Show HN: Flares – CloudFlare DNS backup tool

Flares is a CloudFlare DNS backup tool: every time it runs, dumps your DNS table to the screen || optionally exports it into nicely formatted files.

Docker Quick start (painless)
# CloudFlare auth key is here: https://dash.cloudflare.com/profile ->
# Global API Key -> View
$ docker run -it –rm
-e CF_AUTH_KEY=””
-e CF_AUTH_EMAIL=””
lfaoro/flares domain1.tld domain2.tld
Quick start (full control)
Golang must be installed: https://golang.org/dl/
# flaredns
$ go get -u github.com/lfaoro/flares/cmd/flaredns
$ cd $GOPATH/src/github.com/lfaoro/flares/
# flarelogs (coming soon)
# $ go get -u github.com/lfaoro/flares/cmd/flarelogs
Fill the .env with your CloudFlare and Git credentials
# Provide a .env file in your project with the following variables or export them.
# Check .env.example
$ cat > .env


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/VF6TWbXOng4/flares

Original article

Decision Tables

I really like decision tables but they’ve fallen out of common knowledge. Let’s fix that.

A decision table is a means of concisely representing branching and conditional computations. In the most basic form, you have some columns that represent the “inputs” as booleans and some columns that represent outputs and effects. It looks like this:

A
B
C
f(A, B, C)
T
T
T
1
T
T
F
3
T
F
T
7
T
F
F
“cucumber”
F


NullError
– means that it doesn’t matter what the value is. If you’re feeling saucy you can add enumeration inputs, too, as long as the enumerations cover all the possible values for that input. For a decision table to be “consistent”, all possible inputs must map to exactly one row. There can’t be any inputs that aren’t covered, and a two rows can’t overlap in what inputs they cover. Two rows may, however, map to the same output.

As an example, here’s fizzbuzz:

n % 3
n % 5
f(n)
T
T
“FizzBuzz”
T
F
“Fizz”
F
T
“Buzz”
F
F
n
That’s all there is to decision tables. That’s what I like


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/o1Fc4kOTuMY/

Original article

Distributed Filesystems for Deep Learning

Published by dowlingj on October 17, 2018

More training data gives predictable gains in prediction accuracy
tl;dr When you train deep learning models with lots of high quality training data, you can beat state-of-the-art prediction models in a wide array of domains (image classification, voice recognition, and machine translation). Distributed filesystems are becoming increasingly indispensable as a central store for training data, logs, model serving, and checkpoints. HopsFS is a great choice, as it has native support for the main Python frameworks for Data Science: Pandas, TensorFlow/Keras, PySpark, and Arrow.
Prediction Performance Improves Predictably with Dataset Size
Baidu showed that the improvement in prediction accuracy (or reduction in generalization error) for deep learning models was predictable based on the amount of training data. The decrease in generalization error with increasing training dataset size follows a power-law distribution(as seen by the straight lines in the log-log graph below). This astonishing result came from a large-scale


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/jJSMrXhws4k/

Original article

OpenBSD 6.4 Released

The 45th version of the OpenBSD project has been released, bringing more hardware support (Radeon driver updates, Intel microcode integration, and more), a virtualization tool that supports the disk format qcow2, and a network interface where you can quickly join and switch between different Wi-Fi networks. Root.cz also notes that audio recording is now disabled by default. If you need to record audio, it can be enabled with the new sysctl variable. An anonymous Slashdot reader first shared the announcement. You can download it from any of the mirrors here.

Read more of this story at Slashdot.


Original URL: http://rss.slashdot.org/~r/Slashdot/slashdot/~3/u-A4NHs2JV8/openbsd-64-released

Original article

Building a better gov.uk, step by step

The ‘Apply for your first provisional driving licence’ step by step navigation pageSince GOV.UK launched 6 years ago it has been the home of government’s online content and the starting point for online services.
Every week millions of people use GOV.UK to do complex and sometimes life-changing tasks, such as learning to drive, registering a birth or starting a business.
We want to make these tasks as easy as possible – by making content simple and user journeys intuitive. This is good for users because it makes it quicker to get things done. And it’s good for government because it reduces unnecessary contact and casework.
Why we need to look at end to end services
We know users face challenges carrying out tasks. There’s a lot of information to find for a start. Depending on the task, there might be appointments to book, forms to fill in, applications to be made and tests to


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/hqw0Va9x1wE/

Original article

Ubuntu Linux 18.10 ‘Cosmic Cuttlefish’ Arrives

Ubuntu 18.10 Cosmic Cuttlefish, the latest version of Ubuntu, is now available to download. From a report: Under the hood, the Cosmic Cuttlefish boasts the 4.18 Linux Kernel. This updates comes with better support for for AMD and Nvidia GPU, USB Type-C and Thunderbolt, a way for unprivileged users to mount Filesystem in Userspace (FUSE) can be mounted by, and CPUfreq performance improvements. On top of this, you’ll find the freshest version of GNOME 3.30. You can, of course, use other desktops, but GNOME, since Ubuntu 17.10, is Ubuntu’s default desktop. You’ll be glad to know that GNOME is faster than it has been for a while. That’s because some nasty memory leaks have been patched. Canonical has also added some performance tweaks that didn’t make it into the GNOME 3.30 upstream. Ubuntu 18.10 also comes with a new desktop theme, the Yaru Community theme installed by default, for your


Original URL: http://rss.slashdot.org/~r/Slashdot/slashdot/~3/vlUaWWOlC-k/ubuntu-linux-1810-cosmic-cuttlefish-arrives

Original article

2018-10-18 14:28:13

Terry Heaton sent a note yesterday reminding me that the original BloggerCon was held in October 2003, in other words, exactly 15 years ago this month. We missed the actual days, which I can infer from my blog was October 4 and 5. Unfortunately the first archive.org snapshot of the BloggerCon home page was in 2004, when were getting ready for BloggerCon III at Stanford. I’m going to keep looking on my servers and backups for data about the first BloggerCon. I’m going to archive that as I started archiving the old userland.com websites last month.


Original URL: http://scripting.com/2018/10/18.html#a142813

Original article

Document editor Coda adds third-party integrations with G Suite, Slack, Twilio and more

Coda, the smart collaborative document editor that breaks down the barriers between documents, spreadsheets, databases and presentations, is today launching one of its most important updates since its launch in 2017. With this update, users will be able to pull in data from third-party sources and send out messages to their teams on Slack or by SMS and email. With this, the company’s take on building living documents that are essentially small apps is now really taking shape.
“Coda is a new type of documents,” Coda co-founder and CEO Shishir Mehrotra told me. “It combines the best of documents, spreadsheets, presentations, applications into a new surface. The goal is to allow anybody to build a doc as powerful as an app.” That means you can use your inventory spreadsheet to build a small inventory management app, for example, that lives entirely in a tabbed Coda document. Mehrotra noted that many businesses essentially run on documents


Original URL: http://feedproxy.google.com/~r/Techcrunch/~3/SdlI8uws65U/

Original article

Proudly powered by WordPress | Theme: Baskerville 2 by Anders Noren.

Up ↑

%d bloggers like this: