to a smaller cohesive community.
It's really weird to me that they would have a feature "above the fold" on the main menu of the platform that seems so dusty and under-utilized. What's the deal? Do they not have enough engineers over there or something?
Just because you can, doesn't mean you should. "Better" is subjective anyway. Any change with make some people mad, and no change will make others mad. If your product is stable, that's about the best you can hope for. No need to keep tweaking it out of boredom.
Its like a car, you don't keep pressing the accelerator further and further to travel down the road. You find the optimum location to maintain a safe speed. If something changes, like the road turns, you adjust the accelerator until you are once again safe. If you stayed static, you'd probably fly into a ditch. If you are unnecessarily pumping the pedal, your speed would oscillate unpredictably, and any number of consequences would occur.
sure you can pull a digg and expect more money, but actually lose your users to a competitor. i 'd definitely quit twitter if it starts doing things like what facebook did to FB-page owners by holding their audience hostage for ransom. I think twitter, like reddit, inherently don't have high monetization potential because their format doesn't appeal to the massive audiences which flock to fb / instagram
That is also true.
I did agree with the author's position that Dorsey should not be the chief executive of both Square and Twitter. That's probably his strongest case towards removal.
But your point about Twitter's uniqueness is also valid, and something I hadn't seriously thought of before. Around the time of their IPO, I think the general analyst consensus was that Twitter would not make money and their stock was not a buy. It makes sense because their product is wholly different from the suites offered by Facebook or Google, the companies this article and the commenter compare it to.
Perhaps the other stakeholders understood this aspect of Twitter and didn't expect large returns from day 1.
I am not sure I agree with the author on how many companies a person should be able to run. Depends on the success of the companies themselves. Look at Musk.
With that attitude why even have CEOs? Company’s been built, it should just run itself. Let’s just have CEOs step down as part of the IPO process. Literally nothing could go wrong. I’m proud of how Apple was able to dig itself out in the 1990s all by itself through the will power of the board and executive team. I mean what with Steve’s many South American getaways.
Jack spends half days at two companies and as such isn’t able to meaningfully administer either. And from that platform he’s going on vacation to Africa for 6 months. The line about the real issue being not that he’s going away but rather that he’s coming back is devastating but not far from the mark.
No, I didn't say they aren't involved at all, but they don't need to be involved with day to day operations once they have consistent flow of revenue coming in. They should be finding ways to unlock the next growth multiplier not sitting at a desk in SF worrying about operations.
If in 6 months he has found that next growth multiplier, people won't care where he was, but in 6 months if there is no meaningful strategy derived from the exercise, there will be cause to rethink his role and he himself will see it.
