I recently watched the AeroPress Movie (https://aeropressmovie.com/), which was a fun look into the history and culture around the AeroPress. Something I enjoyed about it was seeing the inventors workshop in his garage, as well as his opinions on what constitutes a "good cup" of coffee.
Interesting. I've not watched the movie but I disagree with the inventor on the type of coffee he tries to make with the Aeropress.
From what I recall he tries to make a more "espresso" style coffee and then dilute it. I prefer to just go 15g coffee to 250g of water for a single normal cup. No dilution.
yeah, he also doesn't steep, but I find that you need to steep for a few minutes unless you want to under extract your beans.
I also greatly prefer a metal filter over paper as it lets more flavor through, but people who don't like flavor call that "acidity" or "bitterness". ️
The statistics cited by Local Babysitter are ten years old, collected in 2009 by Nielsen. This doesn't reflect how media has changed in the last ten years, along with how children consume it. It would be nice to have a more updated source explaining what has changed in the last ten years, since obviously Nielsen is collecting this data.
Cheap boards like the ESP-12F (ESP8266) allow you to remotely control a device or receive sensor data, like that from a DHT11 temperature-humidity sensor. Instead of paying a lot of money, you can flash your own firmware on a device and make your own custom IoT device.
MQTT is a messaging protocol used to communicate with the device. You can use something like Mosquitto to communicate with your new IoT device. It is lightweight protocol and easy to work with.
You'd use something like an ESP32 – which has Bluetooth – if you want a simple BLE beacon, otherwise if you don't need Bluetooth then an ESP8266 device would likely work better.; If you want to look at related videos, I'd suggest looking at the YouTube channel Andreas Spiess publishes with various devices. He has a few videos on BLE and ESP32 (ex: https://www.youtube.com/watch?v=KNoFdKgvskU) where he goes into detail.
Like you hinted, flashing Sonoff-Tasmota firmware on ESP8266 hardware lets you get A LOT of home automation equipment for fairly little. They make many devices and someone getting started should checkout the Sonoff-Tasmota for supported devices AND check out Sonoff.
If you don't want to use a Sonoff Basic or maybe the bigger 4 channel option, I'd suggest getting some of their wall adapters like the S31 (think Wemo wall plug). There is a confirmed flash for Sonoff-Tasmota & ESPurna for it and would let someone dip their toes into cheap home automation with MQTT.
"They even offered a token bounty of 5000 rupees in BigBasket credits, which we have not accepted."
That's a pittance for such a critical flaw in their purchasing system, especially for a company valued at over 180 million USD. 5000 Rupees is, by the way, worth $77.19.
Is it using Apache Spark maybe? I'm confused, even their logo reminded me of Apache Spark in terms of color at least. There are other "Spark" projects out there as well.
There is no upgrade path from OS Luna (previous release) to Freya. Their own web site states, "elementary OS Luna users should back up their data and perform a clean install". Given that the operating system is built on Ubuntu I should be able to upgrade from Luna to Freya without too many issues.
Disappointed? Yes. I wrote this post on my Cr-48 running Elementary OS Luna. Instead of running Apt to update my laptop now I'm browsing DistroWatch in another tab to see what else has come out.
You can install Pantheon(the eOS DE) on other distros if you want, though you'll have to put in the time and effort to get it to work and configure it.
As a workaround you could probably add their Freya repositories and perform an apt-get upgrade.
Whether or not it would work flawlessly is another question altogether, I would do an image backup of the drive just in case it goes pear shaped.
If you like xargs, but want more flexibility, I'd highly suggest GNU parallel. Such flexibility includes running jobs on multiple computers, running intensive command using all available CPU's (like xargs -P), and creating unique scripts to handle multiple parameters.
Parallel also lets you transparently run stuff on remote servers, automatically handling stuff like copying files back and forth. When I have some heavy ad-hoc data processing to do my new favorite trick is spinning up 50-100 ec2 spot instances, point GNU parallel at them and just fire and forget.
I heard Pete's story on NPR a while back and how it helped your company grow. It stuck out since when the reporter tried to verify the story they found no results. The reporter comes short of accusing you or your co-founder of telling a fake story to sell a product. I'd like to hear you or Pete's take on this story (URL below).
Thanks man. I remember that interview (it was a headache). The interviewer seemed to have an agenda going in. Before the interview he kept saying he thought it was "foolish that people would use Google to research people in the first place" and that "nobody should believe Google results anyways".
When Pete was in college there were several criminals who all shared his name and his results were a mess. One of them included a story about someone suspected of dealing drugs. At the time we never thought to save or bookmark those results because we had no idea we'd be starting a company and being interviewed by NPR years later.
We couldn't find an exact article about a drug dealer (perhaps it actually was taken down). We WERE able to show him several other results with about criminals with his name. We linked him to them. When he asked us about it, we told him we couldn't find it, but we offered to redo the interview and be less specific "Pete was being mistaken for criminals with the same name" since we could show results for that.
In his article he claims we simply did not respond to these requests. When I asked him about it afterwards he apologized and said "he must have missed that email". He didn't update the story though. It seems like he wanted to tell a story about how online reputation management helps people permeate lies and that's why you shouldn't trust Google.
All that said, we learned a valuable lesson. We no longer use the term "drug dealer" we use "criminals" and we link to specific articles we're still able to find so people can't question the validity
If you were having trouble finding articles, what's with the Google screen grab on slide 6? Was this something you created specifically for the slides? If that is the case, showing a fabricated screen grab without noting it as such seems a bit disingenuous. It's a bit like selling a diet with a doctored "before" picture.
I agree, it is disingenuous, especially in this context. That slide was specifically made for an in-person presentation that was later posted online (it was posted by somebody else about a year ago). For the sake of a presentation, we thought that image got the point across much better than putting two hyperlinks on the screen. It was basically a design decision. In retrospect we should have put a disclaimer on the image since it looks so realistic, but we honestly had no idea the presentation would live beyond the room I spoke to. That was a mistake
We used to use a similar type image on our about page, but for the same purpose now only link to exact articles or results we can still find. https://brandyourself.com/info/about.
Thank you for answering my question. I was afraid you'd skip it if there was truth in what NPR had presented.
I'm probably not the only person who felt that you and Pete are misleading people with the story of Pete's online profile after that interview. In some ways it may have done more harm than good overall.
Have you done anything since then to try and get NPR to correct or update the article?
Well it was interesting. He asked us if we had a link to the drug dealer article. We looked and couldn't find one. I was in contact with him the whole time, sent him links to other criminal results and even offered to do the interview over and be less specific and say "criminal" instead of "drug dealer". I told him the merits of the story were still true, we could still show that Pete shared google results with a criminal, and we even put him in touch with the person who worked at the company who found the results. Instead he posted the article implying it might not be true, and said when he asked us about it "we went silent". I followed up with him after and he said that he didn't seem my emails (even though he answered some of them)...
We were worried that people would feel the same way, but in general we only got good responses from the interview. We signed a lot of people up, too.
That said we've been careful to be much more specific in the future. We don't want to mislead people especially since it could hurt our brand. The truth is, Pete's results were a mess and it was hurting him. He shared a name with several criminals. There really was a drug-related article about a Pete Kistler but we can't find it (we never thought to keep it)--so now we only say "criminals" and link to the articles we HAVE found
