e-dard

Replace _is_ with _can be_ and I think the general point still stands.

We are yes - InfluxDB IOx. It will be available in our cloud platform.

I think two things are being confounded here: (1) places most likely to contain people with COVID; and (2) places most likely enable COVID to infect someone.

Many people go to supermarkets; they’re like central hubs in a network. It doesn’t mean that as much transmission happen there as in smaller confined places like houses and bars.

Hi, one of the creators of IOx here. Unlike InfluxDB where you can only do efficient range based queries on one time value, in IOx you will be able to apply predicates efficiently to any column. That means you can have several columns for different times (e.g., time created, forecast time, expire time..).

You will be able to do efficient range queries on any of these. Most use-cases will work best when you partition your data into time chunks, which commonly would be the time you inserted the samples into the database (created time in your case I guess).

90th percentile is about £47K, 95th percentile is about £60K. Assuming gross income.

The £30 Apple Pay limit in the UK is becoming less of a thing. In major retailers that accept Apple Pay I have used my phone to pay for items into the hundreds of pounds without issue.

In the last few months we have made quite a few improvements to data storage and indexing. Features that are available by default in 2.0, which are significant changes from releases earlier than, say, 1.6 include:

  - Significant TSM encoding and decoding performance improvements.
  - The TSI index will be on by default.
  - Queries that use the same tag key/value filters will be answered from the index more quickly using an LRU cache.
  - Field keys will now be indexed in 2.0, making filtering/grouping on field keys more efficient.
  - Improvements to how series are extracted from the index, and points data from the TSM engine, which helps with memory performance for queries.
  - Significant performance improvements to measurement deletion. 

Can you elaborate on that?

Hi, I'm one of the engineers working on the storage side of InfluxDB. Improving the performance of adhoc deletes, as well as import/export (backup/restore) are features that my team will be actively working on in the coming weeks and months.

Hi, just to clarify - Telegraf is still stand-alone. You will not need to run InfluxDB 2.0 on every host that you need Telegraf on.

I don't live in Sweden so I can't claim to be an expert (I like in the UK), but I find it highly unlikely that your claim of a 61% tax rate in Sweden is correct. It's likely the highest tax rate possible (in the UK that's 45%).

If you look at a tax calculator you will see that for a Swedish salary equivalent to $60K, the effective tax rate is actually ~27%.

$60K USD == ~540K SEK, which is 45K SEK a month.

Plugging 45,000 into [1] and choosing the municipality of Stockholm results in a net monthly income of 32,587 SEK.

That's a tax rate of ~27%...

https://statsskuld.se/en-sv/jobs/berakna-nettolon

InfluxData (YC W13) | San Francisco or Remote | https://www.influxdata.com/ | Golang | Full Time | REMOTE OK

We're looking for database engineers to help work on InfluxDB and other products. The current InfluxDB team is spread across four timezones and two continents; we're definitely a remote-first company.

If you love Go, open-source, writing high performance code and solving interesting data/distributed systems problems, then we want to hear from you.

We're hiring for a number of roles: https://www.influxdata.com/careers/

No puzzles and technical interviews for us; if you're an exceptional Go candidate you'll probably stand out already with your previous work or software projects.

Maybe that's just poor documentation on the author's part. If you simply try and go get the package (running Go 1.5 here):

    ~ $ go get -u github.com/otoolep/syslog-gollector
  warning: code.google.com is shutting down; import path code.google.com/p/log4go will stop working
  warning: code.google.com is shutting down; import path code.google.com/p/snappy-go/snappy will stop working
    ~ $

Yes, Go does not have a culture of versioning, but it _does_ have a culture of _vendoring_. You should be getting your reproducible builds by vendoring your dependencies. That way, you don't need to do any hunting or git-bisect shenanigans.

Then have a range? Of course you need to pay different employees different salaries, but that doesn't mean you can't expose potential employees to the salary they might expect to earn.

Just an FYI regarding your documentation, of example GET requests (e.g., https://www.shirts.io/docs/quote_reference/)

You can describe GET requests as follows with cURL, which I find a bit smarter :-)

  curl -G https://www.shirts.io/api/v1/quote/ \
  -d "api_key=APIKEY" \
  -d "garment[0][product_id]=3" \
  -d "garment[0][color]=White" \
  -d "garment[0][sizes][med]=100" \
  -d "garment[0][sizes][lrg]=50"

“correct horse battery staple” — the new hacker’s choice of password.

Without meaning to sound snarky or anything, isn’t there something else at work if you have to externally block sites to stop you procrastinating?

I would have thought getting to the root cause of your “HN addiction” would probably be a better alternative.

Maybe I'm missing something, but do people actually see Twitter Cards in general? I mean, how many regular Twitter users actually use the web interface to interact with Twitter?

I don't know anyone that doesn't use a third-party client. And, given Twitter's clear desire to restrict 3rd party client API access, in order to get more eyeballs on more easily monetizable web-pages, Twitter Cards are never going to be seen by many users.

How many ways are there to draw a cloud?

I'm getting warnings on Chrome regarding your SSL cert expiring.

As Pete Doherty wrote - You know twice as much as nothing at all. It's still nothing at all....

117 is my best Python solution yet, but I'm struggling to get higher than that!

Not always.

For example, if the dozens of papers are all disjointed, or on completely different topics, you can simply bolt them all together and call it a PhD thesis.

Like all good stories, there has to be a solid start, middle and end. This generally also means that one fantastic paper is also not enough, content wise, to warrant the award of a PhD.

This is almost exactly how it's already done in the UK, with exception that the loans are not provided by the university, but by an external loans service (The Students Loans Company).

You pay for your tuition by having x% of your salary deducted from future pay packets.

This does not “fix” anything. All it does is make the cost of education higher for those who earn less, like teachers and nurses. Why? Because they take longer to pay off their loans than Engineers, Lawyers, and those in Finance.

An let’s not even get on to the point that Universities are really there for research as well as teaching. We shouldn't be incentivising them to solely churn out future Software Engineers, etc...

she

Well, Myna https://mynaweb.com/ uses multi-armed bandit algorithms for A/B testing (or A/B/../N testing if you prefer).

An n-armed bandit problem is a type of reinforcement learning problem, and the associated algorithms for solving these problems are considered machine learning.

Hi, machine learning PhD here - one way you can start is by brushing up on some fundamentals. A book such as Machine Learning, by Thom M. Mitchell is a reasonable start.

Also, in terms of applying ML, you could scoot through Andrew Ng's Machine Learning Coursera course (not sure if it's running at the moment, though).

Finally... One tip – typically I have always found that when you want to start apply ML to real-world problemss, start simple and only iterate when the results of your approach are not satisficing. This is usually because all the bleeding edge ML research/techniques don't consider a shit-load of real-world issues, like scaleability, applicability to wide-range of problem, unstructured or noisy data and so on.

Yes, but elchief's point is obviously that if the data has a positive or negative skew, the median is going to be more helpful than the mean.

I get annoyed when I read things like: "A performed 4.9% better than B", when the comparison is between two proportions.

Just say it like it is, without putting a slant on it - A's response rate was 0.6% higher than B's.

I don't care if the results are significantly different, if the difference between the two samples is so small.

No! This crappy iOS app has a glitch and only showed the “cheapest way to store 6TB online” part!

That’s amused me somewhat chuckles

I haven't checked, but I would think Amazon Glacier, assuming you only need to long-term archive the data.